THE EDINBURGH

COMPANION TO
CONTEMPORARY
LIBERALISM

General Editor
Mark Evans

Edinburgh University Press

The Edinburgh Companion to
Contemporary Liberalism
This page intentionally left blank
The Edinburgh Companion to
Contemporary Liberalism

General Editor
Mark Evans

Edinburgh University Press

# in this edition Edinburgh University Press, 2001.
Copyright in the individual contributions
is retained by the authors.

Edinburgh University Press

22 George Square, Edinburgh

Typeset in Goudy
by Hewer Text Ltd, Edinburgh, and
printed and bound in Great Britain by
Bookcraft Ltd

A CIP record for this book is

available from the British Library

ISBN 0 7486 1359 5 (hardback)

The right of the contributors to be

identified as authors of this work has
been asserted in accordance with the
Copyright, Designs and Patents Act 1988.
 CONTENTS

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
List of Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Part One: The Liberal Trajectory

1. Issues and Trends in Contemporary Liberalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Mark Evans

2. Twentieth-Century Liberal Thought: Development or Transformation? . . . . . . . . . . . . . . . . . . . 21

Michael Freeden

Part Two: Citizenship: Universalism and Particularism

3. Human Rights and Ethnocultural Justice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Will Kymlicka

4. Liberalism and Citizenship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Andrew Vincent

5. The Sword of Faith and the Shield of Fear: Liberalism and the Power of the Nation. . . . . . . . . . 63
Margaret Canovan

6. Liberal Citizenship and Feminism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

Andrea Baumeister

Part Three: Justice: Identity and Distribution

7. Liberalism and the Politics of Recognition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Jonathan Seglow

8. The Essential Indeterminacy of Rawls's Difference Principle . . . . . . . . . . . . . . . . . . . . . . . . . . 101

Rex Martin

9. Rawlsian Theory, Contemporary Marxism and the Difference Principle . . . . . . . . . . . . . . . . . . 113

Rodney G. Peffer

Part Four: Problems of Liberal Justification

10. Disenchantment and the Liberalism of Fear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

Peter Lassman

11. Pragmatist Liberalism and the Evasion of Politics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

Mark Evans
 CONTENTS

12. Liberalism and Contingency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

Bruce Haddock

Part Five: Liberalism versus Republicanism?

13. Back to the Future: Pluralism and the Republican Alternative to Liberalism . . . . . . . . . . . . . . . 175
Richard Bellamy

14. Accommodating Republicanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188

David Rasmussen

Part Six: The `Autonomous Individual': Feminist Critiques and Liberal Replies

15. Liberalism, Feminism, Enlightenment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

Kate Soper

16. Feminism and Women's Autonomy: The Challenge of Female Genital Cutting . . . . . . . . . . . . . 208
Diana Tietjens Meyers

Part Seven: Liberalism Beyond the Nation-State

17. Civil Association: The European Union as a Supranational Liberal Legal Order . . . . . . . . . . . . 225
Robert Bideleux

18. The Idea of a Liberal-Democratic Peace. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Howard Williams

19. Constructing International Community:

Liberal Theory, Developmental Communitarianism and International Ethics . . . . . . . . . . . . . 253
Peter Sutch

Part Eight: New Directions for Liberal Thinking

20. Liberalism and Post-communism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

Richard Sakwa

21. Liberalism, Ecocentrism and Persons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287

Brian Baxter

22. Prolegomenon to a Liberal Theory of the Good Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299

Mark Evans

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323

vi
 ACKNOWLEDGEMENTS

I would like to thank the University of Wales Academic Support Fund for a generous
grant that financed several conferences and seminar sessions at which early versions
of most of these chapters were aired and discussed. Thanks are also due to the
following for permission to include revised versions of articles they have published: to
the editor of the Review of Constitutional Studies, published by the Centre for
Constitutional Studies at the University of Alberta, for Chapter 3; to the editor
of Journal of Political Ideologies for Chapter 5; to Kluwer Academic Publishers and
Larry May and Marilyn Friedman, editors of Rights and Reason, for Chapter 8; and to
the editor of Environmental Values for Chapter 21. My colleagues in the Department
of Political Theory and Government at the University of Wales, Swansea have been
very supportive of this project, not least in relieving me of teaching and admin-
istration at a crucial time in its gestation. I would particularly like to thank David
Boucher, Bruce Haddock and Rex Martin for their advice and encouragement.
Nicola Carr and the staff of Edinburgh University Press have been patient, cheerful
and immensely helpful. Having taken on such a large volume as my first collaborative
project, I could hardly have wished for a better team. My family, as ever, has been
immensely supportive, too, and Anne in particular has never flagged in her
encouragement during my many distractions and absences throughout the produc-
tion of this book.

vii
This page intentionally left blank
 LIST OF CONTRIBUTORS

Andrea Baumeister is Lecturer in Politics at the Rex Martin is Professor of Philosophy at the Uni-

University of Stirling. versity of Kansas and was, until 2000, Professor of

Political Theory at the University of Wales, Swansea.
Brian Baxter is Lecturer in Politics at the Uni-

versity of Dundee. Diana Tietjens Meyers is Professor of Philosophy

at the University of Connecticut.

Richard Bellamy is Professor of Politics at the

University of Reading. Rodney G. Peffer is Professor of Philosophy at the

University of San Diego.

Robert Bideleux is Reader in Politics at the Uni-

versity of Wales, Swansea. David Rasmussen is Professor of Philosophy at

Boston College, Massachusetts.

Margaret Canovan is Professor of Politics at the

University of Keele. Richard Sakwa is Professor of Politics at the Uni-

versity of Kent at Canterbury.

Mark Evans is Lecturer in Political Theory at the

University of Wales, Swansea. Jonathan Seglow is Lecturer in Politics at Royal

Holloway and Bedford New College.

Michael Freeden is Professor of Politics at the

University of Oxford and a Fellow of Mansfield Kate Soper is Professor of Philosophy at the Uni-

College. versity of North London.

Bruce Haddock is Reader in Politics at the Uni- Peter Sutch is a Tutorial Fellow in the School of

versity of Wales, Swansea. European Studies at the University of Wales, Cardiff.

Will Kymlicka is Professor of Philosophy at Andrew Vincent is Professor of Politics in the

Queen's University, Ontario. School of European Studies at the University of

Wales, Cardiff.
Peter Lassman is Senior Lecturer in Politics at the

University of Birmingham. Howard Williams is Professor of Politics at the

University of Wales, Aberystwyth.

ix
This page intentionally left blank
 Part One
THE LIBERAL TRAJECTORY
This page intentionally left blank
 1
ISSUES AND TRENDS IN
CONTEMPORARY LIBERALISM
Mark Evans
Introduction
For most contemporary liberal political theorists, it is
a defining tenet of their historical self-consciousness
that political theory was `reborn' in 1971 with the
publication of John Rawls's magisterially weighty A
1
Theory of Justice. After many years in which the grip
of linguistic analysis reduced it to the supposedly
impartial role of clarifying the meaning of political
concepts (and which led one commentator to de-
clare the more substantive activity of normative
position-taking in political philosophising to be
2
`dead' ), this book triumphantly demonstrated to
liberals that intellectual validity could, after all, be
claimed for a philosophical theory that proposed
what the content of political morality should be
and how it might be justified. No longer need the
grand theorising of the just or good political society
be taken up only by liberalism's enemies ± particu-
larly, in the years immediately before Theory's ap-
pearance, those to its left whose thinking shared
none of its hesitation over position-taking (a point
which demonstrates the overly parochial nature of
liberals' `death of political theory' claims).
Rawls's importance lies not only in his restoration
to liberals of a confidence that they were indeed
justified in using philosophical argument for the
substantive end of defending the liberal order. It lies
also in the way in which both the specific principles
of his theory and the way in which he defends them
have generated a truly enormous debate and a vast
3
literature. Although it would be an exaggeration to
say that the `Rawls industry' has swamped all other
dimensions of liberal academic argument, it is never-
theless obvious that liberalism's theoretical revival in
the last quarter of the twentieth century was largely
powered by the critical response to Theory. Even
non-liberals have found it difficult to address the
central questions of political theory without being
expected to orient themselves with respect to `the
Rawlsian position'. The influence of Rawlsianism ± if
only as a critical springboard ± can be detected in
most breaths of the resurrected tradition of political
theorising.
For anyone working in this tradition, it is thus
profoundly disturbing to encounter the charge that
the `revitalisation' of liberal political theory as any-
thing more than an `academic' preoccupation (in-
cluding the pejorative sense of that term) is actually
illusory. In a series of excoriating critiques, John
Gray charges that contemporary liberal theorising
`has done little more than articulate the prejudices of
an Anglo-American academic class that lacks any
3
understanding of political life in our age.' In its
contemporary philosophical guise liberalism lamen-
tably mistakes its own parochial, historical and cul-
tural specificity for an ahistorically and universally
applicable mode of political discourse and analysis.
Further, its concepts and concerns ± indeed, the very
methods by which it conducts its theorising and the
forums in which it takes place ± are peculiarly
divorced from the actual politics even of those
 MARK EVANS
western societies in which liberalism is said to be the
dominant political tradition. So for all of their
hubristic pretensions to speak to the pressing poli-
tical problems of our times, liberal theorists ± who
have encouraged a wholly delusive perception that
these can be all addressed within liberal terms of
4
reference ± `are reduced to talking with each other,
and to no one else, about topics of interest to no one
else, least of all in the liberal democracies they are
5
supposed to be addressing.' Rather than a `rebirth'
6
political theory suffered a tragic `still birth' in 1971.
Given this conclusion, Gray proposes that perhaps
political theory is not suited to substantive norma-
tive position-taking after all. Maybe it would be best
for it humbly to retreat once again to the insular
world of the ivory tower and engage, if not in the
dextrous niceties of linguistic philosophy, then to the
`modest exegesis of texts ± like Mill's Liberty, say ±
whose charm is in their distance from the world, or
7
any world we are likely to find ourselves in.'
Now such a devastating critique is hardly likely to
be accepted with alacrity in the massed ranks of
contemporary liberal political theorists, who are
much more likely than some of their linguistic-
philosophical predecessors to be wounded by the
charge of `practical irrelevance'. Doubtless there is
a temptation to treat it with an instantly dismissive
contempt. Yet despite the critique's intemperance,
there may be a lurking fear that there is something to
its claims. For example, liberal thinkers who deal
with what they regard as issues that affect all of their
fellow citizens (and are therefore concerned to keep
in view the `relevance' of their ideas to the wider
world) need little perceptiveness to sense that, out-
side of the academic community, their work may be
8
completely inaccessible. Whatever their intentions
to the contrary, they are indeed effectively talking
only to each other.
If `inaccessibility' is simply a question of the
technical difficulty of their writings, then there
may be no great cause for concern. After all, political
problems do tend to be extremely complex. We
should not necessarily expect there to be any easy,
publicly comprehensible ways of theorising them and
their possible resolution. The actual business of
politics requires no less technical understanding
and expertise, unavailable to the citizenry as a whole,
4
than does the academic study of it. Moreover, pol-
itics is hardly alone in being a concern whose full
comprehension is beyond all bar those few who have
the time, resources and aptitude to master its detail
even when it impacts greatly upon everyone. So
when liberal political theorists decide to take, say,
a Kantian approach to a concrete political problem ±
the selection and justification of basic constitutional
principles, for example ± they are obviously not
immediately vulnerable to the charges of unworldli-
ness and irrelevance simply by virtue of using ideas
unknown to, and probably not fully appreciable by,
those to whom the problem nevertheless applies.
We might, of course, justifiably challenge the
appropriateness of the approach taken and/or the
proposed solutions to it. Certainly, this is what Gray
attempts to do and there is obviously a serious debate
to be had about the applicability of contemporary
liberalism's methodology and recommendations. But
to reach any conclusion on this question substantive
engagement with such theory is required; its rele-
vance to the practical problem at stake must be put
to the test rather than denied straight away. And
even if the results of such assessment are ultimately
negative, it is still possible that the approach has
thrown valuable light upon the concrete problem.
There is no argument here, then, for abandoning
contemporary liberal political theory's normative
ambitions.
Gray's critique, however, amounts to much more
than just the ineffective charge that liberalism's
theoretical approaches are remote from the `real
world of politics'. It urges that the very problems
taken up by liberal theorists and those critics who
debate with them in the same philosophical tradi-
tions are simply not the problems (or at least the
prime or only problems) of the real world, either of
liberal-democratic societies or of human commu-
9
nities more widely. Gray finds it deeply absurd that,
for example, the journal Ethics ± which is one of the
foremost vehicles for contemporary liberal academic
moral and political theory ± should devote an entire
issue to a discussion of the merits and drawbacks of
10
market socialism. The contributors, he says, `might
have done better to discuss the prospects of the
restoration of monarchy in Russia ± far less of an
exercise in anachronism, and just conceivably a topic
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
of some interest to those whose fates it might
11
affect.' In more measured moments, Gray is happy
to grant that some of liberal theory's concerns do
connect with the practice of liberalism. But he still
wants to insist that there is much more to humanity's
political condition than liberal theory has allowed.
The purpose of the Edinburgh Companion to Con-
temporary Liberalism is to demonstrate just how far
liberal theory can stand from the kind of abstract,
unworldly, trivial philosophical navel-gazing
picted in this caricaturising critique. I do not wish
to deny that the Rawls-inspired strand of liberal
theorising in particular has perhaps tended towards
narrowness, both in terms of the issues discussed and
the range of authors who have engaged with them.
Without in any way wishing to belittle the value and
importance of this body of thought, it has probably
become all too easy to mistake its prominence as
meaning that it signifies all that to which liberalism
necessarily amounts today. `Contemporary liberal-
ism' is often reduced to what Rawls and a select band
of thinkers ± such as Dworkin, Nagel, Scanlon and
Raz ± have to say. This phenomenon may well have
softened liberalism up in the face of the kind of
critique rehearsed above. Yet not only is it still unfair
to say that Rawls, his interlocutors and their con-
cerns are divorced from what really matters in po-
litical practice, it is simply wrong to think that
contemporary liberalism does not and cannot range
far beyond the `Rawlsian ambit'. This Companion,
then, will certainly engage with liberal theory's `usual
suspects', but it will also show how liberalism can
tackle issues that Gray's critique might fool us into
thinking are somehow out of its reach.
To this end, the book does not simply present
`contemporary liberalism' in terms of a survey of what
its main protagonists have said over the past thirty
years or so. Instead, its guiding principle is to con-
ceptualise it as a set of issues and approaches that
might be said to constitute liberalism's current prac-
tical challenges and hence its current research agen-
da. After all, the importance of any one theorist
ultimately rests upon the importance of the political
problem she or he is addressing; only once that is
recognised, perhaps, is the value of their work fully
assessable. An `issue-led' account of contemporary
liberalism shows how unilluminating it is to think of
it as nothing other than the genteel exchanges of
philosophical claims between professional aca-
demics. By highlighting the issues, we can see di-
rectly how liberalism engages with the problems of
practical political life today. And with the emphasis
upon agenda-setting ± the topics that can and should
be addressed as well as what are currently being
addressed ± contemporary liberalism can be seen
to be live: relevant, proactive, continuously engaged
de- and adaptable.
If the Companion does not ultimately address all
the issues that Gray thinks to be so strikingly absent
from contemporary western political theory, it may
not be for want of space alone. Despite the claim
favoured by some that liberalism is becoming a global
doctrine and despite, too, the fact that its theorising
of international relations touches matters of concern
for everyone in the world, many contemporary lib-
erals are very wary (much more so than their pre-
decessors typically were) of any hubristic idea that all
of their concerns are matters for humanity in its
entirety. They recognise that `politics' may not be
exhausted by their account of it alone, so that some
of the (sometimes rather bizarre) issues whose ab-
sence Gray observes (such as theism, theocracy,
`Byzantinism' and monarchism) are indeed best left
to other traditions of political thought to address.
Nor need it be thought that the `contemporary
agenda', wide-ranging though I believe it should
be, is completely exhaustive of the issues that could
legitimately warrant liberals' concern now and in the
future. Some issues in practice come to matter more
than others at certain times (a feature which will be
evident in the selection of themes covered in this
volume). But precisely because its fluidity and flex-
ibility is emphasised, we need not think ± as many of
its critics seem implicitly prone to believe ± that
liberalism is somehow conceptually and/or morally
incapable of addressing the topics with which it is
not currently engaging.
In short, the Companion hopes to present what will
hopefully be compelling evidence that contemporary
liberalism has not misidentified the relevant political
concerns and that its approaches to their analysis are
not naõÈvely misguided. Even if debates about such
concerns may still rage and even if ± as some of the
chapters will show ± it may be justifiably doubted
5
 MARK EVANS
whether liberal theory is always the best-equipped to
deal with them, there is still much theoretically and
practically useful insight to be gained from it.
Contesting the Nature of `Liberalism'
One feature that the discussion thus far might seem to
share with much contemporary political theory is the
apparent assumption that `liberalism' needs no expli-
cit definition, as if it is enough simply to name it to
convey what kind of politics is intended. Those who
accept any version of Francis Fukuyama's belief that
liberalism's ideals have morally triumphed over all
others and that ideological conflict in the realm of
ideas is now over may be particularly inclined to hold
12
this assumption. Perhaps in earlier times, when this
battle was still being fought, it was always necessary for
liberals to be explicit and detailed about that for which
they were fighting. When positions no longer need
defending, they may no longer need demarcating.
In fact, of course, liberalism's meaning has always
been contested by liberals themselves and even if we
accept some version of Fukuyama's controversial
claim we should not delude ourselves into thinking
that this particular contest is also over. We should be
especially wary of two opposing tendencies in con-
temporary political theory's definitional handling of
`liberalism'. One, which may be particularly encour-
aged if one accepts that it now has no serious
competition in the Enlightenment tradition of pro-
gressive politics (as socialism, say, once provided),
equates liberalism with a few very general Enlight-
enment moral commitments centring around some
notion of `equal individual freedom'. On this ac-
count, if it is going too far to say that we are `all
liberals now', then it is still the case that non-liberal-
ism ends up being defined very narrowly in terms of
doctrines such as fascism, ethnic nationalism and
certain types of religious fundamentalism. In other
words, `liberalism' is little more than a label to cover
all adherents to the general progressive standpoint
opposed to these. From this perspective, it might
therefore be thought difficult for any `reasonable'
person to argue against it. But with such a capacious
definition `liberalism' would no longer be able to
stake specific political standpoints within the broad
range of `progressivism'.
6
The other tendency is a product of Rawls's pro-
minence in contemporary debates and has already
been alluded to. It identifies liberalism with the
specifically anti-perfectionist politics of constitu-
tional-democratic justice that his work has popu-
larised and which has subsequently engaged so many
in liberal political theory. Thus, it has become very
easy to think it axiomatic that liberalism is about the
`neutrality of the state', `the priority of the right over
the good', `the politics of public reason' and other
leitmotifs of Rawlsian theory. Now we have already
conceded that Rawls has inspired a very influential
and perhaps attractive conception of liberalism. And
I will shortly sketch how it frames many of the
predominant themes in liberalism's `post-Rawls' re-
search agenda. But as Michael Freeden forcefully
demonstrates in Chapter 2, it offers a very histori-
cally and culturally specific interpretation of liberal-
ism (and one whose Kantianism is, moreover, not
nearly as faithful to its cultural roots as it likes to
think).
One consequence of this narrowing of liberalism's
definition has been to render it vulnerable to criti-
cisms that would comfortably bypass other members
of the liberal tradition. Indeed, it often gives the
impression of being blind to the existence of such a
diverse tradition behind it. Chapter 2 therefore
serves to remind contemporary liberals of its exis-
tence. In doing so, it suggests that one research task
of liberal theory now is to open up possibilities of
redefinition and perhaps retrieval of certain liberal
self-images and commitments, lost from sight in the
fog of Rawlsianism, which may be valuable resources
for the doctrine's defence and development.
Neither tendency serves us well, then, in trying to
characterise liberalism. But this claim does not rest
upon any assumption that we can describe an ade-
quate, generally acceptable typology for it that is at
once sufficiently expansive to cover all liberalisms
but also sufficiently detailed not to become indis-
tinguishable from related but distinct political doc-
trines. In fact, it is probably best to treat liberalism as
in essence internally contested. The `liberal label'
denotes a political identity to be regarded as one
that is not permissively open-ended but which oper-
ates within certain conceptual and historical con-
straints, yet whose diversity within these boundaries
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
should be celebrated rather than regretted. As most
of the chapters implicitly demonstrate, `liberalism'
remains an invaluable shorthand term. But, while its
various usages cannot be totally antithetical to each
other without at least one of them being a misuse, we
should not make the mistake of thinking there must
be any thoroughgoing identity of views and commit-
13
ments behind them all.
Leaving it, then, to the individual contributors to
use the term as they wish, I will here confine myself
to the identification of a few general themes and
tendencies in definitions of liberalism, to preface the
14
arguments in the rest of this introduction. It is
important to note that the term can be used to
characterise quite distinct topics or areas of concern
in particular ways and that confusion between them
may contribute to the vagaries of debating liberal-
ism's nature. Three areas in particular are worthy of
note: (1) liberalism as a political morality; (2) lib-
eralism as a form of political regime; (3) liberalism as
a form of culture, or social order.
Under (1), liberalism posits the individual as the
fundamental unit of moral and political concern.
Although this need not be incompatible with a
strong commitment to a notion of community and
the common good (and should not be thought to
imply an unavoidable reliance upon an atomistic
methodological individualism at the heart of its
social-explanatory theory), liberal morality does
not deliberately subsume the individual in, or totally
subordinate it to, any collective unit. The liberal
individual is held to be `equally free' and the purpose
of liberal moral and political society is, in the first
instance, to protect and promote the interests of
individuals as equally free beings. This is at the heart
of the justifications for whatever constraints such a
society will wish to place upon its members. How
precisely this notion of equal freedom is to be inter-
preted is part of what fuels the ongoing contest over
liberalism's nature. The concept of `justice' is often
used nowadays as the peg upon which to hang this
debate. And in their accounts of what justice de-
mands of people to facilitate the equal freedom of all,
liberals can range from very individualistic moral-
ities, which emphasise political `non-interference', to
highly proactive theories of redistributive social jus-
tice that posit substantial degrees of individual inter-
7
dependence and strong ties of mutually supportive
obligations between people. Given the accelerating
spread of what Richard Rorty calls the `human rights
15
culture', however, it is fair to say that liberals in
general have increasingly employed the notion of
`individual human rights' as the conceptual device by
which to express at least part of their moral commit-
ment. They are not, however, conceptually bound to
do so; there are other ways of formulating the general
commitment to equal individual freedom. Nor
should it be thought that liberal individual rights
are necessarily characterised and prioritised in a
manner that would negate or trump any official
support for social justice.
Area (2) concerns how liberalism's moral commit-
ments are cashed out in terms of political institutions
and practices. As we might expect given the ob-
servations above, the liberal tradition has ranged
from support for minimal or `night-watchman' states
presiding over loosely regulated free-market capitalist
economies (sometimes now grouped under the head-
16
ing of `libertarianism' ) to the kind of economically
interventionist welfarist politics sometimes thought
17
to be the preserve of socialism or social democracy.
Once again, though, we can identify certain wide-
spread tendencies across the liberal spectrum. For
example, liberals tend to support some version of
representative democracy as being (a) the most ideal
and/or efficient way in which citizens can be given a
say in how they are ruled; (b) the most ideal and/or
efficient way in which power can be held accoun-
table by those over whom it is exercised; (c) the best
means by which the state's power can be limited to
those specific areas of social life in which liberal
politics is deemed to be rightfully concerned.
The argument for liberalism's equal individual
moral freedoms is reinforced by the commitment
to democracy, for it emphasises how the exercise
of political choice requires freedoms of expression
and association (for example) to be rendered mean-
ingful. The idea that politics is also a limited activity,
with realms of life being left free from direct political
regulation, is a further consequence of the liberal
account of freedom. Though the extent to which it is
to be regulated is significantly variable within liberal
thinking, liberals tend to support the general idea of
a market economy in which private property is
 MARK EVANS
allowable (if not necessarily to the exclusion of any
form of public ownership). The capitalist character
of the liberal order will obviously inform the nature
of area (3), in which we can isolate two prominent
features of liberal society.
The first feature is civil society, the recognition that
a significant part of public social life should be left
free from direct state control, instead being the realm
in which people may freely associate themselves for
their various ends. This includes not just (significant
parts of) economic life, but other institutional asso-
ciations, some perhaps devoted to providing services
(private health-care providers, for example), others
tied to particular projects or commitments
(churches, charities, working people's associations,
and so on).
The second feature consists of the cultural char-
acteristics of `liberal-democratic individuality' .
phrase is designed to capture two aspects of the
character-ideal that liberalism tends to hope, expli-
18
citly or implicitly, will flourish among citizens.
is the disposition to act towards others in ways that
uphold the institutions and the ethos of liberal
society: to be, for example, respectful, tolerant of
disagreement and alternative lifestyles, prepared to
cooperate with others to live under mutually accep-
table terms of social and political accommodation.
Second is the idea that the interests of the individual
are (within terms laid down by justice) to be cele-
brated. The fact that `people's lives are their own to
lead' is something that liberal politics is designed to
facilitate, proactively and/or through non-interfer-
ence. Many liberals (including Rawls, his `anti-per-
fectionism' notwithstanding) favour the idea that
individuals should freely develop their own, unique
characteristics and capacities as the central project in
the pursuit of the good life.
This is not to say that liberalism's `cult of the
individual' in principle favours egoism at the expense
of other-directed social concerns. As contemporary
liberals have arguably become more sensitive to what
they regard as the expanding range of permissible
differences in human forms of life, cultural pluralism
has come to play a more prominent role in their
thinking. Hence, liberals now tend to attach great
significance to cultural (that is, collective) identities
as crucial social resources in the constitution of
individual identities and projects. But, again, they
tend to want to resist the possibilities of such iden-
tities swallowing up individual opportunities not to
conform to them, and certainly to oppose any iden-
tity that fundamentally compromises the commit-
ments of liberal citizenship.
The above is not meant to be an exhaustive
description; it merely indicates some of liberalism's
prominent general characteristics. But this is enough
to orient the rest of this introduction. For next, I first
identify the main research interests in liberal think-
ing that followed in Rawls's wake. We will then be in
a position to see how the chapters of this volume
might build upon that legacy and address the kinds of
concerns that Gray's critique would have us regard as
alien to it. Finally, I suggest a further line of argu-
ment in support of the claim that there are still
This crucial issues with which contemporary liberal poli-
tical theory may substantively engage. Liberals, I
contend, have nothing to be complacent about.
First There are good reasons to think hard about the
longer-term viability of the liberal order.
Issues for a
Contemporary Liberal Research Agenda
In outlining the main themes in contemporary lib-
eral political theory to date, I don't pretend to be
offering an exhaustive account of liberal scholarship
in the years since Theory's publication. I hope instead
simply to offer a sketch to place in context the
directions in which this book suggests liberal thought
may run hereafter. Bearing in mind the caveat that
not everything even in today's liberalism is reducible
to `Rawls and his critics', it is nevertheless easy for a
schematic account such as this to tease out these
themes in terms of debates raised by and generated
from Rawls's work. Two in particular can be traced in
their current form back to Theory. The first concerns
the content of justice: what is it that a just state
should seek to secure for its citizens and by what
principles does it distribute what G. A. Cohen has
19
called justice's `currency'? In defending a concep-
tion of distributive social justice, Rawls's answers to
these two questions were that `social primary goods'
were the raw material of justice, and that they should
be distributed within the institutions and relation-
8
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
ships of society's `basic structure' according to his
20
`two principles of justice'.
Perhaps the most politically resonant attack on his
theory came from Robert Nozick's Anarchy, State and
Utopia. Capturing the ethos of the New Right
(which in numerous western states during
1970s and 1980s electorally eclipsed the kind of
politics many believed Rawls to be defending), No-
zick attacked the very concept of distributive justice
and defended a rights-based, purely procedural ac-
count of the just state that largely served to protect
individual property holdings from redistributive po-
21
licies. At the same time, there was also renewed
interest in the work of F. A. Hayek, who provided a
far more intellectually rigorous defence of procedural
22
against distributive social justice. Together, these
authors contributed much to the revival of what
many thought was harking back to a `classical liberal'
political theory (as we have already noted, people on
both sides of the debate felt sufficiently embarrassed
by this family connection to parcel off this revanchist
classical liberalism under the heading of `libertarian-
ism').
It is a measure of where many academics' sympa-
thies tended to lie that there was more extensive and
innovative intellectual work done on the other side
of the political spectrum. This embraced rather than
jettisoned the notion of distributive social justice,
and sought to contest Rawls's specific characterisa-
tion of it. Indeed, it was precisely because of the fact
that political ideas to the centre and left of the
political spectrum were in electoral retreat that many
thinkers became particularly concerned to bolster
their philosophical credibility. Here, then, the de-
bate was posed around how best to pattern the
distribution of whatever it was that justice demanded
should be equalised. Amartya Sen's seminal 1979
Tanner Lecture `Equality of What?' set the tone for
the debate, and helped to generate a plethora of
rivals for Rawls's claim that social primary goods
23
were the proper focus for egalitarian policy.
The second issue raised by Theory concerned the
justification of political principles: how are they to
be derived and what reasons could be given for
accepting them? Committed to the idea that political
principles should respect all citizens as equals and be
at least the potential object of their agreement, we
have already observed how Rawls propagated the
image of liberalism as committed to a politics of
impartiality or neutralism. For him, liberal politics
was fundamentally about justice alone. Justice em-
bodied principles of `right', to be justified largely
the upon the basis that they were acceptable by all to
whom they applied. These were distinguishable from
`conceptions of the good', the various forms of life-
style from which individuals should be allowed to
choose and which were not therefore the proper
objects of political favouritism.
Much interest was generated over Rawls's revival
of contractarianism as a method of derivation and
justification. His use of the `original position' and its
assumptions about individual rationality and moral
motivation generated an extensive literature, seeking
to test the plausibility of the general idea that
political principles could be derived from idealised
bargaining between equally free, rational indivi-
24
duals. From the early 1980s, however, what be-
came known as the `communitarian' critique of
liberalism developed in reaction against this whole
tendency, epitomised by Michael Sandel's attack on
25
Theory in his Liberalism and the Limits of Justice .
Communitarians argued that contractarianism fun-
damentally misunderstood the nature of moral
agency given its tendency to use `abstractive' meth-
odologies, attempting to generate moral and political
principles from hypothetical conceptions of `rational
moral agents' drawn in abstraction from any parti-
cular identities, commitments and social contexts.
By ignoring the crucial fact that individuals were
essentially social creatures, with their identities and
moralities being the products of particular cultures,
such theories were predestined to fail to generate
conceptually and substantively adequate moralities.
Hence, as exemplified by Michael Walzer's Spheres of
Justice, communitarians argued for what amounted to
a hermeneutic political theory, deriving its moral
principles from an interpretive analysis of concrete
social understandings. Acknowledging the diversity
of societies and social meanings, this approach con-
sequently leant towards (even if it often wished to
resist a total capitulation to) a normative as well as
26
an epistemological relativism.
The debate between liberals and communitarians
also broke off into separable questions about the
9
 MARK EVANS
27
content and justification of political morality. Part
of the communitarian argument over political mor-
ality's content attacked the impartialism cham-
pioned by Rawls, denying that the `right' can be
conceptually separated from the `good'. One sub-
stantive consequence of this was to give politics
responsibility for promoting rather than avoiding
the `good' and, in some versions, communitarism
was happy to bring into political thinking a direct
concern with, for example, cultivating virtues of
28
character. Although many communitarians did
not reject liberal morality in toto, they tended to
criticise the narrowness of what they perceived to be
liberalism's overly individualistic moral stance, con-
tending that it is ill-equipped to accommodate the
importance of group identities and collective goods
to political morality.
Much liberal writing in recent times has thus been
devoted to showing that liberalism is not as blind to
such concerns as the communitarians believe. The
complaints that it is insufficiently responsive to the
demands of community have led, for example, to an
increasing interest in what is often called `civic
liberalism'. This frequently draws upon republican
traditions of thought to postulate interdependence
between liberal individuality and `good citizenship',
which rests upon the willingness and ability of
citizens to participate in and uphold certain shared
29
practices in liberal-democratic society. And, as his
contribution to this volume shows, Will Kymlicka
has been at the forefront of the attempt to establish
the coherence and desirability of `group-based' con-
cerns in liberal politics.
Liberals have also been prompted to respond to
the communitarian assault on their favoured justifi-
catory strategies. Rawls himself has led liberalism's
response; in his Political Liberalism he effectively
localises his political theory in that he now believes
that political justifications for liberalism need to be
drawn from the shared ideas in the public political
culture of a constitutional democracy (a strategy
sometimes called `liberal communitarianism', pro-
posing the dissolution of the antithesis). His recent
work has given much prominence to the theme of
`public reason' as the source of justifications for
liberal principles. Strenuous efforts are now made
to explicate its notion of `reasonable agreement' in a
10
manner that, on the one hand, doesn't rest upon
dubious claims about actual universal consensus in
liberal societies on liberal principles yet, on the other
hand, still holds out as plausible the possibility of
genuine agreement. Particularly in so far as many
(though emphatically not all) liberals have accepted
the need for a Rawlsian `communitarian turn', this
debate has partially substituted for the more abstract
notions of rational calculation favoured in the earlier
30
contractarian debate.
One of the possible limitations of the `commu-
nitarian' school is that, ironically enough, some of its
exponents have tended to use abstractly unspecific
conceptions of `community' which are, in fact, im-
plicitly homogenising. They suppress full recognition
of diversity and conflict over values and practices,
overly simplifying the project of hermeneutically
obtaining a singular political morality for a commu-
nity. Many liberals have therefore paid close atten-
tion to theorising this diversity and the relevant
sense of `community' for the politics of a pluralistic
society. Often, they now raise their concerns within
the terms set by `multiculturalism' and `the politics of
difference.' Feminist thinkers have been at the fore-
front of the project to expose and jettison from
political theory concepts that they argue to be
`difference-blind' to the extent that they utterly fail
to spot their own partiality. (Iris Marion Young's
Justice and the Politics of Difference is representative of
31
this trend. ) Many of liberalism's critics in this
regard tend to argue that the whole Enlightenment
heritage, which liberalism is held to exemplify, un-
successfully masks its own particularity (western and/
or masculine and/or bourgeois) in a cloak of cosmo-
politan universalism. They want to displace from
political thinking any abstractly conceptualised mor-
al concerns with `humanity' (those which treat
objects of moral concern only as `equal rational
moral agents', for example) and any other concepts
which in fact represent a very particular, partial way
of seeing the moral and political world.
In reaction, we find a greater engagement by
liberals with other cultures, seeking to understand
and appreciate their differences in a quest to con-
ceptualise what might be necessary to establish a
political order that can justly accommodate `differ-
ence'. As the chapters by Will Kymlicka, Andrea
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
Baumeister, Jonathan Seglow and Kate Soper show,
for example, this has resulted in liberal arguments
which are much more sympathetically responsive to
the kinds of concerns that critics may rightly have
complained to have gone unrecognised in certain
strands of liberalism but who wrongly charged lib-
eralism in general to be incapable of addressing. We
also find a rejection of the apparent normative
localisation of some communitarians and `difference
theorists' in the recent upsurge of interest in norma-
tive international relations theory.
Concluding this brief summary, we should note
how the heightened sensitivity to diversity contri-
butes to a further problematisation of liberalism's
justificatory project. Where, not so long ago, con-
tractarianism and `morals by rational and reasonable
agreement' seemed to be predominant themes, the
debate has widened and fragmented as the `fact of
pluralism' (itself variably interpreted) has suggested
to some that only a Hobbesian politics of modus
vivendi is possible and, to others (the postmoder-
nists), that the whole philosophical project of liberal
justification is redundant. Here again, there is a
tendency to question liberalism's supposedly Enlight-
enment-bound conceptions of morality and moral
justification.
From this overview, the following seem to emerge
as contemporary liberalism's main preoccupations:
1. the content of `justice';
2. the justification of liberal political principles;
3. the conception of `citizenship' and its relation-
ship to individual freedom, civic virtue and the
requirements of civil order;
4. the conception of a liberal political commu-
nity as applicable to a pluralistic, multicultural
society;
5. the defence of liberal (`Enlightenment') poli-
tical concepts and concerns in the wake of the
challenge posed by pluralism and `difference'.
We are now in a position to summarise the con-
tributions to this volume, seeing how they pick up,
advance and supplement these themes.
As already indicated, Michael Freeden's opening
chapter takes us through numerous academic char-
acterisations of liberalism's nature to confirm its
11
essential diversity. It contends that the Anglo-Amer-
ican philosophical liberalism, whose pre-eminence
in contemporary liberalism has already been identi-
fied, has not displaced the myriad other forms of
liberal thought and discourse. Future research on
liberalism would do well to bear this in mind.
Next, Will Kymlicka demonstrates how liberalism
is capable of addressing many of the concerns raised
by those critics who contended that it was concep-
tually incapable of catering for group- rather than
individual-centred demands. He develops a case for
`group rights' as the centrepiece of an account of
what he calls `ethnocultural justice'. The following
four chapters analyse issues of the nature of liberal
community and citizenship, and the particular un-
derstanding of `justice', that emerges from his dis-
cussion. Andrew Vincent argues that liberalism's
account of citizenship ± the centrepiece of many
liberals' arguments against the communitarian cri-
tique ± sits extremely uneasily astride two contra-
dictory currents: a universalist cosmopolitanism
which underpins liberalism's moral principles and
a highly particularist nationalism which actually
grounds the criteria for citizenship in liberal states.
By explaining the nature and implications of these
currents, he puts this issue firmly on liberalism's
research agenda as requiring urgent attention. Mar-
garet Canovan's chapter throws light on this pro-
blem from a different angle. She identifies two
ostensibly very different liberalisms: the `Crusaders'
(those liberals who prioritise a universalist human
rights agenda) and the `Sceptics' (much more con-
tent with the conduct of a limited politics inside
their own national boundaries). She argues that,
their highly distinctive commitments notwithstand-
ing, both rely tacitly but crucially upon notions of
nationhood. The suspicion that they share of `na-
tionality' in politics may therefore be misplaced;
again, it is shown that there is an intimate yet
contentious relationship between liberalism and
`the nation', the nature of which requires much
closer analysis than has been hitherto given.
Andrea Baumeister then presents a comprehen-
sive discussion of various feminist critiques that have
condemned liberal conceptions of citizenship for
their apparent blindness to the political significance
of difference. She argues, however, that Kantian
 MARK EVANS
liberalism may not fall foul of this deficiency and
that, hence, feminist politics may not need to aban-
don paradigmatic liberal ideas. Jonathan Seglow's
chapter provides an account of one of the newer
additions to liberalism's reflections upon justice,
which is implicit in Will Kymlicka's notion of
`ethnocultural' justice: the idea that `recognition'
of one's identity as well as `redistribution' of resources
is part of what justice requires.
Having raised the question of `justice', the follow-
ing two chapters take us back to its distributive
aspect. Their analyses are crucially informed by
Rawls's `difference principle', but they draw notice-
ably divergent conclusions over its import. Rex
Martin agrees that it has been absolutely pivotal
in liberalism's debates about justice. He offers a
rigorous account of how it is intended to work
and then seeks to demonstrate that it is critically
indeterminate in specifying particular institutional
and policy recommendations. To the extent that it
still takes the difference principle as one of its
benchmarks, this conclusion indicates that liberal
discourse about justice may have reached a theore-
tical logjam through which it is urgent for liberals to
think. Rodney Peffer, on the other hand, is happy to
argue that Rawlsian theory ± with only a few mod-
ifications ± can lead us in a definite political direc-
tion. However, he believes it is one that
justifiably be called Marxist and he sets out to defend
the application of this label to his Rawlsian theory.
He would stridently disagree with Gray's claim,
noted earlier, that reflections upon market socialism
and the like are exercises in irrelevance and fantasy.
But the striking point is the argument that this is
where liberal justice may well lead us: reason enough
for radicals not to be dismissive of it.
The following three chapters turn us to the debate
about the justification of liberal values. It is not
implausible to suggest that the increased interest
in, and perplexity of, the justification debate reflects
a wider cultural anxiety about the status of moral and
political principles, in a world of such plural values
and moral outlooks that scepticism seems to be the
irresistible conclusion even when one would so
desperately wish otherwise. Peter Lassman's chapter
confirms that pluralism constitutes the distinctive
problem of modern politics. But the problem is one
which, for liberals in particular, is compounded at
the heart by controversy over how to understand `the
fact of pluralism' in the first place. His suggestion is
that the project of liberal justification may best be
advanced by the argument Judith Shklar puts for-
ward for an apparently modest `liberalism of fear'. In
`Pragmatist Liberalism and the Evasion of Politics', I
evaluate Richard Rorty's rather more laid-back and
sanguine approach to the project, which is prepared
to dispense with philosophical argument in support
of liberal principles altogether. I argue that the
`Rortyean view' can and should be elaborated more
fully than Rorty himself musters, to grasp just what
consequences his notion of `contingency' might have
for liberal theory and politics. Bruce Haddock con-
tends, however, that the Enlightenment-inspired
aspiration of achieving a universal morality justified
by reason, challenged in the previous two chapters,
can be kept alive. He proposes that a neo-Kantian
practical reason can come into play whenever people
are faced with moral dilemmas. It is upon this basis
that we may be able to construct a case at least for
some of liberalism's fundamental principles. The
perturbing uncertainties of `contingency' in morals
± principles ungrounded by reason ± can be avoided.
Pluralism, of course, poses practical problems of
political organisation as well as theoretical ones of
can philosophical justification. For Richard Bellamy,
liberal theory and practice as it stands is not well
suited to resolve them. He forwards a robust case for a
more participatory form of governance that draws
upon the venerable republican political tradition. In
light of this proposed revival of `republicanism',
David Rasmussen comments upon the recent debate
between John Rawls and Ju
È rgen Habermas, which
can be seen as an exploration of the possible accom-
modation by Rawlsian liberalism of republican po-
litical insights.
As described earlier, many feminists have shared
with postmodernists a pluralistic, `difference-in-
spired' critique of the Enlightenment and hence
the forms of moral and political thinking informed
by it, of which liberalism is a paradigm. To combat
some of the more drastic conclusions drawn from
this, Kate Soper argues against a wholesale rejection
of the Enlightenment tradition. She shows that it
was a good deal more sensitive to feminist causes
12
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
than many contemporary postmodernists in particu-
lar allow. To persist with an `anti-Enlightenment'
theme in championing such causes today may there-
fore be to do violence to what is valuably progressive
in that tradition. This is not to deny that cultural
diversity, among other considerations, should rightly
prompt liberals into a re-evaluation of some of their
concepts. Diana Tietjens Meyers's chapter is pre-
sented as an example of how this critical self-analysis
may proceed. She provides a powerful, innovative
argument for the need to rethink many standard
conceptions of autonomy. She deals with exactly the
kind of question upon which so many liberals stum-
ble: how to conceptualise `autonomy' in light of the
fact that women in some cultures appear autono-
mously to affirm the practice of female circumcision,
which western liberals usually cannot imagine being
supported at all by anybody in a genuinely autono-
mous way. Drawing upon material rarely utilised in
liberal debate, this chapter indicates some of the new
directions to be taken and sources to be pursued in
liberal research.
In the post-Cold War world, many traditional
understandings of politics are becoming redundant
as power increasingly slips from nation-states to
regional and global sites. As an example of how
political theory must innovate conceptually to keep
pace, Robert Bideleux substantially adapts the liberal
conception of civil association to characterise the
institutional form being taken by the European
Union, still very much in the process of evolution.
Provocatively, he argues that in the context of
European politics we must recognise that liberalism
is decisively shedding its connections with democ-
racy. Howard Williams's chapter switches our atten-
tion to the question of the global order and the
normative goal of securing principles by which states
may live together in peace. He assesses the utility of
Kant's thinking in this regard as a classical and
quintessentially liberal starting-point from which
contemporary reflections may proceed. A flavour
of the latter is provided in Peter Sutch's contribu-
tion, which goes on to articulate the concept of
`developmental communitarianism' as the justifica-
tory strategy that a liberal `law of peoples' (to borrow
Rawls's phrase) may usefully adopt.
The Companion ends with three chapters high-
13
lighting some issues fairly new to liberal research but
which may become established preoccupations. In
another example of new seams for liberals to mine,
Richard Sakwa shows how strikingly varied liberal
discourse has become in post-communist states, as
the doctrine attempts to define and orient itself in a
social and political context deeply unfamiliar to
those which provide the backdrop of much liberal
thought to date. Brian Baxter takes up the charge
that liberalism is ill-equipped to face the challenges
of ecologism and environmental politics. He con-
tends that `ecocentric' theories are indeed likely to
be inimical to liberal concepts and politics, but
argues that these are not in fact desirable as they
stand. He then presents a `personcentric' argument
which shows how liberalism may be reconciled to
ecologist concerns. Finally, in the last chapter I
address a question that many liberals once had no
qualms in tackling but which the `Rawlsian ten-
dency' has tried to push out of liberalism's concerns:
the articulation of personal ideals of how it is good to
live, with a view to sanctioning their promotion by a
liberal state under certain conditions. I note that this
`perfectionist' liberalism has been rehabilitated in
Rawls's wake, but it remains peculiarly reluctant
to articulate any particular ideal of the good, an
abstinence which I claim is neither necessary nor
always desirable for liberals. Accordingly, I begin to
sketch some of what I call `candidate conceptions of
the good' in an effort to reconnect liberal politics and
the classical practice of ethical reflection on the
nature of the good life.
Amid all this it is difficult, I think, to spot too
much `irrelevance' and `triviality' in the subjects
covered, as we might be led to expect given Gray's
critique. They engage with real issues concerning the
defence and promotion of the kind of politics that,
after all, most people in the West ± and, increasingly,
beyond ± wish to support. The argument that this
Companion wishes to promote is that political theory
is a vital component of the political project to defend
or contest the liberal order. This claim is based upon
the idea that the problems faced by the liberal order
are sufficiently complex and perplexing to demand
serious, rigorous philosophical reflection. That is
what contemporary liberal political theory attempts
to deliver.
 MARK EVANS
To end this introductory chapter and to supple-
ment the research agenda already mapped out above,
I propose a further demonstration of the potential for
relevant, serious-minded analysis in liberal political
theory. This is designed to test its capacity for critical
self-reflection and to put on its own menu the most
fundamental challenges that liberal practice could
actually face. It will outline yet more avenues for
liberal theory to take, dispelling even more vigor-
ously the illusion that it is complacently trivial.
Prospects for Liberalism
and the Liberal Order
Many of the contributions to this volume suggest or
imply that, though the challenges it faces may indeed
be perplexing, it is conceivable that liberalism has
the resources to address them not only theoretically
but also practically. That is to say, the challenges can
be understood from within a recognisably liberal
conceptual framework and the institutions and prac-
tices of a liberal social and political order need not
have to undergo a revolutionary transformation for
them to be tackled adequately. On this view some
may argue that, though liberalism may not have all
the answers to every problem or even the best answer
in certain instances, it informs what is, on balance,
the best standpoint from which to engage with
politics in the modern world.
It is hoped that this Companion will help readers to
draw their own conclusions about the validity of this
claim, ranging as it does across many of the issues by
which liberalism should be judged. However, the test
or thought-experiment I wish to sketch hypothesises
deep internal tensions within the liberal `settlement',
which might plausibly threaten to destroy the very
coherence of liberalism's commitments. My proposal
is that there is no reason why liberals themselves
cannot describe such scenarios, which are not ne-
cessarily exercises in fantasy but which point to real
potentialities. If they can face up to such possibilities
and think hard about how to address them, then it is
difficult to think of anything more serious for them to
tackle. I hasten to note that I emphatically do not
support the gloomily anti-Panglossian view that I
shall construct to show how this test might be
conducted, not least because no such exercise could
14
proceed too far without indulging in the speculative
realm of crystal ball-gazing. (It would also obviously
be perverse to persevere with contemporary liberal-
ism, as many chapters in this book argue we might, if
one believed it to be so pitifully doomed.) But
forcing liberals to analyse such negative arguments
could strengthen their thinking and convictions,
perturbing to the liberal imagination though they
may be.
Here, at the beginning of the twenty-first century,
it is particularly apposite to ponder the permanence
of the liberal achievement. Confidence in liberal-
ism's ability to adapt in the face of modernity's
problems, or at least to be able to cope with them
better than any possible alternative, doubtless re-
flects the widespread belief (epitomised in Fukuya-
ma's work) that it represents the pinnacle of a
successful struggle by humanity to work out the most
commodious form of political system. But I urge that
contemporary liberal theory should not begin with
this belief as an automatic presumption. Instead, it
should commit itself at most to a continuous testing
of such a proposition; it can never be taken for
granted, for circumstances are too fluid to have blind
faith in the `historical vindication' of liberalism.
So let us consider the argument that, though it
may have beaten off many rivals to date, liberalism
remains fundamentally flawed. There is no guaran-
tee, it says, that liberalism's intellectual apparatus
will be able to help us cope with all of the practical
problems that the liberal political order may face.
And those problems could threaten to overwhelm
liberal practice. Note that this scenario can be
imagined without believing that there is an alter-
native doctrine that, all things considered, will cope
any better with liberalism's predicament. For who
can say whether the victor in the `battle of the
ideologies' will not still be torn apart as its panoply
of moral ideals and political prescriptions founder in
this deeply suboptimal social world of contradictions,
complexities and antagonisms?
To substantiate this idea, we can reflect upon how
a historical awareness should jolt us out of any
sanguine complacency concerning the fruits we
might now think we can forever harvest from liberal
society. A hundred years ago, the optimistic faith in
the direction of rational human development had yet
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
to be darkened by the slaughter of the Somme, the
bureaucratically managed genocide of Stalin's purges
and the Nazis' Final Solution, and the spectre of
global nuclear destruction. No such underestimation
of the human capacity for evil is now likely to be
accepted without grave reservation. Moreover, glo-
bal capitalism is producing such profound institu-
tional and cultural changes at an ever-accelerating
rate that it is increasingly difficult to think of many
of the norms, artefacts and `certainties' of our civi-
lisation as settled. Change is too dizzying, too dis-
orientating for us to feel `at home' for long in any one
manifestation of modernity; as Marx and Engels so
32
presciently put it, `all that is solid melts into air.'
Given all this, with what reason, exactly, could we
hold firm to the beliefs that, whatever happens,
liberalism's ideals will be capable of remaining our
moral inspiration? What follows is a selection of
perhaps numerous ways in which `anti-liberal' pos-
sibilities might be postulated.
Where once the liberals' `bogeymen' were primar-
ily fascists and communists of various hues (and they
may still threaten them in places, of course), it is
nowadays more often the figures of the authoritarian
nationalist and the religious fundamentalist who are
seen as either the main threats to established liberal
orders or as the actually-existing obstacles to the
spread of liberalism in, say, post-communist societies.
The basic themes of their anti-liberalism, however,
have largely remained the same as their predecessors
in liberalism's gallery of rogues: the denial of basic
liberties as the prime political concern, the prepa-
redness to subordinate individual interests to (illu-
sory) `common goods', an overt discrimination
between groups of people in a shameless abandon-
ment of equal respect, and a willingness to prefer
coercion over persuasion in pursuit of such repressive
political ends.
It is obviously right to confront the challenges
that movements such as nationalism and religious
fundamentalism pose to the liberal order. A recog-
nition that liberalism may not be appropriate in all
four corners of the world should not necessarily
mean that we should automatically forego the mak-
ing of the case for liberalism in at least some parts of
the globe where anti-liberal nationalisms and fun-
damentalisms currently hold sway. Nor should we
15
be blithely ignorant of the problems these political
currents cause within actually-existing liberal
democracies (as manifest, for example, in the per-
sistence of racism and other forms of violent `hate
politics'). But let us for the sake of argument deny
that they constitute ± or are likely in the foreseeable
future to constitute ± genuine systemic threats to
33
the latter. They alone do not appear to have the
strength to destroy liberal democracy as did fascism
in many European states did during the inter-war
years.
A more relevant and hence perplexing theoretical
challenge is this. Even if we grant the argument that
liberalism does not now seem to have seriously
credible, morally and practically superior rivals, it
is possible that liberal commitments and practices
may be torn apart from within. There are two ways in
which this might happen:
1. At the level of ideals, some of its moral goals
may be realisable only at the expense of some
of its others. All of the goals may be thought to
be valuable and we might have hoped that the
victor in the battle of the ideologies might
have been able to preserve them together. But
our destiny is not so rosy; we cannot but
endure significant moral loss because of the
practical uncombinability of liberalism's aims.
Whatever doctrine emerges from the choices
we will ultimately have to make may not
deserve the name of `liberal'.
2. At the level of practice, the kind of society that
liberal doctrine supports ends up inadvertently
subverting at least some of its aspirations.
There is no way to fit the latter with the
socio-economic order that the doctrine sup-
ports. Again, in crude terms, something will
have to give way and what results may not
warrant the liberal label at all.
To illustrate just a few of the ways we might render
vivid this conclusion in our thought-experiment, we
can return to our division of liberal concerns into (1)
a moral theory; (2) an account of the ideal political
regime; (3) an account of the ideal social culture. In
so doing, we can see how on this account some of the
faults of which, it was argued above, liberalism's
 MARK EVANS
principles are incorrectly convicted may yet emerge
as unintended consequences of its practice.
Challenges to the Morality of Liberalism
Although I have already indicated that liberalism is
by no means incompatible with the robust concep-
tions of community favoured by `communitarians'
(loosely defined), its central moral focus remains the
individual and some notion of the interests of in-
dividuality. And the more pluralistic one's concep-
tion of human nature, the more individualistic (that
is, less communitarian) that notion may turn out to
be. It could be argued (as I will suggest below) that
modern capitalist society further `individualises' its
subjects by undermining the culture and indeed the
physical constitution of traditional communities
while creating no new `community' in their places.
Liberalism, it might then be contended, tends to
reinforce such developments by favouring the inter-
ests of individual freedom over human collectives.
Here, I shall mention just two moral issues that could
place this doctrine ± and its reinforcement of `in-
dividualisation' ± under intolerable strain.
The first is the ongoing scandal of international
distributive injustice: the billions who live in the
most desperate, degrading poverty, a tragedy that
seems to render them all the more prone to the other
terrible catastrophes brought by war, weather and
disease. We need not believe that redistribution of
wealth from the rich nations alone would be enough
to solve these problems altogether to recognise that
they could be substantially alleviated if such redis-
tribution were to take place. But to the extent that
liberalism seeks not to swallow up all individual
projects and concerns is the extent to which it
34
may not be sufficiently responsive. It may be that,
to meet even fairly minimal demands of justice (such
as satisfying basic needs), the richer nations would
have to transfer such enormous amounts of resources
that many of the kinds of lifestyle that liberalism is
valued for accommodating would not be sustainable.
Maybe justice would then require far more than the
voluntary acts of charity, necessitating a coercive
limitation upon people's freedoms to live such com-
paratively expensive lives. Here, then, is the possi-
bility that liberalism's accommodation between the
16
moralities of individual freedom and social justice
may not be sustainable in the kind of world in which
we find ourselves.
Perhaps liberalism's notions of `individuality' and
`individual projects' can be preserved in the face of
justice so conceived only if we in the rich nations
undergo a radical reconceptualisation of our pre-
ferred or ideal individual modes of good living. If
liberals become more prepared to lead lives that are
much less exorbitant in their use of resources, they
may be more capable of reconciling themselves to
justice. I raise this point in the final chapter to
indicate a possible twist in my argument about
perfectionist liberalism's theorisation of the good
life. There I link it to the second concern that
can be raised under the present heading: what needs
to be done to preserve the global natural environ-
ment. It may be the case that Brian Baxter's inge-
nious melding of liberal and ecological concerns will
ultimately not demand enough of individuals (in
fundamentally changing the way they live and the
resources they use to do so) to do what is necessary to
stave off the destruction of the precarious ecological
balance between humanity and the earth. The con-
sequences of this catastrophe (even from the most
`personcentric' point of view) would be much more
difficult to ignore in the way that many in the rich
world are presently capable of physically and psy-
chologically isolating themselves from the demands
35
of distributive justice.
Challenges to the Political Principles
and Institutions of Liberal Democracy
It is as close as one gets to received wisdom in
political science that the `formal' constitutional
paraphernalia of modern liberal democracy do not
house the only, or even the main, sites of power.
These are located instead in state and party bureau-
cracies, organised interest groups and other socio-
economic elites. Socialists in particular have always
attacked liberalism for a blindness with regard to the
power that capitalists wield over the liberal-demo-
cratic state. In contemporary politics, this charge has
taken on a heightened perspicacity with the phe-
nomenon of `globalisation'. Alongside the develop-
ment of regional and genuinely global institutions,
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
there is a growing realisation that the nation-state's
increasing dependence upon the support of capital is
undermining its very existence as a meaningful
political entity in so far as that capital is acquiring
an increasingly multinational character.
Now we could agree that a liberalism which is
wholly ignorant of these facts, stubbornly clinging to
the obvious fiction that real power rests solely with
the `superficial' structures of the nation-state, should
not be entertained at all seriously. But the mistake
which liberalism's critics may continue to make is to
believe that liberals necessarily peddle such naõÈvety.
In the same way that the defence of liberal politics
need not be based upon the kind of philosophical
beliefs that the communitarians rightly argue to be
discreditable, so it need imply no ignorance of the
realities of power and the limitations imposed upon
its favoured institutions. Max Weber's `competitive
elitism' offers one, albeit noticeably unenthusiastic,
defence of liberal-democratic politics while conced-
36
ing many such limitations.
However, it is always possible that these limita-
tions will profoundly undercut the possibility of
liberal politics. The economic and institutional con-
straints of the contemporary capitalist world may
place too much strain upon it to keep its aspirations
alive. Take, for example, its commitments to uphold
an extensive range of civil liberties and to promote
social justice for its citizens. We could envisage
something like a `legitimation crisis' crippling its
ability to meet them. Sketching this thesis briefly:
the capitalist state is ever more beholden to the need
to attract rather than repel multinational capital
which can roam the world in search of the most
profitable sites to invest. Forced into ever more
exclusively `business-friendly' policies and progres-
sively sapped by capitalism's economic cycles of
`boom and bust', the state finds itself increasingly
unable to fund its other commitments adequately,
primarily those most directly demanded by citizens in
the form of public services and welfare provisions.
The citizens thus become disenchanted and embit-
tered. As part of its increasingly frantic attempts to
control expenditure by reducing the number of in-
stitutions with the authority to spend at their own
discretion, the state is forced to centralise political
control (even where it promises the extension of
17
democracy). Democratic institutions thus begin to
wither. Further, as social unrest intensifies (through
industrial militancy and/or extra-systemic political
protest and/or increases in crime rates due to the
widening gaps between rich and poor, and the
destruction of meaningful economic and social op-
portunities for the latter), the state becomes more
authoritarian, undermining civil liberties as it at-
tempts to impose civil order (or at least contain the
37
effects of disorder).
In light of this, my suggestion is that in western
states it is likely that the biggest threat to liberal
democracy does not come from liberal theory's tradi-
tional nemeses of `fascism' or `religious fundament-
alism'. It lies instead in the drift towards an
authoritarian form of economic libertarianism, as
the competitive pressures of globalising capitalism
increasingly subordinate politics to the satisfaction of
its own exacting needs. (It might be suggested that
certain South-East Asian capitalist states provide the
none-too-enticing models for this.) As David Held
points out, the New Right's rise from the mid-1970s
was a direct response to precisely the kinds of
intractable pressures from which liberalism in its
38
Keynesian welfare-state form suffered. It retrieved,
as observed above, an earlier form of classical liberal-
ism as its inspiration, cutting away at least part of its
later connections to democracy and certainly its
commitments to social justice. The paradox of this,
of course, was that it sanctioned a politics that also
threatened the commitments even to its more mini-
mal conceptions of equal freedom, as the inequities
of power in capitalism were exacerbated by economic
`deregulation' and the power of the state was
strengthened vis-a
Á -vis the basic civil liberties of in-
dividuals. It is, then, hard to suggest that even
classical liberalism was being fully realised in this
situation.
The idea of `legitimation crisis' prompts the
thought that this drift to authoritarian economic
libertarianism may not be entirely led `from the
top', either. If citizens become frustrated at the
perceived failings of the liberal-democratic system
and, in particular, disturbed by its inability to control
rising crime rates and the other objects of the `moral
panics' that periodically seize such societies, they
themselves may be tempted to support altogether
 MARK EVANS
more illiberal politics in the name of law and order.
This fear, whose origins can be traced as far back as
Plato's critique of democracy in The Republic, finds its
most disturbing expression in the spectre of the
Weimar Republic's fall and the willingness among
many German citizens to accept (at least in its early
stages) the Nazi dictatorship.
The ultimate crisis of progressive politics on this
view, though, may lie in an inability plausibly to
recommend any of the radical alternatives to capit-
alism that might be necessary to secure and build
upon the values that the present critique suggests
liberal democracy might hope in vain to uphold in its
midst. Socialism, for example, may just not be
thought to be practicable. If some form of capitalism
is regarded as unavoidable, and liberalism remains
the progressive's best `match' for it, then progressi-
vism has nowhere else to turn ± and yet even a
shrunken neoclassical liberalism may not be able to
withstand the intensification of capitalist practices.
The relationship of liberal-democratic politics to the
exigencies of capitalism will, I suspect, become one
of liberal theory's most urgent concerns in the years
ahead.
Challenges to the Culture
of Liberal-Democratic Individuality
The economic realities of capitalism impact as much
upon culture as they do upon political institutions.
The third category suggests, then, that liberalism's
support for capitalism ± which, as indicated in the
previous section may be far from unalloyed enthu-
siasm even though there is no more preferable
alternative ± undermines its cultural ideals.
For example, there are those who think that
capitalist commodification tends to turn the project
of `individual conceptions of the good' into shallow
simulacra of flourishing individuality, reducing it to
mere hedonism and the shaping of one's identity
purely with respect to the fashions that are just
insidious ways of manipulating the tastes of indivi-
39
duals as consumers. If it is not almost a new kind of
slavery, these critics would certainly regard it as a
travesty of liberalism's ideal of free human develop-
ment. Many who argue in these general terms sup-
port a conception of power that ranges well beyond
18
what is thought to be the traditional liberal idea of it
as a (political) capacity to influence others. They
contend that power is the effect of all manner of
social practices and relationships, `disciplining' or
even constituting subjects in society. As critics of
capitalist society, they attack the forms of subjectiv-
ity produced by the social practices that liberalism
40
upholds.
Others, perhaps not always as prepared to give off
the scent of cultural elitism that attaches to some
versions of the previous argument, nevertheless urge
that there is something about the way in which social
space and time is nowadays constructed that turns
individuals into withdrawn, privatised selves. Such
people are increasingly unwilling and unable to reap
those fruits of genuinely communal relationships
that once were so central to ideas of the good life.
Fukuyama himself resurrects the image of the
Nietzschean `last man' as the ominously possible
paradigmatic character of modern liberal capitalist
society, content with mediocrity now that funda-
41
mental social struggles for improvement are over.
But this development need not be due simply to a
lack of character and lazy contentment with every-
day life in capitalist consumer society. Such are the
pressures of this life that the hedonistic retreat into
the private realm becomes a perfectly understandable
and, perhaps for some, existentially necessary reac-
tion.
As republican thinkers remind us, the privatisa-
tion of the self may have direct political conse-
quences too. As Tocqueville influentially noted,
democracy depends upon a culture of participation.
If individuals are not prepared in significant numbers
to engage in public political life at least to some
degree, then the practices of liberal democracy will
wither, leaving a vacuum to be filled, if only by
default, by non-democratic institutions. In modern
times, the phenomenon of power slipping to une-
lected bureaucratic `quangos' may well be testimony
to this (as well as to the anti-democratic tendencies
42
cited in the previous section).
More widely, such privatisation contributes to an
increasing unwillingness to embrace social obliga-
tions, narrowing if not individualising altogether our
43
sense of to whom our responsibilities lie. As men-
tioned above in indicating how liberalism has re-
 ISSUES AND TRENDS IN CONTEMPORARY LIBERALISM
sponded to the communitarian critique, there has
been a renewed interest in ideas of `civic liberalism'
which are designed to address such fears. But there
may be forces at work that would entirely undermine
the stability of any such ideal in practice. Within
civil society, therefore, an egoistic mentality is fos-
tered that causes the decline of many social institu-
tions and practices which formerly encouraged co-
operation and mutual support. Given the obvious
unwillingness of many major liberal-democratic po-
litical parties to advocate an increased acceptance by
the state of this social burden, this trend means that
more and more social obligations go unfulfilled.
Conclusion
It is not perverse to end this introduction to a further
twenty-one chapters on liberalism with such dire
anti-liberal possibilities. As I have insisted, painting
such dystopian pictures does not commit one to
affirming them. But by taking full measure of the
problems it may face, I believe that liberalism can
become more emboldened to analyse its own con-
dition honestly, and articulate, defend and refine
itself as robustly as possible. As much as anything,
this exposes the limitations of Gray's critique: unless
we are blithely indifferent to the fate of liberal
civilisation it is surely strange to think of these
questions as somehow divorced from our real-life
political concerns. Taking them as seriously as they
are stated in the hypotheses above may be overly
melodramatic ± but it is up to liberalism to argue that
Notes
1. Note that many wish to distinguish `political theory'
and `political philosophy' and would doubtless stress
that it is the fate specifically of the latter which such
claims announce. For my purposes in this discussion,
nothing significant hangs upon the distinction, which
is far from universally recognised.
2. P. Laslett, `Introduction', in Laslett (ed.) (1956), pp.
vii, xiii. Cf. Berlin (1969), pp. 118±19.
3. Gray (1995a), p. 1.
4. Ibid., p. 16.
5. Ibid., p. 3.
6. Ibid., p. 2.
7. Gray (1995a), p. 17.
8. See Bertram (1997) for a discussion of the problems of
19
case. In so doing, I think this can orient liberal
thinking towards a maximisation of its own depth,
strength and self-confidence.
In setting an agenda for contemporary liberal
research, this Companion is a contribution towards
that orientation. Even those liberals who already feel
confident enough to accept a Fukuyama-style verdict
on liberalism's moral superiority should not, then,
regard this book as a final, definitive pronouncement
of that victory. Liberal political theory does not
come to an end even if one thinks that `history'
in Fukuyama's sense has. There are always vexing
issues to be addressed, some of which may move to
the fore and supplant some of those emphasised here.
And we must constantly remain alert to the possi-
bility that liberalism does not constitute the end of
our ideational and institutional development.
Conversely, however, awareness of liberalism's
problems ± both actual and potential ± need not
commit one to the belief that continued support for
liberalism necessarily represents some failure of ima-
gination by today's political thinkers. If any of the
anti-liberal possibilities described above do threaten
us, it may take great effort to sustain liberal ideas and
practices ± and this is no small, meek undertaking.
For liberalism's morality and politics has inspired and
sustained much hope through even the darkest
periods of recent human history. Critical but con-
structive engagement with its problems is vital to the
preservation of its legacy ± and that is the project this
book aims to keep alive.
theoretical complexity and public comprehension in
contemporary political theory.
9. Gray (1995a), pp. 3±4, 11±13.
10. See Ethics, 102 (April 1992).
11. Gray (1995a), p. 4.
12. See Fukuyama (1989), (1992).
13. For a comprehensive discussion of how to characterise
liberalism illuminatingly and substantively while at
the same time doing justice to its diversity, see Freeden
(1996), parts I and II.
14. By `general themes' and `tendencies' I refer to features
which occur in some form to a greater or lesser extent
in most versions of liberalism; I intend nothing more
specific than that.
 MARK EVANS
15. Rorty (1998), p. 170.
16. See Freeden (1996), chapter 7.
17. For examples of this type of liberalism, see ibid.,
chapter 5 and pp. 456±80.
18. A `character-ideal' specifies certain dispositions, ap-
titudes and tastes which constitute what is deemed to
be the ideal form that a person's character could take.
Anti-perfectionist liberals may be thought to eschew
such notions, but in fact they tend not to: Rawls, for
example, has no qualms over talking about the ideal
characteristics of citizens. However, he attempts to
avoid idealising forms of character outside of the
political realm. Perfectionist liberals, as is evident
in Chapter 22 of this book, do not accept this
restriction.
19. See Cohen (1989).
20. See Rawls (1972), especially part one.
21. Nozick (1974).
22. See Hayek (1993).
23. A. Sen, `Equality of What?', in Sen (1982), pp. 353±
73. Representative pivotal works in this debate in-
clude Dworkin (1981a) and (1981b), Cohen (1989)
and Arneson (1989).
24. Other representative contractarian works include
Scanlon (1986) and, for a more libertarian position,
Gauthier (1986).
25. Sandel (1982).
26. See Walzer (1986) and, for a comprehensive discus-
sion of the liberal-communitarian debate, Mulhall and
Swift (1996).
27. As Charles Taylor has argued, `the liberal-communi-
tarian debate' was often presented in a manner which
conflated the distinguishable questions of `ontology'
(such as those on the nature of the self and its
relationship to society), which were centrally relevant
to the justification debate, and `advocacy', which
specifically contested the content of political morality.
See his `Cross-Purposes: The Liberal-Communitarian
Debate', in Taylor (1995), pp. 181±203.
28. See, for example, MacIntyre (1985).
20
29. For an example, see Dagger (1997).
30. See Rawls (1996) and, for commentary on this type of
liberalism, Evans (1999).
31. Young (1990).
32. Quoted in Kamenka (ed.) (1983), p. 207.
33. Having warned against complacency, I don't wish this
point to be taken as ignorance of the activities of neo-
Nazi movements and suchlike in contemporary wes-
tern democracies. Certainly, liberals should steadfastly
resist them in theory and practice, for this is what will
help to resist any future slide into fascist barbarism. I
maintain that the latter is indeed unlikely, but I am
primarily concerned here to divert attention to other
possible anti-liberal trends.
34. Nagel phrases this problem in terms of balancing the
impartial demands of morality with the partial con-
cerns of particular relationships and projects; see Nagel
(1991).
35. See Singer (1993) for a cautiously optimistic assess-
ment about our ability to change our lives to meet
such moral and ecological challenges.
36. See Held (1996), pp. 157±77 for a discussion of
Weber's political views.
37. For a summary of `legitimation crisis' theory, see ibid.,
pp. 240±53.
38. Ibid., pp. 253±63.
39. Classic statements of this kulturkritik, which blames
the processes that liberal society has historically en-
couraged, are found in Marcuse (1964) and `The
Culture Industry: Enlightenment as Mass Deception',
in Adorno and Horkheimer (1986), pp. 120±67.
40. See Hindess (1996) for a discussion of the concept of
`power' in liberal political theory. I suggest in Chapter
22 that a Foucauldian rethinking of `power' may
benefit the concept of a perfectionist liberalism.
41. Fukuyama (1992), part V.
42. For illuminating discussion of Tocqueville in this
regard, see Macedo (1989).
43. See Singer (1993), chapter 1 and passim, for discussion
of this claim.
 2
TWENTIETH-CENTURY LIBERAL
THOUGHT: DEVELOPMENT OR
TRANSFORMATION?
Michael Freeden
In this chapter I seek to investigate how liberalism
was portrayed throughout the twentieth century in
dedicated liberal literature, that is works primarily
devoted to an exposition of the basic tenets of
liberalism. On the surface many, though not all,
of these works present themselves as `second-order'
overviews of liberal theory and ideology. Sometimes,
as with the Rawlsian family of arguments, they
intend both to offer a novel interpretation of liberal
principles and, in parallel, to reflect given cultural
understandings unconsciously and unintentionally.
On the whole, though, writers about liberalism have
tended to elucidate a tradition rather than depart
from it. But this raises a fundamental methodological
problem: when does a `secondary' text become a
`primary' one? The response to that is: when we
interpret it as itself an act of interpretation; and
when we acknowledge that the role of the political
theorist incorporates not only prescription and its
investigation but also interpretation and its investi-
gation. The analysts, chroniclers and popularisers of
liberalism become pivotal to our understanding
when we query its status as a transcendent moral
position and explore it instead as a diverse and
flexible set of readings concerning the epistemology
of the political.
Whatever else liberalism is, it is a cultural artefact,
consciously intended to be adopted by large social
groups. But rather than explore the Ideengeschichte of
21
liberalism in time-honoured conventional manner
on a unidimensional sequence, we need to draw
conclusions from the variable presentation of histor-
ical liberal narratives, or from their juxtaposition
with non-narrative justifications of liberalism. The
history of liberalism in the last century, and indeed in
previous centuries, is not merely the reflection of its
development, or evolution, or change, or perchance
regression. It is not a single story to be told. We now
recognise that all current theories are located in time
as well as in space. But when we say that, we are not
engaging in the simplistic assertion that `context
counts'. Contexts are multi-temporal and multi-spa-
tial even with respect to one family of beliefs such as
liberalism, and they generate both deliberate and
unintentional meta-assumptions quite irrespective of
whether liberalism itself contains views on univers-
alism or relativism or, as we have now come to
appreciate, multiculturalism. Moreover, individual
liberals, and conventional social norms concerning
what liberalism is, exercise choice over what con-
stitutes a context and over which contexts to select
among the many contexts ± existing or yet to be
discovered or invented ± in which liberalism may be
located. A context, like a historical narrative, is not a
reflection of events out there. It is an act of private or
public imagination superimposed upon a set of per-
ceived, if fragmented, facts. This chapter will at-
tempt to identify more precisely not why and when
 MICHAEL FREEDEN
liberalism changes in terms of objective contexts, but
what it is that has changed in the imaginative
presentation of liberalism, and how different em-
phases on various aspects of liberalism reflect and
transform the political understandings of some of its
more salient, or representative, purveyors.
Far too often, recent scholars of liberalism have
confidently and stipulatively approached it as if that
theory or ideology represents a clear and unitary
moral position encompassing knowable and objec-
tive, or at least reasonable, standards of justice and
human rights: a homogeneous model which could
then be incisively contrasted with alternative `mono-
lithic' theories such as communitarianism. This is
not the issue under discussion here; nor is it the most
challenging and rewarding question when examining
liberalism. Rather, liberalism is that semantic field in
which the political understandings of people who
regard themselves as liberals, or whom others regard
as liberals, may be investigated. It is a plastic, chan-
ging thing, shaped and reshaped by the thought-
practices of individuals and groups, and though it
needs to have a roughly identifiable pattern for us to
call it consistently by the same name, `liberalism', it
also presents myriad variations which reflect the
questions posed, and positions adopted, by various
liberals.
In the space available, some samples will be
provided of the conceptualisations of liberalism dur-
ing this century, through examining a number of
books on liberalism which were written as central
textbooks, introductions, surveys, statements, or cri-
tiques. The assumption adhered to is that these
books mirrored, moulded and disseminated broad
understandings of what liberalism was, as well as
carrying some unintentional baggage which has been
left for us to unpack. Ultimately, in order for liberal-
ism to be comprehended we must observe it `at
thought' by exploring the mind-sets of some of its
typical agents. The result, one would hope, would be
an enriched and rather more subtle view than can be
provided by attempting to construct a philosophi-
cally `ideal' or `best-practice' liberal moral theory.
Those philosophical approaches necessarily lose in
breadth and political applicability what they gain in
internal cohesion and ethical persuasiveness. Nor is
this chapter about `mere' history, nor even `mere'
22
intellectual history. Behind it stands a broader agen-
da, whose task it is to provide an analysis of a family
of political theories and ideologies. To do so we
ought to come to terms with liberalism as a complex
and elusive tradition of thinking about politics, and
we may gain from locating the analysis on offer here
squarely among the activities in which political
theorists engage.
Liberal Movement: Evolution and Energy
The twentieth century has witnessed three main
modes of writing about liberalism: as a development
of political thought over time; as a political mani-
festo against extremism, authoritarianism or conser-
vative complacency; and as a philosophical view of
the rational and moral relationship between the
individually unique and the socially common. In
intricate and concrete argumentation, these modes
often overlap. When in 1911 L. T. Hobhouse wrote
his famous volume entitled Liberalism, he was writing
a tract for the times, forged out of the successful
battles fought by the new liberals, but it was a tract
with a moral purpose: that of consolidating the
reforms liberalism had undertaken from within. He
did so by emphasising above all one aspect of liberal-
ism, offering it as an evolutionary body of thought
which not only advocated progress as a social ideal,
but was itself controlled by that same evolutionary
law. This was not a particular liberal view of history,
such as the much-mooted Whig interpretation of
history, but rather a particular historical view of
liberalism, which regarded it as a set of beliefs itself,
crucially, subject to historical processes. In other
words, one defining feature of liberalism, in sharp
contrast to current philosophical liberalism, was that
its theorists held to a specific interpretation of time
as a central conceptual component of liberal ideol-
ogy. Time was not strictly sequential in its trajectory
of human and social improvement, nor totally open-
ended in accompanying human and social matura-
tion, but it nevertheless dictated that liberalism itself
had to undergo steady transformation. In the past
liberalism had experienced a process of growth which
was now reaching its apotheosis by aligning itself
with the development of a newly emerging social
rationality, embodied in an integrated and organic
 TWENTIETH-CENTURY LIBERAL THOUGHT
society and regulated by a benevolent and empha-
tically non-neutral agency ± the state.
Like Mill, and many previous liberals, Hobhouse's
story about liberalism begins as a historical-anthro-
pological narrative. It is one of social harmony and
community undermined by later authoritarian poli-
tical arrangements. Liberalism dawns as a `destruc-
tive criticism', a project of release in order to enable
free, and natural, order and progress. Here are some
of the recognisable liberal core concepts of the
nineteenth and twentieth centuries. But there is a
deeper and ulterior ontology of liberalism in opera-
tion. Hobhouse endows the idea of release with a
specific complexion: `a movement of liberation, a
clearance of obstructions . . . for the flow of free,
1
vital, spontaneous activity'. The idea of `movement'
is subtly and only half-consciously recruited to un-
derpin this specific understanding of liberalism. First,
liberalism is a movement, to be sure: it is an organised
body with, in J. A. Hobson's words, an `active
2
mission' directed towards the attainment of political
objectives. Second, however, liberalism is a carrier of
truth, which moves for Hobhouse in `an expanding
3
circle of ideas'. When in Chapter II Hobhouse lists
the types of liberty embraced by liberals, he con-
ceives them as spatial, embracing parallel activities
such as civil, fiscal, economic and social, and located
in arenas such as the personal, domestic, local,
4
national and international. Similarly, Hobson could
assert that `liberalism will come more definitely to
concern itself with the liberation and utilisation of
the faculties and potencies of a nation and a muni-
cipality, as well as with those of individuals and
5
voluntary groups of citizens.' The liberty at the core
of liberalism is hence conceptualised as undergoing
movement both through space and through time,
and is aligned with the concept of spreading and
contagious progress. Thus it contains seeds of uni-
versalism, not in an analytical or intrinsic sense, but
a historically and spatially contingent universalism,
called forth `by the special circumstances of Western
6 10
Europe'.
Third, as the initial quotation suggests, movement
suggests vital activity. We arrive now at the semantic
kernel. Hobhouse instructs us that `the
movement . . . is coextensive with life'; `Liberalism
is an all-penetrating element of the life-structure of
7
the modern world'. Movement in the first sense, as
`an effective historical force', is sustained by move-
ment in this second animated, living sense. Liberal-
ism is the release of dynamic biological and spiritual
energy, a view reiterated by Guido de Ruggiero in his
seminal The History of European Liberalism, when he
identified liberalism as concerned with the free play
of individual forces which have `vital importance and
8
energising power'. The world is naturally active and
vigorous and liberalism is above all an instrument for
freeing this flowing human essence. This is no mere
9
`exercise concept' in Charles Taylor's phrase, but
the setting free of organic, non-static, life which will
otherwise be suppressed. Unsurprisingly, well-being
is for Hobhouse as important as liberty. Moreover,
the identification of processes as natural indicates
the typical ideological device of insulating funda-
mental ontological claims from the sphere of debate
and contention. In adhering to that thought-practice
liberalism is no different from other ideological
families. Furthermore, although an understanding
of twentieth-century liberalism should not be deter-
mined merely through the yardstick of some of its
later manifestations, the contrast with Rawls here is
striking. For if liberalism is `all-penetrating', it pro-
vides both the common element enabling the ex-
pression of all human cooperative activity, a function
allocated by Rawls to narrow political liberalism, and
the comprehensive liberalism which Rawls firmly
detaches from the political.
Liberal Emotionalism
Hobhouse unequivocally relates liberalism to human
emotion on two levels. On the one he observes that
the philosophies that have driving force behind
them are those that arise . . . out of the practical
demands of human feeling. The philosophies that
remain ineffectual and academic are those that
are formed by abstract reflection without relation
to the thirsty souls of human kind.
Liberalism opens the door by means of the rational
Liberal method `to the appeal of reason, of imagination, of
11
social feeling'. It is not only that human beings
have emotions and imagination that demand respect
23
 MICHAEL FREEDEN
in social arrangements, but that expressivism and
spontaneity are defining features of life, and hence of
the liberalism that enables life. Liberalism contains
core rational attributes, but is nonetheless irreducible
to the contemplative faculty. Reason and emotion
coexist and support each other in a manner ruled out
of court by most current manifestations of philoso-
phical liberalism. Hence Hobhouse's support for
forms of nationalism, for
inasmuch as the true social harmony rests on
feeling and makes use of all the natural ties of
kinship, of neighbourliness, of congruity of char-
acter and belief, and of language and mode of life,
the best, healthiest and most vigorous political
unit is that to which men are by their own feelings
12
strongly drawn.
In addition, if `rational' may be contrasted with
`emotional', it may also be contrasted with `unre-
flective'. Hobhouse also drew attention to these
latter components of liberalism when he contended
that `human progress, on whatever side we consider
it, is found to be in the main social progress, the work
13
of conscious or unconscious co-operation.' Indeed,
in his later work Hobhouse was to associate the
emergence of the ethical precepts of liberalism with
14
fundamental psychological processes. Similar views
on the evolution of liberalism were expressed by
Laski, for whom it was `never direct and rarely
15
conscious.'
From another perspective, most twentieth-century
liberals adopt an emotive idiom when addressing
liberalism. For Hobhouse, liberalism aims at `a spirit
of comradeship'; its vision of justice `kindles a pas-
sion that may not flare up into moments of dramatic
scintillation, but burns with the enduring glow of the
16
central heat' ± hardly the predominant phraseology
of current Anglo-American philosophy. De Ruggiero
put this similarly:
Liberalism possesses that kind of tact or flair . . .
which in its highest manifestations is true poli-
tical sensitiveness, and serves to recognise every-
thing that is human ± human strength and human
weakness, human reason and human passion,
human interest and human morality ± in the
24
relations between rulers and ruled; and is able
to turn this experience to the ends of the eleva-
17
tion of human society.
The liberal spirit is possessed of a `lofty impartial-
18
ity'. A mid-century American review of liberalism
refers to the liberal who `likes to preen himself on the
triumphs of human intelligence and virtue', and to
the liberal tradition as a `call to adventure: the
adventure of growing up, becoming autonomous,
living spontaneously' (and note, incidentally, the
constructive tension between development, freedom
19
and unreflectiveness). In describing liberalism as a
mood, Laski identified a `flavour of romanticism' in
its temper. It tended to be `zealous' for individual
action, `subjective and anarchist, to be eager for the
change which comes from individual initiative, to be
insistent that this initiative contains within itself
20
some necessary seed of social good.' Zeal, eagerness,
insistence ± these features of liberalism are necessa-
rily the attributes of an ideology whose reforming,
even revolutionary, ardour is part and parcel of its
appeal. For the over-critical Laski, the energising
21
power of liberalism had become exhausted. It may
be the case that the fervour of liberalism diminished
upon attaining many of its initial goals, though the
emotional aspects of its beliefs have never quite
disappeared.
Liberal Coherence and Liberal Diversity
Throughout the twentieth century there has been a
curious tension between the static and dynamic
facets of liberalism. This problematic of liberalism
may be redefined by asking whether it is a pliable
structure or heavily rule-bound. The editors of The
Liberal Tradition, published in 1956, stated that `at
first sight, the most striking thing about the Liberal
tradition is its intellectual incoherence . . . This is
the strongest argument in favour of treating Liberal-
ism historically'. Considering liberalism as develop-
ing and changing allows such differences to fall into
22
place. Coherence, however, is not the product of a
narrative alone, because invented narratives are
contingent and selective. It is importantly the pro-
duct either of the formulation of directive rules or
the discovery of unintended patterns (a discovery
 TWENTIETH-CENTURY LIBERAL THOUGHT
which itself may then be fashioned into a set of
directive rules). If liberal history is not accidental or
contingent, if liberalism is not a mathematical ag-
gregate of its historical parts, that is because history is
understood as constituting the ineluctable design of
liberalism. To write about liberalism is for Hobhouse,
and no less so for de Ruggiero, always and necessarily
to write about its history. They both conceptualise
liberalism as the vehicle best encompassing the
dynamic of individual and social life. Liberalism is
a process and it evokes an activity, the activity of
being self-directively spontaneous, inventive and
imaginative. Process entails history; history is the
intellectual coherence of liberalism, rather than
being superimposed to account for it. By contrast,
the liberal projects of Rawls or Ronald Dworkin
prioritise rules as stasis, equilibrium and consensus
over rules of change. The idea of a liberal constitu-
tion, to which they are wedded, is precisely such a
device, creating inviolable space, such as a Bill of
Rights or the two principles of justice, removed from
the ravages of social time. From a slightly different
perspective, the liberalism of foundational rules pri-
vileges mechanics, whereas the liberalism of creative
expressivism privileges organics, both as growth and
as human interrelationship. Hobhouse summed up
his position as follows:
The heart of liberalism is the understanding that
progress is not a matter of mechanical contri-
vance, but of the liberation of living spiritual
energy. Good mechanism is that which provides
the channels wherein such energy can flow un-
impeded, unobstructed by its own exuberance of
output, vivifying the social structure, expanding
23
and ennobling the life of mind.
Current philosophical liberals tend to concentrate
on the channels while professing to have no view on
the consequences of providing them, other than the
tautological attainment of justice which has already
been furnished by the very construction of the
mechanical channels themselves. Hobhouse's agen-
da consisted rather in attaching liberalism to an idea-
environment formed through concepts such as re-
lease, movement, energy and vitality. That is his
`thin' theory of the good, which informs the core
25
liberal concepts of liberty, rationality, progress, in-
dividuality, sociability, a common good, limited and
responsible power ± in other words, a theory that,
according to Hobhouse, undergirds every version of
liberalism.
De Ruggiero's notion of history rested on similar
conceptions of liberal movement, but his Hegelian
historical scheme reshaped the liberal time-space
relationship into a different pattern of coherence.
His comparative treatment of liberalism permitted
the postulation of multiple, interacting liberalisms,
but his reading of the logic of liberalism was not, as
with Hobhouse, through its diffusion in ever-expand-
ing circles but through convergence and conciliation
of opposite liberal currents. Liberalism has to be
understood `in the diversity of its national forms
24
and the unity of its historical organism'. The
progress which is liberalism's lifeblood is not Hob-
house's non-teleological `onward course'. For Hob-
house, a release of energy is not necessarily self-
realisation. For de Ruggiero, `a more comprehensive
Liberalism would recognise the dialectical ground of
the antithesis and would see resistance and move-
ment, conservation and progress, justified and vali-
dated in a higher synthesis which is political life in its
25
concreteness'. The end-result for de Ruggiero is
thus a single liberalism which can be retrospectively
accounted for as the diachronic emergence of a
civilising strain in modern societies, the `irresistible
26
operation of civil society'. Its political persuasive-
ness crowds out rival arguments and secures victory
in a multi-ideational world of really-existing compet-
ing viewpoints.
The study of the historical forms of European
Liberalism has shown us, through all the differ-
ences of the various national minds, a process of
mutual assimilation, gradually building up a Eur-
opean Liberal consciousness pervading its parti-
cular manifestations without destroying their
27
differences.
Without such reconciliation, the liberal enterprise
would be undermined by the damaging potential of
its disparate currents.
That argument was taken even further by Bene-
detto Croce. In a chapter revealingly titled `Liberal-
 MICHAEL FREEDEN
ism as a Concept of Life' he, too, combined devel-
opmentalism with a view of liberalism as coinciding
`with a complete idea of the world and of reality,'
beginning in the world but going `beyond the formal
theory of politics'. The very tolerance built into
liberalism located it as the framework doctrine with-
in which a dialectic competition among ideas could
be resolved. It was characterised above all by this
immanent process which removed it from
authoritarian fiat of otherworldly, externally im-
posed, transcendental beliefs. History as struggle
was endemic to liberalism, and struggle encompassed
28
impulse and spontaneity. Nevertheless Croce was
drawn into joining to that profoundly secular doc-
trine an extra-rational, quasi-religious understanding
of liberty, attached to human `vocations' and `mis-
sions'. He believed that `the liberal mind regards the
withdrawing of liberty and the times of reaction as
illnesses and critical stages of growth, as incidents
and steps in the eternal life of liberty.' As with Hegel,
timelessness engulfed and superseded time in the last
resort; moreover, this afforded a standpoint from
which history manifested pathologies as well as
normalities. The transcendental status of liberty
returned to do the work of faith and underlying
essential meaning, justifying the course taken by a
non-transcendental history. That such liberal his-
tories attained impartiality, as Croce claimed, is
hardly borne out by the language in which that
29
claim is expressed.
A similar viewpoint, with different conclusions, is
evident in Louis Hartz's modern classic The Liberal
Tradition In America. Liberalism has multiple facets,
crucially situated not above politics but in it: `we
know the European liberal, as it were, by the enemies
30
he has made.' Whereas European liberals experi-
enced social diversity and conflict, the American
variant acquired non-contentious invisibility
through a cultural and ideological uniformity, `on
the basis of a submerged and absolute liberal faith'.
Hartz identified a central ideological feature at work
in the manner American liberalism decontested its
basic premises, one involving `silent omissions as
31
well as explicit inclusions'. Hence while American
liberalism pre-empted political and ideational com-
petition, European liberalism had to fight its ground
within a culturally and historically disputed space.
As Hobson noted, reviewing de Ruggiero's book, `in
each country the spirit of liberalism was largely
formed by the practical tasks which the needs and
interests of the dominant classes incited them to
32
undertake'.
Both Hobhouse and de Ruggiero subscribed to a
concrete spatial universalism which is a function of
time (as of course does Marxism) rather than a
the transcendental and spaceless universalism which is
a function of atemporal reason. But de Ruggiero
removed internal pluralism and unstructured open-
endedness from liberalism as it emerged as a tele-
ological world-view in a very real, and irreversible,
sense. Hobhouse, to the contrary, voiced an increas-
ing postwar despair concerning the future of liberal-
ism when he identified `arrest, retrogression or decay'
as part of the historic course of change and con-
cluded: `Hence if progress means the gradual realisa-
tion of an ethical ideal no continuous progress is
33
revealed by the course of history.' There is none of
that pessimism concerning the evolution of the hu-
man mind in de Ruggiero's 1927 deluded prognosis
that `one may conclude that the omens are favour-
able towards the capacity of the German people to
win for itself that liberal education in politics which
34
the old regime denied it'. Consequently, de Rug-
giero was epistemologically unable to criticise liberal
practices and beliefs in their current form because of
liberalism's inherent evolutionary capacity of self-
development and improvement. These distinctions
aside, the liberalism of Hobhouse and de Ruggiero is
primarily about individual and social expression and
development, and only secondarily about pluralism
and difference, especially in view of its allegiance to
the nation-state. By contrast, current liberal plural-
ism, in some of its multicultural forms, applies a pre-
formed, ahistorical, singular liberal morality to con-
crete societies comprised of many components. It
regards structural variety as a service concept for
autonomy, rather than regarding temporal develop-
ment as a service concept for individuality, socia-
bility or welfare. What de Ruggiero noted about
liberty and liberties ought to be said about the
tension between liberalism conceived of as universal
philosophy and as situated ideology. The one is `an
abstraction, a concept intended to express the es-
sence of human personality, exalted above all his-
26
 TWENTIETH-CENTURY LIBERAL THOUGHT
torical and empirical contingency', the other a com-
plex of features `acquired one by one as circum-
stances dictate'. Each on its own is, tellingly,
35
totally dispossessed of the Liberal spirit.
The Renewed Search for Certainty
The experience of totalitarianism in the 1930s and
1940s dramatically changed some interpretative lib-
eral paradigms. One of the most striking reformula-
tions of liberalism to emerge from the Second World
War was John Hallowell's The Decline of Liberalism as
an Ideology. His analysis constitutes a watershed in
twentieth-century liberal self-conceptualisation
mainly through its disentangling and counterposing
of the historical and the ahistorical mode of liberal-
ism. If for Hobhouse and de Ruggiero liberalism
embodies a vital growth impetus, Hallowell reversed
this historical tendency to identify a move from
something live and vigorous to something decadent
and degenerate. Growth may also imply regression,
especially once de Ruggiero's Hegelian dialectic of
progress is abandoned. `As a political ideology born
of a particular historical period in a specific socio-
logical environment [liberalism] is subject, like all
such systems of ideas, to development, decline, and
36
death.' Significantly, and in contradistinction to
Hobhouse's anxieties concerning external impedi-
ments to liberalism, Hallowell attributed the death
of liberal ideology to liberals themselves, for a man-
made belief system is subject to human fallibility.
That fallibility has revealed itself in the forsaking of a
parallel but morally and intellectually superior lib-
eralism, `integral' liberalism, couched in the language
of absolutes which would have prevented the in-
tellectual perversion of liberalism, particularly by
Nazism. Hallowell's version of liberalism embodied
the typical disillusionment of postwar scholars with
the fickle impermanence of legal systems, unless
anchored in dictates of objective reason. The redis-
covery of normative foundationalism became promi-
nent in the American legal tradition, in its renewed
anti-positivist appeal to natural law and its conse-
quent revival of the notion of universal human rights
37
during the 1940s and 1950s. Liberal ontology
therefore sought refuge in invoking new variants
of traditional certitudes: the absolute value and
27
dignity of human personality, and the construction
of impenetrable boundaries which the state may not
pervade. Liberal principles are, or ought to be, un-
touched by history, even by the experimental reason
which many liberals see fit to encourage in people.
Instead, they are to be secured to transcendental
standards and eternal truths, beyond the contami-
nating reach of untrustworthy human judgement.
Hallowell occupies a strange position in relation
to the liberal epistemology of this century, both
regressive and progressive. The notion of progress,
38
so central to liberal values, had in his view become
attached to a theory of evolution which, when
applied to liberalism itself, undermined the absolute
moral value of the individual. It did so by legitimat-
ing humanly directed change, especially through the
German construct of the Rechtsstaat, which estab-
lished positive law ± a law that was made and not
found ± as the guarantor of equality. One of Hallo-
well's inputs into the liberal debate was to re-em-
phasise not its polysemic but its static character.
Another was to distinguish between liberalism as
philosophy and as ideology, a move of importance
associated with the mid-century decline in prestige of
the concept of ideology. However, the decline of
ideology is not articulated in terms of the postwar
debate over undesirable dogmatism versus com-
mended pragmatism, but in terms of undesirable
pragmatism versus incontrovertible truth. Hallowell
approvingly quoted C. E. M. Joad, for whom prag-
matism ministered
to human complacency by assuring human beings
that right and wrong, beauty and ugliness, reality
and unreality, are not external facts, features of
the universe to which human beings must in the
long run subject themselves, but are the products
of human consciousness and, therefore, amenable
39
to human desires.
Hallowell proposed the following ideational com-
pound as liberalism's epistemological base: first, a
humanist doctrine that relinquishes anthropocentr-
ism and rejects intuitive judgement; second, a theory
of human freedom that dismisses the view of indi-
viduals as autonomously capable of exercising their
subjectively free wills; third, a conception of ration-
 MICHAEL FREEDEN
ality that subjugates individuals to the `dictates of
reason' and the external criteria of justice as obliga-
tion to transcendental law, accessed through an
inner conscience. For a liberal, Hallowell imposed
a particularly stringent limitation on value pluralism,
for liberalism relies on `a common knowledge, or
recognition, of values transcending individuals.'
Nevertheless all this is a `theory of political order
based upon individualism', and it is identified as a
41
`comprehensive Weltanschauung'. Contrast this
with the Rawlsian view of rationality emerging
volitionally from individuals and expressed in a
reasonable, overlapping and public consensus over
principles of justice, and moreover a view that now
relegates individualism to a feature of comprehensive
liberalism, a liberalism which is no longer specifically
endorsed as necessary to a just social order. And
contrast this with the Hobhousian faith in the
evolution of social understandings of valued prac-
tices which release the potential and sustain the
well-being of a community as well as those of its
members.
Ontological Minimalism
and the Supra-Political
Works such as Hallowell's which began as manifestos
against totalitarianism have been transformed into a
richer and deeper philosophical position. The heri-
tage and motivation are plainly visible in late twen-
tieth-century liberal theorising, but so are
epistemological and conceptual modifications. One
of the residues of this mode of thinking has been to
reduce liberalism once again to a form of ontological
minimalism, captured in the loose use of the inde-
terminate word `thin'. But one of the modifications it
has undergone is the rediscovery of a Kantian-in-
spired autonomy, an autonomy unavailable or de-
centred in early and mid-century liberal thought.
The two have combined in the revival of the private-
public dichotomy as a moral statement about the
inherent but socially indifferent good sense of in-
dividuals, detached from the impact of sustaining
groups, the latter now perceived as more dangerous
42
than formative. In addition liberalism has been
absorbed into the dominant Anglo-American phi-
losophical practice of model-building and counter-
factual thought experiments which translate the
requirements of coherence and procedure, rather
than growth and energy, into structural fundamen-
tals of liberalism. Is this, intriguingly, a parallel
tendency to what Orton describes as `the queer
double strain in nineteenth century liberalism.
40
The rationalistic, system-making tradition in lais-
sez-faire theory never found common ground with
the personalistic and humanitarian impulse' ± an
inspirational, empathetic and energy-liberating hu-
43
manism? Hobson denied that `coldness and pla-
cidity of purpose belong essentially to Liberalism',
44
rather than `organic purpose' and `free enthusiasm'.
Is it also the case that what some feminists term
the colder philosophical liberalism is a product of the
Cold War that never got round to thawing? Not for
nothing has the end of the Cold War witnessed a
reheating of liberalism under the impact of nation-
alism and the accompanying revalorisation of the
emotional resonances that have always existed with-
in liberal discourse: the language of bonds, alle-
giance, sympathy. What nonetheless sustains the
current philosophical variant is the predominant
and overwhelmingly shared image of the American
constitution as mediating between the presumption
of the reasonableness emanating from human beings
and the requirement of permanent, supra-political
standards. This avoids the language of transcendent-
alism while attaining some of its ends. It is hence
unsurprising to find Rawls describing the Supreme
45
the Court as `the exemplar of public reason'. Is that
another way of asserting, as Hartz did, that `the
absence of ``opposing principles,'' the fact that be-
neath its political heroics the nation was of the
``same mind'' on the liberal formula, settled in ad-
46
vance the philosophic question'? Only instead of
the nation are we not referring to the academic
community, sustained by its Justices with their pre-
sumption of expressing the vox populi? Indeed, Hartz
had given the role of the Supreme Court, and the
liberalism it sustained, a totally different and more
persuasive gloss: `the Supreme Court had always
been the Hebraic expositor of the American general
will, building on the irrational acceptance of Locke
the Talmudic rationality involved in his application
47
to specific cases.' Moreover, Hartz was deeply
cognisant of the emotive aspect of concrete liberal-
28
 TWENTIETH-CENTURY LIBERAL THOUGHT
isms. The American version, he argued, possessed
`the ideologic power of the national irrational liberal-
ism' which, in the case of the New Deal, `was
responsible for its whole pragmatic orientation, for
48
its whole aversion to systematic social thought.'
In view of the above, Croce's reaction to critics of
liberalism, who labelled it as `formalistic', `empty',
`sceptical' and `antagonistic', is edifying. Rather than
see these terms as referring to foundational categories
whose generality enables the integration of all rea-
sonable ways of life, Croce saw them as embodying
the essence of non-dogmatism in the spirit of modern
philosophy. The liberal conception thus denied `first
place to laws, casuistry and charts of duties and
virtues, and places the moral conscience at its cen-
tre.' This was `like modern aesthetics, which refuses
models, categories and rules, and places at its centre
the genius that is good taste, both sensitive and very
strict.' That approach was formalistic only in the
sense that it shied away from imposing philosophical
formulae and principles. The end result was strongly
emotive and parti pris: `the liberal conception is not
meant for the timid, the indolent and the pacifist,
but wishes to interpret the aspirations and the works
of courageous and patient, of belligerent and gener-
ous spirits.' No wonder that Croce could refer to
Hobhouse's opus as `a beautiful English eulogy and
apology for liberalism' ± language more apposite for a
work of art than for the mechanics of a system of
49
rules.
The Rawlsian Project: A Paradigmatic Shift
Looking back at liberalism from our vantage point,
the unity of the Rawlsian project is conspicuous. In
the early years of the twentieth century, as in much
of the nineteenth, the growth model central to
liberal conceptions of human nature was applied
to liberalism itself. The development of liberalism
was tantamount to the development of civilisation.
After the Second World War that very capacity for
change, with its newly highlighted contingency, lost
its attractiveness, and was increasingly replaced with
an appeal to independent immutable standards.
Rawls himself commenced his argument in Political
Liberalism with that problematic: `political liberalism
assumes that, for political purposes, a plurality of
29
reasonable yet incompatible comprehensive doc-
trines is the normal result of the exercise of human
reason within the framework of the free institutions
of a constitutional regime' ± a philosophical anthro-
pology diametrically opposed to Hobhouse's. Rawls's
overriding aim is for a political liberalism which
provides not only justice but also unity in what
would otherwise be a fissiparous, even Hobbesian,
society. The difficulty with the core liberal concept
of non-necessitarian development was that it had to
allow for a choice which could include error. The
risky model of development, even riskier when `ir-
reconcilable conflict' is assumed to be socially la-
50
tent, was replaced with a more advantageous model
of justice, stability and autonomy, which would
51
eliminate unpredictable choice. De Ruggiero,
too, had attempted to eliminate choice on a grander,
supra-individual scale by positing a Hegelian neces-
sitarianism, but that variety had been extinguished as
a fashionable archetype under onslaughts such as
Karl Popper's.
Nor was Hallowell's transcendentalism acceptable
to 1960s and 1970s Anglo-American philosophy.
Hence the foundational assumptions of Rawls's sys-
tem were couched in anthropocentric terms of em-
pirically attainable reasonableness, even though the
outcome of this process was to all effects and pur-
poses equally invariable. Like the contract model,
Rawls's theory simulated open-ended liberal choice
without, of course, offering it, through assuming both
a pre-discursive rationality (the veil of ignorance)
and a convergent reasonableness (the overlapping
consensus). Rawls's political liberalism was presented
as a western paradigm, but one whose rightness would
gain it universal recognition, moving from historicity
to ahistoricity. Its universal appeal would not be
forged out of competition over legitimacy with other
ideological arrangements, as most liberals have mea-
sured their successes, earning their spurs on the
battlefield of ideas, but as a non-pluralist procedural
and moral necessity.
Put differently, this liberalism employed a second-
order argument not over substantive goods, but over
justifiable rules. Justification depended not on the
nature of the political benefits available, but ± and
here lay its novelty ± on the shared acceptance of
52
those rules by all the members of the polity. Locke's
 MICHAEL FREEDEN
liberalism by contrast, had sought a limited general
acceptability only as a prior arrangement to estab-
lishing relations of mutual and substantive trust,
anchored in the norms of natural rights. Late twen-
tieth-century liberalism has, however, been charac-
terised as being that system of ideas which requires
and attains general acceptance, even prior to the
political implementation of those ideas. The only
ideas which comply with this scheme are those
concerning the public rules of moral engagement
and interaction. In earlier versions of liberalism,
acceptance had to be won through the substantive
benefits that would accrue from the liberal way of life
in the struggle with other political options.
Moreover, the Rawlsian procedure was con-
structed to create boundaries and hence moral spaces
in which individuals were responsibly reflective.
Intentionally or otherwise, that accomplished three
ends. First, it restored liberalism's emphasis on the
inviolability of individual space, rather than on the
controlled permeability of space which organicist,
communitarian liberals had encouraged. This was
closely linked to a thinning of the sociability com-
ponent of liberalism and a resurrection of individu-
53
alism as an attribute of separate personal existence.
Second, it focused on equality ± a concept to
which evolutionary liberal traditions could not fully
subscribe because of the unpredictable dynamics of
human relationships. The a priori rule-following of
Rawlsian liberalism had to assume not only temporal
stasis but relational stasis: equality as the participa-
tion of each and every member in decision-making
and hence in justifiable and responsible action.
Strikingly, this limited notion of equality was a
non-voluntarist view of participation. Participation
and sharing in responsibility had become the be-all
and end-all of liberalism, rather than adjacent con-
cepts that could be put into practice by those who
54
wished to participate. As an ethical exercise, this
liberalism offers no space to opt out or to refrain from
opting in. Obtaining the considered and uncon-
strained consent of all ± a mechanical view of politics
± becomes the heart of the liberal process rather than
the attainment of a package of values. The extreme
view of this position, outside current exemplars of
the liberal family, was expressed by non-liberals such
as Leo Strauss in his book Liberalism Ancient and
30
Modern: `liberal education is the necessary endeavour
to found an aristocracy within democratic mass
society.' The point, however, as Mill recognised,
was that liberalism was shaped by a few and then
disseminated to the many. Some of its substantive
values might precede the full evolution of demo-
cratic consent and might not be entirely sustained by
that evolution. As is the case with all ideologies,
liberals wish to protect the values they most cherish.
The problem with liberalism is that some of those
values (such as self-determination or participation)
are intrinsically dependent on the exercise of choice
by all, and may be catered to instantly, and some
(such as individuality or progress) are dependent on
the exercise of good choices by as many as possible,
but by less than all if unavoidable, and are time-
sensitive.
Third, it disposed of change in the arena of
political liberalism. Gone was the experimental at-
titude of Hobhouse and his colleague J. A. Hobson
towards some of the decontestations of liberal va-
55
lues. Rawls assumed coherence rather than tenta-
tiveness. `We may', he writes, `reaffirm our more
particular judgements and decide instead to modify
the proposed conception of justice until judgements
at all levels of generality are at last in line on due
56
reflection.' If there is experimentation, it is in order
to iron out kinks in the overlapping public reason-
ableness of political liberalism, not experimentation
that can provide different and contrasting views of
liberalism's values. Compare this with another mid-
century appraisal of liberalism: `as a form of social
thought Liberalism has been empirical, scientific,
57
mundane, and more or less skeptical-minded.'
The sense of paradigmatic shift is palpable.
Gone also were Hobhouse's and Hobson's anxi-
58
eties concerning an unknown liberal future. Raw-
lsian liberalism is curiously unperturbed, not only
because logic, procedure and coherence are by defi-
nition emotionally neutered (though their outcomes
are hardly value-neutral) but because moral conflicts
are politically soluble. Certainty allows one to dis-
pense with some of the affective language in which
liberalism has usually indulged. How can that non-
transcendental certainty be sustained? Because the
precarious nature of progress based on human rea-
59
son is eliminated at a stroke by removing liberalism
 TWENTIETH-CENTURY LIBERAL THOUGHT
from the impact of a temporal perspective.
One pre-eminent consequence of this paradig-
matic change has been the drastic contraction of
the natural habitat of liberalism, the political are-
60
na. The bulk of the liberal conceptual configura-
tion now spills over into a politics-free area, as
61
Dworkin has reminded us, and individuals are
literally abandoned to their own fates in the names
of autonomy and self-determination. The result is a
political liberalism that claims to be compatible with
an indeterminate variety of reasonable comprehen-
sive social and political beliefs, as long as they accept
the procedural `minimum-kit' of democratic consti-
tutionalism. This has significant repercussions for
liberalism. For in universalising political liberalism,
the specific appeal of other doctrines is undercut
either by excluding them as unreasonable or by
reducing them ± through revealing them as congru-
ous with political liberalism. Political liberalism thus
becomes a particularly potent weapon that can out-
bid the attractions of other doctrines/ideologies by
subsuming aspects of them into itself. The shrinkage
of the political is thus both very modest and highly
ambitious. To judge this enterprise as successful
depends on whether one thinks such formulae work,
or on whether one can demonstrate that such phi-
losophical liberals are unaware, or in denial, of their
own surpluses of meaning. As I have argued else-
where, Rawls's political liberalism contains almost all
of the crucial attributions of meaning that typify the
family of liberal concepts. His political liberalism has
gone so far down the road towards a comprehensive
liberalism that there is no turning back, no fork in
the road that could lead on to a comprehensive
socialism, conservatism or reasonable religious doc-
62
trine. It is therefore edifying to find Rawls's most
recent utterances redefining his theory of justice as
fairness ± hitherto understood to be a general theory
Notes
1. Hobhouse (1964), p. 28.
2. Hobson (1909), p. 91.
3. Hobhouse (1964), p. 60.
4. Ibid., pp. 16±29.
5. Hobson (1909), p. 95.
6. Hobhouse (1964), p. 15.
7. Ibid., pp. 28.
31
± as a comprehensive doctrine, and hence bereft of
the reasonable pluralism that a narrower political
63
liberalism requires.
Other theorists, as we have seen, have distin-
guished between good and bad forms of liberalism,
but no one prior to Rawls proposed to regard liberal-
ism as a two-tier edifice. What has this done to
liberalism? It has injected into it a spurious neutrality
and dissociated it from gathering the fruits of the
political process, hitherto a central aim of the liberal
agenda. It has upset the delicate balance between
essence and contingency so typical of liberal argu-
ment. And it has emaciated liberalism by relegating
its comprehensive values to the status of equal
contenders with other socio-political doctrines, thus
undercutting the specific ideological appeal of lib-
eralism which its other proponents have been so
eager to advance. That is another way of arguing that
what are now termed the liberal virtues are excluded
from the static preconditions concerning justice and
pluralism. Maybe we should take Rawls at his word
and stop calling his project liberal in a doctrinal
sense altogether?
Of course, liberalism continues to exist as a devel-
opmental and humanist theory. While universities
have over the past two decades sent liberalism on a
confusing trajectory, the ideology called liberalism is
still a comprehensive and culture-bound set of con-
ceptual decontestations that operates, like any ideol-
ogy, to compete over the legitimate meanings of
political language. Philosophical notions of liberal-
ism have not replaced historical, concrete, polysemic
and hence ideological understandings. The former
too may be seen in that latter light and simply serve
as further evidence of the contingent and manifold
guises liberalism adopts within the broad constraints
of shared family resemblances.
8. De Ruggiero (1927), p. 359.
9. See C. Taylor, `What's Wrong With Negative Lib-
erty?', in Ryan (ed.) (1979), pp. 175±93.
10. Hobhouse (1964), p. 30.
11. Ibid., p. 66.
12. Ibid., p. 72.
13. Ibid., p. 71. See also Freeden (1998).
 MICHAEL FREEDEN

14. See Freeden (1986), pp. 234±7. 46. Hartz (1955), p. 141.
15. Laski (1936), p. 11. 47. Ibid., pp. 208±9.
16. Hobhouse (1964), pp. 90, 127. 48. Ibid., pp. 249, 307.
17. De Ruggiero (1927), p. 390. 49. Croce (1946), pp. 87, 106.
18. Ibid., p. 437. 50. Rawls (1996), p. xxvi.
19. Orton (1945), pp. 78, 303. 51. Rawls (ibid., pp. 72±8) attempts to distinguish be-
20. Laski (1936), p. 14. tween various types of autonomy in a discussion that is
21. Ibid., p. 171. highly contentious. Suffice it to observe that his
22. Bullock and Shock (1956), p. xix. notions of rational and political autonomy, as distinct
23. Hobhouse (1964), p. 73. from full ethical autonomy, are posited on an unten-
24. De Ruggiero (1927), p. 90. able divide between political and comprehensive lib-
25. Ibid., p. 361. eralism, and on a blindness to the thickness of his
26. Ibid., p. 437. notion of an autonomous person in the space occupied
27. Ibid., p. 347. by political liberalism.
28. Croce (1946), pp. 79±81. 52. See Evans (1999).
29. Ibid., pp. 84±5. 53. See M. Freeden, `Liberal Community: An Exercise in
30. Hartz (1955), p. 16. Retrieval', in A. Simhony and D. Weinstein (eds),
31. Ibid., pp. 7, 10. Non-Individualist Liberalism (Cambridge: Cambridge
32. Hobson (1928). University Press, forthcoming).
33. Hobhouse (1924), p. 30. 54. This is notably the case for Locke, who in a little-noted
34. De Ruggiero (1927), p. 274. comment on the transition from the state of nature to
35. Ibid., p. 348. civil society notes: `This any number of Men may do,
36. Hallowell (1946), p. 1. because it injures not the Freedom of the rest; they are
37. For a discussion of these developments within the left as they were in the Liberty of the State of Nature'
American legal profession, see Primus (1999), pp. (Locke, 1991, section 96). This historically astute non-
177±97. universalism is inconceivable in the Rawlsian project.
38. In the extreme case, progress is seen as the central 55. See J.A. Hobson, `Character and Society', in Parker
structural principle of liberalism: `Progressivism is in- (ed.) (1912), p. 96.
deed a better term than liberalism for the opposition to 56. Rawls (1996), p. 45.
conservatism. For if conservatism is, as its name in- 57. Neill (1953), p. 18.
dicates, aversion to change or distrust of change, its 58. Hobhouse (1964), pp. 115±17; Hobson (1909), pas-
opposite should be identified with the opposite posture sim.

toward change, and not with something substantive 59. Neill (1953), p. 21.
like liberty or liberality' (Strauss, 1968, p. vii). 60. Contrast this with Laski's complaint that liberalism
39. Hallowell (1946), p. 88. has been far too political, in the sense of espousing the
40. Ibid., p. 109. ends of political, but not economic, democracy. For
41. Ibid., pp. 21, 108±9. Laski, `it was inherent in the Liberal idea that men
42. Kymlicka attempts to attach the (cultural) properties should use their political power for the improvement
of groups to the promotion of individual identity-cum- of their material position' (Laski, 1936, pp. 158±9).
autonomy alone. See Kymlicka (1995). 61. R. Dworkin, `Liberalism', in Dworkin (1985), pp.
43. Orton (1945), p. 308. 181±204.
44. Hobson (1909), p. 92. 62. See Freeden (1996), chapter 6.
45. Rawls (1996), pp. 231±40. 63. Rawls (1999), p. 614.

32
 Part Two
CITIZENSHIP:
UNIVERSALISM AND PARTICULARISM
This page intentionally left blank
 3
HUMAN RIGHTS AND
ETHNOCULTURAL JUSTICE
Will Kymlicka
I Introduction
It is an honour for me to be giving this Lecture in the
1
memory of John Rees, whose influential book on
John Stuart Mill helped to clarify my understanding
2
of liberalism and liberal democracy. Liberal democ-
racy rests on principles of freedom and equality, but
as Rees noted, it is not easy to specify or define these
principles. He suggested, however, that these prin-
ciples are perhaps best illustrated or reflected in the
doctrine of human rights ± that is, the doctrine that
each person has an intrinsic moral worth; that each
person's interests must be taken into consideration
by the state; and that each person should receive
certain inviolable protections against mistreatment,
3
abuse and oppression.
The articulation and diffusion of the doctrine of
human rights is indeed one of the great achievements
of postwar liberalism. This year represents the fiftieth
anniversary of the Universal Declaration of Human
Rights, and it is impressive how far we have come in
those fifty years in ensuring the protection of human
dignity and human rights. Yet amid the anniversary
celebrations there are an increasing number of scep-
tics and critics who attack the doctrine of human
rights for neglecting the realities of ethnocultural
diversity, and for being unable to deal with the
conflicts which arise from this diversity, either with-
in a society or between societies.
This is an issue which Rees did not foresee, and
perhaps could not have foreseen. But it is an urgent
35
issue for anyone who wishes to defend and promote
the liberal project in the twenty-first century. But
what exactly is the challenge which ethnocultural
diversity raises for human rights? We can distinguish
two separate challenges. The first is foundational.
Some critics argue that the conception of human
personhood and human needs underlying the doc-
trine of human rights is culturally biased. More
specifically, it is `Eurocentric', and in particular it
exhibits a European commitment to individualism,
whereas other non-western cultures have a more
collectivist or communitarian understanding of hu-
man identity. On this view, given the depth of
cultural differences around the world, the very idea
of developing a single set of universal human rights is
inherently ethnocentric, and involves imposing one
culture's view of human personality and human
identity on other cultures.
The second challenge is more modest. Some
critics say that the idea of universal human rights
is acceptable in principle, but that the current list of
human rights is radically incomplete. In particular, it
fails to protect minority cultures from various forms
of injustice, and so needs to be supplemented with an
additional set of what are sometimes called `collec-
tive rights' (or `group rights', `minority rights' or
`cultural rights').
Most discussions of cultural diversity and human
rights have focused on the first, foundational, chal-
lenge. Here, however, I will focus on the second,
since it is less explored. Many commentators tend to
 WILL KYMLICKA
confuse and conflate the two challenges; they assume
that any group which is demanding `collective rights'
must be doing so because they are `collectivist' in
their cultural outlook and attitudes, and so are,
implicitly at least, raising the foundationalist chal-
lenge to the very idea of universal human rights. If
we conflate the two debates in this way, it will seem
that we need to revisit the foundationalist issue every
time an ethnocultural group demands collective
rights.
This tendency to conflate the two debates is, I
think, profoundly unhelpful. It is a source not only of
philosophical misunderstanding, but also, and per-
haps more importantly, of political confusion which
has prevented liberals from understanding and effec-
tively responding to ethnocultural conflicts around
the world. So my aim is to discuss the second issue on
its own terms. I will try to show why current con-
ceptions of human rights leave serious issues of
ethnocultural injustice unaddressed, and why these
issues of injustice are not reducible or dependent on
the foundationalist challenge to human rights. I
hope to show that these claims of injustice are
relevant and urgent even for those of us who reject
the foundationalist critique of human rights. Indeed,
I think that we will only be in a position to effec-
tively address the foundationalist critique of human
rights if we have first made sure that our conception
of human rights deals satisfactorily with these issues
of ethnocultural injustice.
The discussion is divided into two main sections.
In the first section, I discuss why human rights are
insufficient for ethnocultural justice and may even
exacerbate certain injustices, and hence why human
rights standards must be supplemented with various
minority rights. In the second section, I ask whether,
if human rights are supplemented with minority
rights, we can hope or expect to get greater agree-
ment on the transnational application of human
rights. I argue that we can indeed hope for greater
agreement on the principles of human rights, but that
there will still be very difficult issues remaining about
appropriate institutions for the enforcement of these
rights. I will conclude with some broader reflections
on the relationship between cultural diversity and
human rights.
36
Individual Rights and Group Rights
According to many commentators, human rights are
paradigmatically individual rights, as befits the in-
dividualism of western societies, whereas non-Eur-
opean societies are more interested in `group' or
`collective' rights, as befits their communalist tradi-
tions.
I think this way of framing the debate is quite
misleading. For one thing, individual rights have
typically been defended within the Western tradi-
tion precisely on the grounds that they enable
various group-oriented activities. Consider the para-
digmatic liberal right ± namely, freedom of religion ±
which Rawls argues is the origin and foundation for
all other liberal rights. The point of endowing in-
dividuals with rights to freedom of conscience and
freedom of worship is to enable religious groups to
form and maintain themselves, and to recruit new
members. And indeed according individuals rights to
religious freedom has proven very successful in en-
abling a wide range of religious groups ± including
many non-western religions ± to survive and flourish
in western societies.
Based partly on this example of religious toler-
ance, many commentators have argued that indivi-
dual rights provide a firm foundation for justice for
all groups, including ethnocultural minorities. In-
deed, this was explicitly the argument given after
the Second World War for replacing the League of
Nations' `minorities protection' scheme ± which
accorded collective rights to specific groups ± with
the UN's regime of universal human rights. Rather
than protecting vulnerable groups directly, through
special rights for the members of designated groups,
cultural minorities would be protected indirectly, by
guaranteeing basic civil and political rights to all
individuals regardless of group membership. Basic
human rights such as freedom of speech, association
and conscience, while attributed to individuals, are
typically exercised in community with others, and
so provide protection for group life. Where these
individual rights are firmly protected, liberals as-
sumed, no further rights needed to be attributed to
the members of specific ethnic or national mino-
rities:
 HUMAN RIGHTS AND ETHNOCULTURAL JUSTICE
The general tendency of the postwar movements
for the promotion of human rights has been to
subsume the problem of national minorities under
the broader problem of ensuring basic individual
rights to all human beings, without reference to
membership in ethnic groups. The leading as-
sumption has been that members of national
minorities do not need, are not entitled to, or
cannot be granted rights of a special character.
The doctrine of human rights has been put for-
ward as a substitute for the concept of minority
rights, with the strong implication that minorities
whose members enjoy individual equality of treat-
ment cannot legitimately demand facilities for the
4
maintenance of their ethnic particularism.
Guided by this philosophy, the United Nations
deleted all references to the rights of ethnic and
national minorities in its Universal Declaration of
Human Rights.
There is of course much truth in this claim that
individual rights protect group life. Freedom of as-
sociation, religion, speech, mobility and political
organisation enables individuals to form and main-
tain the various groups and associations which con-
stitute civil society, to adapt these groups
changing circumstances, and to promote their views
and interests to the wider population. The protec-
tion afforded by these common rights of citizenship is
sufficient for many of the legitimate forms of group
diversity in society.
However, it is increasingly clear that the list of
common individual rights guaranteed in western
democratic constitutions, or in the UN Declaration,
is not sufficient to ensure ethnocultural justice,
particularly in states with national minorities. By
national minorities, I mean groups which formed
functioning societies, with their own institutions,
culture and language, concentrated in a particular
territory, prior to being incorporated into a larger
state. The incorporation of such national minorities
is usually involuntary, as a result of colonisation,
conquest or the ceding of territory from one imperial
power to another, but may also occur voluntarily,
through some treaty or other federative agreement.
Examples of national minorities within western
democracies include the indigenous peoples, Puerto
Âbe
Ricans and Que  cois in North America, the Cat-
alans and Basques in Spain, the Flemish in Belgium,
the Sami in Norway, and so on. Most countries
around the world contain such national minorities,
and most of these national minorities were involun-
tarily incorporated into their current state ± a testa-
ment to the role of imperialism and violence in the
formation of the current system of `nation-states'.
Very few if any of these national minorities are
satisfied merely with respect for their individual
human rights, and it is easy to see why. I will discuss
three examples where individual rights do not ade-
quately protect their interests: decisions about inter-
nal migration/settlement policies, decisions about
the boundaries and powers of internal political units,
and decisions about official languages.
In each of these examples, and throughout the
analysis, I will be using the term `human rights' in an
imprecise way. I am not referring to any particular
canonical statement or declaration of international
human rights, but rather to the constellation of
individual civil and political rights which are for-
mulated in western democratic constitutions, and
which many advocates of human rights would like to
see entrenched and enforced as transnational stan-
to dards of human rights. Some of these rights are
included in the original Declaration, others in sub-
sequent conventions (such as the 1966 Covenant on
Civil and Political Rights), others are still being
debated. In short, I am referring more to a particular
public and political discourse of `human rights', than
to the actual list of human rights in any particular
document.
5
Internal Migration/Settlement Policies
National governments have often encouraged people
from one part of the country (or new immigrants) to
move into the historical territory of the national
minority. Such large-scale settlement policies are
often deliberately used as a weapon against the
national minority, both to break open access to their
territory's natural resources and to disempower them
politically by turning them into a minority even
6
within their own traditional territory.
This process is occurring around the world, in
Bangladesh, Israel, Tibet, Indonesia, Brazil, for ex-
37
 WILL KYMLICKA
7
ample. It has also happened in North America. The
nineteenth-century Canadian Prime Minister, Sir
Âtis that `these
John A. MacDonald, said of the Me
impulsive half-breeds . . . must be kept down by a
strong hand until they are swamped by the influx of
8
settlers.' And the same process occurred in the
American Southwest, where immigration was used
to disempower the indigenous peoples and Chicano
populations who were living on that territory when it
was incorporated into the United States in 1848.
This is not only a source of grave injustice, but is
also the most common source of violent conflict in
the world. Indigenous peoples and other homeland
minorities typically resist such massive settlement
9
policies, even with force, if necessary. One would
hope, therefore, that human rights doctrines would
provide us with the tools to challenge such policies.
Unfortunately, there is nothing in human rights
doctrine which precludes such settlement policies
(so long as individual members of the minority are
not deprived of their civil and political rights). There
are other elements of international law which might
be of some help, in exceptional circumstances. For
example, UN Resolution Number 2189, adopted in
December 1967, condemns attempts by colonial
powers systematically to promote the influx of im-
migrants into their colonised possessions. But this
only applies to overseas colonies or to newly con-
quered territories, not to already incorporated na-
tional minorities and indigenous peoples. So it is of
Âtis or Tibetans.
no help to the Me
Human rights doctrines are not only silent on this
question, but in fact may exacerbate the injustice,
since the UN Charter guarantees the right to free
mobility within the territory of a state. Indeed,
ethnic Russians in the Baltics defended their settle-
ment policies precisely on the grounds that they had
a human right to move freely throughout the territory
of the former Soviet Union. It is important to
remember that most countries recognised the bound-
aries of the Soviet Union, and so the UN Charter
does indeed imply that ethnic Russians had a basic
right to settle freely in any of the Soviet republics,
even to the point where the indigenous inhabitants
were becoming a minority in their own homeland.
Similarly, human rights doctrines, far from prohibit-
ing ethnic Han settlement in Tibet, imply that
38
Chinese citizens have a basic human right to settle
there.
To protect against these unjust settlement poli-
cies, national minorities need and demand a variety
of measures. For example, they may make certain
land claims ± insisting that certain lands be reserved
for their exclusive use and benefit. Or they may
demand that certain disincentives be placed on
in-migration. For example, migrants may need to
pass lengthy residency requirements before they can
vote in local or regional elections. Or they may not
be able to bring their language rights with them ±
that is, they may be required to attend schools in the
local language, rather than having publicly funded
education in their own language. Similarly, the
courts and public services may be conducted in
the local language. These measures are all intended
to reduce the number of migrants into the homeland
of the national minority, and to ensure that those
who do come are willing to integrate into the local
culture.
These are often cited as examples of the sort of
`group rights' which conflict with western individu-
alism. They are said to reflect the minority's `com-
munal' attachment to their land and culture. But in
fact these demands have little if anything to do with
the contrast between `individualist' and `commun-
alist' societies. Western `individualists societies also
seek protections against in-migration. Take any wes-
tern democracy. While the majority believes in
maximising their individual mobility throughout
the country, they do not support the right of in-
dividuals outside the country to enter and settle. On
the contrary, western democracies are typically very
restrictive about accepting new immigrants into their
society. None has accepted the idea that transna-
tional mobility is a basic human right. And those few
immigrants who are allowed in are pressured to
integrate into the majority culture. For example,
learning the majority language is a condition of
gaining citizenship, and publicly funded education
is typically provided only in the majority's language.
Western democracies impose these restrictions on
immigration into their country for precisely the same
reason that national minorities seek to restrict in-
migration into their territory ± namely, massive
settlement would threaten their society and culture.
 HUMAN RIGHTS AND ETHNOCULTURAL JUSTICE
The majority, like the minority, has no desire to be
overrun and outnumbered by settlers from another
culture.
To say that the desire of national minorities to
limit in-migration reflects some sort of illiberal com-
munalism is therefore quite hypocritical. When the
majority says that mobility within a country is a basic
human right, but that mobility across borders is not,
they are not preferring individual mobility over
collective security. They are simply saying that their
collective security will be protected (by limits on
immigration), but that once their collective security
is protected, then individual mobility will be max-
imised, regardless of the consequences for the col-
lective security of minorities. This is obviously
hypocritical and unjust, but it is an injustice which
human rights doctrines do not prevent and may even
10
exacerbate.
The Boundaries and
Powers of Internal Political Subunits
In states with territorially concentrated national
minorities, the boundaries of internal political sub-
units raise fundamental issues of justice. Since
national minorities are usually territorially concen-
trated, these boundaries can be drawn in such a way
as to empower them ± that is, to create political
subunits within which the national minority forms a
local majority, and which can therefore be used as a
vehicle for meaningful autonomy and self-govern-
ment.
In many countries, however, boundaries have
been drawn so as to disempower national minorities.
For example, a minority's territory may be broken up
into several units so as to make cohesive political
action impossible (for example, the division of
France into 83 `departments' after the Revolution
which intentionally subdivided the historical regions
of the Basques, Bretons and other linguistic mino-
rities, or the division of nineteenth-century Catalo-
nia); conversely, a minority's territory may be
absorbed into a larger political subunit so as to ensure
that they are outnumbered within the subunit as a
whole (such as the Hispanics in nineteenth-century
11
Florida).
Even where the boundaries coincide more or less
39
with the territory of a national minority, the degree
of meaningful autonomy may be undermined if the
central government usurps most or all of the sub-
unit's powers and eliminates the group's traditional
mechanisms of self-government. And indeed, we can
find many such instances in which a minority nom-
inally controls a political subunit but has no sub-
stantive power, since the central government has (1)
removed the traditional institutions and procedures
of group self-government; and (2) arrogated all im-
portant powers, even those affecting the very cultural
survival of the group ± for example, jurisdiction over
economic development, education, language. (Con-
sider the plenary power of the American Congress
over Indian tribes in the United States.)
This usurpation of power is, I believe, a clear
injustice, particularly when it involves seizing powers
or undermining institutions which were guaranteed
to the minority in treaties or federating agreements.
Yet here again, it seems that human rights doctrines
are inadequate to prevent such injustice. So long as
individual members maintain the right to vote and
run for office, then human rights principles pose no
obstacle to the majority's efforts to gerrymander the
boundaries or powers of internal political subunits in
such a way as to disempower the national minority.
This is true even if the arrogation of power violates
an earlier treaty or federative agreement, since such
internal treaties are not considered `international'
agreements (the minority which signed the treaty is
not seen under international law as a sovereign state,
and so its treaties with the majority are seen as
matters of domestic politics, not international law).
Not only do human rights doctrines not help
prevent this injustice, but they may exacerbate it.
Historically, the majority's decisions to ignore the
traditional leadership of minority communities and
to destroy their traditional political institutions have
been justified on the grounds that these traditional
leaders and institutions were not `democratic': they
did not involve the same process of periodic elec-
tions as majority political institutions. The tradi-
tional mechanisms of group consultation,
consensus and decision-making may well have pro-
vided every member of the minority community with
meaningful rights to political participation and in-
fluence. However, they were swept away by the
 WILL KYMLICKA
majority in the name of `democracy' ± that is, the
right to vote in an electoral process within which
minorities had no real influence, conducted in a
foreign language and in foreign institutions, and
within which they were destined to become a per-
manent minority. Thus the rhetoric of human rights
has provided an excuse and smokescreen for the
subjugation of a previously self-governing minor-
12
ity.
To avoid this sort of injustice, national minorities
need guaranteed rights to such things as self-govern-
ment, group-based political representation,
rights over issues that directly affect their cultural
survival, and so on. Again, these demands are often
seen as conflicting with western individualism and as
proof of the minority's `collectivism'. But in reality,
these demands simply help to redress clear political
inequalities. After all, the majority would equally
reject any attempt by foreign powers to unilaterally
change its boundaries, institutions or self-govern-
ment powers. So why shouldn't national minorities
also seek guarantees of their boundaries, institutions
and powers?
In a recent paper, Avigail Eisenberg details how
the debate over Aboriginal political rights in the
Canadian north has been seriously distorted by the
focus on western `individualism' versus Aboriginal
`collectivism'. This way of framing the debate misses
the real issues, which are the ongoing effects of
colonisation ± the political subordination of one
people to another, through the majority's unilateral
efforts to undermine the minority's institutions and
13
powers of self-government.
Official Language Policy
In most democratic states, governments have typi-
cally adopted the majority's language as the one
`official language' ± that is, as the language of gov-
ernment, bureaucracy, courts, schools, and so on. All
citizens are then forced to learn this language in
school, and fluency in it is required to work for, or
deal with, government. While this policy is often
defended in the name of `efficiency', it is also
adopted to ensure the eventual assimilation of the
national minority into the majority group. There is
strong evidence that languages cannot survive for
long in the modern world unless they are used in
public life, and so government decisions about offi-
cial languages are, in effect, decisions about which
14
languages will thrive, and which will die out.
Just as the traditional political institutions of
minorities have been shut down by the majority,
so too have the pre-existing educational institutions.
For example, Spanish schools in the American
Southwest were closed after 1848, and replaced with
English-language schools. Similarly, French-lan-
guage schools in Western Canada were closed once
veto English-speakers achieved political dominance.
This can be an obvious source of injustice. Yet
here again, principles of human rights do not prevent
this injustice (even when, as in the Southwest, there
were treaties guaranteeing Hispanics the right to
their own Spanish-language schools). Human rights
doctrines do preclude any attempt by the state to
suppress the use of a minority language in private,
and may even require state toleration of privately
funded schools which operate in the minority lan-
guage. But human rights doctrines say nothing about
15
rights to the use of one's language in government.
On some interpretations of more recent interna-
tional conventions which include minority rights,
public funding for mother-tongue classes at elemen-
tary level may under some circumstances be seen as a
`human right'. But this remains a controversial de-
16
velopment. Moreover, mother-tongue education at
the elementary level is clearly insufficient if all jobs
in a modern economy require education at higher
levels conducted in the majority language. Indeed,
the requirement that education at senior levels be in
the majority language creates a disincentive for
minority parents to enrol their children in minor-
17
ity-language elementary schools in the first place.
To redress the injustice created by majority at-
tempts to impose linguistic homogeneity, national
minorities may need broad-ranging language poli-
cies. There is evidence that language communities
can only survive intergenerationally if they are nu-
merically dominant within a particular territory, and
if their language is the language of opportunity in
that territory. But it is difficult to sustain such a
predominant status for a minority language, particu-
larly if newcomers to the minority's territory are able
to become educated and employed in the majority
40
 HUMAN RIGHTS AND ETHNOCULTURAL JUSTICE
language (for example, if newcomers to Quebec are
able to learn and work in English). It may not be
enough, therefore, for the minority simply to have
the right to use its language in public. It may also be
necessary that the minority language be the only
18
official language in their territory. If immigrants, or
migrants from the majority group, are able to use the
majority language in public life, this may eventually
undermine the predominant status, and hence via-
19
bility, of the minority's language. In other words,
minorities may need not personal bilingualism (in
which individuals carry their language rights with
them throughout the entire country), but rather
territorial bilingualism (in which people who choose
to move to the minority's territory accept that the
minority's language will be the only official language
in that territory). Yet this sort of territorial bilingu-
alism ± which denies official language status to the
majority language on the minority's territory ± is
often seen as discriminatory by the majority, and
indeed as a violation of their `human rights'.
These demands for extensive language rights and
territorial bilingualism are often described as evi-
dence of the minority's `collectivism'. But here again
the minority is just seeking the same opportunity to
live and work in their own language that the ma-
jority takes for granted. There is no evidence that the
majority attaches any less weight to their ability to
20
use their language in public life.
One could mention other issues where human rights
are insufficient for ensuring ethnocultural justice
(public holidays, school curriculum, national sym-
bols, dress-codes and so on). But enough has been
said, I hope, to make the general point. Moreover, it
is important to note that the three issues I have
examined ± migration, internal political subunits
and language policy ± are all connected. These are
all key components in the `nation-building' pro-
grammes which every western state has engaged
21
in. Every democratic state has, at one time or
another, attempted to create a single `national iden-
tity' among its citizens, and so has tried to undermine
any competing national identities, of the sort which
22
national minorities often possess. Policies designed
to settle minority homelands, undermine their poli-
tical and educational institutions and impose a single
41
common language have been important tools in
these nation-building efforts. There is no evidence
that states intended to relinquish these tools when
they accepted human rights conventions, and indeed
no evidence that states would have accepted a
conception of human rights which would preclude
such nation-building programmes.
Of course, human rights standards do set limits on
this process of nation-building. States cannot kill or
expel minorities, or strip them of citizenship or deny
them the vote. But human rights standards do not
preclude less extreme forms of nation-building. And
if these nation-building measures are successful, then
it is not necessary to restrict the individual civil and
political rights of the minority. Where nation-build-
ing programmes have succeeded in turning the in-
corporated group into a minority even within its own
homeland, and in stripping it of its self-governing
institutions and language rights, then the group will
not pose any serious threat to the power or interests
of the majority. At this point, there is no need to
strip minority members of their individual rights.
This is not necessary in order to gain and maintain
effective political control over them.
In short, human rights standards are insufficient to
prevent ethnocultural injustice, and may actually
makes things worse. The majority can invoke human
rights principles to demand access to the minority's
homeland, to scrap traditional political mechanisms
of consultation and accommodation, and to reject
linguistic policies which try to protect the territorial
viability of minority communities.
In these and other ways, human rights have in-
directly served as an instrument of colonisation, as
various critics have argued. However, I would not
agree with those critics who view this solely as a
problem of `western imperialism' against non-Eur-
opean peoples. After all, these processes of unjust
subjugation have occurred between European groups
(such as the treatment of national minorities by the
majority in France, Spain, Russia), and between
African or Asian groups (such as the treatment of
the Yao minority by the Chewa majority in Malawi,
or the treatment of the Tibetan minority by the Han
majority in China), as well as in the context of
western colonisation of non-western peoples. These
processes have occurred in virtually every state with
 WILL KYMLICKA
national minorities, and to ascribe it to western
individualism is to seriously underestimate the scope
of the problem.
If human rights are not to be an instrument of
unjust subjugation then they must be supplemented
with various minority rights ± language rights, self-
government rights, representation rights, federalism
and so on. Moreover, these minority rights should
not, I think, be seen as in any way secondary to
traditional human rights. Even those who are sym-
pathetic to the need for minority rights often say that
we should at least begin with human rights. That is,
we should first secure respect for individual human
rights, and then, having secured the conditions for a
free and democratic debate, move on to questions of
minority rights. When national minorities oppose
this assumption, they are often labelled as illiberal or
anti-democratic. But as I've tried to show, we cannot
assume that human rights will have their desired
consequences without attending to the larger con-
text within which they operate. Unless supplemen-
ted by minority rights, majoritarian democracy and
individual mobility rights may simply lead to min-
ority oppression. As we've seen, various forms of
oppression can occur while still respecting the in-
dividual rights of minorities. As a result, the longer
we defer discussing minority rights, the more likely it
is that the minority will become increasingly wea-
kened and outnumbered. Indeed, it may over time
become so weakened that it will become unable even
to demand or exercise meaningful minority rights
(for example, it may lose the local predominance or
territorial concentration needed to sustain its lan-
guage or to exercise local self-government). It is no
accident, therefore, that members of the majority are
often loudest in their support for giving priority to
democracy and human rights over issues of minority
rights. They know that the longer issues of minority
rights are deferred, the more time it provides for the
majority to disempower and dispossess the minority
of its land, schools and political institutions to the
point where the minority is in no position to sustain
itself as a thriving culture or to exercise meaningful
self-government.
This is why human rights and minority rights must
be treated together, as equally important compo-
nents of a just society. Of course, it would be an
42
equally serious mistake to privilege minority rights
over human rights. In questioning the priority of
traditional human rights over minority rights, I am
not disputing the potential for serious rights viola-
tions within many minority groups, or the need to
have some institutional checks on the power of local
or minority political leaders. On the contrary, all
political authorities should be held accountable for
respecting the basic rights of the people they govern,
and this applies as much to the exercise of self-
government powers by national minorities as to
the actions of the larger state. The individual mem-
bers of national minorities can be just as badly
mistreated and oppressed by the leaders of their
own group as by the majority government, and so
any system of minority self-government should in-
clude some institutional provisions for enforcing
traditional human rights within the minority com-
munity.
It is not a question of choosing between minority
rights and human rights, or of giving priority to one
over the other, but rather of treating them together
as equally important components of justice in eth-
noculturally plural countries. We need a conception
of justice that integrates fairness between different
ethnocultural groups (via minority rights) with the
protection of individual rights within majority and
minority political communities (via traditional hu-
23
man rights).
The Enforcement of Human Rights
Assuming that we can come up with some new
theory which combines human rights and minority
rights, would the existing level of opposition to
transnational human rights standards diminish?
Would we then get consensus on the enforcement
of international standards of human rights?
We can expect, I think, that the elites of some
groups will continue to say human rights principles
contradict their cultural `traditions'. I will return to
this possibility in my conclusion. But my guess is that
much of the current opposition to human rights
would fade away. As I noted earlier, human rights
are not inherently `individualistic' and do not pre-
clude group life. They simply ensure that traditions
are voluntarily maintained, and that dissent is not
 HUMAN RIGHTS AND ETHNOCULTURAL JUSTICE
forcibly suppressed. To be sure, self-serving political
elites who want to suppress challenges to their
authority from within the community will continue
to denounce human rights as a violation of their
`traditions'. This explains the recent criticisms of
human rights doctrines by the Indonesian and Chi-
nese governments. But my guess is that if human
rights doctrines are no longer seen as a tool for
subordinating one people to another, but rather as
a tool for protecting vulnerable individuals from
abuse by their political leaders, then such opposition
to human rights will increasingly be seen simply as a
self-serving defence of elite power and privilege.
So I would hope that we could gain greater
international consensus on the principles of human
rights. But this is not to say that we are likely to get
consensus on the appropriate enforcement mechan-
isms of human rights/minority rights, either at the
international level or even in the domestic setting.
There are at least two major difficulties here. First, it
is difficult to see how minority rights can be codified
at the international level. Minorities come in many
different shapes and sizes. There are `national' mino-
rities, indigenous peoples, immigrants, refugees,
guestworkers, colonising settlers, descendants
slaves or indentured labourers, Roma, religious
groups, and so on. All of these groups have different
25
needs, aspirations and institutional capacities.
ritorial autonomy will not work for widely dispersed
groups, and even territorially concentrated groups
differ dramatically in the sort of self-government
they aspire to or are capable of. Similarly, language
rights (beyond the right to private speech) will not
be the same in India or Malaysia (which contain
hundreds of indigenous languages) as in France or
Britain.
This is why international declarations of minority
rights tend to waver between trivialities like the
`right to maintain one's culture' (which could simply
mean that respect for freedom of expression and
association, and hence add nothing to existing de-
clarations of human rights) and vague generalities
like the `right to self-determination' (which could
mean anything from token representation to full-
26
blown secession). Minorities are not going to
accept the lowest common denominator, which even
the smallest or most dispersed group seeks, but
majorities are not going to give all groups the max-
imal rights demanded by the largest and most mo-
bilised groups (which may include secession).
I see no way to overcome this problem. While
minority rights are indeed essential, the solution is
not to add a detailed list of minority rights to human
rights declarations in international law. Instead, we
must accept that traditional human rights are in-
sufficient to ensure ethnocultural justice, and then
recognise the need to supplement them within each
country with the specific minority rights which are
appropriate for that country. As I will discuss later,
24
international bodies can play a useful role in adju-
dicating minority rights conflicts, but this role is
unlikely to take the form of adjudicating or enforcing
a single codified international list of minority rights.
This then leads to the second problem. If human
rights and minority rights must be integrated at the
domestic level rather than through a single interna-
tional code, can we find an impartial body to ad-
judicate and enforce these rights at the domestic
level? Many people will naturally assume that these
rights should be listed in a single national constitu-
tion which is then adjudicated and enforced by a
of single supreme court. Certainly most liberals have
assumed that the supreme court in each country
should have final jurisdiction regarding both human
Ter- rights and minority rights.
But in fact we find strong resistance to this idea
among some minority groups, even if they share the
principles underlying the set of human rights and
minority rights listed in the national constitution.
Consider the situation of Indian tribes in the United
States. American constitutional law protects both
certain minority rights for Indian tribes (they are
recognised as `domestic dependent nations' with
treaty-based rights of self-government) and also a
general set of individual human rights (in the Bill of
Rights). This could be seen as at least the beginnings
of an attempt to fairly integrate minority rights and
27
human rights at the domestic level.
But who should have the power to enforce these
constitutional provisions regarding individual and
minority rights? American liberals typically assume
that the federal Supreme Court should have this
power. But many American Indians oppose this idea.
They do not want the Supreme Court to be able to
43
 WILL KYMLICKA
review their internal decisions to assess whether they
28
comply with the Bill of Rights. And they would
prefer to have some international body monitor the
extent to which the American government respects
their treaty-based minority rights. So they do not
want the federal Supreme Court to be the ultimate
protector either of the individual rights of their
members or of their minority rights.
Needless to say, Indian demands to reduce the
authority of the federal Supreme Court have met
with resistance. The American government has
shown no desire to accept international monitoring
of the extent to which the treaty rights of Indians are
respected. Indeed, the American government has
jealously guarded its sovereignty in these matters,
and so refused to give any international body jur-
isdiction to review and overturn the way it respects
the treaty rights, land claims or self-government
rights of indigenous peoples.
And the demand to have internal tribal decisions
exempted from scrutiny under the Bill of Rights is
widely opposed by liberals, since it raises the concern
in many people's minds that individuals or subgroups
(such as women) within American Indian commu-
nities could be oppressed in the name of group
solidarity or cultural purity. They argue that any
acceptable package of individual rights and minority
rights must include judicial review of tribal decisions
by the American Supreme Court to ensure their
compliance with the Bill of Rights.
Before jumping to this conclusion, however, we
should consider the reasons why certain groups are
distrustful of federal judicial review. In the case of
American Indians, these reasons are, I think, ob-
vious. After all, the federal Supreme Court has
historically legitimised the acts of colonisation and
conquest which dispossessed Indians of their prop-
erty and political power. It has historically denied
both the individual rights and treaty rights of Indians
on the basis of racist and ethnocentric assumptions.
Moreover, Indians have had no representation on
the Supreme Court, and there is reason to fear that
white judges on the Supreme Court may interpret
certain rights in culturally biased ways (such as
democratic rights). Why should Indians agree to
have their internal decisions reviewed by a body
which is, in effect, the court of their conquerors?
44
And why should they trust this court to act impar-
tially in considering their minority and treaty rights?
For all these reasons, to assume that supreme courts
at the national level should have the ultimate
authority over all issues of individual and minority
rights within a country may be inappropriate in the
case of indigenous peoples and other incorporated
29
national minorities. There are good reasons why
American Indians do not trust federal courts either
to uphold the minority rights needed for ethnocul-
tural justice between majority and minority, or to
fairly determine whether the minority is respecting
human rights internally.
It is quite understandable, therefore, that many
Indian leaders seek to reduce the role of federal
judicial review. But at the same time they affirm
their commitment to the basic package of human
rights and minority rights which is contained in the
US constitution. They endorse the principles, but
object to the particular institutions and procedures
that the larger society has established to enforce
these principles. As Joseph Carens puts it, `people
are supposed to experience the realisation of prin-
ciples of justice through various concrete institu-
tions, but they may actually experience a lot of
30
the institution and very little of the principle.'
This is exactly how many indigenous peoples per-
ceive the Supreme Court in both Canada and the
United States. What they experience is not the
principles of human dignity and equality, but rather
a social institution which has historically justified
their conquest and dispossession.
What we need to do, therefore, is to find impartial
bodies to monitor compliance with both human
rights and minority rights. We need to think crea-
tively about new mechanisms for enforcing human
rights and minority rights that will avoid the legit-
imate objections which indigenous peoples and na-
tional minorities have to federal courts.
What would these alternative mechanisms look
like? To begin with, many Indian tribes have sought
to create or maintain their own procedures for
protecting human rights within their community,
specified in tribal constitutions, some of which are
based on the provisions of international protocols on
human rights. It is important to distinguish Indian
tribes, who have their own internal constitution and
 HUMAN RIGHTS AND ETHNOCULTURAL JUSTICE
courts which prevent the arbitrary exercise of poli-
tical power, from ethnocultural groups which have
no formal constitutions or courts, and which there-
fore provide no effective check on the exercise of
arbitrary power by powerful individuals or traditional
elites. We should take these internal checks on the
misuse of power seriously. Indeed, to automatically
assume that the federal courts should replace or
supersede the institutions which Indians have them-
selves evolved to prevent injustice is evidence of an
ethnocentric bias ± an implicit belief that `our'
31
institutions are superior to theirs.
Indian tribes have also sought to create new
transnational or international procedures to help
monitor the protection of their minority rights.
The international community can play an important
role, not so much by formulating a single list of
minority rights that applies to all countries (for that
is impossible), but rather by providing an impartial
adjudicator to monitor the extent to which domestic
provisions regarding minority rights are fairly nego-
tiated and implemented.
From the point of view of ethnocultural justice,
these proposals might be preferable to the current
reliance on the federal Supreme Court. But it would
be even better, I think, to establish international
mechanisms which would monitor both the indivi-
dual rights and minority rights of Indian peoples.
While the internal courts and constitutions of tribal
governments are worthy of respect, they ± like the
courts and constitutions of nation-states ± are im-
perfect in their protection of human rights. So it
would be preferable if all governments ± majority and
minority ± are subject to some form of international
scrutiny.
Many Indian leaders have expressed a willingness
to accept some form of international monitoring of
their internal human rights record. They are willing
to abide by international declarations of human
rights, and to answer to international tribunals for
complaints of rights violations within their commu-
nities. But they would only accept this if and when it
is accompanied by international monitoring of how
well the larger state respects their treaty rights. They
accept the idea that their tribal governments, like all
governments, should be accountable to international
human rights norms (so long as this isn't in the court
45
of their conquerors). But they want this sort of
external monitoring to examine how well their
minority rights are upheld by the larger society,
not just to focus on the extent to which their
own decisions respect individual human rights. This
seems like a reasonable demand to me.
These international mechanisms could arise at the
regional as well as global level. European countries
have agreed to establish their own multilateral hu-
man rights tribunals. Perhaps North American gov-
ernments and Indian tribes could agree to establish a
similar multilateral tribunal, on which both sides are
fairly represented.
My aim here is not to defend any particular
proposal for a new impartial body to monitor the
protection of individual rights and minority rights.
Rather, it is to stress again the necessity of treating
individual rights and minority rights together when
thinking about appropriate enforcement mechan-
isms. On the one hand, we need to think about
effective mechanisms which can hold minority gov-
ernments accountable for the way individual mem-
bers are treated. I see no justification for exempting
minority self-government from the principles of hu-
man rights ± I believe that any exercise of political
power should be subject to these principles. But we
need to simultaneously think about effective me-
chanisms for holding the larger society accountable
for respecting the minority rights of these groups. As
I argued in the previous section, minority rights are
equally important as individual human rights in
ensuring ethnocultural justice, and so should be
subject to equal scrutiny. Moreover, focusing exclu-
sively on the latter while neglecting the former
is counter-productive and hypocritical. Minority
groups will not agree to greater external scrutiny
of their internal decisions unless they achieve greater
protection of their minority rights. And since exist-
ing institutional mechanisms are typically unable to
meet this twin test of accountability, we need to
think creatively about new mechanisms that can
deal impartially with both individual human rights
and minority rights.
I should emphasise again that in questioning the
role of federal courts in reviewing the internal de-
cisions of tribal governments I am not trying to
diminish the significance of human rights. On the
 WILL KYMLICKA
contrary, my goal is to find fairer and more effective
ways to promote human rights, by separating them
from the historical and cultural baggage which makes
federal courts suspect in the eyes of incorporated
peoples. History has shown the value of holding
governments accountable for respecting human
rights, and this applies equally to Indian tribal
governments. But the appropriate forums for review-
ing the actions of self-governing indigenous peoples
may skip the federal level, as it were. Many indi-
genous groups would endorse a system in which their
self-governing decisions are reviewed in the first
instance by their own courts, and then by an inter-
national court, which would also monitor respect for
minority rights. Federal courts, dominated by the
majority, would not be the ultimate adjudicator of
either the individual or minority rights of Indian
peoples.
The human rights movement has always set itself
against the idea of unlimited sovereignty and pushed
for limitations on the power of governments to
mistreat their citizens. But we need to do this in
an even-handed manner. We should hold the min-
ority accountable for respecting the human rights of
its members, but we need also to hold the majority
accountable for respecting minority rights, and we
need to find impartial enforcement mechanisms
which can do both of these together.
Conclusion
I have argued that our aims should be twofold: (1) to
supplement individual human rights with minority
rights, recognising that the specific combination will
vary from country to country; and (2) to find new
domestic, regional or transnational mechanisms
which will hold governments accountable for re-
specting both human rights and minority rights.
If we manage to solve these two (enormous) tasks,
then I believe that the commitment to universal
human rights need not be culturally biased. Indeed, if
we resolve these issues satisfactorily, then the idea of
human rights can become what it was always in-
tended to be ± namely, a shield for the weak against
the abuse of political power, not a weapon of the
majority in subjugating minorities.
If the present arguments are at all valid, then it
46
suggests a number of new avenues for future research
± avenues which would depart dramatically from the
existing patterns of research and debate. At the
moment, wherever there is a conflict between `local
practices' and `transnational human rights standards',
commentators tend to locate the source of the con-
flict in the `culture' or `traditions' of the group, and
then look for ways in which this culture differs from
`western' culture. This tendency is exacerbated by
the rhetoric of a `politics of difference' or a `politics of
identity' which encourages groups to press their
demands in the language of respect for cultural
`difference'.
My suggestion, however, is that we should not
jump to the conclusion that cultural differences are
the real source of the problem. Rather, in each case
where a group is objecting to the domestic or trans-
national enforcement of human rights principles, we
should ask the following questions:
1. Has the majority society failed to recognise
legitimate minority rights? If so, has this cre-
ated a situation in which the implementation
of human rights standards contributes to the
unjust disempowerment of the minority? I
have discussed three contexts or issues where
human rights standards can exacerbate ethno-
cultural injustice if they are unaccompanied by
minority rights, but it would be interesting to
come up with a more systematic list of such
issues.
2. Is there any reason to think that the existing or
proposed judicial mechanisms for adjudicating
or enforcing human rights are biased against
the minority group? Have these judicial me-
chanisms treated minorities fairly historically?
Is the minority group fairly represented on the
judicial bodies? Were these judicial mechan-
isms consensually accepted by the minority
when it was incorporated into the country,
or is the imposition of these judicial mechan-
isms a denial of historical agreements or trea-
ties which protected the autonomy of the
group's own judicial institutions?
My guess is that in many cases where minority
groups object to transnational human rights stan-
 HUMAN RIGHTS AND ETHNOCULTURAL JUSTICE
dards, it will be for one of these reasons, rather any
inherent conflict between their traditional practices
and human rights standards. Where these problems
are addressed, I expect that many minority groups
will be more than willing to subscribe to human
rights standards.
I don't mean to deny the existence of illiberal or
anti-democratic practices within minority commu-
nities or non-western societies. But it's important to
note that at least in some cases the existence of such
practices is itself the consequence of some prior
ethnocultural injustice. That is, many minorities feel
compelled to restrict the liberties of their own mem-
bers because the larger society has denied their
Âaume has
legitimate minority rights. As Denise Re
noted, part of the `demonisation' of other cultures is
the assumption that these groups are naturally in-
clined to use coercion against their members. But in
so far as some groups seem regrettably willing to use
coercion to preserve group practices, this may be due,
not to any innate illiberalism, but to the fact that the
larger society has failed to respect their minority
rights. Unable to get justice from the larger society
in terms of protection for its lands and institutions,
the minority turns its attention to the only people it
does have some control over, namely its own mem-
32
bers.
This tendency does not justify the violation of the
human rights of group members, but it suggests that
before we criticise a minority for imposing such
Notes
1. I'd like to thank David Boucher for inviting me to
give the Rees Memorial Lecture and for his hospi-
tality during my stay in Wales. I'd also like to thank
David Schneiderman for encouraging me to write this
paper, and for his helpful comments. Thanks also to
John McGarry for his comments, and to Thomas
Nagel and Ronald Dworkin for inviting me to discuss
this paper in their Colloquium in Law, Philosophy
and Political Theory at New York University in
September 1997. (Editor's note: Professor Kymlicka's
chapter was delivered as the J. C. Rees Memorial
Lecture, commemorating the founding professor of
the Department of Political Theory and Govern-
ment, on 23 February 1998 at the University of
Wales Swansea. The opening remarks of that occa-
sion have been preserved.)
47
restrictions on its members, we should first make
sure we are respecting all of its legitimate minority
rights. In short, the current conflict between local
practices and transnational standards may not be the
result of a deep attachment to some long-standing
`tradition' in the local community, but rather the
(regretted) result of some new vulnerability which
has arisen from the denial of their minority rights.
To be sure, there will be some cases where mem-
bers of a group really do object to the very content of
the human rights standard on the grounds that it is
inconsistent with their cultural traditions. Even if we
solve the problem of minority rights and enforce-
ment mechanisms, we will still find some people
rejecting `western' notions of human rights. They
will say that restricting the liberty of women or
suppressing political dissent is part of their `tradition',
and that human rights theories reflect a biased `euro-
33
centric' and `individualistic' standard. These claims
may come from minority groups, or indeed from large
and powerful majority groups or governments, as in
Indonesia or China.
I do not want to enter into that debate and the
issues of cultural relativism which it raises. As I said
earlier, we are all familiar with that debate and I have
little to add to it. My aim, rather, is to insist that this
is not the only debate we need to have. On the
contrary, we may find that such conflicts are fewer
once we have properly dealt with the issues of
ethnocultural justice.
2. Rees (1985).
3. Rees (1971), p. x.
4. Claude (1955), p. 211.
5. There are now several attempts to define a theory of
ethnocultural justice in the literature, for example
Minow (1990), Young (1990), Kymlicka (1995). It
would take a separate paper to explain or defend any
particular one. But for the purposes of this paper we
can use a minimalist definition of ethnocultural justice
as the absence of relations of oppression and humilia-
tion between different ethnocultural groups. A more
robust conception of ethnocultural justice could be
developed by asking what terms of coexistence would
be freely consented to by the members of different
ethnocultural groups in a Habermasian/Rawlsian set-
ting where inequalities in bargaining power have been
 WILL KYMLICKA
neutralised. For example, a Rawlsian approach to
ethnocultural justice would ask what terms of coex-
istence would be agreed to by people behind a `veil of
ignorance', who didn't know whether they were going
to be born into a majority or minority ethnocultural
group. Such a Rawlsian approach is likely to produce a
more demanding conception of ethnocultural justice
than the mere absence of oppression and humiliation,
but the main claim of this chapter is that human rights
are insufficient even to ensure this minimal compo-
nent of ethnocultural justice.
6. See McGarry (1998).
7. See Penz (1992) and (1993). I discuss these cases in
Kymlicka (1996).
8. MacDonald, quoted in Stanley (1961), p. 95.
9. Gurr (1993).
10. It would not be hypocritical to criticise minority
demands to limit in-migration if one also criticised
state policies to limit immigration ± that is, if one
defended a policy of open borders. But such a policy
has virtually no public support, and is certainly not
endorsed by most of the people who criticise minority
demands. However, this raises an important limita-
tion on my argument. I am discussing what justice
requires for minorities in the world as we know it ±
that is, a world of nation-states which retain signifi-
cant control over issues of migration, internal poli-
tical structures and language policies. One could
(with difficulty) imagine a very different world ± a
world without states, or with just one world govern-
ment. The rights of minorities would clearly
different in such a hypothetical world, since the
power of majorities would be dramatically reduced,
including their ability to impose relations of oppres-
sion and humiliation. My focus, however, is on what
ethnocultural justice requires in our world. See also
note 20.
11. Cf. Kymlicka (1998). In cases where national mino-
rities are not territorially concentrated different me-
chanisms of disempowerment are often invoked.
During the period of devolved rule in Northern Ire-
land (1920±72), for example, the Catholics were
disempowered not so much by the gerrymandering
of boundaries (although this occurred), but by the
adoption of an electoral system (with single-member
constituencies and plurality rules) designed to ensure
unity within the Protestant majority while ensuring an
ineffective Catholic opposition. This is another ex-
ample of how a rhetorical commitment to democracy
and human rights can coexist alongside the oppression
of a national minority.
12. A related example is the law which existed in Canada
prior to 1960 which granted Indians the vote only if
they renounced their Indian status, and so abandoned
any claim to Aboriginal political or cultural rights. In
order to gain a vote in the Canadian political process
(a process they had no real hope of influencing), they
had to relinquish any claims to participate in long-
standing Aboriginal processes of self-government.
This transparent attempt to undermine Aboriginal
political institutions was justified in the name of
promoting `democracy'.
13. Avigail Eisenberg, `Individualism and Collectivism in
the Politics of Canada's North', paper presented at the
annual meeting of the Canadian Political Science
Association, May 1995. A revised version of this paper
will appear in Painting the Maple: Essays on Race,
Gender and the Construction of Canada, edited by Joan
Anderson, Avigail Eisenberg, Sherril Grace and Ver-
onica Strong-Boag (Vancouver: UBC Press, forthcom-
ing).
14. On the necessity of extensive language rights for the
survival and flourishing of linguistic minorities, see
Kymlicka (1995), chapter 6.
15. The view that language rights are not part of human
rights was explicitly affirmed by the Canadian Su-
preme Court in MacDonald v. City of Montreal [1986] 1
SCR 460; Socie
 te
 des acadiens du Nouveau-Brunswick v.
Minority Language School Board no. 50 [1986] 1 SCR
549.
16. For a comprehensive review of the current status of
language rights in international human rights law, see
de Varennes (1996).
17. Low enrolment is then often (perversely) cited by
majority politicians as evidence that most members of
the minority are not interested in preserving their
be language and culture, and that it is only a few ex-
tremists in the minority group who are the cause of
ethnic conflict.
18. This is called the `territorial imperative', and the trend
towards territorial concentration of language groups is
a widely-noted phenomenon in multilingual western
countries. For a more general theoretical account of
the `territorial imperative' in multilingual societies, see
Laponce (1987) and (1993).
19. This is obviously the rationale given for requiring
immigrants to Quebec to send their children to
French-language schools.
20. Here again, it would not be hypocritical to criticise
minority demands regarding self-government rights
and language rights if one applied the same standard
to majorities. For example, one could imagine letting
the United Nations determine the boundaries and
language policies of each state. Imagine that the
UN, in a free and democratic vote, decided to merge
all countries in the Americas (North, South and
Central) into a single Spanish-speaking state. If the
anglophone majority in Canada or the United States
would accept such a decision ± that is, if they were
willing to abandon their own self-government powers
and language rights ± then it would not be hypocritical
to criticise the demands of Francophone, Hispanic or
48
 HUMAN RIGHTS AND ETHNOCULTURAL JUSTICE
Aboriginal minorities in North America. But I don't
know any English-speaking Canadians or Americans
who would agree to amalgamate into a single Spanish-
speaking state, even if this merger was supported by
most countries in the Americas (and/or by most people
living in the Americas). In reality, the anglophone
majorities in both the United States and Canada
zealously guard their right to live in a state where
they form a majority, and their right to have English
recognised as a language of public life. This defence of
the boundaries and linguistic policies of existing na-
tion-states is as `collectivist' as the demands of mino-
rities for protection of their self-government and
language rights.
21. And, in a different way, in the Communist bloc. See
Walker Connor's (1984) account of how Communist
leaders dealt with the issues of settlement policy,
gerrymandering and linguistic policies, all of which
were key policy tools in the Communist approach to
national minorities.
22. For the ubiquity of this process, see Gellner (1983).
23. See the related analysis in de Sousa Santos (1996).
He argues that attention to the claims of indigenous
peoples and ethnic minorities can help develop a
new `non-hegemonic' conception of human rights
which would retain its commitment to protecting
the weak and vulnerable without serving
instrument of western colonialism (for example on
p. 353).
24. As a general rule, we should be wary about the claims
of elite members of a group to speak authoritatively
about the group's `traditions'. Some individuals may
claim to speak for the group as a whole, and may say
that the group is united against the imposition of
`alien' ideas of human rights. But in reality, these
people may simply be protecting their privileged
position from internal challenges to their interpreta-
tion of the group's culture and traditions. In other
words, debates over the legitimacy of human rights
should not necessarily be seen as debates
whether to subordinate local cultural traditions to
transnational human rights standards, although this is
how conservative members of the group may put it.
Instead, debates over human rights are often debates
over who within the community should have the
authority to influence or determine the interpretation
of the community's traditions and culture. When
individual members of the group demand their `hu-
man rights', they often do so in order to be able to
participate in the community's process of interpreting
its traditions.
25. For a typology, see my `Ethnocultural Minority
Groups', in Encyclopedia of Applied Ethics, edited by
Ruth Chadwick, (San Diego, CA: Academic Press,
forthcoming).
26. For an acute criticism of existing minority rights
declarations, see Donald Horowitz, `Self-Determina-
tion: Politics, Philosophy and Law', in Kymlicka and
Shapiro (eds) (1997), pp. 421±63.
27. It is, at best, an imperfect beginning, in large part due
to the Plenary Power which Congress arbitrarily asserts
over Indian tribes. See Kronowitz (1987).
28. Indeed, tribal councils in the United States have
historically been exempted from having to comply
with the federal Bill of Rights, and their internal
decisions have not been subject to Supreme Court
review. Various efforts have been made by federal
legislators to change this, most recently the 1968
Indian Civil Rights Act, which was passed by Con-
gress despite vociferous opposition from most Indian
groups. American Indian groups remain strongly op-
posed to the 1968 Act, just as First Nations in Canada
have argued that their self-governing band councils
should not be subject to judicial review by the Ca-
nadian Supreme Court under the Canadian Charter of
Rights and Freedoms. They do not want their mem-
bers to be able to challenge band decisions in the
courts of the mainstream society.
29. See also the analysis in David Schneiderman, `Human
Rights, Fundamental Differences? Multiple Charters
in a Partnership Frame', in Beyond the Impasse, edited
by Guy Laforest and Roger Gibbins (Montreal, In-
as an stitute for Research on Public Policy, forthcoming).
30. Joseph Carens, `Citizenship and Aboriginal Self-Gov-
ernment' (paper prepared for the Royal Commission
on Aboriginal Peoples, Ottawa, 1994).
31. To be sure, some Indian tribal constitutions are not
fully liberal or democratic, and so are inadequate from
a human rights point of view, but they are forms of
constitutional government, and so should not be
equated with mob rule or despotism. As Graham
Walker notes, it is a mistake to conflate the ideas
of liberalism and constitutionalism. There is a genuine
category of non-liberal constitutionalism which pro-
vides meaningful checks on political authority and
over preserves the basic elements of natural justice, and
which thereby helps ensure that governments main-
tain their legitimacy in the eyes of their subjects. See
Walker, `The Idea of Non-Liberal Constitutionalism',
in Kymlicka and Shapiro (eds.) (1997), pp. 154±84.
Âaume (1995).
32. Re
33. As I said earlier, I do not think that the substantive
interests protected by human rights doctrines are
either individualistic or Eurocentric. However, it
may well be that to talk of these interests in terms
of `rights' is a specifically European invention which
does not fit comfortably with the discourse or self-
understandings of many cultures. I don't think we
should get hung up on `rights talk'. What matters,
morally speaking, is that people's substantive interests
in life and liberty are protected, but we should be
open-minded about what institutional mechanisms
49
 WILL KYMLICKA

best provide this protection. There is no reason to of the language of rights as Eurocentric, see Turpel
assume that the best way to reliably protect people's (1990). On ways to protect the substantive human
basic interests will always take the form of a judicially interests underlying human rights without using the
enforceable constitutional list of `rights'. For a critique language of `rights', see Pogge (1995).

50
 4
LIBERALISM AND CITIZENSHIP
Andrew Vincent
The central argument of this chapter is that liberal-
ism has a strong conceptual relationship with `citi-
zenship' yet citizenship, in turn, has an ambiguous
conceptual relationship with nationalism. `Liberal
citizenship' is thus at the interstice between two
potentially contradictory forces.
There are two theses implicit in the issue of
liberalism and citizenship, which need to be exam-
ined. The first concerns the cosmopolitan implica-
tion of liberal citizenship; the second concerns the
criteria for citizenship in liberal states.
First, intuitively, we tend to see a close conceptual
relation between liberalism and citizenship. This
relation has deep philosophical roots, bound up with
ideas of self-determination, autonomy and individu-
alism. The citizen is conceived as markedly different
from the subject by dint of possessing, intrinsically,
agency. Citizenship denotes choice, an opportunity
or space to act. It thus links conceptually with
freedom and agency. This intrinsic connection also
combines with other fundamental ideas in liberal
thought ± conscience, rights, equality, free speech
and so forth. Thus, there is a more than fortuitous
connection between citizenship and liberalism. In
the last two centuries, citizenship has been integrally
part of a web of liberal values and beliefs.
Second, citizenship has a complex and diverse
etymological history which will be analysed here.
In light of this chapter's critical focus upon the link
with liberalism, it might nevertheless be claimed that
since liberalism, above all other social and political
theories, embodies a belief in moral and political
individualism and agency, and citizenship is intrin-
51
sically committed to agency and individualism, then
liberalism is the most succinct modern bearer of
citizenship claims. This argument is problematic,
as we shall see. If, however, we can accept the
compound `liberal citizenship', then there are further
implications which follow from the intrinsic logic of
both citizenship and liberal beliefs.
These can be summed up in the concept of
`cosmopolitanism'. The argument for this claim is
as follows: the core values of liberal citizenship ±
autonomy, agency, individualism, liberty, rights,
equality, justice and so forth ± are universalising
values. They contain an intrinsic logic which shows
that when dealing with the justice, rights or free-
doms of human beings there are no rational grounds
to stop at state boundaries. In other words, to think
within the ontology of liberal citizenship is to think
oneself into cosmopolitanism. This logic would
account for the stress often laid on human rights,
international justice, cosmopolitan democracy and
the like by many liberal theories. Boundaries be-
come morally arbitrary in this context, since we are
dealing with humanity as a single moral category.
This consideration leads to the concept of `inter-
national' or `global citizenship'. Now there are
many problems here, not least the legal, moral,
political and social implications of this concept.
However, my central claim is that it is `concep-
tually' natural for us (`natural' in the sense that it
appears to be an unforced implication of the con-
cepts we use) to read cosmopolitanism into liberal
citizenship. The impact of Kant's thought upon the
modern western intellectual psyche has been pro-
 ANDREW VINCENT
found here. It is, therefore, a short step from being a
good liberal citizen to being a good member of
Amnesty International or being a proponent of
international human rights.
My own response to this is deeply ambivalent. The
core intuition concerning cosmopolitanism is im-
mensely compelling, but it also contains some quite
fundamental problems. The deepest problem relates
to the link between nationality and citizenship.
My discussion will first flesh out the cosmopolitan
intuition. Secondly, it will examine the ambiguity of
liberalism itself. Thirdly, it will review the diverse
senses of citizenship and the ambiguous relation with
nationalism. Overall, the chapter draws attention to
a complex of ideas which form an unresolved subtext
to much current, legal, moral and political thought.
Hence it seeks to illuminate the dimensions of the
problem and indicate an important topic for further
research, one which liberals have tended to overlook
but which exposes considerable tensions at the heart
of their political thinking.
A Cosmopolitan Intuition
As suggested, the central premise of this argument is
that citizenship embodies a strong conception of
agency which is, in turn, a central value for liberal-
ism. Citizenship ± although a concept with a broader
ambit ± can therefore be seen as a very important
element of the liberal perspective, a characteristic
liberal motif. The `citizen' can literally (and for-
mally) denote the individual person (agent) with
basic legal or moral rights to life, freedom, con-
science and association.
However, the central purpose of this section is to
substantiate the argument that liberalism embodies,
in turn, an underlying cosmopolitan impetus and
that the logic of liberal citizenship would appear to
be directed towards a global perspective. One of the
first western philosophies to expound this cosmopo-
litan perspective was Stoicism. It involved a quite
conscious rejection of the singularity and bounded
citizenship of the Greek city-state. The citizen/bar-
barian distinction, common to such Hellenic
thought, was anathema to Stoics. As Marcus Aur-
elius put it, `my city and my country, so far as I am
Antonius, is Rome; but so far as I am a man, it is the
52
1
universe.' This argument alone made Stoicism at-
tractive to Christian thinkers like St Augustine. For
Stoics, we dwell both in the community of our birth
and the community of human argument, reason and
aspiration. Our decisive moral values come from the
latter, which is a moral, not a legal community. This
does not deny the separateness of peoples, but it
argues that such `separateness' is relatively unimpor-
tant in comparison with the wider moral community.
Thus, world citizenship should be our ultimate moral
focus.
The cosmopolitan view, developed under the
aegis of Christianised natural law thinking, surfaces
again most significantly (for the last century) in the
2
work of Kant. For a modern Kantian, like Thomas
Pogge, most cosmopolitan arguments contain, firstly,
a commitment to individualism as against commu-
nities or tribes; secondly, a belief in universalism,
namely that all humans are valued equally; and,
thirdly, a sense of generality, that is that cosmopo-
litan concerns are not limited to specific regimes.
Persons are of concern to everyone, wherever they
are. Our fellow countrymen are no more or less
important than anyone else in the world. Pogge also
distinguishes types of cosmopolitanism. `Legal' cos-
mopolitanism is committed to a concrete juridical or
governmental order under which people will have
equal basic rights, as in international courts on hu-
man rights or international codes and covenants.
Most commentators see the central problem here of
jurisdiction, namely, will the courts be able to en-
force their rulings? `Moral' cosmopolitanism, on the
other hand, holds that all persons stand in certain
moral relations to each other ± mutual concern or
respect is incumbent upon us. Every human being is
3
therefore an ultimate unit of moral concern. Moral
cosmopolitanism can, however, vary in its content
with different moral positions. It can thus require
simply leaving individuals alone and not harming
them (which might equate with a more passive sense
of civil citizenship), or taking a definite responsi-
bility for others, even distant individuals (which
might entail strong civic duties).
The basic `logic of citizenship' argument would be
this: citizenship, historically, has been used within
states to equalise, generalise and regularise treat-
ment, thus preventing exclusionary practices within
 LIBERALISM AND CITIZENSHIP
states. If citizenship, within the state, is used against
exclusionary practices, why not extend it to `exclu-
sionary practices' beyond state boundaries? An ex-
clusionary practice remains an exclusionary practice
wherever it appears. What is so significant about
state boundaries?
Citizenship is thus embroiled in the logic of a
universalising ethics which does not necessarily have
to stop at the national boundary. Rights before the
law, rights of participation and freedom, justice, the
requirement for consent, and duties to promote a
social good (all implicit within citizenship), are all
potentially universalisable. Thus, for Andrew Link-
later, recent theorists like Charles Beitz and Ju
È rgen
Habermas who appear to be engaging in this kind of
reasoning `universalise ideas about consent and dia-
logue which are intrinsic to citizenship in the domes-
tic domain' and `enlarge the meaning of citizenship by
conferring rights of participation on every member of
4 7
the species.' Moral universalism thus arises automa-
tically out of domestic liberal citizenship.
One comparatively recent example of the above
argument is deployed by Martha Nussbaum. Using a
Stoic analogy, she contends that we should think of
ourselves as surrounded by concentric circles: the
self, the family, the neighbourhood, the city, the
country and humanity. We cannot ignore our lo-
cality, but we must work to make all humans part of
our community, drawing all the circles to the centre.
For Nussbaum, we should therefore `recognise
humanity wherever it occurs, and give its funda-
mental ingredients, reason and moral capacity, our
5
first allegiance as respect.' She puts the point well
when she states that
I believe that . . . [the] emphasis on patriotic
pride is both morally dangerous and, ultimately,
subversive of some of the worthy goals patriotism
sets out to serve . . . These goals . . . would be
better served by an ideal that is any case more
adequate to our situation in the contemporary
world, namely the very old idea of the cosmopo-
litan, the person whose allegiance is to the world-
6
wide community.
Even good nationalism ultimately subverts the values
it upholds.
53
Nussbaum therefore contends that we should
automatically educate ourselves and our children
in a cosmopolitan mode of thought. In so doing,
first, we will learn more about ourselves. Education
which focuses within national boundaries en-
courages irrationality. To be ignorant of humanity
is to be ignorant about ourselves. Second, we can
only make headway solving international problems if
we acknowledge global issues. The air that citizens
breathe and the water they drink, for example, are
not confined by national boundaries. Third, we
should be troubled by the fact that our high living
standards are not more widely available. We should
therefore train children in a basic universalising
`Kantian awareness', otherwise we will nurture only
hypocrites. Nussbaum remarks that if `we do believe
that all human beings are created equal and endowed
with certain inalienable rights, we are morally re-
quired to think about what that conception requires
us to do.' We should work to acquire knowledge
that will enable us to consider these rights seriously.
Fourth, nationalist writers often seek to encourage
nationals to join hands across racial or economic
divisions; why, asks Nussbaum, stop at the arbitrary
boundary of the nation ± `what is it about the
national boundary that magically converts people
towards whom we are both incurious and indifferent
into people to whom we have duties of mutual
8
respect?' Multicultural respect within a nation must
lead to wider world respect. The `respect argument',
in nationalist writers, must carry us in this direction.
These arguments of Nussbaum's clearly illustrate the
logic-of-citizenship argument, which Linklater as-
sociates also with Beitz, Pogge and Habermas.
Some of these cosmopolitan ideas have been
translated into deeply practical concerns. The liberal
rights of the citizen to freedom, conscience, proce-
dural justice, speech, or to dissent, can also be seen as
universal `human' rights. Although there are pro-
blems with ratification and application, human
rights have also been codified and translated into
positive legal rights. So far the main obstacle to the
full realisation of such rights has been persuading
states to agree and work with the United Nations
and International Courts. The global citizen is there-
fore one who simply acknowledges universal human
rights and believes, either on a personal level or on
 ANDREW VINCENT
the institutional level of international courts and
conventions, that he or she should accept the duties
correlative to such rights.
There is a major problem here, namely that state-
based citizens have a rich panoply of well-established
positive rights, whereas the global citizen's rights are,
in comparison, thin. One way out of this dilemma for
some theorists would be to argue for an extension of
international legal organisations, expanded interna-
tional courts of human rights and at the most ex-
treme end, an elected parliament at the United
Nations. In the latter scenario, states which recog-
nised cosmopolitan citizenship would provide full
diplomatic and economic support to the UN, pro-
mote arms control, uphold human rights, and pro-
vide generous foreign aid. These ideas are
speculative, although not overly so. The fact that
we have international commissions, like that on
Global Governance with their report Our Global
Neighbourhood, indicates that many take such ideas
very seriously. As such the logic of global citizenship
certainly shows little sign of diminishing in world
politics.
The discussion now turns to three crucial, over-
lapping ambiguities, which are often glossed over, in
the above arguments. These are, first, the ambiguity
of liberalism itself; second, the ambiguity of citizen-
ship qua liberalism; and thirdly, the ambiguous rela-
tion of nationalism, liberalism and citizenship. By
explaining and analysing each, the nature of liberal-
ism's problem here will become more fully apparent.
Liberalism
Is liberalism a clear doctrine such that we could
postulate a particular understanding of a potentially
cosmopolitan citizenship, as outlined in the previous
section? The quickest response is `no': differing
conceptions of liberalism provide divergent re-
sponses to the question. There have been many
typologies of liberalism and so, to survey such re-
sponses, the following argument draws a distinction
between three types of liberalism: classical liberalism,
social liberalism and cultural liberalism.
First, the classical view (a predominantly proce-
dural view, certainly in its modern guise), which
insists that individual rights must always come first
54
9
and must take precedence over collective goals.
Michael Walzer characterises this procedural liberal-
ism as
committed in the strongest possible way to in-
dividual rights and, almost as a deduction from
this, to a rigorously neutral state, that is, a state
without cultural or religious projects or, indeed,
any sorts of collective goals beyond freedom and
10
physical security.'
This is a view that one finds, according to Charles
Taylor, in writers like Rawls, Dworkin and Acker-
11
man. This conception of liberal society has no
substantive view about the ends of human life.
Rather, society is united behind an idea of equal
respect. For Taylor, the roots to this procedural view
are very deep. Kant is probably the single most
important figure articulating this perspective. Hu-
man dignity focuses on autonomy and the ability of
the individual to determine their own notion of the
good life. Taylor has his own thoughts on the
problems that this notion of liberalism has caused
in the Canadian situation. For him, procedural
liberalism enshrines a politics of equal respect which
is hostile or indifferent to difference, because it
insists on the uniform application of rights and is
thus suspicious of any collective goal.
Second, social or `new' liberalism appeared in
Britain in the 1880s and 1890s, although variants
also developed in America with John Dewey, in Italy
with writers such as Guido de Ruggiero, and in
Germany in the writings of Albert Lange and Karl
Ènder. Social liberal thought was committed to
Vorla
collective welfare goals pursued through the state. It
was thus statist, but not overtly nationalist. Essen-
tially, social liberals were reacting to certain themes
present within classical liberalism, notions like ato-
mised individualism, the negative conception of
liberty, the radically free market economy and mini-
mal state theory. They wished to replace these with a
more socialised, holistic and developmental under-
standing of the individual, a more `positive' concep-
tion of liberty linked to notions like self-realisation
and self-development, a conception of a mixed
economy and a more responsive, collectivised and
ethical conception of the state. State intervention
 LIBERALISM AND CITIZENSHIP
was premised upon the common good and the full
realisation of human personalities. It implied a de-
licate balance between self-government and corpo-
rate government action. The new liberals were, in
effect, developing the embryonic form of the Wol-
fartstaat. Their arguments were rooted in evolution-
ary theory, social utilitarianism and philosophical
12
idealism.
Thirdly, cultural liberalism, which developed in
reaction particularly to procedural liberalism. Cul-
tural liberals promote the conception of collective
cultural goods, such as the French language for the
Quebeckers. This view of liberalism has been devel-
oped by Will Kymlicka, among others and it claims
that classical, procedural liberalism overemphasises
individual rights against collective goals. For classical
liberals, communities (and cultural goals) have no
moral existence. For Kymlicka, however, group and
individual rights are not necessarily opposed. He thus
tries to defend group rights within the framework of
cultural liberalism. The crux of his move is to suggest
that when we exercise our liberty it `is determined by
13
our cultural heritage.' Agency requires culture.
Culture provides the content of choice, its norms
and historical heritage. Thus, `the primary good
being recognised is the cultural community as a
14
context of choice.' Contrary to his perception of
communitarians, Kymlicka thinks that we can ab-
stract ourselves from ways of life and criticise them.
We can be thus `unencumbered'. We do not just
discover views and we are not haplessly trapped by
them. Cultural liberalism, therefore, distinguishes
fundamental liberties from privileges and immu-
nities. Provided the exponents of collective goals
and cultural goods respect the fundamental rights of
minorities and individuals, they can still be consid-
ered liberal. This liberalism is seen as more sensitive
to `cultural survivals' and, for Taylor, is the most
relevant model for contemporary porous and cultu-
rally diverse societies. Liberalism has to do more than
let differences exist. It has to recognise the `worth' of
different cultures.
Some very general claims can be made concerning
liberalism and cosmopolitanism from this brief sur-
vey. First, cultural liberalism is the least politically
distinctive of the three in that it tells us little about
the nature of the state or welfare. Thus, it might or
55
might not share beliefs with social liberals about, say,
the market. It only affirms that culture is immensely
important for choice and that groups, as well as
individuals, have rights. There is therefore nothing
inherently to stop cultural liberalism from being
assimilated with either an older variant of classical
liberalism (one less overtly procedural, like J. S.
Mill's liberalism) or, alternatively, social liberalism.
Cultural liberalism might thus legitimately share a
classical liberal antipathy to distributive justice and
the welfare state ± especially if the existent cultures
embody that antipathy. It therefore seems to have no
specific economic or political content.
Secondly, cultural liberalism arrests, but does not
undermine, the argument for citizenship leading to
cosmopolitanism. Because cultural liberalism tends
to be more sympathetic to liberal nationalism, its
universalising appeal is stunted. Thirdly, because
social liberalism is statist and centralist in content,
it is ideologically antipathetic to classical liberalism.
It would also have little sympathy with cultural goods
and group rights. Further and crucially, because it
sees citizenship in the context of the state, it would
also inhibit the force of any cosmopolitan argument.
Fourthly, classical liberalism is intrinsically critical of
the statism of social liberalism. It also feels deep
unease with the cultural goods and group rights being
pursued at the expense of individual rights and
procedural justice.
Thus classical liberalism is, oddly, most susceptible
to the cosmopolitan thesis outlined in the previous
section. It is, literally, much easier to universalise the
generic negative individual rights of classical liberal-
ism ± life, liberty, property. Yet, at the same time,
most recent postcolonial writers see classical liberal-
ism as the expression, par excellence, of a particular
historically oppressive community. In this latter
scenario, minimal generic rights are no longer con-
sidered through the lens of cosmopolitanism, but
rather through imperialism. This critical argument
has, paradoxically, been applied to the treatment of
internal minorities, even within the USA, by Iris
Marion Young. Rights-talk and notions of neutrality
are conceived of as modes of group and cultural
domination.
 ANDREW VINCENT
Citizenship
The upshot of the previous section is the suggestion
that, given the profoundly contestable character of
liberalism, the initial intuition concerning cosmo-
politan citizenship looks less secure. Indeed, some
senses of liberalism might be at odds with cosmopo-
litanism. The same observation also applies to citi-
zenship itself, as this section will now demonstrate by
surveying examples of both its theory and empirical
practice.
On the theoretical level, there are certain con-
ceptual parameters which reveal the dimensions of
the concept. These parameters are active and passive
notions of the citizen, the character of rights and
duties and their correlativity. The origin of the terms
`active' and `passive' is debatable, but they can be
 Sieye
found in the writings of the Abbe Ás. (Sieye
Ás's
use is not quite my present usage, although there are
sufficient parallels to warrant discussion here.) For
Ás, passive citizenship referred essentially to the
Sieye
natural and civil rights of individuals. Such rights
concern the protection of all persons equally in their
Ás, despite the fact that
liberty and property. For Sieye
all persons within a state possess passive rights, not
all can take part in the `formation of public power'.
Passive citizenship is thus available to all. Active
citizens, on the other hand, are those who `contri-
bute to the public establishment' and can be likened
to `true shareholders in the great social enterprise'.
At the time, active citizenship excluded women,
foreigners and children, who were all regarded as
`passive citizens'. However, within the class or group
of `active citizens' no one has more rights than
anyone else.
It is worth noting here that active citizens are not
Ás primarily in the context of rights.
discussed by Sieye
Active citizenship, although it incorporated concep-
tions of political voting rights, also included a sense
of higher vocation. Taking part in the `great public
Ás in terms of the
enterprises' is seen more by Sieye
performance of public duties. He thus remarks that,
`it is a major error to envisage the exercise of public
power as a right; it is . . . a duty. The officers of the
nation are distinguished from other citizens only in
15
having more duties.'
In terms of the present discussion I wish to pick up
56
Ás's use of the term `active citizen'.
on aspects of Sieye
However, the active sense of the citizen also refers
(in my use) to a more rigorous conception of political
involvement, which was revealed at certain mo-
ments during the French revolution, and harks back
to an older civic tradition. In my usage, the active
conception conceives of the citizen as politically
involved and identified with the wider community.
Such an idea can be found ± in differing formats ± in
the doctrines of Aristotelianism, some civic repub-
licanism and the later Rousseauist and Hegelian
traditions. It can also be found implicitly in some
recent expressions in communitarian thought. The
active/passive divide bears some conceptual parallels
with the distinction, made by Benjamin Constant,
between ancient and modern liberty. Another way of
formulating it is in terms of `civic citizenship' and
`civil citizenship'. The civic idea corresponds with
the activist vision and the civil with the more passive
tradition.
The category of civic or active citizenship is the
oldest view. The good of the individual citizen was
seen as closely tied to the common good of the polis.
It implied, say, for Aristotle, minimally, membership
of a city-state, the possession of certain claims and
duties, the duty to participate in the adjudication
processes of the polis (at least in Athens) and, finally,
an inner capacity for rational virtue, entailing the
internalisation of communal norms ± an art of both
ruling and being ruled. The themes of political self-
discipline, public spirit, a conception of the common
good, unease with private gain and self-interest, and
a belief in simplicity and piety of life were reasserted
in the later civic republican tradition. Again, the
active citizen was one who (mythically or not) gladly
and intelligently participated in the promotion of
the public good. The civic world of the small city-
state was more overtly small-scale and defensive in its
stance. Its citizenry, despite being mobilised to a
common task were, nonetheless, premised upon a
large body of non-citizen slaves and foreign workers.
The spectre of civic citizenship developed once more
with the rise of nationalism in the nineteenth cen-
tury. Its ethos, though, was `more creative and
protective, stressing liberation rather than de-
16
fence.'
However it is couched, one major criticism of
 LIBERALISM AND CITIZENSHIP
`civic citizenship' is that, at its extreme, the citizen
will be asked to participate continuously in every
aspect of political life. The public realm would
totally absorb the private realm of the individual.
Liberals, from Constant to Isaiah Berlin, have been
profoundly worried by such arguments as potentially
undermining the value of individual liberty and
privacy. Ralf Dahrendorf has referred to this com-
17
munal conception as the `total citizen'. The total
citizen and the total state would be seen, in this
argument, as two sides of the same coin.
The civil citizen, on the other hand, is a product of
the era of developing individualism and embryonic
liberal thought. The citizen is understood as separate
and, at least partially preformed, with desires, inter-
ests and often basic (natural) rights. The function of
a constitutional legal order is to protect and uphold
these fundamental interests or rights. Citizenship is
thus conceived more negatively in terms of the legal
protection of pre-existing rights to life, liberty and
property. The private moralities of the individual are
distinct from the formal but minimal public ethics of
citizenship. For classical liberal theorists, although
individuals may have lost some of the benefits of
close communal life, nonetheless they gained from
the privacy, modern liberty and new-found prosper-
ity of commercial liberal society. Later nineteenth-
and twentieth-century versions of this understanding
of citizenship extends further into features like po-
litical representation.
Civil citizenship, in my usage, is thus more con-
ventionally associated with classical liberalism, im-
plying negative rights (understood largely as `side-
constraints') to protection of one's person, property
and liberty. The classical liberal view of citizenship
thus favours limited political rule, a framework of
laws and minimal welfare. Markets are the preferred
technique for allocation of resources. Essentially, this
is a more procedural view of politics. Politics does
not pursue any particular social or moral goals. The
consequences of this for citizenship are that indivi-
duals secure maximum negative liberty. They are left
alone and free from coercion. The good citizen
therefore upholds the rule of law and equal liberty
for all. This is the conventional liberal Rechtstaat.
However, this is not a purely passive conception
within classical liberal thought. In my own reading,
57
it also incorporates political rights to vote or stand
for office. The argument about political rights de-
pends upon how significant one regards them; it is
arguable that such rights can still be viewed as quasi-
passive notions. As one recent writer has noted, `the
occasional visit to the polling booth has seemed to
many a pathetically inadequate definition of citizen-
18
ship.' Some writers, like Schumpeter, have of
course made a virtue of this quasi-passivity. How-
ever, the intuition to grasp here is the point that
political citizenship, as it has been conventionally
understood, can be viewed as constrained, passive
and perfectly in accord with classical liberal theory.
In sum, citizenship in classical liberal thought is
generally confined to a relatively narrow field, being
defined in terms of constrained political rights to
political involvement, combined with legal or civil
rights of persons, their liberty and their property,
which prohibit mutual coercion. Classical liberal
thought on citizenship generally excludes rights
and entitlements to economic and social resources.
A well-established criticism of civil citizenship is
that it is too minimalist. It ignores the socially and
historically embedded (or encumbered) character of
individuals and abandons any clear sense of perfec-
tionism, community and nationality. This is a char-
acteristic criticism made of deontological and
consequentialist liberal individualism, to the present
day, by communitarians. The citizen of deontological
liberal individualism could be described as a `mini-
mal citizen', essentially being the other side of the
minimal state.
Turning now to the parameters of rights and
duties, the older tradition of civic or active citizen-
ship did not have such a rich vocabulary of rights at
its disposal. Aristotelian-influenced proponents of
active citizenship tended to speak more in terms
of the duties of the citizen or what was due to the
citizen. The assertion of the importance for citizens
to perform their civic duty was not therefore neces-
sarily premised on any correlative right, rather what
was due to and from the common good. Later ex-
ponents of a more active vision of citizenship, like T.
H. Green, transformed this form of argument into
the vocabulary of rights, although, characteristically,
it was usually the vocabulary of social (ethicised)
rights, not natural or human rights. Such socio-legal
 ANDREW VINCENT
rights were also usually conceptually tied ± in the
case of Idealist thinkers ± to a conception of the
common good.
The civic citizenship tradition also embodied a
strong correlativity thesis. Rights were premised on
the common good. Such rights implied that the
citizen had correlative duties to that good. Citizen-
ship was not just a juridical category or an indication
of nationality. It implied rather the consciousness of
the ends of human life as embodied in the institu-
tional forms of the public life. It was thus an ethical
disposition, where the individual developed to a
level of self-consciousness and ethical awareness
inclusive enough to be identified with the public
sphere of the whole community.
The conception of definite individualistic rights
was more characteristic of civil citizenship. Rights
expressed, ontologically, the separate or independent
identity of individuals. They were claims asserted
against other individuals, groups or states. The more
general of such rights ± natural and human rights ±
were fundamental claims to life, liberty, happiness,
property, speech, opinion, association and so forth.
Such rights correlated loosely with duties of forbear-
ance or non-interference, usually with the proviso
that such rights did not overtly injure or harm others.
They did not, characteristically, demand positive
public action by any individual ± except in the duty
to discharge one's debts and fulfil contracts. They
were usually more negatively protective rights, im-
plying passive and relatively costless duties. This
latter conception of rights is, to some extent, en-
capsulated in the categories of civil and political
citizenship, discussed by T. H. Marshall in his locus
classicus account of citizenship, Citizenship and Social
Class. For Marshall, civil citizenship, which he saw
developing from the early eighteenth century, im-
plied a comprehensive equality of rights to civil
freedoms, which might broadly be called `constitu-
tional' freedoms. Political citizenship, which devel-
oped gradually over the nineteenth century,
concerned the extension of suffrage rights, namely
the right to participate in the political process which
determines the condition of one's life.
The problematic category in the Marshall schema
is `social citizenship'. It was still deeply individualis-
tic and rights-orientated, but it also embodied col-
58
lective goals and claims to collective resources.
Social citizenship, as it developed in the twentieth
century, was more characteristic of a different species
of liberalism ± social liberalism. In many ways social
liberalism was a hybrid of the civic and civil forms of
citizenship. The social liberal case is essentially that
individual citizens are morally justified in claiming a
particular economic status and a right to resources,
via the welfare state and social justice. The social
liberal theorist therefore wants to extend arguments
beyond civil and political rights to social rights.
Marshall recognised that this created immense pro-
blems and strains for a liberal capitalist society ±
problems which he never claimed to resolve.
The more empirical approach to the citizenship
issue raises a different range of problems. First,
citizenship ± despite all that has been said earlier
± empirically is a technique of exclusion. The state is
a state of a bounded citizenry. It is about closure.
Citizenship creates the concept of a foreigner, an
alien. Second, the character of citizenship changes
between state traditions, even liberal-democratic
states. This change is one that crucially cross-cuts
the civic/civil distinction outlined above.
The two best-known ways of conceiving the citi-
zen has been the jus soli and the jus sanguinis con-
ceptions. Jus sanguinis associates citizenship with
direct descent and blood relations. If one can trace
direct blood or birthright descent then citizenship is
automatic. In jus soli citizenship follows prolonged
residence or birth within a territory. There are
usually combinations of these in most countries.
Thus, in the UK since 1981, where one parent must
be lawfully resident and settled, then birth in the
territory gives citizenship. France and Germany offer
purer examples of the two conceptions.
Two phenomena affected the German case ± the
Prussian state and romanticism. The Prussian state
view of the citizen was a corollary of the absolutist
state in intrinsic conflict with the older, medieval
Standestaat. Territorial residence took precedence in
the Prussian conception. The romantic movement
provided more of a background `pattern of thought'.
The aesthetic socio-historical idiom of romanticism
was ideally suited for the misty formulations of
ethnocultural ideas, namely conceptions of unique-
ness, inwardness, feeling over reason and organic
 LIBERALISM AND CITIZENSHIP
growth over conscious artifice. These latter ideas
provided a frame for jus sanguinis, developing be-
tween 1871 and 1913. Before 1913, German citizen-
ship was a partly contradictory affair. The older
Prussian absolutist model of the citizen ± as member
of territorial community ± jostled with the ethno-
cultural `community of descent' idea. It was the latter
which won out, largely in 1913 with citizenship
legislation which established the concept of Aus-
landsdeutsche. This legislation aimed at the preserva-
tion of the pure German Volkstum. Citizenship was
severed from questions of residence. Blood and des-
cent now became critical.
The difference between the Wilhelmine and the
National Socialist ideas of citizenship was that in the
latter there was a definite ideology of ethnocultural
dominance, coupled with the desire to deprive groups
of citizenship. In the Wilhelmine case, the ideology
was not so well-developed and the desire was rather
to prevent immigrants acquiring citizenship. As Ro-
gers Brubaker comments, `remarkably, German citi-
zenship today remains governed by a law of the
Wilhelmine period.' He continues:
As a result of this continuity across two world
wars, three regime changes, and the division and
reunification of the country, the marked restric-
tiveness of citizenship law towards non-German
immigrants was carried over from the Wilhelmine
Germany into the Federal Republic and, in 1990,
into the new German nation-state. The 1913
system of jus sanguinis, with no trace of jus soli
continues to determine the citizenship status of
19
immigrants and their descendants today.
It is important to note here (in my reading) that we
are still looking at a liberal state in Germany, but its
citizenship is deeply exclusive and measured by blood
descent. Something does not seem to be quite right
here.
The French case is more interesting yet for my
purpose, namely the question concerning the uni-
versalising aspect of citizenship. In the French case,
jus soli is predominant. The drive for citizenship was
always state-centred and assimilationist. It was part
of a revolutionary republican tradition. In its purest
universalist reading, the only foreigner in France
59
would be the bad citizen. Self-determination (in
individuals or nations) was conceived in terms of
the universal demand for liberty. The doctrine of the
`citizen' and the `citizen-nation' moved hand-in-
hand. Republican ideology usually emphasised mili-
tary service as a key factor here. Schools and the
army were engines of assimilation. Second- and
third-generation immigrants in France became citi-
zens and were expected to save the nation militarily.
National purpose and assimilation were reinforced in
the Third Republic by the civilising mission of the
French nation. French citizenship was therefore al-
ways a peculiar brand of chauvinism and universal-
ism. Before the 1870s it was very much a liberal and
radical push; after the 1870s it migrated to the right,
reaching an ascendancy in the Dreyfus Affair. The
left still, however, promulgated universalism through
imperialism.
Ethnocultural criteria did appear in France in the
20
mid-1800s. As they developed, the French idea of
the citizen lost some of its universalising aspect. In
Arthur de Gobineau's work citizenship and nation-
alism began to take on racial dimensions, a theme
reflected in Taine, Renan and Thierry, particularly
after the Franco-Prussian war. Still, the issue of
citizenship has exercised the French Republican
tradition, even to the present day. Whereas the
Dreyfusard ethnocultural line of thought is still
represented in the Front Nationale, republican
thought remains motivated by an underlying, if
semi-dormant, cosmopolitanism.
Nationality, Liberalism and Citizenship
How does citizenship link up liberalism and nation-
ality? The French case is particularly instructive here.
French citizenship traditionally involved a core of
universal human or natural rights. The declaration of
1789 speaks stridently of the rights of `man' and
`citizen'. The rights involved are `imprescriptible'
and `inalienable', and governments are instituted
to uphold them. Thus, `the end of all political
associations, is the preservation of the natural and
imprescriptible rights of man'. The French National
Assembly `recognises such ``sacred'' rights. It does not
create them. All men are all born free and equal, with
respect to these rights'; thus `civil distinctions can be
 ANDREW VINCENT
founded only on public utility.' Yet, at the same time
as these assertions, the nation is also conceived of as
sovereign. Thus, no individual or group are `entitled
to any authority which is not expressly derived from
it.'
So what is the relation between the nation and the
universal cosmopolitan rights of man? The pro-
foundly direct (but also equivocal) answer is that
the nation is the will of the whole community of
citizens and it embodies the imprescriptible rights of
humanity within its own prescriptions. Law and
government, as an expression of the national will,
are only there to give `security to the rights of man
and of citizens'. Thus, `the acts of the legislative and
executive powers of government, [can be] . . . every
moment compared with the ends of political institu-
21
tions.'
The above forms the substance of my analysis of
this question and my key theme. Particularistic
nationalism and cosmopolitanism coexist, simulta-
neously, in liberal citizenship. Strikingly, it is actu-
ally possible to assert simultaneously an extreme
chauvinism and an extreme cosmopolitanism. In
the case of jus sanguinis, the particularism and inclu-
siveness comes to the fore unproblematically. Eth-
nocultural citizenship links easily with racism,
nationalism and cultural liberalism. The liberal ci-
tizenship/nationalism connection is thus more ob-
vious here. However, jus soli, particularly in the
French republican tradition, reveals more clearly
the intrinsic ambiguity of liberal citizenship, namely
its continual alternation between cosmopolitanism
and particularism.
Liberalism, generically, often represents itself as a
form of historical emancipation from older political
traditions ± the conventional portrayal of status to
contract or the release from the medieval Standestaat.
Liberal citizenship sees itself as removing feudal
shackles. It casts off the `subjecthood' status of in-
dividuals under absolute monarchy or feudalism.
Liberal rights theory thus rejects conceptions of
status based upon birth or position. There is, there-
fore, an insistent `progressive Whig' logic or dialectic
to ideas of equality and freedom. Such ideas, it is
suggested, cannot be checked by any hierarchical
regime. Liberal citizenship identified equal rights
with individuals as individuals. Citizenship is, there-
60
fore, a central motif within liberalism, and liberalism
has always believed in equality of individual rights,
regardless of race, gender, class or birth. In the
French Republican tradition particularly, in the
classical liberal tradition, there has been a strong
leaning to some form of cosmopolitanism. Yet, co-
existing with this universalising dialectic is the na-
tion ± usually read through the state. As Rogers
Brubaker has put it:
Almost all [states] claim to derive state power
from and exercise it for (and not simply over) a
nation . . . A state is a nation-state in this
minimal sense insofar as it claims (and is under-
stood) to be a nation's state: the state `of' and `for'
a particular, distinctive, bounded nation . . .
How the state-bearing and state-justifying nation
is culturally or legally bounded is irrelevant; that it
22
is bounded is what matters.
One has only to think of the commonplace usage of
`citizenship law', as directly synonymous with `na-
tionality law', to see the truth of this point.
The above observation creates an oddity in the
liberal argument outlined above. In the context of
the nation, the citizen has rights which are premised
on the natural arbitrariness of birth, even in the jus
soli and liberal settings. In one sense, closure is
pragmatically necessary, for the purposes of military
service, taxation and public administration. Yet the
nation, birth, citizen relation is stronger than simple
pragmatism. Given the history that liberalism re-
lates, and given its lurking cosmopolitanism,
something strange is taking place. As Brubaker
comments:
[Citizenship] represents a striking exception to
the secular trend away from ascribed statuses. And
it is difficult to reconcile with a central claim ±
perhaps the central claim of liberal political the-
ory: the idea that political membership ought to
23
be founded on individual consent.
The ascription of rights based on birth or blood
descent is at odds with the panoply of liberal rights
and liberal agency, in general, and yet they all
coexist within liberal citizenship.
 LIBERALISM AND CITIZENSHIP
Conclusion
We can now appreciate exactly how liberal citizen-
ship is at the interstice between contradictory forces,
incorporating both an embracing cosmopolitanism
and an exclusive nationalism. There are, though, a
number of forms of exclusion and some are more
problematic than others. But still the exclusions of
citizenship do not rest easily with the cosmopolitan
dialectic. The nationalist/birth connection of citi-
zenship does not readily conjoin with the liberal
narrative. The two theses discussed in this chapter ±
the cosmopolitan implication of liberal citizenship
and the potentially illiberal criteria for liberal citizen-
ship ± present us with a dilemma. Another, possibly
simpler, way of framing the unease here is through
Veit Bader's recent reflections on transnational ci-
tizenship. Bader, reflecting procedural and cultural
liberalism, remarks that procedural liberalism `has
underestimated the importance of communities and
cultures.' But cultural liberalism, which he describes
as `communitarian liberalism', has actively taken
culture and nation on board, but to Bader has
`not yet fully addressed the exclusionary effects of
24
communities and cultures.' I would add to this the
point that even procedural liberalism subsists with
radical exclusion.
A number of thinkers appear to want to address
this dilemma. Bruce Ackerman, for example, has
25
described himself as a `rooted cosmopolitan'. Lib-
eral universalism is seen to be virtually implicit
within the United States Constitution. Conse-
quently, Ackerman is therefore provisionally per-
suaded to `embrace the American Constitution'. A
Notes
1. Heater (1990), p. 12.
2. Kant argued for the moral autonomy of states within a
voluntary league of states committed to maintain peace.
A liberal constitutional republic would thus be inclined,
in his view, to international peace. Liberal theory, qua
Kant, conceives of citizenship in terms of contract and
consent. The citizen comes to the state ideally or actually
with entitlements. The citizen will agree to limitations
for the common good. Liberal citizenship is thus a bearer
or vehicle of rights within the state.
3. Thomas Pogge proposes two forms of moral cosmo-
politanism: `institutional' and `interactional'. Institu-
61
more cynical appraisal of this would be Gray's nicely
sour observation of `Kantianism in one country'. For
Ackerman, the poor Europeans would, however, be
condemned to be `rootless cosmopolitans'.
But note the problem in Ackerman's view of the
precise character of the national mission and liberal
universalism. There are a number of liberal states
where global citizenship can be seen as an extension
of their own rights structures. Yet, different inter-
pretations of both liberalism and cosmopolitanism
are inevitably involved. North Americans might
well embrace generic liberal rights (on a civil level)
but not widespread distributive justice (social and
social-democratic rights) let alone specific cultural
rights. Social rights and demands for widespread
redistribution would be a universalisation of civil
citizenship. Others might see distributive justice and
social rights as a logical extension of human rights to
life. If this happened, it would imply large transfers of
wealth across the globe. However, since this latter
conception has been to date reliant upon social
liberal or social-democratic sovereign states, it is
an unlikely international scenario. France, on the
other hand, would give full legal rights within its
Republican notion of citizenship. These would in-
clude civil, political and social rights. But this would
not necessarily allow any special or cultural rights
and would also imply an unacceptable level of im-
perial control.
Thus the implications of `liberalism' and `citizen-
ship' seem to be inevitably linked ± but the tensions
and countervailing currents exposed here show just
how the nature of this common fate remains pro-
foundly unclear.
tional cosmopolitanism assigns moral responsibility
(for example, maintaining and promoting human
rights) to institutional schemes. It is dependent upon
the existence and maintenance of human rights
through institutions. Some argue, in this scenario,
for supranational state-like institutions to be set up.
Interactional cosmopolitanism is `ethical' in the sense
that it assigns moral responsibility to individual or
collective agencies. Each person is equally a subject of
moral concern. Note that there is no necessary link
here between the institutional and interactional/ethi-
cal types. One could therefore believe in the institu-
 ANDREW VINCENT

tional conception and not adhere to the ethical. accounts of the new liberalism can be found in
Equally, one could be an ethical cosmopolitan and Vincent (1990).
not believe in an institutional setting or, at least, a 13. Kymlicka (1989), p. 165.
central world state. Therefore, even if one rejects 14. Ibid., p. 172.
world government, it does not lead to a rejection of 15. Quoted in Forsyth (1987), pp. 117±19.
moral cosmopolitanism. See T. Pogge, `Cosmopolitan- 16. Kelley (1995), p. 91.
ism and Sovereignty', in Brown (ed.) (1994), pp. 89± 17. Dahrendorf (1974).
122. 18. Heater (1990), p. 96.
4. Linklater (1992), p. 32. 19. Brubaker (1994), p. 165.
5. M. Nussbaum, `Patriotism and Cosmopolitanism', in 20. Ibid., p. 99: `The legal nationality conferred by the
Cohen (ed.) (1996), pp. 2±17, at p. 7. state and the ethnocultural nationality invoked by the
6. Ibid., p. 4. ``principle of nationality'' are of course different
7. Ibid., p. 13. things; the former may be conferred in utter disregard
8. Ibid., p. 14. of the latter. Yet the thrust of the principle of nation-
9. This is close to Michael Sandel's use of the term ality was precisely to connect the two ± not
`procedural republic' for this liberalism. See Sandel directly . . . but indirectly via the redrawing of po-
(1984). litical boundaries so as to make legal and ethnocultural
10. Walzer, `Comment,' in Gutmann (ed.) (1994), pp. nationality converge.'
99±103, at p. 99; see also Habermas in the same 21. Tom Paine's translation of the `Declaration of the
volume, p. 123ff. Rights of Man and of Citizens' by the National
11. For Taylor, the `procedural' view is best encapsulated Assembly of France (1789), in Ritchie (ed.) (1954),
in Dworkin's early paper on `Liberalism'; see Taylor pp. 291±2.
`The Politics of Recognition', in ibid., p. 56. 22. Brubaker (1994), p. 28. Debates about both nation-
12. The term `new liberal' is immensely complex and ality and citizenship are also closely associated with
contested. I have not attempted to examine it, but distinct state traditions, although this is an even bigger
have largely taken for granted in this chapter that we issue.
can speak of its ideology and social theory. This is by 23. Ibid., pp. 31±2.
no means an uncontroversial position. My own at- 24. See Bader (1997), p. 772.
tempt to summarise and assess some of the diverse 25. Ackerman (1994).

62
 5
THE SWORD OF FAITH AND THE
SHIELD OF FEAR: LIBERALISM
AND THE POWER OF THE NATION
Margaret Canovan
When the terms `Liberal' and `Liberalism' first ar-
rived in English politics in the early nineteenth
century, they indicated support for national libera-
1
tion in Greece and South America. The struggle of
oppressed nations for freedom continued to be a
liberal cause, so much so that liberalism has over
the past two centuries been as closely associated with
2
nationalism as internationalism. Revulsion against
fascism after the Second World War weakened this
association except in the case of anti-colonial move-
ments; liberals became more wary of nationalism,
and the dominant strand in liberal political thinking
during the past half-century has been cosmopolitan.
Even so, we have recently seen a revival of liberal
nationalism among philosophers who argue that
liberal respect for individuals and their rights entails
respect for the national identities those individuals
3
value.
Despite obvious tensions, therefore, liberalism and
nationalism do have affinities. I shall argue that the
connections go deeper than one might suppose and
that even versions of liberalism that are explicitly
hostile to nationalism rely for much of their plausi-
bility on tacit assumptions about the presence and
efficacy of nationhood. I shall try to show that this is
equally true of two sharply opposed contemporary
versions of liberalism: on the one hand, the crusading
liberalism of universal human rights, and on the other
the sceptical liberalism of limited government. The
63
former is cosmopolitan in its appeal as well as its
aspirations, though it is particularly well represented
among political thinkers in the USA. The latter has
always had a narrower academic base. On this side of
the Atlantic it is particularly (though not exclu-
sively) associated with a school of political thinkers
4
influenced by Michael Oakeshott. To provide a
frame for these two different sorts of liberalism, I
shall begin with some of Oakeshott's own reflections
on two contrasted styles of politics, which corre-
spond neatly to the versions of liberalism with which
we shall be concerned.
The Politics of Faith
and the Politics of Scepticism
Oakeshott was fond of dichotomous analyses of
political styles. His celebrated contrast between `ra-
tionalism in politics' and the practical activity of
5
`attending to the arrangements' of a society was
followed by the more elaborate analysis in On Hu-
man Conduct of two opposed understandings of the
modern state, as an `enterprise association' (universi-
6
tas) and in terms of `civil association' (societas).
Each of these contrasts carried the message that
the former of the two types of politics was dominant
in the modern world but that the latter was to be
preferred. The contrast that concerns us here is a
third variant on the same theme, elaborated in an
 MARGARET CANOVAN
essay (published posthumously) on `The Politics of
Faith and the Politics of Scepticism'. For the past five
hundred years (according to Oakeshott), European
politics has been marked by a tension between the
two opposed styles named in the essay's
Although the two have mingled in practice, they
7
can in principle be characterised as opposites. These
two ideal types bear obvious resemblances to Oake-
shott's more familiar pair of antagonists, but the
characterisation is more vivid than in the case of
universitas and societas and the treatment consider-
ably more even-handed than in the case of political
`rationalism' and its rival.
Those who practice `the politics of faith' (so
Oakeshott would have us believe) approach the
business of government with ambition and confi-
dence, for within this style `the activity of governing
is understood to be in the service of the perfection of
8
mankind.' Political action (it is assumed)
achieve such perfection in this world, though Oake-
shott stresses that the `perfection' aimed at can be
understood in many different ways. Seventeenth-
century projects for using the state to establish a
godly commonwealth stand alongside the secular
projects of the last two hundred years under the
9
banner of the politics of faith. Interestingly, the
projects Oakeshott had earlier called `rationalist' fall
within this category: `faith' should be seen as the
opposite not of reason but of scepticism. The most
important political implication of this faith in the
capacity of government to save mankind is that it
justifies great increases in power. No legalistic re-
strictions can be allowed to stand in the way of such a
crusade, with the result that government becomes an
`unlimited activity' and even a `godlike adventure.'
The `politics of scepticism' differs from this in ways
that any reader of Oakeshott might expect. Political
sceptics have lower expectations of government,
together with greater suspicion of power. Instead
of looking to power to save the world, they deny
that it has any single overriding purpose. Rather, it is
a matter of maintaining order, moderating the con-
flicts that continually arise in human affairs and
trying to minimise concentrations of arbitrary power.
Rule according to settled law is crucial, as is the
gradual development and maintenance of institu-
tions that check power. Although such concerns
may seem unglamorous, `the sceptic understands
order as a great and difficult achievement never
11
beyond the reach of decay and dissolution,' so that
a political system that can properly carry out these
title. modest tasks is doing very well.
Although these two styles of politics can be pre-
sented in ideal-typical form, Oakeshott maintains
that they have been continuously intermingled with-
in European political traditions, and he even suggests
that neither is sufficient without some infusion of the
12
other. Despite this formal even-handedness his
own preference for the politics of scepticism is clear.
He preserves impartiality by arguing that in a poli-
tical climate dominated by the politics of faith, a
modern `trimmer' will add his intellectual weight to
13
the beleaguered cause of scepticism.
Following this long tug-of-war, our political vo-
cabulary is systematically ambiguous because it has
can been `obliged for nearly five centuries to serve two
14
masters.' We inherit terminology marked by both
styles and capable of being plausibly used for opposite
purposes. Oakeshott himself chooses to focus on the
ambiguities of democratic discourse to illustrate his
15
point, but if we now turn back to liberalism we can,
I believe, identify two sharply contrasted versions
(both current within contemporary political thought
and each boasting a proud liberal ancestry) which
vividly illustrate Oakeshott's contrasted styles. These
are not the only sorts of liberalism there are, and I do
not wish to imply that the rest could be fitted onto a
spectrum of which these two are the ends. But I hope
it will be agreed that both sceptical liberalism and the
liberalism of faith ± which I shall call crusading
liberalism ± are familiar presences not only in the
10
liberal tradition but within the complex contempor-
ary discourse of liberalism. My characterisations of
the two will verge on the ideal-typical, though
illustrated with examples from specific thinkers.
Crusading Liberalism
A large proportion of all those who have ever
thought of themselves as `liberal' could without
undue distortion be assigned to this camp, and its
roots are older than the term itself. But the first true
manifesto of crusading liberalism was the Declaration
of the Rights of Man and of the Citizen, approved by
64
 THE SWORD OF FAITH AND THE SHIELD OF FEAR
the French National Assembly in 1789, and its most
characteristic twentieth-century document is the
Universal Declaration of Human Rights, adopted
by the United Nations in 1948. Although there is
more to crusading liberalism than the discourse of
human rights, that discourse vividly illustrates the
salient features of this kind of liberalism. In the first
16
place, crusading liberalism is a political project.
However abstract discussions of human rights may
at times become, they are more or less closely linked
to a movement with practical political aims. The
liberalism of human rights is a crusade today, just as it
was in the days of the nineteenth-century campaign
against slavery.
Secondly, crusading liberalism is a universalist
creed. The notion of human rights clearly expresses
the conviction that liberal principles do not lose
their relevance at the borders of states or at the
fringes of cultures. Despite the worries about cultural
17
variation provoked by such claims, the dominant
assumption is that a single moral and political order
is right for all human beings: for those suspected of
crimes in Sweden or in Saudi Arabia; for religious
minorities in Ireland or in Iran; for political dissi-
dents in Canada or in China; for women in Australia
or in Afghanistan. Thirdly, besides being a univers-
alist project, crusading liberalism is radical in its
confidence that the world must and can be changed
to conform to this universal order. Institutions and
practices that contravene human rights have no
justification, whatever the weight of established
tradition and cultural sanctity supporting them. If
the institutional traditions and culture of the US
armed forces contravene the rights of women and
gays, then they must be changed, and if the entire
historic culture of China or of Islam clashes with
human rights, then they must all be changed as well.
It is clear enough that crusading liberalism manifests
the ambition and confidence described in Oake-
18
shott's account of the `politics of faith', and (as
we shall see later) it also harbours the quest for power
that he found there.
To shift one's focus from crusading liberalism to
sceptical liberalism is to enter a different (and very
much smaller) world. Unlike the former, sceptical
liberalism has never been a vast political movement
but the business of a handful of intellectuals.
65
Whereas the academic representatives of crusading
liberalism are linked more or less closely to politi-
cians, campaigners, UN agencies and NGOs, scep-
tical liberalism is much less conspicuous within the
contemporary political arena and owes its continuing
significance to the intellectual stature of its expo-
nents. Oakeshott himself, Hayek, Shklar and others
stand in a distinguished line of sceptics that runs
back through Tocqueville and Constant to Mon-
19
tesquieu. This sort of liberalism is essentially con-
cerned with limits: limits on power and on ambition
in the uses of power. Mindful of the long history of
misuse of political power, it is concerned (as Judith
20
Shklar says) with `damage control.' Whatever their
views on the universality or relativity of moral and
political principles, sceptical liberals share the view
that the liberal institutions of civil society are a fragile
achievement, one that can be easily endangered by
hubristic political projects. Like Oakeshott himself,
they see themselves swimming against a political tide
fed by torrents of crusading enthusiasm and trying to
maintain the crumbling dikes that hold it in check.
In doing so they lay particular emphasis upon the
rule of law (according to Hayek, liberalism is `the
same as the demand for the rule of law'; similarly for
Shklar, `the rule of law' is `the original first principle
21
of liberalism' ) and they seek to remind their con-
temporaries that it cannot be taken for granted.
There is little love lost between crusaders and
sceptics within the liberal camp. To the former, the
22
latter seem mere conservatives, while sceptics fear
that their cherished heritage of liberal institutions is
23
put at risk by the hubris of crusaders. Interestingly,
however, there is one topic on which they are agreed:
nationalism is anathema to both of them. Should a
crusader and a sceptic find themselves stranded
together in a snowstorm, they could generate a warm
glow of mutual solidarity by denouncing the perni-
cious political influence of national feeling. True,
their hostility has different grounds, but on both
sides the hostility is pronounced. Crusading liberals
object to the particularity of nations, which builds
barriers against the universality of human rights. This
is in a sense an argument against any kind of political
demarcation, but non-national boundaries may at
least be regarded as temporary matters of practical
convenience. National boundaries are more formid-
 MARGARET CANOVAN

able obstacles, buttressed by the widespread belief
that they have moral significance, so that our fellow
nationals have special claims that trump the claims
24
of other human beings. Similarly, the notion that
national cultures and traditions have value in them-
selves often seems (from the crusading point of view)
like an excuse to deny the universal rights of those
25
traditionally victimised.
Sceptical liberals have quite different reasons for
disliking nationalism, which they associate with the
26
populist enthusiasm and `ideologies of solidarity'
that put liberal institutions at risk. In their view, it is
a fundamental mistake to expect political institu-
27
tions to express the `identity' of their subjects. By
chasing the mirage of `self-government' for members
of a nation, nationalism undermines established
political systems that may in themselves have been
relatively benign, and replaces them at great political
cost with others that exercise arbitrary power in the
name of the nation. For the sake of national self-
determination, many people who had enjoyed a
relatively high degree of liberty under law as subjects
of the Habsburg or of the British Empire, found
themselves `liberated' into subjection to local tyr-
ants. In any case, the mixed societies of the late
twentieth century are inappropriate sites for nation-
alist fantasies. To quote John Gray:
The aspiration to make states correspond to com-
munities cannot be other than a dangerous illu-
sion. Where moral solidarity is lacking, where (as
in all modern societies) there is cultural diversity
rather than seamless community, the role of
government is first and last that of preserving
28
liberty in civil association under the rule of law.
So far I have been attempting to establish two
contrasted ideal types of liberalism corresponding to
Oakeshott's `politics of faith' and `politics of scepti-
cism', and to show that one of the few areas of
common ground between them is that both are
(for different reasons) hostile to nationalism. The
remainder of the chapter will try to show that despite
this hostility, crusaders and sceptics are both much
more dependent on national attachments than they
would like to admit. Each version of liberalism has a
hole at its centre where a theory of power ought to
be, and in each case the nation has to be smuggled in
to stuff the gap. With convenient symmetry, the way
this is done illustrates Oakeshott's observation that
neither of his two styles of politics is self-sufficient,
for we shall see that sceptics need nationhood to
provide an infusion of enthusiasm, while crusaders
need it to give them a stiffening of realpolitik.
Sceptical Liberals and the Power Deficit
Sceptical liberals are deeply suspicious of power and
of grand political schemes that rely on expanding the
role of government and mobilising the people be-
hind a common cause. But that concern with the
dangers of excessive power leads them to overlook a
power deficit within their own style of politics. They
tend to underestimate the power that a limited
government needs and to overlook the problem of
where such power might be found. At first sight their
political ambitions seem modest ± deplorably so to
those whose politics is concerned with righting
wrongs across the world. They have no grand
schemes and do not expect perfection. Virtually
all they ask is that arbitrary power should be re-
strained by known rules, and that established rights
to person and property should be protected by an
impartial judiciary and by fair and effective enforce-
ment of the law. In the words of John Gray, what
29
they look for is a state that is `strong but small'.
This may not seem much to ask. In criticising the
excessive ambitions of crusaders, sceptics often imply
that their own aims are much less demanding, some-
times even that they are really quite easy to attain.
(Hayek says, for example, that his kind of liberalism
30
is `an essentially modest creed'. ) But this is an
illusion. The crucial point to bear in mind is that
(despite their proximity to the more sceptical ver-
sions of conservatism) our sceptics are liberals and it is
their liberal commitments that expose the power
deficit in their thinking. If they were simply Hobbe-
sian conservatives, asking nothing from politics but
peace and order, then that `strong but small' state
might be easier to come by. Peace and order have on
occasion been achieved by a wide variety of regimes
in many different kinds of society: by traditional
monarchies and military juntas, by oligarchic repub-
lics, and even by communism in its phase of Brezh-

66
 THE SWORD OF FAITH AND THE SHIELD OF FEAR
nevite immobilism. Neither social cohesion nor
democratic politics is required. Indeed, if peace
and order are really what you are looking for, then
the power to enforce them may in some circum-
stances be more effectively achieved by non-demo-
cratic than by democratic means, whether
coercion, by traditional authority, or through a
negotiated modus vivendi between powerful
31
tions.
But although sceptical liberals may be tempted by
the Hobbesian politics of peace, in so far as they
remain liberals they cannot accept such solutions,
simply because most ways of establishing peace and
order are not compatible with the rule of equal law.
John Gray was deluding himself when, daringly
dismissing the notion that democracy is necessary
to `the practice of liberty', he claimed that the `civil
condition' so much valued by sceptical liberals `is
fully compatible with a variety of political regimes',
although it `presupposes always the rule of law and
32
equality before the law.' Very few kinds of political
regime and political society are in fact compatible
with that `civil condition', because the equal rule of
law implies two very demanding conditions. The first
is that the state enjoys a fund of consensual power,
enabling it to enforce the law without employing a
degree of coercion which would be inconsistent with
the civil condition. The second is that it presides
over a society sufficiently united to allow it to
enforce equal law across the whole country. What
this amounts to is a high level of identification of the
polity with the population and vice versa, which in
the condition of modern politics normally means
nationhood. Let us look more closely at these two
conditions.
Consensual Power
For the `strong but small' state to generate its
strength in a manner compatible with the rule of
law, it must be recognised as legitimate by the mass of
its subjects. Without that support the minimal liberal
state is impossible; impartial justice is scarcely to be
expected from rulers who have seized power in a
military coup, or from a conquering army sitting on
rifle butts. Even a state that is recognised by most of
its subjects as their legitimate representative may find
grave difficulties in providing the rule of law in
particular regions where that legitimacy is not gen-
erally recognised, as the sorry history of British rule in
Ireland demonstrates. This is not to say that regimes
enjoying this recognition will necessarily provide
by equal justice ± far from it. But to have even a sporting
chance of enjoying the rule of law, we do need a
fac- government with legitimate power.
Followers of Hayek or Oakeshott will object that
the authority of the law should not be confused with
popular support for the regime; all that is required is
recognition of the authority of a system of rules and
33
of judicial decisions in accordance with it. Oake-
shott himself might cite as an analogue the authority
to adjudicate on the rules of cricket gradually ac-
34
quired by the MCC. But this implies popular
acceptance of the system of rules and, crucially, trust
in the judicial processes by which they are adminis-
tered and enforced. And that begs a vital question:
under what political conditions is that impartial rule
of law most likely? For this is the Holy Grail sought
by republican thinkers from Cicero to Rousseau, who
wanted `the government of laws, not men', and who
knew that such a government required a strong sense
of shared, inherited public good and its accompany-
ing public spirit. Under modern conditions that
public spirit, like legitimacy and consensual power,
is most likely (though far from certain) where the
polity is supported by a strong sense of collective
identity ± where it belongs to a people. Much as
sceptical liberals may deplore populism, they can
hardly deny that impartial law enforcement with
minimum coercion is normally easier where the state
enjoys the legitimacy of being generally regarded as
35
our state.
It may still be objected that my argument confuses
a contingent historical cause with a necessary con-
dition: that although as a matter of historical fact the
establishment of the rule of law in England owed a
lot to collective national attachment to `the laws of
36
England', once established it became a practice
that could be freely transferred elsewhere. In other
words, while a strident national pride may have been
crucial in enabling Coke, Hampden and their heirs
to defend law and limited government against the
Stuarts, the system has since become self-sustaining:
wherever a state does provide the benefits of impar-
67
 MARGARET CANOVAN
tial justice, then popular consent will follow. Bhikhu
Parekh argues, for example, that nothing by way of
national cohesion is required because `the modern
state is a self-sufficient institution. It has a commonly
accepted structure of public authority, a set of pro-
cedures for taking collectively binding decisions, and
a body of widely shared politico-legal values', notably
`the rule of law, equality before the law, respect for
human dignity and common citizenship.' Although
this `self-sufficient state' is a contingent product of
history, it does not now need the support of national
sentiment. Being a member means recognising its
`politico-legal values' and abiding by the laws. `Noth-
ing more is required of its members in order for it to
37
remain united and stable.'
For the increasing number of cosmopolitan indi-
viduals who move easily between different western
liberal democratic countries, expecting much the
same legal protections and offering the same minimal
allegiance in each, this account may seem reasonable
enough, and the national identity of the local po-
pulation may seem irrelevant. But we should not be
misled by talk of `the modern state' into supposing
that there is something about the advent of the
millennium that makes such `modernity' easy to
achieve without a background of nationhood. Most
modern states (in the sense of states actually in
existence today) are not `modern' in this sense. As
a matter of empirical observation, most UN members
do not provide anything near enough to the rule of
law to earn the consent of a sceptical liberal. Cor-
ruption, violence and arbitrary power are common-
place, even in many of the states that have formally
liberal constitutions. Evidently a properly `modern'
state as understood by Parekh is one that does provide
equal law, but we need to consider the role of
nationhood or the lack of it in making that more
or less likely.
Equal Law
If taken seriously, the sceptic requirement of equality
before the law seems to demand a substantial degree
of social unity. This is the case even if we leave aside
all the problems of equal justice between rich and
poor and men and women that would take us into
crusading liberal territory, and consider only the
68
obstacles a plural society poses to the achievement
of impartial judicial treatment of (say) felonies.
Sceptics like to claim that the minimalist state they
favour is particularly suited to ethnically divided
societies, because its modest ambitions allow con-
38
tentious issues to be depoliticised. But this is an
implausible claim, perhaps betraying the confusion
we noticed earlier between the conservative insis-
tence on peace as the political goal and the more
demanding aims of liberals. A sceptical conservative
could argue that, where a political entity (such as
Bosnia) is sharply divided between hostile commu-
nities, then the best that can be hoped for is a
minimal state based on negotiation between leaders
of the contending factions, themselves holding
power within their own patches as best they can.
But that is not a solution any liberal could be content
with, for it would certainly not imply equality before
the law, especially for minorities in the various
39
enclaves. Even in societies that are less radically
divided, the Hobbesian concern for peace and order
may well clash with the rule of law. For despite the
sceptics' image of the law as an umpire presiding
impartially over communal divisions, in such cir-
cumstances everything to do with the legal process
becomes politicised. Apart from the problems of
staffing juries and making judicial appointments,
the outcome of every contentious trial becomes a
victory or defeat for different sections of society. The
tensions surrounding the trial of O. J. Simpson in the
USA is perhaps sufficient comment on the notion
that the rule of equal law can comfortably coexist
with communal divisions, even in a deeply legalistic
polity. For the equal rule of law to function, it is vital
that the apparatus of justice should not be seen by its
recipients as their law rather than ours.
The notion of the law as impartial umpire above
the fray may in some cases be a piece of unconscious
post-imperial nostalgia, a regretful glance back to the
days when relatively upright British officials admi-
nistered justice between the castes and religious
groups of India, or when relatively professional im-
perial bureaucrats dealt with the multifarious nation-
alities of the Habsburg Empire. If so, two
observations are apposite. In the first place, imperial
systems have never been models of impartial justice,
as anyone who fell foul of the imperial authorities
 THE SWORD OF FAITH AND THE SHIELD OF FEAR
quickly discovered. Secondly, imperial regimes have
in this century proved quite unable to attract enough
popular consent to maintain peace and order, let
alone the rule of law.
In other words (as John Stuart Mill pointed out
40
long ago ) the ideal venue for the kind of politics
favoured by liberals is actually a polity belonging to a
people bound together by the mutual sympathies of
nationhood. A state in which there is a long-estab-
lished sense of collective identification with the
polity, and in which state, government and laws
are seen as ours rather than an alien imposition,
provides a particularly favourable basis for strong but
consensual government, a sense of public interest
41
and support for the fair and impartial rule of law.
This is an ideal never fully realised in practice, like
the republican ideal of the polity of virtuous citizens
united in their devotion to the patria. But the
historical connection between liberalism and nation-
hood is not accidental. Sceptical liberals are inclined
to take this background of nationhood for granted
without reflecting upon its contribution to the in-
stitutions they value.
Some critics might concede this last point but
object that I am ignoring a vital distinction between
nationhood and nationalism. They would argue that
one can acknowledge a debt to nationhood while
deploring nationalist ideologies and movements.
This is (for example) the position that John Gray
attributes to Isaiah Berlin. Unlike most other liberals
of his time, according to Gray, `Berlin perceives that
a liberal civil society cannot rest upon abstract
principles or common rules alone, but needs a com-
mon national culture if it is to be stable and com-
mand allegiance.' But (he continues) `in common
with all liberal thinkers, Berlin repudiates the
42
pathology of nationalist sentiment.' This position
is attractive but simplistic. It may seem easy in
principle to affirm nationhood while condemning
`the pathology of nationalism', but the line is not so
easy to draw in particular cases (including Zionism,
43
which Berlin supported). It is significant that the
word `nationalism' can itself refer not only to move-
ments and an ideology but also to the sense of
nationhood itself, perhaps indicating the difficulty
44
of distinguishing them. In practice, where a strong
but civilised national sentiment exists it is normally
69
the product of a much more strident history. There is
a good deal to be said for Michael Ignatieff's distinc-
45
tion between `hungry' and `sated' nations, with the
caveat that even the satisfied nations sometimes
wake up feeling peckish. Sceptical Liberal distrust
of nationalism is reinforced by the fear that even
apparently innocuous forms of the virus are liable to
flare up if given the slightest encouragement.
Nevertheless, if the argument of this section holds,
sceptical liberals cannot afford to be so fastidious, for
they cannot easily do without some measure of
national feeling to underpin the rule of law with
the power of popular legitimation. Instead of dream-
ing of a polity untroubled by any such emotional
stirrings, the far-sighted sceptical liberals would
choose a polity that belongs to a `sated' nation where
the `pathologies of national sentiment' are buried
several generations deep in the past, and where the
rule of law is cherished as a national heritage.
Crusading Liberalism and the Power Deficit
If sceptical liberals find it hard to do without political
faith in nationhood to support the power needed by
their minimal state, a complementary point can be
made about crusading liberalism. For crusaders can-
not find salvation through faith alone, and their
good works take them into a world of realpolitik in
which they find themselves relying to a degree not
normally acknowledged on the power generated by
nationhood. Power is not something that crusading
liberal political theorists like to think about; as their
critics (including sceptics) have observed, their pre-
occupations often seem more moral than political,
and questions about the sources, limits and costs of
power are rarely raised in their writings. But liberal-
ism of this kind is more than a set of moral injunc-
tions addressed to individuals; it is a political project.
Crusaders want to see human rights recognised and
protected across the world, and questions of political
46
agency inevitably follow. Seeking to make the
Marxist political project effective, Lenin hit on
the notion of the powerful party: what collective
actor or actors can (by analogy) bear the project of
human rights?
Having as little instinctive trust in professional
politicians as sceptical liberals have in popular move-
 MARGARET CANOVAN
ments, crusading liberals might be inclined to look
first to the non-governmental organisations that
have to such an extraordinary degree furthered hu-
manitarian projects across the world since the Sec-
ond World War. But however good they may be at
famine relief, NGOs are not well suited to protect
the human rights of individuals under threat from
tyrannical regimes or ethnic militias. The former can
simply refuse to admit them, while in the times of
civil war and anarchy that breed the latter, NGOs
themselves need protection, as events in Bosnia and
elsewhere have shown. It is clear that political power
is required. Calling for `humanitarian interven-
47
tion', crusaders will turn next to international
organisations that sprung up (like NGOs) in the
second half of the twentieth century. Organisations
of this kind (established by the more liberal nation-
states) have indeed transformed the conduct of
international politics, and the ways in which they
have modified older notions of state sovereignty
48
should not be underestimated. Nevertheless, their
inability (so far) to take the sort of decisive action
crusaders often want has been cruelly exposed by
recent events. If action is to be taken to guard human
rights where they are most seriously under threat, it
can be taken only by a political agent able to project
power effectively, which means a powerful state or
49
alliance of such states.
It seems, then, that crusaders must compromise
with political reality to the extent of calling on the
power of states to further their cause. What is more
interesting for our present purposes, though, is that
only certain sorts of state will do. Power alone is not
enough; it must be power of a certain kind, based not
on the coercion of its own subjects but their support.
So the state in question will need to enjoy popular
consent (with the implications already indicated in
relation to sceptical liberalism), and will need to be
one in which public opinion supports the cause of
international human rights to the point of being
willing to send `our boys' to fight for them. Under
what circumstances are these conditions most likely
to be fulfilled? History offers an answer: the kind of
state most likely to function as the secular arm of
crusading liberalism is one in which liberalism of that
kind is perceived as a national mission, implying that
the state and people concerned have a special re-
70
sponsibility to act in defence of rights across the
world.
The experience that supports this claim goes back
nearly three hundred years. The role of global stan-
dard-bearer for what later came to be called liberal
principles was first played by Britain, in a secularisa-
tion of the responsibilities borne since the sixteenth
century by the Protestant English as God's chosen
50
people. We can see the political mission evolving
out of its more religious precursor in the poems of
Thomas Tickell, who wrote in the days of Queen
Anne:
Her guiltless glory just Britannia draws
From pure religion and impartial laws
Her labours are to plead the Almighty's cause
51
Her pride to teach the untamed Barbarian laws.
During the French Revolution Britain's mission as
the universal shield of liberty was challenged by
France. In the consciousness of the revolutionaries
who carried the banner of the Rights of Man across
Europe, their universal mission of liberation became
inextricably entangled with national pride. Interest-
ingly, this entanglement was not a foregone conclu-
sion, for at the start of the Revolution no one could
have been more critical of nationalism than the
Jacobins. Istvan Hont has shown, however, that
despite Robespierre's insistence that the Republic
`was acting only on behalf of the peoples of the
world . . . nothing was more striking than the
turning of Jacobin republicanism into the most dra-
52
matic form of national patriotism.' In our century a
similar fusion of power, patriotism and messianic
fervour has made the USA the favoured agent for
the projects of crusading liberalism from Woodrow
53
Wilson to the present.
Some may argue that the USA is peculiarly suited
to the task of defending universal human rights
precisely because it is not a nation, but instead a
polity united by subscription to universal liberal
principles. This case has been eloquently stated by
John Schaar, who defends American patriotism as
something quite different from the `parochial and
primitive fraternities of blood and race'. It is instead a
`covenanted patriotism', which is ethically superior
 THE SWORD OF FAITH AND THE SHIELD OF FEAR
to nationalism because it is `guided by and directed
toward the mission established in the founding
covenant' of the Republic. Being committed to
universal principles of liberty, equality and self-gov-
ernment, this kind of patriotism `is compatible with
the most generous humanism' and indeed `assigns
54
America a teaching mission among the nations.'
But this is a red herring. Nations vary a great deal
and are by no means exclusively defined in terms of
55
ethnic unity. What is much more crucial is the
collective consciousness that enables a nation to act
as a single body and, for all its internal divisions, the
sense of being American is something that the USA
has in abundance. That collective identity gives the
country a unique capacity to act on the world stage
and to mobilise power proportionate to its size and
resources. It is true that the USA's peculiar suit-
ability as a vehicle for the projects of crusading
liberalism comes from the conjunction of national
power with a universal mission. It may even be true,
as Schaar implies, that belief in human rights is part
of what it means to be an American. But crusaders
inclined to argue that this is a redeeming feature
should take note of the other side of the coin, which
is a proprietary attitude towards such beliefs. Liberal
principles tend to be experienced as American prin-
ciples ± seen by Americans as our principles, a
legitimate source of national pride. Reflecting on
the ideological quality of American identity, Benja-
min Barber remarks that `the American trick was to
use the fierce attachments of patriotic sentiment to
bond a people to high ideals. Our ``tribal'' sources
from which we derive our sense of national identity
are the Declaration of Independence, the Constitu-
tion and the Bill of Rights.' He admits, however, that
`cosmopolitanism too has its pathologies and can also
56
breed its own antiseptic version of imperialism.'
The dilemma for those who favour `humanitarian
intervention' by the `international community' is
that nations cast as the representatives of humanity
57
always have their own axes to grind.
The unpalatable truth facing crusading liberals, in
other words, is that their universal mission gains
maximum political clout when it is in effect captured
by a powerful nation and becomes part of that
nation's own identity and national pride. Such are
the contortions of national identity that nations may
71
even take a possessive national pride in overcoming
the limitations of nationalism, like the ingenious
nineteenth-century Frenchman who congratulated
his fellow countrymen on their cosmopolitanism: `it
is we who taught the nations of Europe to detach
themselves from a narrowly national ideal and to
58
march resolutely towards a human ideal.' A less
strident example of the same phenomenon is the
national pride Swedes feel in being (collectively) a
shining example of internationalism. While this is
(from a liberal point of view) an enormous improve-
ment on more obviously chauvinistic versions, it
remains a form of nationalism.
To sum up this section, then, in so far as crusading
liberals leave the realms of pure faith and consider
the conditions of political effectiveness, they must
realise that despite the real difficulty of reconciling
universalist humanitarianism with the exclusiveness
of national attachments, liberal nations with a strong
sense of their national mission as representatives of
the `world community' are their most useful allies.
Conclusion
I have argued that liberals are a great deal closer to
national enthusiasms than most of them like to
think, and that even those who are actively hostile
to the nation and all its works find it hard to do
without the political power it can offer. Nationhood
offers sceptical liberals their best chance of a polity
able to conduct its affairs according to the rule of law,
while the most effective support crusading liberals
can summon up in defence of human rights around
the world is a sense of national mission ± the US
Cavalry, in a manner of speaking. It might perhaps
be thought that this conclusion adds weight to the
arguments of the new `liberal nationalists' I men-
tioned at the beginning of the chapter, and that
liberals of all kinds should join them in celebration of
(some versions of) nationalism. But that would be
too hasty an inference, for the argument I have been
presenting is of a quite different kind. One way of
highlighting the difference is to say that the present
argument is concerned with contingent political
considerations to do with power and feasibility,
whereas defenders of liberal nationalism normally
argue their case on moral and philosophical grounds.
 MARGARET CANOVAN
The standard arguments for liberal nationalism try
to show that liberalism and nationalism are morally
compatible, and even that liberal principles entail
rights to national self-determination. These are not
positions defended here. On the contrary, the reluc-
tance of so many liberals to acknowledge their debts to
nationhood is entirely understandable, for the nation
makes an uncomfortable bedfellow for either of the
versions we have been looking at. Many nationalisms
are, after all, explicitly hostile to liberalism in any of its
forms; more crucially, even the most liberal version of
nationalism is too irrational and populist really to suit
the fastidious sceptics, and too narrow in its identity
and commitments for the visions of crusaders. Re-
membering the tensions (as well as the links) between
liberalism and nationalism is important, for two rea-
sons. Fittingly, one of these reasons recalls Oakeshott's
`politics of scepticism', the other his `politics of faith'.
The first is that both kinds of liberals need to come
to terms with the political cost of their aspirations.
The cost of the liberal freedoms valued by sceptics is
a greater or lesser degree of the nationalist enthu-
siasm they deplore. For crusaders, the costs of effec-
tiveness under present conditions are considerably
higher, amounting (if crusading aspirations were to
be seriously pursued) to a new imperialism to be
undertaken not by the `world community' but by the
United States. Whether or not the goal is worth such
a price is another matter, but the question deserves
59
more attention than it currently gets. But sceptical
`realism' about the genuine and serious political
Notes
1. Blease (1913), p. 12.
2. On mid-Victorian liberal enthusiasm for national self-
determination (and on tensions between this and the
Cobdenite tradition of a non-interventionist foreign
policy), see Bradley (1980), pp. 97±104, 133.
3. See, for example, Miller (1995), Tamir (1993), N.
MacCormick, `What Place for Nationalism?', in Ca-
ney, George and Jones (eds) (1996).
4. Friedrich Hayek's `Anglo-American' as opposed to
`continental' liberalism belongs to the wider family
of sceptical liberalism, as does Judith Shklar's `liberal-
ism of fear'.
5. Oakeshott (1962), pp. 1±36, 123±7.
6. Oakeshott (1975a), pp. 200±6, 313±16.
7. Oakeshott (1996), p. 21.
72
dilemmas that nations pose for liberals in the con-
temporary world is not the whole story. There is a
second reason for bearing in mind the tension as well
as the link between nationhood and liberalism.
I have argued that sceptics and crusaders alike rely
on the nation's capacity to mobilise and institutio-
60
nalise consensual power: power that can support
equal laws at home and provide a launching pad for
universalist projects. But although this dependence
on national power is important and too little no-
ticed, it is a matter of political contingency rather
than philosophical necessity. In principle, it may
some day turn out to be the case that consensual
power can be mobilised and institutionalised in other
ways. It may be significant that some of the inter-
national institutions launched from national bases in
the past half-century are showing signs of acquiring a
legitimacy that enables them to some extent to
mobilise power of their own. While deeply depen-
dent on nation-states (and especially on a few states
with strong national identities and political and
economic clout to match), these institutions are
already far from negligible, since they constrain
the activities of those same states. In the long term
(who knows?) crusading liberalism may free itself
from its national launching pad and build a home of
its own, a political space station able to circle the
globe in serene detachment from the blood and soil
61
of nationality. At any rate, as crusaders look for-
ward in faith to that day, sceptics are in no position
62
to declare it impossible.
8. Ibid., p. 23.
9. Ibid., p. 59.
10. Ibid., pp. 27±9.
11. Ibid., p. 32.
12. Ibid., p. 91.
13. Ibid., p. 128.
14. Ibid., p. 21.
15. Ibid., pp. 130±1. I have pursued some of the intima-
tions of that discussion in Canovan (1999).
16. Judith Shklar acutely observes that, whereas her own
`liberalism of fear' is prompted by `memory' ± memory
of the awful things that people have done to one
another ± the `liberalism of natural rights' is prompted
by `hope'; Shklar (1989), pp. 26±7.
17. See, for example, Renteln (1990).
 THE SWORD OF FAITH AND THE SHIELD OF FEAR
18. Cf. Donnelly (1989), Waldron (1993), pp. 23±4;
Bobbio (1996), pp. 12±14, 30, 61±70. The volume
edited by Martha Nussbaum and Amartya Sen (Nuss-
baum and Sen (eds), 1993) forms a particularly inter-
esting illustration because its contributors are acutely
aware of many of the objections to which this sort of
project is open, but nevertheless mostly endorse it.
19. The most conspicuous recent British defender of
sceptical liberalism has been John Gray; see, for ex-
ample, Gray (1993), especially pp. 314±20 (though, as
we shall see below, in his most recent work Gray seems
to have gone beyond liberalism altogether). See also
O'Sullivan (1989) and (1997), especially pp. 748±54.
20. Shklar (1989), p. 27.
21. Hayek (1967), p. 165; Shklar (1989), p. 37.
22. Sceptical liberalism does indeed share a border with
the more sceptical versions of conservatism, with a
number of notable political theorists (including Mon-
tesquieu and Tocqueville) inhabiting the disputed
territory. Even Hayek, who actually wrote an essay
entitled, `Why I am Not a Conservative', can be
plausibly treated as one in some contexts. See, for
example, Devigne (1994), p. 33. Oakeshott himself is
most often categorised as a conservative, which is
justified, but only with the qualification that he is
an unusual kind of `sceptical liberal conservative,
lacking both the emotional attachments of Burkean
conservatives and the faith in capitalism of the ``New
Right''.' Cf. Gray (1989), p. 199; Franco (1990). As
we shall later see, however, where nationalism is
concerned, there is a test that can distinguish between
the more liberal and the more conservative aspects.
23. For example, Gray (1993), pp. 326±7.
24. For a classic critique of this last belief, see Goodin
(1988).
25. See Donnelly (1989), pp. 119±20.
26. Shklar (1989), p. 36.
27. On the dangers of atavistic `tribal' loyalties, see Hayek
(1993), volume II, p. 143. In Hayek's view, nation-
alism shares with socialism the dubious distinction of
being one of `the two greatest threats to a free civilisa-
tion,'; ibid., volume III, p. 111.
28. John Gray, `The Politics of Cultural Diversity', in Gray
(1993), pp. 253±71.
29. Ibid., p. 270.
30. Hayek (1967), p. 161.
31. The classic view of nationalism from a sceptical con-
servative perspective can be found in Kedourie
(1993); see, for example, Kedourie's comments on
Mazzini at p. 92.
32. Gray (1993), p. 318.
33. Hayek (1993), volume I, chapter 4.
34. Oakeshott (1975a), p. 154.
35. Cf. Beetham (1991), p. 127. On the relation between
national identity and the `generalised trust' required by
modern civil society, see Seligman (1992), pp. 147±
73
82. Though strong, the connection is historically
contingent. Bruce Haddock has drawn my attention
to the counter-example of contemporary Italy, where
long-standing dissatisfaction with the performance of
the Italian state expresses itself in a popular support for
EU institutions that seems to indicate a wish not to be
ruled by one's `own' state. Might this be a harbinger of
a new form of empire-by-consent?
36. Helgerson (1992), p. 65, 81; Pocock (1957), pp. 30±
55.
37. Parekh (1995), pp. 139±40.
38. Gray (1993), p. 265; O'Sullivan (1997), p. 753.
39. John Gray seems recently to have recognised this
point, and has taken to describing his position as
`pluralist rather than liberal', seeing the appropriate
arrangement for deeply plural societies not as a suppo-
sedly impartial legalism but as a kind of millet system
with different laws for different communities related to
one another by a process of political bargaining; Gray
(1995a), pp. 126, 136±7.
40. Mill (1972), pp. 391±4.
41. Scruton (1990), p. 322; Miller (1995), pp. 90±3.
42. Gray (1995b), pp. 99±100.
43. Ibid., p. 117.
44. Hutchinson and Smith (eds) (1994), p. 4.
45. Ignatieff (1994), p. 189.
46. Consider, for example, Robert Goodin's assertion that
since we all have `general duties' to all other people,
regardless of nationhood or citizenship, `if some states
prove incapable of discharging their responsibilities
effectively, then they should either be reconstituted or
assisted.' The choice of passive verbs conveniently
dodges the tricky question of who is to do the `recon-
stituting'; Goodin (1988), p. 685.
47. Cf. Ramsbotham and Woodhouse (eds) (1996), Har-
riss (ed.) (1995).
48. For example, states that desire acceptance by Western
clubs such as the Council of Europe or the EU find
themselves under pressure to respect minority rights.
This point was drawn to my attention by Richard
Sakwa.
49. Bhikhu Parekh suggests that `morally concerned citi-
zens' from around the world might move into such
situations and `stage well-publicised and carefully
planned acts of a Gandhian type of non-violent
resistance', such as placing themselves between the
warring parties in an ethnic conflict. This is a heroic
idea, but could be effective only if it mobilised public
opinion inside powerful states to pressure those states
themselves to take action; Parekh (1997).
50. Greenfeld (1992), pp. 60±6.
51. `On the Prospects of Peace', in Tickell (1822), pp.
135±6.
52. Hont (1994), pp. 207, 219, 222.
53. See, for example, Shue (1980).
54. Schaar (1981), p. 293.
 MARGARET CANOVAN
55. On the nature and variability of nations, see Canovan
(1996), pp. 50±80.
56. B. Barber, `Constitutional Faith', in Cohen (ed.)
(1996), pp. 30±7, at pp. 32±3. Cf. C. Taylor, `Alter-
native Futures: Legitimacy, Identity and Alienation in
Late Twentieth Century Canada', in Cairns and Wil-
liams (eds) (1985), p. 215. As in the British and
French cases, secular American universalist national-
ism owes a lot to a pre-existing sense of religious
mission. See O'Brien (1988).
57. Parekh (1997), p. 58.
58. Quoted in Newman (1987), p. 49. For a more sophis-
ticated contemporary restatement of similar senti-
ments, see Schnapper (1994).
74
59. Cf. Miller (1995), pp. 77±8.
60. So successfully, indeed, that such power has come to
be taken for granted in nation-states, as if it existed by
nature. This point is more extensively discussed in
Canovan (1996).
61. For a vision of this kind, see Linklater (1998).
62. I have benefited from the criticisms and observations
of the other participants at the `Liberalism at the
Millennium' conference organised by the Department
of Political Theory and Government in the University
of Wales Swansea. I am also indebted to John Horton
for his characteristically incisive and helpful com-
ments.
 6
LIBERAL CITIZENSHIP
AND FEMINISM
Andrea Baumeister
Introduction
One of the most striking features of contemporary
feminist discourse has been the increasing disen-
chantment with traditional liberal conceptions of
citizenship. Whereas early reforming feminists ap-
pealed to the liberal principles of equality and uni-
versality in their attempts to include women in the
realm of citizenship, many contemporary feminists
have criticised liberalism for failing to acknowledge
the significance of difference and particularity. In the
1
eyes of critics such as Pateman and Young, the
traditional emphasis upon formal equality and uni-
versality encourages liberals to ignore the extent to
which our identities are irrevocably shaped by the
particularities and contingencies of our existence.
Thus, whereas liberal conceptions of citizenship
typically relegate difference, particularity and affec-
tivity to the non-political private sphere, many
contemporary feminists have sought to undermine
this liberal public/private distinction in their attempt
to assert the political significance of our unique and
particular characteristics.
This regard for difference and particularity has led
some feminists to be sceptical of all universal claims.
Thus writers such as Spelman, Harris, hooks and Flax
not only reject liberal universalism but also question
2
the validity of a general, unified women's politics.
For these feminist anti-essentialists gender is intrin-
sically bound up with other aspects of identity such
as class and race. In the face of such multifaceted
75
differences, which are also more fluid than fixed in
character, an appeal to a unified women's politics is
liable to prove just as exclusionary as the traditional
liberal preoccupation with universality. Yet, while
the dangers of exclusion are widely acknowledged, a
number of feminists have recently expressed concern
about the implications of such a rejection of all
universal claims and norms. Not only does the
anti-essentialist commitment to `radical otherness'
potentially undermine the idea of an effective fem-
inist politics, it also fails to recognise the important
role appeals to universality and formal equality have
historically played in the struggle by oppressed groups
to be included in the realm of citizenship. In the light
of these concerns feminists like Susan Moller Okin
and Martha Nussbaum have reasserted the impor-
tance of women's shared experiences and the con-
tinued relevance of universal standards in the
3
struggle for equality for women worldwide.
Faced with this lively debate among feminists, a
number of writers have attempted to strike a balance
between the claims of difference and the concerns
which have traditionally given rise to the liberal
emphasis on universality. One of the most influential
examples of such an approach is Iris Marion Young's
4
model of a `heterogeneous public'. Although Young
stresses that each social group has its own unique
experiences, histories and social life, which can
never be entirely understood or adequately repre-
sented by outsiders, she nonetheless recognises the
potential appeal of liberalism's commitment to uni-
 ANDREA BAUMEISTER
versality and the equality as a justification for the
inclusion of all in moral and social life. Yet, while
Young's model of a heterogeneous public overcomes
many of the problems associated with earlier feminist
reconceptualisations of citizenship, this chapter will
argue that her model of group representation remains
at odds with both her dynamic conception of the
social collective and her commitment to deliberative
democracy. Nonetheless, the manner in which
Young combines an emphasis on universality at
the point of participation with the recognition of
the claims of difference in the allocation of rights and
resources is, from a feminist perspective, rather at-
tractive.
However, an attempt to strike such a balance need
not imply a rejection of the liberal tradition. It is a
rich and complex body of thought and practice
which, in the final analysis, may not be as hostile
to diversity as is frequently assumed. Indeed, careful
reading reveals that questions of difference and
particularity play a significant role in the Kantian
strand of liberalism. The distinction Kant draws
between perfect and imperfect obligations and the
open-ended, context-responsive manner in which a
Kantian approach determines specific obligations
allows for the recognition of the complex relation-
ship between the public and the private and is
sensitive to the demands for more effective repre-
sentation on the part of traditionally marginalised
5
groups. These traditional Kantian concerns have
continued to inform the work of contemporary
6
Kantians such as Ju
È rgen Habermas. This strand of
liberalism, therefore, can provide a framework for
citizenship which potentially enables feminists to
resolve one of the most troubling tensions at the
heart of contemporary feminist theorising.
Women and Liberal Citizenship
In their battle for equal rights for women many early
feminists were clearly inspired by the liberal Enlight-
enment ideals of reason, progress, individual freedom
and equal rights. For example, the notion of uni-
versal reason, central to the thought of liberal En-
lightenment philosophers such as Kant, provided
early feminist thinkers with powerful arguments to
challenge those social conventions which excluded
76
and marginalised women. If, due to their capacity for
rational thought, all human beings deserve equal
rights and respect, then women, as beings capable of
reason, are surely entitled to the same rights and
privileges as men. Thus first-wave feminists, such as
7
Macaulay and Wollstonecraft, employed liberal no-
tions of universal reason and formal equality to show
that sexual differences should be seen as merely
contingent and, in the final analysis, inconsequen-
tial. Ultimately men and women shared the same
faculties and capacities. Women's apparent failure to
develop their rational faculties did not rest upon
deep-seated biological differences between men and
women but simply reflected the social norms and
pressures placed upon them. Not surprisingly, early
feminists regarded education as a vital stepping stone
in the campaign for the enfranchisement of women.
If granted access to equal education, women would
realise their rational faculties and develop the capa-
city for autonomous action. Once these goals had
been attained women could not reasonably be de-
nied the vote. After all if the notion of self-sover-
eignty is to be meaningful it must surely include the
right to self-governance in the political sphere. Yet,
while many first-wave feminists actively advocated
liberal ideals and endorsed the liberal vision of a
common citizenship, the attainment of equal legal
and political rights did not bring about the radical
improvement in the position of women first-wave
feminists had so confidently anticipated. In the light
of this failure many second-wave feminists have
questioned the values which underpin the liberal
notion of individual rights in general and the liberal
conception of citizenship in particular. They have
argued that these commitments blind liberals to the
significance of difference in their conception of both
individual and citizen.
At the level of the self, feminist theorists such as
Pateman have drawn attention to the manner in
which liberals have attempted to abstract the in-
dividual from all social, economic and biological
8
contingencies. Subsequently, the individual be-
comes disembedded and disembodied, giving rise
to a unitary vision of the self ± a self that is the
same for all humanity. Hence morality is equated
with impartiality, with recognising the claims of the
other who is just like oneself. Thus, we are invited to
 LIBERAL CITIZENSHIP AND FEMINISM
view one another as abstract, autonomous beings,
unencumbered by the particularities of our existence.
To act morally is to follow the norm of formal
equality enshrined in a system of justice based on
a network of formal rights and duties. However, for
many feminists such an account of the individual
ignores the extent to which our identity is irrevoc-
ably shaped by the particularities and contingencies
of our existence. Here, a number of feminists have
highlighted the impact our physical being has on our
identity. As Pateman notes, if the individual is to be
a universal figure, liberalism must ignore that `hu-
9
mankind has two bodies, female and male.'
however, glosses over the political implications of
women's bodily capacities and passions. For feminists
such as Pateman, the tendency to regard the ato-
mised individual as the norm has led liberal theorists
either to view women and children as `deviant' cases
or to ignore or deny the interdependence between
10
women and children. To many feminists, liberal-
ism's failure to acknowledge women's experiences
has given rise to a discourse in which what consti-
tutes a proper person, a true individual, is represen-
tative of men's experience and reasoning specifically,
not women's. Paradigmatically, the liberal individual
becomes synonymous with the independent, proper-
tied male head of a household.
Since this conception of the individual shapes
liberal notions of citizenship, the exclusion of the
perspective of all but one group has important
political consequences, not only for women, but also
for marginalised groups in general. Just as liberals
seek to abstract the individual from all social, eco-
nomic and biological contingencies, traditional lib-
eral conceptions of democratic citizenship require
the exclusion of particularity. For Young, liberal
citizenship is therefore characterised by an appeal
11
to an ideal of `generality'. Citizens are expected to
transcend their differences and act on the basis of the
common good or the `general will'. This appeal to
generality entails a sharp distinction between the
public and the private sphere. Here the public realm
is identified with normative reason and the pursuit of
shared interests, whereas difference, particularity and
affectivity are firmly relegated to the private non-
political sphere. Yet, just as the liberal notion of an
abstract, disembedded, disembodied self excludes
women's experiences, feminists fear that the liberal
appeal to `generality' and the subsequent public/
private distinction denies women an effective voice
within the political arena. In this context they have
drawn attention not only to the fact that historically
women have been assigned to the private sphere, but
also to the point that this genderisation of the public/
private distinction formed for many centuries the
basis for excluding women from citizenship. Women
were identified with the sphere of the particular and
the affective, with nurture, reproduction, love and
care, and, consequently, were seen as lacking in the
This, qualities required for public life. Yet, if particularity is
assigned to the non-political private sphere, then
once women enter the `male' public sphere the way
in which they differ from men is seen as deviating
12
from the norm. As Susan Mendus points out,
equality then becomes defined in terms of the re-
moval of women's `disadvantage' or `disability', with
`disadvantage' being determined by a model which is
intrinsically male.
In the eyes of critics such as Young, this margin-
alisation of women's voices in the public realm is
indicative of the limitations of the very ideal of
`generality'. Given that different groups have differ-
ent experiences, histories and perspectives on social
life, no one group can entirely understand the ex-
periences of another. Hence no one group can speak
for another. To adopt, in the face of such profound
diversity, a conception of citizenship based upon
formal equality and impartiality is merely to privilege
the dominant group. If all are given equal rights, but
no one can speak for the other, the interests of the
dominant group will prevail as the members of the
dominant group will be able `to assert their experi-
ences of and perspectives on social events as im-
13
partial and objective.' So not only does the denial
of difference allow privileged groups to ignore their
own group specificity, it also disadvantages groups
whose experience, culture and socialised capacities
differ from this allegedly neutral standard. Given
that, on Young's analysis, the ostensibly neutral
standards of the liberal public sphere merely reflect
the experiences of the dominant group ± white,
middle-class, western males ± the liberal appeal to
formal equality and impartiality prevents the mem-
bers of groups such as women, whose experiences and
77
 ANDREA BAUMEISTER
perspectives differ significantly from those of the
dominant group, from effectively participating in
the public sphere.
Radical Diversity
and the Dangers of Fragmentation
In order to remedy this marginalistion of women's
perspective many feminists have attempted to dis-
place the traditional liberal emphasis upon univers-
ality and formal equality with a notion of citizenship
which recognises the political significance of differ-
ence and particularity. Initially this led them to focus
on the differences between men and women in an
attempt to highlight the political significance of
women's traditional experiences as mothers and
carers. Writers such as Elshtain, Dinnerstein and
14
Ruddick argued that the virtues of love, empathy,
compassion and emotional sensitivity
with women's traditional role as mothers and carers
have potentially important political implications and
are indeed capable of laying the foundations for a
better society. However, while such attempts to stress
the political consequences of women's experiences
provide a forceful critique of liberal universalism and
the liberal public/private distinction, an increasing
number of contemporary feminists have argued that
such attempts to formulate a `women's politics' con-
tinue to endorse a unitary vision of individual iden-
tity and ultimately still fail to take
seriously.
This regard for the wider implications of difference
is informed not only by an increasing awareness of
the experiences of African-American
minority feminists, but reflects the general scepticism
regarding all universal claims typical of postmodern
thought. For writers like Spelman, Harris, hooks and
16
Flax, concepts such as `gender' and `woman' are
profoundly problematic unless they are `qualified by
and seen in the context of race, class, ethnicity and
17
other differences'. From this perspective the em-
phasis a `women's politics' places on the experience
and identity of `women as such' implies an unsus-
tainable essentialism. As Harris notes, the claim that
the `biological and social implications of mother-
hood shape the selfhood of all or most women'
upon two key assumptions: the supposition of a deep
unitary self that is relatively stable and unchanging
and the belief that, although there are significant
differences between men and women, this self is the
same for all women and for all men regardless of class,
race or sexual orientation. For anti-essentialists such
feminist essentialism is as unsustainable as earlier
liberal claims to universality: just as liberalism's fail-
ure to acknowledge the significance of gender differ-
ences led theorists to presume that the experience of
male adults can be taken as normative for humanity
at large, so the focus on `women as women' has given
rise to a discourse in which the experiences of
western, white, middle-class women have been con-
flated with the experiences of all women. Conse-
quently, just as liberal universalism defines difference
in terms of a deviation from a standard that is
essentially male, so feminist essentialism views dif-
ferences among women as a divergence from a
associated standard that is defined by the experiences of Wes-
15
tern, white, middle-class women. Hence in Spel-
man's opinion `the phrase ``as a woman'' is the
19
Trojan horse of feminist ethnocentrism.' In the
eyes of these writers, feminist essentialism treats the
experiences of women who are subject to multiple
forms of oppression as merely `addition' problems:
black women suffer from sexism plus racism, while
working-class women are oppressed by sexism plus
class structures. For Spelman and Harris such an
approach not only forcibly fragments the experiences
difference of black, poor and lesbian women, but also leads
feminists to presume that the oppression women face
`as women' is best identified by studying the position
of women who are not subject to other forms of
and other oppression. This privileges the experiences of wes-
tern, white, middle-class women. For example, char-
acterisations of the dominant feminine stereotype as
passive and dependent failed to recognise the ex-
periences of black women who have struggled against
images of matriarchy and sexual permissiveness. In a
similar vein the claim by many second-wave femin-
ists that the solution to women's oppression was to be
found in work outside the home ignored the experi-
ences of millions of women who had always worked
outside the home, but for whom work had been a far
from `liberating' experience.
18
rests Spelman and Harris conclude that instead of
holding onto unsustainable notions of a deep, uni-
78
 LIBERAL CITIZENSHIP AND FEMINISM
tary and stable women's self, feminists must recognise
that women are enmeshed in many and often contra-
dictory discourses of sexuality, race and class. There-
fore, in place of the objective, rational and universal
standpoint central to the liberal disembedded and
disembodied self, Spelman and Harris posit a vision
of a radically embedded, subjective self. From such a
perspective identities are seen as multiplicitous, con-
tingent and context-bound. Differences are always
relational not inherent. Thus `the social agent is
constituted by an assembly of subject positions that
can never be totally fixed in a closed system of
20
differences.' Since identity is always defined in a
specific context Á -vis
vis-a specific others, gender
should not only be viewed as a relational concept,
whose characteristics and attributes can only be
identified by comparing the situation of women with
that of men. The construction of gender attributes
will also vary according to race, class and nationality.
The task of feminist theorising is not to attempt to
construct essences, but to explore these contingent
relationships. Such an approach clearly has far-
reaching implications with regard to the political
sphere. Given that what constitutes a woman differs
across cultural and social contexts, a common wo-
men's perspective cannot simply be regarded as a
metaphysical given. On the contrary, `while some
women share some common interests and face some
common enemies, such commonalties are by no
means universal; rather they are interlaced with
21
differences, even with conflict.' Feminism, there-
fore, constitutes a complex network of different
strands, which link discourses of gender to those
of class, race, ethnicity and sexual orientation.
Rather than a unitary political movement, it is
therefore best regarded as the product of a complex,
shifting and often difficult `patchwork of overlapping
22
alliances'.
While this vision of deep and radical diversity has
been highly influential, not all second-wave femin-
ists have endorsed such an outright rejection of
liberal principles. Indeed a number of feminists have
expressed profound disquiet about the political im-
plications of a commitment to `radical otherness'.
The basic worry here is that in the absence of any
appeal to universally shared standards, an emphasis
upon radical diversity will give rise to a process of
79
fragmentation which may not only threaten the
viability of a wider feminist politics, but may ulti-
mately undermine the very notion of democratic
23
citizenship. As Young notes, whereas theorists such
as Spelman have tended to regard the categories of
class, race and nationality as relatively fixed, the
stability of these categories is just as doubtful as the
appropriateness of the category of `women'. One
possible response to this complex interaction be-
tween race, class, nationality and gender is to dis-
tinguish further between subcategories. Thus one
could, for example, subdivide working-class women
further according to race, religion, nationality, eth-
nicity, region and sexuality, or distinguish between
the different attributes of an African-American gen-
der in relation to African-American men on the one
hand and white men on the other. However, given
that `any category can be considered an arbitrary
unity', such a strategy ultimately gives rise to an
infinite regress which dissolves all `groups into in-
24
dividuals'. In the eyes of critics such a process of
fragmentation is liable to undermine the wider sym-
pathies essential to a shared sense of citizenship.
While cooperation for the common good and gen-
eral solidarity are widely regarded as important ele-
ments of democratic citizenship, a commitment to
`radical otherness' is liable to give rise to a focus on
25
sectional interests. Indeed, in the absence of shared
norms and standards, understanding and cooperation
across group lines is likely to prove difficult, if not
impossible. From the perspective of traditionally
marginalised groups this failure to generate wider
sympathies is potentially rather worrying. In the
absence of a general sense of solidarity, not only
may we lose any reason to hope that the dominant
groups within society could cultivate feelings of
obligation to take into account the perspective
and needs of less powerful groups, but the possibility
of solidarity between the latter may also recede.
Indeed, the process of differentiation could lead to
the fragmentation of wider political movements such
as feminism.
In the light of these fears a number of feminists
have reasserted the importance of women's shared
experiences and the continued relevance of universal
standards. In contrast to the anti-essentialist denial
of a `generalisable, identifiable and collectively
 ANDREA BAUMEISTER
26
shared experience of womanhood', feminists such
as Okin assert that sexism does constitute `an iden-
tifiable form of oppression many of whose effects are
27
felt by women regardless of race and class.' Thus,
she claims that, if the findings of Anglo-American
feminists about justice within the household are
compared with the burdens and benefits in poor
households, the injustices of gender are revealed
to be quite similar: `in both situations, women's
access to paid work is constrained both by sex
segregation in the workplace and the assumption
that women are ``naturally'' responsible for all or
28
most of the unpaid work of the household.' Not
only does feminist anti-essentialism fail to recognise
the significance of women's shared experiences, it
also underestimates the potential appeal of the lib-
eral universalist project. As Anne Phillips ob-
29
serves, historically the standards of impartiality
implied by the liberal notion of a common humanity
have been employed by many oppressed groups,
including women, in their struggle for equality. Thus
liberal notions of universality and formal equality
have enabled oppressed groups to show that the
manner in which they differ from the dominant
group ± be it in terms of gender, race or religion ±
does not constitute a legitimate ground for excluding
them from citizenship. For Okin and Martha Nuss-
30
baum, such an appeal to universal standards is
today still likely to be the most effective tool for
31
securing equality for women worldwide. A crucial
problem is that the victims of oppression and dom-
ination frequently not only lack the intellectual and
economic resources to challenge these injustices, but
have often internalised the very values which op-
press them. According to Nussbaum and Okin, then,
such deeply entrenched discrimination can only be
challenged effectively by an appeal to universal
standards. These can be employed to hold local
government, aid organisations and international
bodies such as the UN to account for their failure
to improve the position of women.
The `Heterogeneous Public'
This awareness of the potential strength of the
traditional liberal emphasis upon universality and
equality has been reflected in some of the most
80
influential contemporary feminist conceptions of
citizenship. For example, in Iris Marion Young's
model of `the heterogeneous public' the commitment
to difference and diversity is tempered by a recogni-
tion of the appeal of liberal values and preoccupa-
32
tions. Although Young stresses that each social
group has its own unique experiences, histories and
perspectives on social life, which can never be
entirely understood or adequately represented by
outsiders, she nonetheless retains a commitment to
universality and the equality of moral worth of all
persons. While the principle of equal moral worth
provides a powerful argument in favour of the in-
clusion of all members of society in social and
political life, the liberal commitment to formal pro-
cedural rules and basic rights safeguards minorities
against the whim of the majority by setting limits to
democratic deliberation and outcomes. However,
although Young appeals to the value of universality
at the point of inclusion, she stresses that `univers-
ality in the sense of the participation and inclusion of
everyone in moral and social life does not imply
universality in the sense of the adoption of a general
point of view that leaves behind particular affilia-
33
tions, feelings, commitments and desires.' For her,
the liberal ideal of `impartial reason' (the `general
point of view') fails to recognise the significance of
difference and particularity on at least three counts:
it insists that all situations are treated according to
the same moral rule and thus `denies the particularity
of situations'; it tries to `master or eliminate hetero-
geneity in the form of feeling' by abstracting the
individual from the needs, inclinations and feelings
which accompany particular situations; and, most
importantly, it reduces the `plurality of moral judge-
ments to one subjectivity' by abstracting the subject
34
from the particularities which individualise her.
Thus, while acknowledging the potential appeal of
liberal values and preoccupations, Young contends
that the scope and nature of these values must be
reconceptualised to account for the political claims
of difference and particularity. Since each group
differs with regard to its experiences, history and
perspective on social life, equality in terms of the
participation and inclusion of all groups requires a
specific set of rights for each group and for some
groups a more comprehensive system than for others.
 LIBERAL CITIZENSHIP AND FEMINISM
In place of the liberal emphasis upon formal equality,
Young proposes a differentiated citizenship which
guarantees `the effective recognition and representa-
tion of the distinct voices and perspectives of those
of its constituent groups that are oppressed or dis-
35
advantaged.' Since dominant groups within society
will always attempt to shape the public sphere in the
light of their experiences and perspective, tradition-
ally marginalised groups, such as women, the poor,
ethnic minorities and the old, should be granted
additional rights to group-specific representation at
the various levels of government, thereby increasing
their opportunities for political participation. Op-
pressed groups should be provided with the resources
to organise themselves, should be invited to analyse
and formulate social policy proposals and should
have the right to veto specific policies which affect
groups directly. Young contends that, given the
different experiences and perspectives of social
groups, such measures are vital to promote justice.
Thus group-specific representation for traditionally
marginalised groups ensures greater procedural fair-
ness in the setting of the public agenda, allows for the
recognition of all needs and interests in democratic
deliberations and adds to democratic accountability
by providing an `antidote to self-deceiving interest
36
masked as an impartial or general interest.'
Young's heterogeneous public clearly offers a so-
phisticated model for a feminist reconceptualisation
of citizenship which combines a recognition of the
political significance of difference and particularity
with a commitment to universality and equality of
moral worth at the point of inclusion. Yet, despite its
apparent strength, her model of group representation
remains at odds with both her dynamic conception
of social collectives and her commitment to delib-
erative democracy.
Young offers a fluid and multi-layered conception
of social collectives, which seek to acknowledge the
worries expressed by feminist anti-essentialists while
safeguarding the possibility of a women's politics
based upon collective action. Women should not,
in the first instance, be conceived of as a unified
social group but are best viewed as a series. While
groups are self-conscious, mutually acknowledging
collectives with a shared sense of purpose, a `series'
refers to `a social collective whose members are
81
unified passively by the objects around which their
actions are orientated or by the material effects of the
37
actions of the other.' While the members of a series
do not see themselves engaged in a common en-
terprise and hence do not identify with one another,
their actions are constrained by the same set of
material objects and collective habits which provide
the background for the actions of each member of
the series. For Young, the series of `women' is con-
stituted by the `rules, practices and assumptions of
institutionalised heterosexuality' and the accompa-
38
nying sexual division of labour. While these gender
structures both enable and constrain action, they do
not define it and different individuals may adopt a
whole range of strategies to deal with gender struc-
tures. Hence, although `no individual woman's
identity . . . will escape the markings of gender . . .
39
how gender marks her life is her own.'
On Young's account, it is this serialised existence
which provides the basis for women's collective
action. Groups of women are formed when women
take up and reconstruct `the gendered structures that
40
have passively unified them'. Such groups are
unlikely to incorporate all women, but are usually
socially, historically and culturally specific. Here she
is careful to distinguish her conception of a social
group from both interest and identity groups. Thus
social groups are neither merely an `aggregate or
41
association of persons who seek a particular goal',
nor are they the expression of an essential set of
characteristics. Rather, social groups define them-
selves in relation to one another. Since group iden-
tities are the product of this process of
differentiation, they are fluid and often shifting.
Group membership, therefore, does not signify a
shared set of attributes; it is the product of a sense
of identification and affinity. But, while social groups
are partially constitutive of their members' identities,
an individual is not wholly determined by her group
membership. In many ways she remains independent
from the group's identity and can transcend and
reject group membership. In a complex, plural so-
ciety this sense of fluidity is further reinforced by
multiple group membership, which ensures that
every social group has group differences cutting
across it.
However, while Young's fluid, multi-layered con-
 ANDREA BAUMEISTER
ception of social collectives may allay many of the
fears of anti-essentialists, her sophisticated under-
standing of collective identity and action does ap-
pear to be at odds with her notion of
representation. The notion of a guaranteed right
to group representation implies that groups can be
readily identified, have a stable membership and are
sufficiently homogeneous to be able to formulate a
group response. A system which guaranteed op-
pressed groups the right to representation would
hardly encourage them to recognise the contextual
nature and fluidity of their boundaries. On the
contrary, it would be in the interests of groups to
remain distinct and to attempt to build a loyal, fixed
membership. Therefore, group representation may
not only be divisive, but it could also encourage
precisely the `essentialism' Young is so keen to avoid.
The degree of division associated with group
representation will in part depend on the wider
mechanisms for democratic decision-making.
this context Young is keen to distinguish her model
of communicative democracy from interest-group
pluralism. Rather than simply assert their predeter-
mined interests, democratic deliberation requires
that `participants discuss together the issues and
come to a decision according to principles of jus-
43
tice.' Therefore the participants in democratic
discourse must express their demands not as wants
but as entitlements of justice. However, given the
specificity of group experience, the various partici-
pants will not always be able to recognise the claims
of others and a certain amount of conflict is inevi-
table. The manner in which such conflicts are to be
resolved will largely depend upon the nature and
character of the public standards that govern demo-
cratic deliberation. Yet here Young remains vague.
At times she suggests that these standards are local
and do not entail an appeal to impartiality and the
`general point of view'. Thus she contends that `the
rationality of norms can be grounded only by under-
standing them as the outcome of discussion includ-
44
ing all those who will be bound by them.'
these standards are strictly local, then once negotia-
tion has broken down the decision-making process
comes to a standstill. However, Young's own appeal
to universality at the level of participation indicates
that such conflicts should be resolved by an appeal to
principles which are not merely local. She herself
appears to recognise this when she maintains that
only those claims `are normatively valid which are
group generalisable in the sense that they can be recognised
without violating the rights of others or subjecting
45
them to domination.' This suggests that the stan-
dards by which such claims are to be assessed are not
merely local. Yet, although such an emphasis upon
`generalisability' would offer an effective route for the
resolution of group conflict, it entails a commitment
to impartiality that is at odds with the regard for
particularity and partiality central to the notion of
group representation. While Young could potentially
overcome this tension by basing her defence of group
representation upon an appeal to impartiality, this
would place her conception of a `differentiated citi-
zenship' back within the liberal tradition she rejects.
However, such a return to liberal ideas should
arguably neither alarm nor surprise feminists. Liberal-
42
In ism is a rich and complex tradition which in the final
analysis may not be as hostile to diversity as feminists
frequently assume. Here, the distinction Kant draws
between perfect and imperfect obligations and the
open-ended, context-responsive manner in which a
Kantian approach determines specific obligations are
of particular interest. These traditional Kantian con-
cerns have continued to inform the work of con-
temporary Kantians such as Habermas. His context-
responsive conception of impartiality allows for the
recognition of the complex relationship between the
public and the private and is sensitive to the de-
mands for more effective representation on the part
of traditionally marginalised groups.
Liberal Citizenship,
Difference and Particularity
From a Kantian perspective the imperative to treat
humanity as an `end' and never merely as a `means'
requires of us not only to abstain from infringing the
rights of others, but also to take, as far as possible, an
But, if active interest in the well-being of others. This
concern is central to the Kantian conception of
imperfect duties. Therefore, while perfect obligations,
such as non-coercion and non-deceit, address ques-
tions of justice, imperfect duties, like benevolence
and charity, focus upon the specific circumstances of
82
 LIBERAL CITIZENSHIP AND FEMINISM
46
particular individuals. For Kant, such imperfect
obligations are not supererogatory; they constitute
a vital aspect of a balanced moral approach. How-
ever, whereas perfect obligations can be discharged by
simply abstaining from certain actions, such as, for
instance, coercion or deceit and are, therefore, uni-
versally applicable, imperfect obligations require
time, presence, resources and an understanding of
the particular situation and circumstances of the
other. Consequently, imperfect obligations are con-
text-specific. Once the principle of universalisability
is combined with the recognition of particularity,
indifference to the specific circumstances and needs
of others is not permissible since it may constitute a
threat to the agency of the other.
This regard for difference and particularity is
further underlined by the context-responsive reading
of the categorical imperative advocated by modern
47
Kantians such as O'Neill and Herman. While the
categorical imperative aims to rule out fundamental
principles of action which could not consistently be
adopted by all, it does not apply to ancillary prin-
ciples. After all since the particular considerations
which inform our actions in particular circumstances
vary widely, the more specific principles upon which
human beings may act cannot be universally acted
upon. While the categorical imperative rules out
certain fundamental principles, it cannot, by itself,
discriminate among more specific principles of ac-
tion. Consequently, the adoption of ancillary mo-
tives which could not be consistently adopted by all
is not prohibited. Thus the categorical imperative
merely provides us with a criterion against which to
measure principles. It does not tell us which princi-
ples to select in specific situations and how to
implement them. To work out what an obligation
entails in a specific situation requires careful evalua-
tion of the various elements which influence the
situation. The main purpose of the categorical im-
perative, therefore, is to provide a starting-point for
this process of critical reflection.
Kant's concern with the specific circumstances of
particular others and context sensitivity in the ap-
plication of norms have continued to inform the
work of contemporary Kantians. Thus, for example,
in his recent work Ju
È rgen Habermas has argued that
if the autonomy of all citizens is to be protected, it is
83
important to ensure de facto rather than mere de jure
48
equality. While a formal system of rights may grant
citizens formal equality, equal rights and liberties can
be used differently and may impact upon citizens
differentially depending upon their circumstances.
Thus genuine equality can only be guaranteed if the
specific needs and circumstances of particular others
are taken into account. These considerations are
akin to the concerns Kant addresses via the notion
of imperfect duties. Furthermore, just as a context-
responsive reading of the categorical imperative
differentiates between ancillary and fundamental
principles of action, Habermas stresses the need to
distinguish between the formulation and the appli-
cation of norms. While norms can be justified by
reference to universal principles, the impartial jus-
tification of a norm requires careful attention to the
specific circumstances of a given situation. To apply
a norm impartially we must determine whether, in
the light of all relevant facts, a specific norm is the
most appropriate one in these circumstances.
Within the context of citizenship the sensitivity of
the Kantian tradition to specific circumstances and
particular needs provides a potentially powerful the-
oretical framework for voicing the concerns that have
preoccupied many contemporary feminists. For in-
stance, the distinction between perfect and imperfect
duties enables a Kantian approach to be sensitive to the
problems highlighted by feminists such as Young.
While the emphasis on our common humanity which
informs the Kantian notion of perfect duties provides a
powerful justification for the extension of citizenship
to all members of society and offers a clear set of
standards for resolving potential conflicts between
the demands of various citizens, the idea of imperfect
duties ensures that a Kantian conception of citizenship
will be sensitive to the specific needs and circum-
stances of particular citizens. From a Kantian perspec-
tive citizenship is not merely a question of granting
everyone the same formal rights. It also implies an
obligation to ensure that all individuals are in a posi-
tion to effectively fulfil their role as citizens. In the light
of these considerations Habermas (and other contem-
porary Kantians) have stressed the need for greater
political participation on the part of traditionally
49
marginalised or socially disadvantaged groups. Just
as imperfect obligations highlight human vulnerability
 ANDREA BAUMEISTER
and particularity, Habermas highlights the complex
relationship between the public and the private
sphere. Given that persons are individuated through
the processes of socialisation, personal autonomy can
only be safeguarded if the life-context within which an
individual's identity is formulated is also protected.
This can only be achieved if all citizens participate in
the formulation and interpretation of the rights and
norms that safeguard individual autonomy. After all,
unless all citizens participate in public debate and
express their specific needs, there is a real danger that
the specific needs of socially disadvantaged groups will
be misunderstood. This in turn may undermine the
capacity of such groups to pursue effectively their own
conception of the good life. Thus for Habermas gen-
uine equality between men and women cannot be
achieved unless women take part in the formulation of
the criteria which determine what kind of treatment is
to count as equal. Yet, since traditionally marginalised
groups may require support to organise themselves and
to effectively participate in the political process, Ha-
bermas acknowledges that some citizens may require
special cultural and social rights. However, in contrast
to Young, Habermas not only conceptualises such
cultural and social rights as individual rights, he clearly
grounds them in a liberal appeal to impartiality. There-
fore while, like Young, Habermas advocates a discur-
sive model of democracy, his approach avoids the
tensions inherent in Young's `heterogeneous public'.
Although modern Kantians such as Habermas
invoke the language of rights, feminists, in their
struggle to secure greater participation for women,
may find it advantageous to formulate their demands
in Kant's language of obligations. Since on Kant's
account those in power have a clear duty to assist
those less fortunate, an obligation-based approach
which firmly places the onus on those in positions of
influence to take all steps in their power to ensure
that every citizen is capable of effectively exercising
her citizenship is liable to strengthen the position of
traditionally marginalised groups such as women.
Notes
1. Pateman (1988); Young (1990).
2. Spelman (1988); Harris (1990); hooks (1981); Flax
(1995).
84
Rather than placing the burden on marginalised
groups to claim their rights, the language of obliga-
tions reminds the powerful of the moral requirement
to fulfil their duties. Consequently the powerful
cannot simply ignore the needs and aspirations of
traditionally marginalised groups. Furthermore, the
open-ended manner in which an obligation-based
approach establishes specific obligations is well-sui-
ted to capture the emphasis contemporary feminists
have placed on the complex and multiplicitous
nature of women's identities. Whereas a rights per-
spective requires the prior establishment of a set of
rights in order to function, an obligation-based
approach can allow for obligations to be established
as situations arise. All an obligation-based approach
needs is a formula, such as the categorical imperative,
which can be applied to specific situations in order to
establish which obligations flow from these situa-
tions. A Kantian perspective therefore recognises
that the nature and extent of the obligation to ensure
the effective exercise of citizenship can only be
established in response to the particular needs and
circumstances of given individuals.
The regard for particularity and human vulner-
ability already evident in Kant's work suggests that
this strand of liberal thought is potentially capable of
providing a framework for a richer and more ba-
lanced account of citizenship. Undoubtedly the re-
sponses to the claims of difference and diversity
advocated by contemporary Kantians remain proble-
matic. Habermas, for example, has quite rightly been
criticised for advocating a model of communication
which places an undue emphasis upon critical, ra-
tional argumentation, marginalising the speech pat-
terns of traditionally disadvantaged groups such as
50
women. Nonetheless, the manner in which such
an approach tempers its commitment to `universal-
ity' with a recognition of `diversity' suggests that such
an account of liberal citizenship may enable feminists
to resolve one of the most troubling tensions at the
heart of contemporary feminist theorising.
3. Okin (1994); M. Nussbaum, `Human Capabilities,
Female Human Beings', in Nussbaum and Glover
(eds) (1995), pp. 61±104.
 LIBERAL CITIZENSHIP AND FEMINISM
4. Young (1990).
5. Kant (1963b).
6. Habermas (1993).
7. Ferguson (1985); Wollstonecraft (1989). See, also, Kate
Soper's chapter in the present volume for a close analysis
of early feminism's relationship with the Enlightenment.
8. Pateman (1988).
9. Ibid., p. 8.
10. Frazer (1996).
11. Young (1989).
12. Mendus (1992).
13. Young (1989), p. 259.
14. Elshtain (1981); Dinnerstein (1976); Ruddick (1989).
15. Many of these early attempts to formulate a `politics of
maternal thinking' were influenced by Carol Gilligan's
work on moral development (Gilligan, 1982). For
Gilligan not only the self but also the other person
towards whom one is acting has to be viewed as
radically situated and particularised. Thus, the gen-
eralised other of liberal theory is replaced by the
notion of the concrete other.
16. Spelman (1988); Harris (1990); hooks (1981); Flax
(1995).
17. Okin (1994), p 6.
18. Harris (1990), p. 603.
19. Spelman (1988), p. 13.
20. Mouffe (1992), p. 28.
21. Fraser and Nicholson (1988), p. 391.
22. Ibid.
23. Young (1995). As Young notes, it is, for example,
simply misleading to assume that a working-class
woman's gendered experiences can only be properly
identified by comparing her situation to that of work-
ing-class men. After all, gendered experiences such as
sexual harassment cut across class lines.
24. Ibid., p. 195.
25. See, S. Mendus, `Time and Chance: Kantian Ethics
and Feminist Philosophy', Morrell Discussion Paper
(York: Department of Politics, University of York,
1990).
26. S. Benhabib and D. Cornell, cited in Okin (1994), p.
9.
27. Ibid., p. 7.
28. Ibid., p. 9.
29. Phillips (1993).
30. S. Okin, `Inequalities Between the Sexes in Different
Cultural Contexts', in Nussbaum and Glover (eds)
(1995), pp. 275±97; M. Nussbaum, `Human Capabil-
ities, Female Human Beings', in Nussbaum and Glover
(eds) (1995), pp. 61±104.
85
31. For Nussbaum the commitment to the `radical other-
ness' of different cultures has led some theorists to
defend or at least condone many traditional cultural
practices which systematically discriminate against
women. Here Nussbaum cites two vivid examples of
the approaches to which she objects. The first one
refers to an American economist who cites the ex-
tension of the idea that menstruating women pollute
the kitchen to the workplace as an instance of the
integration of the values that prevail in the workplace
with those that shape home life, an integration which
he regards as lacking in Western countries. The
second one refers to a French anthropologist who
`expresses regret that the introduction of smallpox
vaccination in India by the British eradicated the cult
of Sittala Devi, the goddess to whom one used to pray
in order to prevent smallpox,' (Nussbaum, `Human
Capabilities, Female Human Beings', in Nussbaum
and Glover (eds) (1995), p. 65).
32. Young (1990).
33. Ibid., p. 105.
34. Ibid., p. 100.
35. Ibid., p. 184.
36. Ibid., p.185.
37. Young (1995), p. 199.
38. Ibid., p. 204.
39. Ibid., p. 209.
40. Ibid., p. 210.
41. Young (1990), p. 186.
42. For an interesting discussion of the relationship be-
tween Young's commitment to group representation
and her model of communicative democracy see J.
Squires, `Group Representation, Deliberation and the
Displacement of Dichotomies', in M. Saward (ed.),
Innovations in Democratic Theory (London: Routledge,
forthcoming).
43. Young (1990), p. 190.
44. Ibid., p 116.
45. Ibid., p 107.
46. Kant (1963b).
47. O'Neill (1986); Herman (1985).
48. Habermas (1993).
49. Ibid.
50. See I. M. Young, `Communication and the Other:
Beyond Deliberative Democracy', in Benhabib (ed.)
(1996), pp. 120±35. Feminists such as Warnke have
also questioned whether it is possible to draw a sharp
a distinction between the application and formula-
tion of norms as Habermas advocates (see Warnke,
1995).
This page intentionally left blank
 Part Three
JUSTICE: IDENTITY
AND DISTRIBUTION
This page intentionally left blank
 7
LIBERALISM AND THE
POLITICS OF RECOGNITION
Jonathan Seglow
I
Liberals, it is said nowadays, are having trouble com-
ing to terms with the concept of recognition. They are
not sure what it is and hence do not know how to
evaluate it, whether to incorporate it into the liberal
canon or eschew it altogether. And this is odd because
recognition is always familiar to us. Or at any rate, less
glibly, the concept of recognition is not a difficult one.
It is the universal human need to have one's identity
esteemed and valued by others. It is a vital need
because in order to be an effective agent in a liberal
society, with a secure sense of one's own identity, you
require recognition ± positive affirmation ± from
others. Politically speaking, recognition is the need
to have one's group identity publicly appraised and
affirmed. But political recognition is in short supply
among some groups in liberal societies. Many mem-
bers of minority cultures, in particular, do not receive
the public recognition they deserve. And one reason
for that is a lacuna in liberal theory itself: it does not
recognise the importance of recognition in forming
individual and social identities. In this chapter, then, I
will argue for an agenda: that recognition should enter
the normative constellation of liberalism (of some
sort). I say `of some sort' because there are certainly
conflicts between recognition and orthodox liberal
values such as freedom, opportunity and impartiality.
But, as this volume amply demonstrates, the liberal
identity is itself contested so there is no reason that
conflicts between recognition and these values should
89
automatically be resolved in favour of the latter.
Rather, there must be some interpretive compromise
on both sides so that recognition can be incorporated
into the liberal canon.
II
We should begin by acquainting ourselves with the
kinds of claims to recognition actually made by
minorities, of different sorts, in a liberal society.
The sorts of cases I have in mind include the
following:
. Should the children of recent immigrants be
entitled to schooling in their mother tongue on
the grounds that without it their language
would swiftly die away, depriving the second
generation of an important sense of orientation
and identity?
. Should racist (`hate') speech against specific
minorities be outlawed on the grounds that it is
not just offensive but threatens their fairly
fragile public presence in a wider community
not of their making?
. Should a gay couple be entitled to a legally
recognised same-sex marriage on the grounds
that they should enjoy public recognition of
their mutual commitment just like a mixed
couple?
. Should religious minorities enjoy exemptions
and privileges with respect to dress codes at
 JONATHAN SEGLOW
work (for example, Sikh policemen who wear
turbans or Muslim girls who wear the chador in
place of a conventional school uniform)?
. Should public resources be directed towards
measures which benefit minority communities
and only them (such as building cultural or
religious centres)?
. Should there be quotas which make it easier for
under-represented groups such as ethnic mino-
rities and women to gain university places and
certain jobs?
. Should curricula in state schools be revised to
promote a positive image of minority commu-
1
nities' history and cultural achievements?
These cases are all quite different, and involve
`cultures' in the very broadest sense of the term to
include ethnic groups, religions and genders. None-
theless, at the risk of distorting some particular cases I
shall consider them all together as claims for recog-
nition made by `collectives' or `groups', or cultures in
this broad sense. Now all these recognition claims are
claims for justice too because each group suffers from
some sort of unchosen disadvantage and that is what
social justice aims to correct. They are not mere
wants as when, for example, I demand that you
recognise my `need' for a million pounds. However,
as claims of justice, they are all highly contentious
from a liberal perspective. For they are claims for
special treatment, for something extra in the way of
rights or resources which other people do not have.
This goes against the central liberal idea that the
state should be impartial among different citizens
whatever their particular identities and allegiances.
The impartial liberal state does offer respect for each
person's abstract moral worth, but for it to recognise
particular concrete identities appears to be a viola-
tion of impartiality. Put another way, the liberal state
allows people to be different, but why should it
actively sponsor such difference?
To address these questions we need first to ex-
amine the theory of social justice of which liberal
political theory typically avails itself. I shall sketch a
simplified version of the account of equal opportu-
nities developed by John Rawls and Ronald Dwor-
kin, and then we can assess whether it can address
2
these sorts of claims to recognition.
90
I mentioned just now that those who claim
recognition suffer from some sort of unchosen
disadvantage and this idea is crucial to the liberal
account of social justice. The crucial distinction is
between unchosen circumstances, which people are
not responsible for, and their own freely made
choices, which they are. While we quite properly
must bear the cost of our choices, we are not
responsible for the unchosen circumstances which
diminish our capacity to choose in the first place.
Some people in society ± the poor, the ill, the
handicapped, the unemployed, for example ± la-
bour under disadvantages which other members of
society do not face and which prevent them from
choosing how to live their lives. The basic prin-
ciple of the liberal theory of social justice is,
therefore, that unchosen circumstantial disadvan-
tages should be compensated for, and in a way that
enables each person to determine for herself how
she should live her life.
Two critical normative assumptions are made by
this liberal account. First, all human beings are
regarded (for the purposes of justice) as morally
identical choice-making agents. Looking beyond
the fact that someone is confined to a wheelchair,
for example, we see that she is a person who wants to
be able to determine how she lives her life. It is just
the bad luck of being disabled which prevents her
from doing this. The second assumption is that the
circumstantial disadvantages which people face ±
such as illness, handicap and poverty ± are all
contingent and not essential aspects of a person's
identity. By this I mean that people regard them-
selves fundamentally as choice-making agents. Very
often, this is evidenced by the fact that a person
would much rather be without the disabling aspect of
her identity. Hence the compensation individuals
receive in order to overcome their disadvantage has,
whatever its specific form, the same generic purpose:
to enable them to regain the capacity for being a
choice-making agent.
Can this account of social justice address the
claims for recognition I listed earlier? There are good
reasons for thinking it cannot. To begin with, the
kinds of people who make claims for recognition do
not regard themselves first and foremost as choice-
making agents. Rather it is a particular identity they
 LIBERALISM AND THE POLITICS OF RECOGNITION
want recognised. Immigrant ethnic groups who de-
mand a more positive public evaluation of their
culture, for example, want the culture to be recog-
nised, not their choice-making capacity. While they
are in a disadvantaged position because of their
cultural membership, they do not regard their culture
as a disabling circumstance which merits compensa-
tion like the wheelchair-bound person. On the
contrary, they want their culture positively affirmed
and promoted. More generally, the liberal theory of
social justice seems ill-equipped to deal with what we
might call `ideational' issues that relate to ethnicity
and culture, gender and sexual orientation, religion
and language, and so on. It seems better suited to
addressing material disadvantages such as poverty,
illness and handicap. Indeed, many ideational in-
justices may not appear to be injustices at all for the
liberal theory. Is it an injustice for the children of
recent immigrants not to be entitled to schooling in
their mother tongue, for example? Is it unjust (as
opposed to unfortunate) to have overwhelmingly
negative images of some minority ethnic group dis-
played in the media? It is hard to see how the liberal
theory can view these as injustices.
This leads us naturally to ask whether the idea-
tional realms of culture, ethnicity and so on are really
fit subjects for justice at all. If they were not, there
would be no need to extend liberal social justice to
accommodate them. But I think there are two very
good reasons which show that ideational issues are
injustices which liberalism must address. The first
reason, which I discuss in the next section, hinges on
the phenomenon of self-esteem. Drawing on Axel
Honneth's theory of recognition, I will argue that it
is only through having one's substantive traits and
attributes positively appraised by others, that one can
esteem them oneself, and only yet only through self-
esteem that one can become a properly functioning
agent. I then discuss in section IV how the public
culture in a liberal society systematically privileges
certain peoples' identities and derogates others. We
simply do live in societies that appraise some people's
identities more positively than others. This gives
some groups much greater collective esteem than
others which, since esteem is necessary for identity
and agency, is an injustice. In the final sections of the
chapter I will ask how liberalism might address this,
91
looking at Charles Taylor's essay, `The Politics of
3
Recognition'.
III
The notion of recognition takes as central a person's
own view of what she is about, her identity. Identity
involves taking a normative attitude towards one's
own substantive traits and abilities, what following
Taylor I will call an agent's self-interpretations.
These self-interpretations are basic and constitutive
of who I am: there is no prior autonomous self that
can assess and revise them. To assume otherwise just
pushes the question back a stage of what this auton-
omous self is. Thus as Taylor puts it, `to ask what a
person is, in abstraction from his or her self-inter-
pretations, is to ask a fundamentally misguided ques-
tion, one to which there couldn't in principle be an
4
answer.' Now one's self-interpretations are, in a
fundamental way, formed by others. I have the
self-interpretations that I do because I have inter-
nalised how others see me. The basic idea is that only
through others' recognition can one acquire an
identity. But of course if that were the only way
of acquiring an identity the individual person would
merely be a locus for external social forces. So we
must add a second, more individual, moment to
personal identity which appropriates the internalised
attitudes of others for its own purposes, and, no less
critical, looks outwards and takes attitudes towards
others. These two moments ± self and others ±
synthesise in the identity of a psychologically fully
formed person. Hence identity is fundamentally a
negotiated achievement. There is no pure self-making
(that ignores the need for others' recognition), but
neither are persons constituted by external social
5
forms. Others' attitudes shape my very self-inter-
pretations, but at the same time I can revise them
myself and project them outwards in the form of
attitudes to others which in turn help to shape them
and so on. From this complex and ever-shifting
matrix, different persons' identities are formed.
In his important recent book on recognition Axel
Honneth argues that there are three distinct dimen-
sions to the phenomenon: love, rights and solidar-
6
ity. The human infant who is loved by her parents
develops basic self-confidence; she has faith in her
 JONATHAN SEGLOW
ability to express her needs and desires without being
abandoned as a result. This capacity to trust one's
needs and desires is, Honneth argues, a precondition
for self-realisation in any human community. I shall
not discuss it here. When a person is recognised by
her community as a bearer of rights, by contrast, she
is recognised to share those universal features of
personhood, such as agency, rationality and respon-
sibility, which make her able to assert claims upon
others. Others' recognition help enable people to
become claimers of their own rights. Honneth calls
this self-respect. Finally, self-esteem connotes a posi-
tive relation to one's own substantive, concrete traits
and abilities. Against the background of a shared, yet
open and plural, value horizon, individuals recognise
each other as beings with positive traits and abilities
which enable them to contribute to the common
good. This final form of recognition enables indivi-
duals, operating in a nexus of approving others, to
realise their own concrete identities.
Honneth's central insight is that to become a
person with a fully formed, integrated, fulfilled iden-
tity requires recognition by others and this must
include respect, esteem and love. As he summarises
it:
It is only due to the cumulative acquisition of
basic self-confidence, of self-respect, and of self-
esteem ± provided, one after another, by those
three forms of recognition ± that a person can
come to see himself or herself, unconditionally, as
both an autonomous and an individuated being,
7 9
and to identify with his or her goals and desires.
These three forms of recognition are not an extra
bonus for agents who are already autonomous any-
way. Rather, they help create the self-interpretations
which constitute any person's identity. Thus:
The connection between the experience of re-
cognition and one's relation-to-self stems from
the intersubjective structure of personal identity.
The only way in which individuals are constituted
as persons is by learning to refer to themselves as,
from the perspective of an approving or encoura-
ging other, beings with certain positive traits or
8
abilities.
92
This quotation highlights the fact that we cannot
disassociate recognition as respect from the more
substantive recognition which brings self-esteem.
Both aspects (along with self-confidence) are neces-
sary to produce a person with a fully formed identity.
Substantive recognition of one's traits and abilities is
not simply a welcome endorsement of what you
think you are about. It is fundamentally necessary
to become a properly functioning agent capable of
forming intentions and pursuing projects in the first
place. It is not an intrusion upon one's autonomy,
but rather a precondition of it.
Let us consider the idea of self-esteem more
closely. It is a central self-interpretive attitude and
takes as its object the substantive empirical traits that
a person has. Someone with a healthy sense of self-
esteem regards her concrete identity as morally valu-
able, as something worth having and developing. As
a precondition for pursuing one's ends, an adequate
sense of self-esteem is a universal human need. Of
course, a crucial source of self-esteem is success in
one's own projects. If I do well as a painter or a
footballer, developing certain skills and attributes in
the process, I generally significantly raise my self-
esteem. But here I want to focus on the converse, the
way that self-esteem is a precondition, as much as a
result, of engagement in projects. For a person will
only engage in her own projects, developing her
traits and abilities, if she thinks they are things worth
developing, in other words if she has a degree of self-
esteem to begin with. As Honneth shows, this is not
the same as self-respect which liberal theory is more
familiar with. Self-respect involves recognition of
one's general capacity to make claims upon other
people. An agent with self-respect is confident that
she is entitled to pursue her own projects when
others disagree that she should do so. But this
confidence in one's abstract moral worth is quite
distinct from the value one places on one's own traits
and abilities, and projects as externalisations of those
abilities. Self-esteem involves a positive self-evalua-
tion of the particular person one is, self-respect that
one shares the inherent dignity of all rights-bearing
persons.
When we claim that a person with self-esteem
believes her own traits, abilities and projects are
valuable and worth having we are making a claim
 LIBERALISM AND THE POLITICS OF RECOGNITION
about the prior and independent value of those traits,
abilities and projects. The more immediate and less
significant decisions an agent takes are genuinely
matters of taste where we can say of a person that,
because she wills it, a goal is valuable for her. But in
order to summon the motivation to pursue longer-
term projects and goals, which are externalisations of
her particular abilities, a person must perceive that
they have independent value. I could not esteem
myself as a good parent or a good doctor, for example,
unless I believed that parenting and medicine were
morally valuable goals.
Some putatively valuable features of persons that
are significant sources of self-esteem are individual,
such as a talent at music or the virtue of generosity,
but many others have an irreducible collective di-
mension which refers to (for example) gender and
sexuality, ethnicity and religion, and linguistic and
national aspects of one's identity. These are all
features of persons who belong to the relevant group
and from which they take pride or, in our terms,
10
esteem. Of course, like other largely unchosen
aspects of identity these collective memberships
may be more or less salient in a person's life. But
for many people a thriving culture or religion, say, is
an important aspect of their self-interpreted identity
and will contribute significantly to their self-esteem.
On Honneth's view, that self-esteem will in turn
help fulfil an individual's own identity and develop
her capacity for agency. It follows, therefore, that if
people need recognition in order to develop their
identity and capacity for agency, they require it of
these collective aspects as well as of their more
individual traits and abilities. In other words, for
people who do identify with these collective groups,
a positive evaluation of them by others enables them
better to be agents, that is effective pursuers of their
own projects. This is true even where people are
respected as rights-bearing citizens by the rest of the
community. If ethnic minority X is generally re-
garded to be an uncivilised, backward tribal group,
for example, then members of X who identify with it
will suffer a lack of self-esteem and have demeaned
individual identities. If the general view of women is
that they should do nothing but raise children and
(to use Mill's phrase) `beautify and adorn the home'
then, independent of their legal status as citizens,
93
they will suffer a deficit in their practical capacity to
be agents in the public realm. In sum, the public
evaluative status of collective identities has a crucial
bearing on their members' self-esteem and in turn on
their ability to become properly functioning agents.
IV
I now want to consider the role of the public culture
in providing an orientating medium within which
citizens conduct their lives. Sometimes liberals write
as if individual choice and agency takes place merely
within a thin `framework' of rights and liberties with
no further relevant values. This may be an ideal
inspired by the `fact of pluralism', but as a descriptive
claim it is certainly incorrect. Even the most liberal
societies contain thick, substantive public cultures
which articulate a great many normative values and
ideals. Notwithstanding the fact of pluralism, liberal
citizens live by a set of shared ideas and values about
how to live that help bind and integrate their
community and serve as points of reference for
individual members making their own choices about
how to live. Because this culture is implicit and in
the background we do not always realise its great
power in orientating our ethical identities. But con-
sider the attitudes in this country to, say, the role of
the Christian religion in public life (including its
implications for free speech), the value and status of
the family compared to other modes of human
relationship, the importance of publicly funded
health care, our history of Empire and conquest,
the place of languages other than English in public
life, and more broadly the degree to which immi-
grants must adapt themselves to our way of life. I
hope that even this (almost) random list reveals the
extent to which our public life is substantially nor-
matively encoded. Shared values and attitudes,
transmitted through many sorts of social relation-
ships and amplified by the media, help maintain a
nation as an `imagined community'. Like all com-
munities, it takes a stand in moral space. The point is
that whatever scope people have for individual
choice in liberal societies they cannot but take their
bearings from the public moral consciousness ± the
public culture ± of the nation. Of course, this public
culture is not some hypostatic entity standing over
 JONATHAN SEGLOW
and above individual agents. It is woven from the
fabric of our own attitudes. My claim is simply that
its normative encoding means that a huge number of
values and attitudes are privileged from the start and
this unfailingly structures whatever choices people
are formally free to make. Political societies which
are conscious of themselves as such are integrated by
shared norms and values which cannot but have a
certain moral mass.
Public cultures are, of course, revisable and re-
vised. They change in an evolutionary way. But the
relevant distinction is between a conservative,
theistic or nationalistic society where the public
culture is seen to be an objective and unchangeable
moral system and a liberal society where individuals
can participate in the progressive redefinition of
their public culture. (I will return to the latter
ideal.) The relevant distinction is not between
the former ultra-communitarian ideal and a purely
procedural republic constituted merely by abstract
values of freedom and equality in which individuals
can choose to be whatever they like with no moral
`externalities'. A perfectly competitive free market
for values and ways of life is an ideal not yet realised
and perhaps unrealisable. (Analogously, consider
how supposedly value-neutral natural and social
scientific research works through dominant para-
digms which orientate practitioners' questions and
findings.) In sum, a liberal public culture is sub-
stantive though revisable over time. Norms or
values are not fixed and immutable but nonetheless
they seep into the crevices within which individuals
exercise freedom of choice.
V
When the theory of recognition is linked to the
sociological truth about the normative encoding of
the public culture we can more clearly see that
members of certain collectivities have a case in
justice for political recognition, and more specifically
the kinds of measures I sketched at the beginning.
The public culture contains a negative appraisal of
certain religious, ethnic, cultural and sexual identi-
ties which can lower the self-esteem of those for
whom that collective identity is important. That in
turn has a negative impact on their own capacity for
94
personal agency. A theory of justice which addresses
the problem of misrecognition, then, instead of
merely respecting people's abstract moral worth,
should focus directly on the public evaluative status
of significant collective identities. Policies of recog-
nition which raise that status should improve the
collective self-esteem of members by legal means and
by gradually shifting the attitudes of the public
culture.
As part of a liberal theory of justice, policies of
recognition must of course be justifiable not only to
the minority community which benefits from them
but also to the majority of citizens who do not. The
problem is that policies of recognition will very often
involve a redistribution of freedom and opportunity
between different groups, and those who lose out will
demand some justification. The sorts of redistribu-
tions I have in mind include affirmative action
policies which reduce the opportunity for those
who do not benefit from them to gain jobs and
places in education; restricting property ownership
in a territory to those whose ethnic homeland it is
(which erodes others' rights of movement); abridging
freedom of speech in order not to offend minority
religions; and more general redirections of resources
to fund minority activities (their own schools, cul-
tural events, etc.) which of course take away from
resources that could be spent on everyone.
The basic justification undergirding all these mea-
sures of recognition is an ideal of equality which aims
to achieve a rough parity of the social bases of self-
esteem, as expressed in law and the public culture, for
different social groups. Rather than set out and
defend how that justification would work in full,
which is beyond the scope of this chapter, I would
like here merely to explore one aspect of it, and that
is the issue of moral value. As I argued above, what
disadvantaged groups want recognised is the moral
value of their collective identities. And the majority
who do not benefit from the kinds of policies of
recognition I've just listed would be more willing to
assent to them if they too appreciate the value of the
way of life the policy is designed to benefit. Hence
the value question is at the heart of the justification
of a recognition-sensitive liberalism.
We might argue that only objectively valuable
collective identities merit official recognition. On
 LIBERALISM AND THE POLITICS OF RECOGNITION
this view, collective ways of life that bear great moral
value but which, nonetheless, are systematically
devalued by the public culture should be recognised
in the ways I've catalogued. Well, we might, but it
isn't promising because whatever moral consensus
exists in a political community about the value of
different ways of life is expressed by the public
culture and yet this, as we've seen, denigrates them.
The objective view would have to appeal to a more
abstract, ideal account of values in order to generate
the moral authority for recognition. But even sup-
posing this could be done ± which I doubt ± it would
suffer from a further drawback: it would begin to lose
touch with the self-interpretations of those groups
who it seeks to benefit. For it is only by positively
evaluating their identity as they perceive it that the
self-esteem of demeaned groups is raised; their self-
interpreted account of the value of their identity may
or may not have any connection with the objective
one. Conversely, then, we could embrace a subjec-
tive account. In this instance any collective whose
members believed that their way of life was insuffi-
ciently valued by the public culture would merit
11
official recognition. This account has the advan-
tage that it perfectly tracks the normative self-inter-
pretations of collectives. Indeed, the justification for
policies of recognition is assimilated precisely to their
subjective feeling. But the subjective account would
no longer be a justification for recognition: it would
commit the state positively to recognise debased,
harmful and valueless ways of life and would remove
the evolutionary power of what Kymlicka calls the
`cultural marketplace' in helping them to wither
12
away.
VI
What we need is an account of the value of collec-
tive identities that does justice to their self-inter-
pretations while at the same time provides a firm
moral basis for justifying policies of recognition.
With these criteria in mind I want in the remainder
of this chapter to examine Charles Taylor's influen-
tial essay `The Politics of Recognition'. Although he
does not use the term, Taylor presents a theory of the
justice of recognition grounded in a particular ac-
count of the value of collective identities. That value
95
is encapsulated in the notion of authenticity and we
would do well to analyse this first.
Authenticity is `the idea that each of us has an
original way of being human: each person has his or
13
her own ``measure'' '. Each person has their own
particular, original accent on the human condition.
Taylor encapsulates the idea of authenticity thus:
There is a certain way of being human that is my
way. I am called upon to live my life in this way
and not in imitation of anyone else's life. But this
notion gives a new importance to being true to
myself. If I am not, I miss the point of my life; I
14
miss what being human is for me.
The authentic person tries to construct for herself
a unique yet coherent and narratable life-project
15
rather than being an unthinking social conformist.
She will of course internalise others' normative
attitudes as her identity cannot but be part-formed
by them. But rather than being captive to these
attitudes she will use them as symbolic material to
construct an identity that is her own.
As a value claim, authenticity is the idea that
particular identities have value just because of their
particularity. They have objective value, but not
because they can be subsumed by a moral universal
(as when we claim that all persons have equal moral
worth). Rather, authentic identities possess exemp-
lary validity; they are valuable just because of their
16
authentic originality. Authenticity, in sum, wrests
moral value from the singular experience of the
subject.
Of course, Taylor's main concern is with cultural
authenticity and cultural recognition. Straight after
introducing the idea of personal authenticity, he goes
on to claim that collective identities such as cultures
also have their own particular way of worth or
measure. Commenting on Herder's notion of
authenticity, Taylor writes, `Herder applied his con-
ception of originality at two levels, not only to the
individual person . . . but also to the culture-bearing
people . . . Just like individuals, a Volk should be
17
true to itself, that is, its own culture.' However,
Taylor says little more to defend cultural authenticity
than this. The value of cultural authenticity is a
18
simple deduction from personal authenticity. The
 JONATHAN SEGLOW
main reason for this, I think, is that Taylor sees the
central constituent of cultures as language. People
who share a culture speak the same language and
thus language defines and demarcates different cul-
tures. Cultures, then, as formed and maintained by
different languages, have their own authentic col-
19
lective identities.
The norms and values of the public culture provide
the medium by which cultural members evaluate their
own collective identities. If the public culture system-
atically denigrates a minority culture then this de-
means its identity internally. Unfortunately, this is
20
very often just what happens. The example of
misrecognition which especially exercises Taylor is
his native province of Quebec. His question is how
Âbe
can the Que  cois maintain their own authentic
21
cultural identity? Taylor catalogues how the Quebec
government introduced various policies of recogni-
tion the most important of which was the requirement
that French-speaking parents send their children to
French-speaking schools. The reason for this was the
overriding good of survivance of the authentic cultural
identity of Quebec, constitutively defined by its being
French-speaking. Taylor does not himself employ the
concept of self-esteem which I borrowed from Hon-
neth. But we can certainly translate his politics of
recognition into those terms and say that the rationale
for promoting Quebec culture is precisely that it gives
Âbe
the Que  cois a secure sense of self-esteem, and in
turn identity and agency.
The question is whether Quebec culture does have
authentic moral value such that its claim for recog-
nition is justified when, clearly, members of other
collective identities (Anglophone Canadians, for
example) may not recognise the exemplary moral
value of Quebec culture. This takes us to Taylor's
larger concern which is when and how different
cultures should recognise each other's moral value
in general. And this is our concern too for we saw
how public cultures are normatively encoded so as to
privilege certain cultural identity. The problem is a
particularly acute one when the `cultural gap' be-
tween two groups is large ± as it can be, for example
with black and Asian cultures (whether immigrants
or indigenous peoples) who demand the recognition
from the white majority in Europe and North Amer-
ica when it is the latter's identity that is privileged.
96
We (white Europeans) simply lack the perspective
from which distortion-free judgements of the worth
of alien cultures can be made. For Taylor, no purely
impartial perspective exists, and the solution instead
is to construct a shared cross-cultural language from
which comparative judgements of the worth of
22
different cultures can be made. By entering into
dialogue with alien cultures, he argues, we enlarge
the moral horizon from which we make judgements
of worth and a more perspicuous understanding of
the other culture becomes available to us. For both
cultures, orientating beliefs which were deep yet
implicit are suddenly brought to consciousness as
they face the challenge of other values. What was
previously strange and unfamiliar becomes clearer as
new values reconfigure old attitudes and beliefs. The
result is a `fusion of horizons' where both sides
develop a new vocabulary of comparison from which
23
to make moral judgements of cultural worth.
In approaching other cultures we presume that,
unfamiliar though they are, their values have some
objective status. `As a presumption the claim is that
all human cultures that have animated whole socie-
ties over some considerable stretch of time have
24
something important to say to human beings.' For
It is reasonable to suppose that cultures that have
provided the horizon of meaning for large num-
bers of human beings, of diverse characters and
temperaments, over a long period of time ± that
have, in their words, articulated their sense of the
good, the holy, the admirable ± are almost certain
to have something that deserves our admiration
and respect even if it is accompanied by much
25
that we have to abhor and reject.
This, then, is the `starting hypothesis with which
we ought to approach the study of any other cul-
26
ture.' If the starting hypothesis is redeemed, so that
from our more enlightened perspective we see that a
culture does possess authentic moral value, its claim
to recognition is then justified.
VII
Taylor's essay has aroused much comment. The two
main arguments made against him are, first, that
 LIBERALISM AND THE POLITICS OF RECOGNITION
there is no such thing as an authentic cultural
identity since cultures are in fact internally plural
and diverse, and second, that the state promotion of
a particular cultural identity entailed by recognition
is unfair since it privileges one form of life at the
expense of others and hence derogates individual
27
freedom and opportunity. These two objections
are, of course, related. If there are no authentic
cultural essences then there is certainly no justifica-
tion for politically recognising one culture at the
expense of others.
I think Taylor can be rescued from these objec-
tions, although not without modifying his position
somewhat. I shall address the second problem and
then the first, but, before doing so, let me mention
briefly a particularly acute version of the second
argument made by K. Anthony Appiah in his com-
mentary on the essay. Appiah agrees that many
collective identities such as members of minority
ethnic cultures do suffer from systematic low collec-
tive esteem. Hence `one form of healing the self that
those who have these identities participate in is
learning to see these collective identities not as
sources of limitation and insult but as a valuable
28
part of what they centrally are.' But, he goes on,
some recipients of recognition might object that it is
not they who are receiving it, but just some attribute
of theirs ± first language, skin colour ± which they
invest with no special significance. There are many
people whose self-interpretations do not have a
central place for an authentic collective identity.
`It is at this point that someone who takes autonomy
seriously will ask whether we have not replaced one
29
kind of tyranny with another.'
The second ± `freedom' ± objection to Taylor
appears to take, therefore, two separate forms. One
is that other people's (non-beneficiaries) freedom
and opportunity is limited by policies of recognition;
the other is Appiah's subtle point that the benefi-
ciaries of political recognition have their identity as
autonomous agents misrecognised by being viewed
through a cultural lens when their fundamental self-
interpretation is simply as free agents.
I have already sketched the general justification
for policies of recognition which very often redis-
tribute freedom and opportunity. Beginning from a
fundamental commitment to the equality of persons
97
a politics of recognition seeks to achieve a rough
parity of the social bases of self-esteem for different
social groups (where self-esteem is crucially depen-
dent on how one's collective identity is appraised by
the public culture). Since a positive evaluation of
people's concrete collective identities affects their
very capacity for agency (the linkage we drew from
Honneth), the former merit recognition to bolster
their appraisive status in the public culture, provided
that, as we have been exploring, the collective
identity possesses exemplary moral value ± a judge-
ment that can only be made post facto after a dialogic
encounter. It is worth recalling too that the liberal
theory of social justice, which I sketched at the
beginning of this essay, also redistributes freedom
and opportunity, as when for example better off
people disproportionately finance the welfare state.
The justification there is that the state should pro-
vide roughly equal opportunity for each person to
exercise their capacities for choice and agency. The
justification for political recognition is very similar,
only that the primary focus is on the constitution and
maintenance of the capacity for personal agency in
the first place, and that the beneficiaries of recogni-
tion are people who esteem their own culture rather
than people whose self-identities are primarily as free
choosers. If we follow Honneth we should not see a
morally great distinction between `free choosers' and
`culturally-encumbered peoples', but whether that is
so must await a full justification of liberal recogni-
tion.
I'm not sure the misrecognition of those who
regard themselves primarily as free agents as cul-
ture-bearing ones is really a kind of `tyranny', as
Appiah claims. For it is arguable that a measure of
misrecognition ± as when, for example, ethnic min-
ority citizens complain that policies of affirmative
action denude them of the opportunity to compete
for jobs or university places on their own merits ± is
simply a price worth paying for the undoubted
benefits which recognition brings to those who do
identify with their collective cultural identity. Ap-
piah's critique might also be put in terms of the
public/private division so that the distinction is
between those who wish publicly to assert their
collective identity (the gay pride movement, for
example) and those who regard their identity (for
 JONATHAN SEGLOW
example, their sexual identity) as essentially a pri-
30
vate, non-political matter. But again, is it really
tyrannical to have ± to continue the example ± the
right to a same-sex marriage, or is it rather an
opportunity which gay couples can take advantage
of if they wish? I have written all along of `claims' for
recognition so perhaps the solution here is a kind of
consent test which would prevent the introduction
of policies of recognition which a majority of mem-
bers of the collective identity in question did not
themselves desire. Finally, it is also worth recording,
against Appiah's critique, that there are already
many other ways that the state misrecognises our
self-interpreted identities which arise from less con-
troversial principles of justice. Unemployment or
disability benefit, for example, was sometimes said
(by the New Right) to demean its recipients by
31
labelling them with a negative stereotype.
Although the argument was over-inflated ± since
the positive relief brought by these benefits vastly
outweighs the negative stereotype ± the point re-
mains a valid one. And if that is true then cultural
recognition of the kind Taylor commends does not
uniquely misidentify its subjects: Appiah's tyrannical
effects may be an unfortunate by-product of all
progressive policies of justice.
It remains true nonetheless that if there are no
authentic collective cultural identities then there is
no case in justice to recognise them. This is the first
objection made against Taylor. And it is also true
that, compared to the brute facticity of being poor or
disabled or unemployed (all collective identities),
cultures are always amalgams, mongrel entities, the
products of internal compromise and contestation
and, indeed, prevailing legal norms. There are no
pure, crystalline cultural identities waiting to be
recognised by the public culture. There are few (if
any) people in liberal societies who construct their
social role from a homogeneous cultural script and
nothing else. But in reply to this objection, we
should ask, simply: what needs to be true in order
that cultural identities merit recognition? The an-
swer is that, first, there are people who recognise
each other as members of a shared cultural identity,
and second, that that cultural identity is disadvan-
taged with respect to the public culture. The second
answer I have already argued for. The first answer is
98
consistent with the truths that (1) people want to be
recognised not just through their culture but often in
other ways besides, and (2) that there is disagreement
among cultural members about the proper interpre-
tation of their collective cultural identity. Problem
instances of (1) occur when a person is politically
recognised as being black, for example, when what
she regards as much more central to her identity is
that she is a doctor. This problem is a version of
Appiah's critique above, but it is worth pointing out
here that there is no obvious inconsistency or in-
justice for the state to recognise doctors (some of
whom are black) for the purposes of collective pay-
bargaining and to recognise, say, a black arts group
which receives government money (and of which
our doctor is not a member). Issue (2) is a problem
because it seems to militate against the authenticity of
cultures. Perhaps authenticity is the wrong term here
for we need only really say that cultures are real and
embody morally valuable human possibilities. In any
case, we must be careful not to go too far in the other
direction and implicitly contrast the inauthenticity
of minority cultures with the authenticity of the
public culture in a way that denies the justice of
claims to recognition. The latter is the hegemonic
strategy of cultural conservatives. In reality, neither
the public nor minority cultural identities are more
authentic or real than the other. Each borrows from
the other and both are constantly reformed and
revised. Demands for recognition made by members
of minority cultures are best understood as challenges
to the current self-interpretations of the public cul-
ture. For these reasons, I think Taylor's dialogue
between different cultures is best understood as
one about the best ± authentic ± self-understanding
of the public culture. This is given a recognition on
all sides that no group of citizens possesses the
authoritative and final interpretation of what their
public culture is about and that minority groups have
the right to participate in the progressive redefinition
32
of the public culture just like everyone else. On this
view dialogue is about reaching a public consensus
on the self-interpretation of the political community
and the value of cultural identities as part of that
political community. The moral challenges which
minority cultures inevitably bring to the public
culture will always provide a rationale for dialogue
 LIBERALISM AND THE POLITICS OF RECOGNITION
and I am suggesting that this process should be
formalised and institutionalised.
This more deliberative-democratic interpretation
of intercultural dialogue in fact supplies two rather
different justifications for rights to recognition. The
first says that official recognition is justified when
there is an appraisive consensus on cultural value.
But since, in a complex, plural society, where the
public culture is ever-revisable, such a consensus can
only ever be regarded as the ideal aim not the
concrete result of public deliberation, this first jus-
tification has its limits. The second justification
explicitly links claims for recognition with public
deliberation by regarding the rights granted to min-
ority cultures as participatory rights which enable
them to enter into collective deliberation with their
fellow citizens about the proper understanding of the
public culture on terms that are fair. Rights to
recognition, in other words, are recursively justified
by the need for public deliberation itself. They help
to foster a more secure sense of identity and status for
minority groups, not just for their own sake, but to
empower them to participate in the collective re-
definition of the public culture. The right to wear
religious dress at work, for example, or the right to
bilingual schooling do not just boost the self-esteem
of members of minority cultures. Rather, they help to
create citizens who feel secure and esteemed as
members of the public culture too and are therefore
better able to co-deliberate on what its identity
should be.
VIII
As I said at the beginning of this chapter my aim has
merely been to argue for an agenda: that claims for
recognition need not be illiberal but rather should
Notes
1. Cf. the lists of `recognition rights' in E. A. Galeotti,
`Neutrality and Recognition', in Bellamy and Hollis
(eds) (1999) and Kymlicka (1995), pp. 30±1.
2. For the theory of justice explaining and underlying the
choice/circumstances division, see Dworkin (1981a)
and Rawls (1972), esp. pp. 83±90, 100±8, and for a
discussion of Rawls and Dworkin, see Kymlicka (1990),
chapter 3. Kymlicka's well-known book Multicultural
99
become part of what a liberal theory of social justice
tries to accommodate. A full justification of liberal
recognition needs to elaborate further the key con-
cepts of recognition, identity, self-esteem, and the
public and minority cultures. I shall conclude briefly
by signalling three further issues which liberal re-
cognition needs to address. First, I have said nothing
about illiberal cultures which deny their members
freedom and choice. Illiberal cultures are especially
problematic because they may be significant sources
of self-esteem for (at least some of) their members.
Should they, then, merit political recognition? I
have argued only that the value of a culture is an
emergent property of dialogic interchange and that
harmful cultures do not merit recognition. But this
takes us to the second problem which is separatist
cultures who refuse to interact with the wider public
culture but demand political recognition (group
autonomy, for example, or separate schooling) none-
theless. Perhaps we should just say that entering into
some kind of dialogue with the wider world is the
price to be paid for political recognition (even if that
transforms a group who wished to use recognition to
insulate their collective identity in the first place).
The third issue is the complex relationship I have
glossed over between material social justice and
cultural recognition. As Nancy Fraser effectively
argues, these two forms of justice can conflict with
33
each other. Race-blind equal opportunities poli-
cies, for example, would appear to conflict with
recognition-based affirmative action policies that
positively esteem a marginalised racial group. Here
I will only say, banally, that a unified liberal theory of
social justice needs to reconcile both sorts of claims.
But my main point in this essay is not, I hope, banal:
recognition should be incorporated into the liberal
identity in the twenty-first century.
Citizenship (1995) offers a very sophisticated defence of
minority rights grounded in the Rawls/Dworkin account
of justice. However, Kymlicka does not view these as
rights which recognise the value of particular cultural
and collective identities. Unfortunately, there is not the
space to discuss Kymlicka's theory in this chapter. But
see the discussion in Seglow (1998), pp. 966±9 and the
secondary literature on Kymlicka cited there.
 JONATHAN SEGLOW
3. Taylor, `The Politics of Recognition', in Gutmann
(ed.) (1994), pp. 25±73. Taylor's is perhaps the best
known, but by no means the only, theory of cultural
recognition on the intellectual market. See, for ex-
ample, the important works by Tully (1995) and
Young (1990).
4. Taylor (1989), p. 34.
5. Hence the recognition view cross-cuts the (false)
dichotomy between the liberal autonomous self and
the communitarian encumbered one.
6. Honneth (1995).
7. Ibid., p. 169.
8. Ibid., p. 173.
9. Rawls, of course, lists self-respect as `the most impor-
tant' of his primary goods: see Rawls (1972), pp. 440±
6. For a brief critique of this see Seglow (1998), pp.
965±6. For a distinction between self-respect and self-
esteem similar to Honneth's see Sachs (1981).
10. Although Honneth has been criticised for glossing
over the more collective media of recognition such as
group identity ± see Alexander and Lara (1996), esp.
pp. 133±5.
11. I. M. Young's very interesting book Justice and the Politics
of Difference (1990) comes close to a wholly subjective
account. She writes, for example, that `groups cannot be
socially equal unless their specific experience, culture
and social contribution are publicly affirmed and re-
cognised' (p. 174) and she calls for a `politics [of
difference which] asserts that oppressed groups have
distinct cultures, experiences and perspectives on social
life with humanly positive meaning' (p. 166).
12. See Kymlicka (1995), pp. 107±9.
13. Taylor, in Gutmann (ed.) (1994), p. 30.
14. Ibid., p. 30.
15. This theme of self-sincerity links Taylor's account of
authenticity to that developed by another contem-
porary proponent, Alessandro Ferrara: see Ferrara,
(1998), chapter 1.
16. For more on the idea of exemplary validity, see ibid.,
chapters 1 and 4.
100
17. Taylor, in Gutmann (ed.) (1994), p. 31.
18. Cf. K. Anthony Appiah, `Identity, Authenticity, Sur-
vival: Multicultural Societies and Social Reproduc-
tion', in Gutmann (ed.) (1994), pp. 149±63, at p. 153.
19. Hence Taylor's focus is somewhat narrower than our
interest in the recognition of gender and sexual and
religious identities.
20. Taylor in Gutmann (ed.) (1994), pp. 32±5.
21. Taylor states that it `is axiomatic for Quebec govern-
ments that the survival and flourishing of French
culture in Quebec is a good': Taylor, in Gutmann
(ed.) (1994), p. 58.
22. For the philosophy behind this, see Taylor (1995).
23. A phrase Taylor borrows from Gadamer. See Gut-
mann (ed.) (1994), pp. 66±73.
24. Ibid., p. 66.
25. Ibid., pp. 72±3.
26. Ibid., pp. 66±7.
27. For examples of the first argument see Ferrara (1998),
p. 17; Rorty (1994); and S. Wolf's `Comment', in
Gutmann (ed.) (1994), pp. 75±85. The best examples
of the second are Cooke (1997) and Habermas's
`Struggles for Recognition in the Democratic Con-
stitutional State', in Gutmann (ed.) (1994), pp. 107±
48.
28. Appiah, in Gutmann (ed.) (1994), p. 161.
29. Ibid., pp. 162±3.
30. This example comes from Miller (1995), pp. 133±4
who argues for a democratic public culture but is
critical of Taylor's politics of recognition.
31. Another example is the way that labelling people as
criminals, by putting those charged with crime
through the judicial process, helps make them into
criminals.
32. A consequence of this interpretation of the public
culture as ever-evolving is that no rights to recogni-
tion are justified forever; they are instead revisable as
the self-understanding of the public culture changes
through public deliberation.
33. Fraser (1995).
 8
THE ESSENTIAL INDETERMINACY
OF RAWLS'S DIFFERENCE
PRINCIPLE
Rex Martin
I 1
John Rawls is probably the leading theorist of justice
in the world today (certainly the English-speaking
world). For him justice is, or should be, a virtue of
society, specifically of the political and social and
economic arrangements in its `basic structure'. In
Rawls's view, what justice requires of these basic
institutions, as a set, is not only that they exhibit (1)
the principle of equal basic liberties and (2) that of
equality of opportunity, but also (3) that they work
together in such a way as to (a) encourage contribu-
tions that (b) increase the production of goods and
services, which in turn are so distributed as (c) to
improve continually the level of income and wealth
of all the various income groups involved, when
considered from the perspective of a representative
person's whole life, a normal life.
We might, following Rawls, describe (1) equal
basic liberties and (2) equality of opportunity as prior
demands of justice and (3) the principle of mutual
benefit as a secondary demand, one that is to be
fulfilled in the context of fulfilling these prior de-
mands and without sacrifice by them. On this view,
then, in a just or well-ordered society, inequalities in
economic or social positions and attendant inequal-
ities in economic or social positions can be allowed ±
indeed, should be allowed ± subject to meeting these
three conditions. Since the third of the conditions
101
allows for differences in income and wealth Rawls
refers to it as the `difference principle'.
A society that met this particular set of standards ±
that is, which satisfied the goals set by equal basic
liberties, equality of opportunity and the difference
principle understood as mutual benefit ± would be
`thoroughly just', in Rawls's view. But to be `perfectly
just' it would also, at some point, have to maximise
the level of income and wealth of the least well-off
2
group in particular ± say, the bottom one-fifth.
This point about the maximising of the minimum
is not intended to identify a benefit for everyone in
the target group. Rather, the relevant effects here are
the effects on a `representative' person; they are
effects on an ideal-type average individual in the
3
bottom group.
In the present chapter I want to focus attention on
Rawls's difference principle. In order to avoid con-
fusion and distraction, I will specify this focal project
rather narrowly. Rawls is concerned principally to
describe normal or standard patterns of just distribu-
tion (without regard to prior maldistributions). In
other words, we are being asked to consider (at least
as a first stage, in order to see what picture would
result) what it would be like for a society simply to
satisfy the difference principle and the other de-
4
mands of justice.
Rawls typically states his difference principle as
being itself closely attached to the principle of
 REX MARTIN
equality of opportunity: the two together make up, in
5
his view, a single principle of justice. I want to
begin, then, by examining the reasoning Rawls offers
6
in support of this single but composite principle.
First, he establishes the importance of the idea of
equality of opportunity. We can assume here, for
purposes of exposition, that a group of people are
thinking seriously about the way societies are ar-
ranged respecting such matters as social and econom-
ic position and attendant income and wealth. They
note that, typically, there is considerable inequality
between persons and between classes in these mat-
ters. We can imagine now the following interior line
of thought going on in the minds of the people in our
group: each realises that inequality in position and
attendant wealth may well be involved for the
society they are in, but no one wishes to accept
an inferior position for oneself, even though it might
prove to be for the common good in a particular
society. Each person sees, then, upon reflection, that
it would be irrational from anyone's point of view
simply to take an inferior position as their lot in life.
Nor could anyone regard it as desirable to be forced
to take such a position. So, granting that inequality
in positions is both highly likely (perhaps inevitable)
and conceivably useful, each would want income-
and wealth-generating positions to be open to all on
some principle of equality of opportunity.
This establishes the point Rawls wants to make,
that equality of opportunity is important and that we
should want a principle specifying it. Then, next, in
a series of further reflections on this very point,
Rawls elaborates a set of ideas that, taken together,
constitute an argument for the composite (or sec-
ond) principle of justice.
There are, in my view, three main thoughts or
ideas advanced here. Let me sketch them briefly and
loosely, just to set the big picture in view.
The first thought is that no person is responsible
for that person's own starting point in life. The
second is that, in view of this first point, it makes
sense for people (acting as members of a single
political economy) to attempt to achieve a higher
index of wealth and income for all representative
persons at the various income levels there (higher,
that is, than existed at some initial point of presumed
equality). And the third thought is that, in doing
102
this, such a society should be concerned to reduce
the inequalities in income and wealth that will arise
between individuals and between groups of people.
We can treat these three thoughts as steps in a
single overall argument designed to justify the com-
posite principle of justice we are here concerned with
and to specify both a precise notion of equality of
opportunity and the content of the difference prin-
ciple. The argument we are examining, as Rawls
initially states it, is an `informal' or `intuitive' argu-
7
ment. After we've canvassed the argument further,
we'll look briefly at what it would take (in Rawls's
view) for it to become a fully formal and conclusive
argument.
The argument starts with the idea of `democratic'
equality of opportunity ± conceived as the taking of
remedial steps, conscientiously, to reduce the initial
inequality in advantages accruing to individuals from
two main sources. The idea that dominates this
particular discussion is that no one is responsible
for ± hence no one deserves ± their own starting
points, their natural endowment or the social posi-
tion into which they are born and reared. Since
equality of opportunity can never be strict or abso-
lute, Rawls introduces a further idea, the difference
principle, to complement equality of opportunity
and complete the initial line of argument. The
difference principle adds two more ideas, two further
remedial steps; it adds the principle of everyone's
continual benefit, which in turn is constrained by
the idea that, where there are several mutually
improving options available, we should choose that
option which most reduces the resultant inequality
in outcomes between the top-most and bottom-most
groups. The object of this three-step process is to
reduce, ideally to minimise, the gap between persons
by taking account of both starting points and end
8
results.
The idea of collective asset, the name Rawls gives
the argument just described, provides the rationale
and justification for the main features of Rawls's
second principle of justice. It underwrites both equal-
ity of opportunity in its `democratic interpretation'
and the principle of distributive economic justice ±
9
that is, the difference principle.
 INDETERMINACY OF RAWLS'S DIFFERENCE PRINCIPLE
II
We get to the usual specification of the difference
principle by reiterating the set of ideas just sketched.
That is, we do so (assuming here a continuing
conscientious effort at achieving `democratic' equal-
ity of opportunity) by repeatedly deploying the
principle of everyone's continual benefit as con-
strained by egalitarianism (thereby reducing differ-
ences in outcome between the top-most and bottom-
most group) up to the optimum point, where this
difference is absolutely minimised and the least well-
off are as well-off as they can be (without making any
other group worse-off in the process).
This particular iterated argument has a singular
virtue; it explains a puzzling fact about Rawls's
exposition of the second principle that holds
throughout his book A Theory of Justice (hereafter
TJ). Repeatedly, Rawls seems to shift between talk-
ing about the difference principle as improving every-
body's prospects and, alternatively, as maximising the
benefits of the least well-off group. But the puzzle
dissolves once we see that both these versions belong
to one and the same story. Each formulation specifies
a single feature of the overall argument; one for-
mulation emphasises the idea that drives the differ-
ence principle ± the notion of everyone's continual
benefit ± while the other emphasises the optimal
result, the goal or end point of the entire argument.
That goal can be stated in either of two distinctive
ways: (1) as absolutely minimising the difference
(measured in terms of income or wealth) between
the top-most and bottom-most group, consistent
with the realisation of everyone's continued better-
ment; or (2) as achieving `the greatest benefit of the
10
least advantaged', that is, the greatest benefit for
the least well-off group. Here we have a distinction
without a difference; the two goal-point formulations
11
(1) and (2) say the same thing.
And here I come to the heart of my analysis, the
main point I want to make. The difference principle
can be represented as an argument that proceeds
through a series of stages, each one of which embo-
dies a conscientious effort at achieving equality of
opportunity and each one of which repeats the same
theme: first satisfy the standard of mutual benefit (or
efficiency) and then reduce differences in income
103
between the top-most and the bottom-most group.
This iterated pattern continues until we reach an
optimum point, at which no further mutually im-
proving moves are possible. At this point we have
minimised absolutely the difference in question, and
those least well-off here have their greatest benefit.
The leading ideas of this argument were sum-
marised (at the end of the previous section) under
the name `the argument from collective asset'. Per-
haps calling it an argument may not convey the
exact picture I have in mind. Let us call it instead,
taking a hint from Rawls's later writings, a `procedure
12
of construction'. A procedure of construction is
more like a constructive proof in mathematics ± say,
a geometrical proof ± in which we get a result by
making certain moves (constructing certain figures,
making certain assumptions) than it is like a classical
syllogism in which certain propositions simply fol-
low, by logical entailment, from ready-made pre-
mises.
So the kind of argument I have in mind is itself
best understood as a procedure of construction. Here
a procedure is identified and then employed, by
reference to certain leading ideas (which give the
point of the procedure), to get a certain result; that
procedure is iterated, repeatedly deployed, until the
identified end point is reached. This point is the
intended or optimum result (given the iterated pro-
13
cedure and its governing assumptions).
What is curious about Rawls's informal (or `col-
lective asset') argument ± as we actually find it in TJ
(the heart of the discussion being at Chapter 2,
section 13) ± is that, while Rawls presents its gov-
erning concerns clearly enough (in what I have
called three main `thoughts') and while he presents
its goal point equally clearly (as the `greatest benefit
14
of the least well-off group' ), he never lays out the
intermediate step (the procedure of construction
itself) whereby, through repeated deployment, we
can get from these concerns to the goal point.
But once we actually `see' the procedure of con-
struction in action, as I have tried to represent it in
the previous section, the move from governing con-
cerns (as given in the three main thoughts) to goal
point seems natural enough and even plausible. One
point of the procedure is to show how in fact Rawls's
conclusion (for example, the maximin result) follows
 REX MARTIN
from his premises, the governing considerations (the
three main thoughts).
In the present account, then, I'm suggesting that
Rawls's argument from collective asset is best under-
stood as constituted by three distinct things. It
involves: (1) certain governing considerations which
together model or specify a procedure of construc-
tion; (2) the repeated deployment of that procedure
up to an identified end point; and (3) the actual
reaching of that point by using that procedure.
Is this argument, so understood, conclusive? I
think not. But it becomes logically conclusive if
we make certain simplifying assumptions.
We must assume, as does Rawls, that we start from
a hypothetical `zero point' of strict equality. And,
second, we must assume that chain connection (or
something like it) holds. Chain connection means,
roughly, that `if an advantage has the effect of raising
the expectations of the lowest position, it raises the
15
expectations of all positions in between.' Or, as
Rawls redescribes the same idea, so long as the `curve'
of the least well-off class shows improvement, all the
16
other classes do so as well. The point is that,
without assuming chain connection (or something
like it), we cannot guarantee ending up at the
optimum or goal point while deploying the three-
step collective asset argument and the procedure of
construction identified in it. But with it we can.
Chain connection is an idea Rawls himself uses
and, with qualification, endorses. The qualifications
have to do with whether chain connection holds in
actual societies or would hold invariably in a just
society modelled on the two principles. So under-
stood, chain connection is a description of actual
social relations. The difference principle is not lo-
gically contingent on whether this actual relation
17
holds or not.
However, we are not in the present account using
our variant of chain connection as a description of
actual social relations. Rather, it is like the assump-
tion of starting from a position of strict equality in
social primary goods. It is a simplifying assumption, a
regulatory constraint designed to keep the collective
asset argument (and the deployment of the proce-
dure of construction) in focus. The constraint re-
quires (1) that so long as the `curve' of the least well-
off group could possibly be higher it is also possible
104
that the curves of the other groups could be higher as
well; and that (2) so long as reciprocal benefit is
possible or feasible it is in fact sustained right on up
to the optimum or goal point.
Clearly, we are not talking here about actual social
relations in a certain pattern but about a `logical'
situation that must be realised whenever such a
pattern actually does hold. We might describe this
situation, then, as the `presupposition' behind chain
connection: the presupposition that, so long as the
benefit of the least well-off group could possibly be
higher, that of the other groups could also be higher,
right on up to the optimum or goal point.
With the presupposition (or, if you will, the
possibility) of chain connection, as we are using
the idea here (as a simplifying assumption belonging
to mathematics, not to social policy or, at least, not
to a description of social reality) in place, as one of
the background assumptions, we have completed our
account of Rawls's informal argument for the second
principle.
How would the argument fare as a formal argu-
ment? How would it do in the original position?
III
Rawls's contractarian method is very complex. I will
be able to mention only a few of its main features
here. One which is often emphasised ± and that
Rawls has continued to include even in his more
recent writings ± is that the `parties' to the contract
are placed, in the original position, behind a thick
`veil of ignorance'. Here they are instructed in their
subsequent reasoning to ignore their own particular
traits (traits that distinguish them from all or most
other people), to be unaware of (or, at least, to
ignore) their actual place in society, to be unaware
of their society's place in history or in institutional
evolution, and so on.
Thus extreme uncertainty (as to starting points
and outcomes for any given individual) would char-
acterise the deliberations in the original position
setting, in which individuals are called upon to
construct and then choose the principles of justice
that they would prefer to govern the society in which
they are to spend their entire lives.
Given this high degree of uncertainty, we find that
 INDETERMINACY OF RAWLS'S DIFFERENCE PRINCIPLE
Rawls's collective asset argument fares rather well.
For example, the transition from the idea that no-
body deserves their starting points to the idea that
people should use their natural endowments and
their social origins (where these things are advanta-
geous) in such a way that everybody benefits would
surely go more smoothly behind the veil of ignorance
than it would where people were already aware of
their own and others' natural endowments and social
origins. This transition would certainly carry more
conviction for the parties in the original position and
would, presumably, be emphatically endorsed by
them there.
And, for a second example, the mutual benefit
part of the collective asset argument would gain
strong endorsement there, especially if we assumed
a starting-point of strict equality. The argument
would go as follows: the parties would have no reason
to give up this equality in their choice of principles
unless there were benefits for each and all, or at least
18
for some of them (and no losses).
Let me make the same argument now in somewhat
different words. In the original position a certain
amount of role-playing is allowed, individuals are
allowed to assume certain standpoints and then to
consider how things would play out in the delibera-
tions of the parties. One could assume, for example,
that one was in a religious minority or in the least
19
well-off economic class. Under such an assumption
and behind the veil of ignorance, no one would
prefer disadvantageous deviations (from equal or even
from unequal shares) were they on the losing end,
and hence would veto such deviations. Thus, only
deviations advantageous to all would survive the
veto (that is, only such advantageous deviations could
achieve the unanimity required of conclusive delib-
erations in the original position construct). But it
would be rational, in the eyes of each, to allow for
mutually beneficial changes, where there were more
benefits for each and all (or at least for some) and no
losses.
It follows, too, where persons have an equal status
(as parties to the deliberation) and each has equal
claim to shares of primary goods, that the parties
would prefer a mutually beneficial outcome that
reduced the difference in income between the top-
most and bottom-most group over one that increased
105
that inequality. The idea is that even after mutual
benefit is assured one should continue to use equal
shares (of primary goods) as a standing constraint on
beneficial options, as a tiebreaker of sorts. Here,
among available options, that efficient and mutually
beneficial outcome which most reduces inequality is
to be selected. To do otherwise would be to opt for a
surplus of inequality, that is an inequality greater
than is required to achieve compensating reciprocal
benefit. It would be unsuitable, given the commit-
ment to equality (an initial benchmark and a con-
tinuing concern), to perpetuate or increase
inequality unnecessarily.
It is precisely where fundamental equality in start-
ing points is not fully and strictly achieved, or cannot
be, that concern for reducing the inequality of
resultant outcomes is in order. Thus, efficiency con-
strained by egalitarianism is a standard that each and
all could appeal to in the original position. None
would veto it. So the parties would endorse the
principle that, where two or more feasible efficient
schemes were available, that option should be se-
lected which most reduced the difference (when
measured in terms of income and wealth) between
20
the top-most and the bottom-most group.
In sum, I think the collective asset argument
would fare well in the original position construct.
And, since that construct is hypothetical and is
defined by a number of ideal features, it seems that
the simplifying assumptions mentioned earlier ± that
we start from a position of strict equality and that a
certain regularity condition (the possibility of chain
connection) holds good ± would be comfortably at
home there.
It important to see that Rawls nowhere repudiates
the informal argument, the argument from collective
asset. Indeed Rawls refers to TJ, section 13 (the very
heart of the informal argument), as the one that
21
defines the difference principle. And the most
prominent argument that we do find in TJ, chapter
3 (the chapter devoted to the original position), the
famous maximin argument, actually presupposes and
builds on the argument from collective asset.
We can put the line of reasoning in this well-
known argument quickly and intuitively as follows.
Behind the veil of ignorance and given the high
degree of uncertainty there, each individual thinks
 REX MARTIN
that, since they don't know how or where they might
end up, it behoves them all ± behoves each of them ±
to set things up (in the principles they select, each
one having a veto there) so that the worst control-
lable outcome (for any one of them) is the best of a
bad lot (the best, that is, of the set of worst out-
comes).
This line of reasoning, which has its home in
rational choice theory (and can be found in eco-
nomic theory and political science) is, as I've already
indicated, sometimes called maximin reasoning (that
is, reasoning literally on the principle of maximising
the minimum). The standard line of interpreting
Rawls's difference principle (at what I have called
its goal or optimum outcome) has been that it was
supported justificatorily and was rationalised by max-
imin reasoning. This maximin line of reasoning was
attributed to Rawls by his initial expositors (and
often, then, they severely criticised him for holding
to it) and more recent critics (such as Will Kymlicka,
Jonathan Wolff and Andrew Williams) have con-
tinued to view maximin as the standard pattern of
justification intended by Rawls for the difference
22
principle. This line has, in the view of most, been
encouraged by Rawls himself (and can be found
clearly enough, they think, in TJ and later works).
But, contrary to what most of his interpreters and
critics suppose, Rawls does not claim that his differ-
ence principle is itself supported specifically by max-
imin reasoning. At least, he makes no such claim in
his published writings. Rather, maximin reasoning
seems to have a secondary role in his account, as
developed in TJ and thereafter. It is brought in only
after his principles of justice have been formulated
and established on other grounds; it is then employed
by him to defend the whole set of these principles.
Here, the main use of the maximin argument in
Rawls's hands is to rule out or to constrain utilitarian
alternatives to his preferred principles of justice.
We can see the force of the contention just made
by noting a simple fact, that the outcomes, which the
maximin argument ranges over, must themselves be
generated somehow. They are in fact generated by
the competing principles under review ± by the two
principles, on the one hand, and by the principle of
average utility, on the other. It is these particular
outcomes, two sets of characteristic outcomes, which
106
the maximin argument then chooses between. It
follows that, in order for the maximin argument
to function at all, the principles which generate
these characteristic outcomes must themselves be
independently formulated and established, for other-
wise they could not give these well-defined results.
Thus the maximin procedure must actually presup-
pose these outcome-giving principles, presuppose
them as independent and presuppose them as
23
well-established.
The maximin argument provides no arguments for
the two principles per se, either for the first principle
(which we've merely taken for granted in this chap-
24
ter) or for the second (the arguments for which we
have already canvassed). Accordingly, it adds noth-
ing new to the already established argument for these
principles. Rather, it is an argument for preferring the
two principles, already established on other grounds,
over the most plausible of the utilitarian theories, the
principle of average utility (a principle already es-
tablished in Rawls's account, on its own grounds, at
TJ, section 27).
It is Rawls's view that utilitarians and others,
especially in the setting afforded by the original
position, would allow the sacrifice or attenuation
of some of the demands of justice, of the prior
demands especially, or would do so for some people
at least. Thus, the maximin argument doesn't even
focus on the second principle, let alone its difference
principle part. Instead, it focuses on the loss of equal
basic liberties of the sort enshrined in the first
25
principle. And it is this fact that marks the primary
ground, in Rawls's view, for preferring the two
26
principles over the principle of average utility.
So much, then, for the usual reading of the
maximin argument. It is based on a sort of optical
illusion: that maximin is the argument for the two
principles, in particular for the difference principle.
On the contrary, the maximin argument is not an
argument for the two principles per se; it neither
formulates nor establishes them. Rather, it presup-
poses the two principles and the results they char-
acteristically give and argues that these results are
preferable (under original position considerations,
where maximin reasoning is at home) to those
characteristically given by the principle of average
utility. Most especially, and for the same reasons, it is
 INDETERMINACY OF RAWLS'S DIFFERENCE PRINCIPLE
not an argument for the difference principle in
particular. That principle is hardly touched upon
at all in the maximin argument, as we find it in TJ,
Chapter 3. Indeed, the difference principle as such
doesn't have a face-off with the principle of average
utility until TJ, Chapter 5, and there (in section 49)
the argument has nothing to do with maximin.
Thus, we must look elsewhere for Rawls's argu-
ment for the second principle, in particular the
difference principle. In Rawls's theory, the argument
specifically and peculiarly designed for the purpose of
supporting the difference principle in particular is the
one based on the idea of collective asset; it is an
argument for reciprocal benefit of the sort that could
27
be developed in the original position. It is this
argument that Rawls later calls the `compelling' one
28
for the difference principle.
A natural reading of Rawls's difference principle,
along the lines set by collective asset, is that (assum-
ing throughout a continuing conscientious effort at
achieving `democratic' equality of opportunity) it
requires mutual improvement and that, where more
than one pattern of mutual improvement is possible,
society should select that one which minimises the
difference in income and wealth between the top-
most and the bottom-most group. A consistent
application of this principle would yield an optimum
or goal situation where the well-being of the least
well-off group (the bottom 20 per cent, say) was as
great as it could possibly be.
The `everybody benefits' principle (as qualified by
the egalitarian constraint) is fully compatible ± in
principle ± with the optimum or goal point of
maximising the well-being of the least well-off group.
Indeed, that principle (through the specification and
use of a procedure of construction) constitutes the
principal argument for reaching that very goal point.
Accordingly, there is always a determinate directive
that can be given to society by the Rawlsian differ-
ence principle: structure things via policies and
suchlike such that mutual improvement leading
ultimately to the greatest well-being of the least
well-off group is achieved.
IV
29
Now consider the problem, set forth in figure 8.1.
107
Fig 8.1
In the figure, x1 is the axis line that measures well-
being for the better-off group (in terms of units of
income, say) and x2 serves similarly as the axis for the
less well-off group. The dotted line (running at a 45 8
angle) is the equality line; it represents the points of
strict equality between the two groups. The space on
and below the curve 0P (correctly read as zero/P)
represents the available economic options in a par-
ticular society, given its resources, reasonable expec-
tations of development and so on (and subject, of
course, to the constraint of meeting the prior de-
mands of justice). The darkened line on 0P is the
Pareto-optimal `zone'; in it, no improvements for any
group are possible without worsening the prospects of
at least one other group. The point a, on the 0P
curve (and in the Pareto-optimal zone), is the point
of maximum well-being for x2 within the available
space.
Now let us suppose, first, that we were on or under
the curve 0a. From any given point within this space,
the mutual-benefit egalitarian solution (if we assume
certain regularity conditions) would intersect the 0P
curve at a (= the point of maximum benefit for the
least well-off group, the Rawlsian goal or optimum
point).
However, suppose next that we were on or under
the curve aP (but not at a or on the dotted line that
runs perpendicular from a to the x1 axis). Suppose,
for example, that we were at point j on the ub line;
here, the mutual-benefit egalitarian solution could
not intersect the 0P curve at a; rather, it would
intersect that curve somewhere in the segment aP
(but not at a).
The only way we could get to a, once we were to
 REX MARTIN
the right of the perpendicular, is to move in such a
way as (at some point) to reduce the well-being of the
members of the x1 class, thereby violating the re-
quirement of everyone's benefit. And this the mu-
tual-benefit solution, as I have portrayed it, would
not allow. Thus, the whole area to the right of the
perpendicular is a `forbidden zone' for that particular
solution.
Indeed, within that area no satisfactory result is
possible. There is no way the determinate directive
(described at the end of the previous section) could
be satisfied. One would have to choose either to
satisfy the mutual-benefit egalitarian solution or the
maximin one. The two are no longer compatible in
principle and hence, to realise the one, we would
have to give up on the other. Here we encounter the
essential indeterminacy of Rawls's difference princi-
ple.
It is possible that a society could find itself in this
zone (and to find itself there without prior injustice
having occurred). Imagine, for example, that a so-
ciety is following a mutual benefit path (for example,
the ub line in the figure). Such a society would be a
`thoroughly just' one, by Rawls's standards; its being
at j on that line would not be unjust, nor would it
reflect past injustice.
There are two rather obvious possible cases here.
First, a society could already be at j (on the ub line,
the line it had been following) when TJ is published
and the inhabitants (members of a thoroughly just
society up to then) start thinking things through in
the light of its arguments. Second, a society (again, a
thoroughly just one up to that point) wanders into
the `forbidden zone' through a typical and non-
culpable ignorance of its full set of economic options
(its own feasibility curve) at the time. It does so
while at the same time staying on the ub line (up to
point j). And we assume that the configuration of
the curve doesn't change significantly after that so as
to redefine the `forbidden zone' in the case at hand.
For any society in the situation just described, the
Rawlsian difference principle could not afford direc-
tion, could not specify a coherent goal for policy. In
short, the difference principle, on Rawls's usual
understanding of it, is here inherently indeterminate.
We can put the problematic now before us in two
distinct ways. We can view it as a theoretical pro-
108
blem or as a practical one. Let us turn first to the
theoretical problem. I will address the practical one
subsequently.
Collective asset (which includes the idea of mu-
tual benefit) is the argument for the difference prin-
ciple, in either its improving or its maximising
version (as we find these versions in TJ at pp. 60
and 83 respectively). This means that it is the
argument for the conclusion Rawls wants to draw
respecting the optimum or goal point, the ultimate
end of the whole process, so to speak. In the situation
we are here contemplating, the preferred conclusion
and the argument for it have come apart. They are
literally incompatible. This is a serious problem for a
theoretical, or ideal, argument to be in.
There are, of course, ways to avoid this problem.
And we are familiar with them from earlier discus-
sion (in section II). Thus, we conduct the argument
starting from a standpoint of strict equality and we
assume that the possibility of chain connection (as
there defined) holds throughout. The so-called for-
bidden zone, then, merely specifies a case where this
argument would not ± logically could not ± be
applicable. It identifies a case where the presupposi-
tion behind chain connection fails to hold and thus
identifies a zone in which the procedure of construc-
tion could not operate to effect its assigned end,
could not so operate because one of the essential
background assumptions for its proper use fails to
hold in that entire zone.
There is, then, a `theoretical' solution of sorts to
the theoretical problem. We achieve it by carefully
seeing to it that our argument conforms to certain
background assumptions; by doing so we necessarily
avoid the problem posed.
The theoretical solution envisioned here is ulti-
mately a very pricey one. It requires us to make some
rather strong assumptions: (1) that we start from
strict equality; (2) that the background presupposi-
tion (that chain connection is possible and thus
realisable) holds good; (3) that we can readily iden-
tify a society's feasibility curve and, on it, the point of
the greatest benefit for the least well-off group and,
thus, readily identify the zone where the presupposi-
tion (in (2)) holds and the zone where it doesn't ± so
that we can stay out of the latter. These are, as I said,
strong assumptions. Even so, we might admit them.
 INDETERMINACY OF RAWLS'S DIFFERENCE PRINCIPLE
But they may be more than mere assumptions. In
Rawls's account the `parties' in the `original position'
are invited to reason with certain assumptions at
hand. And they are invited to reason in earnest. For
people so to reason, I would suggest, they must
believe that these assumptions are applicable to their
own case.
Now take the third of the assumptions. If people
reasoned in the original position, with such an
assumption in place, then they could guarantee that
by staying in the zone where the background assump-
tion held good they could employ the procedure of
construction to reach the assigned end in view. But
so to reason suggests that they know (or believe) the
society they are here constructing (or evaluating)
itself occupies, or can occupy, precisely such a zone.
As a belief about social location this is, given the
veil of ignorance, not the kind of belief that could be
admitted into the original position. Clearly, if one
were to call such a belief a matter of knowledge, it
would be impermissible to entertain it in the original
position.
So we strike the offending belief out; we do not
allow it to operate at all. But then the parties are
forced to reason about matters of justice (matters of
justice for them) that may not even be applicable to
the society they inhabit, or may not be applicable to
its expected or allowable course of development
(allowable, that is, if a society is to be and remain
a thoroughly just one). Indeed, reasoning in this way
(reasoning without a well-founded belief as to sui-
table social location) may prove to be wholly un-
realistic.
By developing the principles of justice under ideal
assumptions we guarantee an impeccable theoretical
result. At the same time we guarantee, should the
assumptions fail to hold in fact, that the so-called
theoretical solution is inapplicable there.
A certain set of working assumptions has become
locked into the original-position argument pattern.
Where these assumptions cannot hold there must be,
for a certain zone of the suboptimal space, no the-
oretical solution available, not even in principle. We
have, in fact, made it systematically impossible to
solve the problem of what to do in that zone.
There will always be such a zone. Thus, the
situation envisioned could in principle arise for
109
any given society. Indeed, it could arise in fact for
many societies in the real social world.
Following good policies (policies of mutual benefit
designed to ensure `thorough' justice), even follow-
ing the best policies (policies of mutual benefit as
constrained by egalitarianism, policies designed to
assure, under a procedure of construction, the max-
imum benefit of the least well-off group), will not
guarantee that a society will not wander off course.
There is no guarantee that it will not, under real
conditions of uncertainty (and risk), enter a zone in
which the procedure of construction cannot operate
to effect its assigned end. The problem, in short,
30
could arise for a thoroughly just society.
A society that found itself in such a situation ± for
non-culpable reasons and without having committed
injustice ± would discover that, so long as the
situation could not be altered or could not realisti-
cally be expected to change in any significant way, it
could no longer follow a coherent set of policy
directives (as based on Rawls's collective asset idea)
for reaching Rawls's preferred goal point. It could
only act arbitrarily here ± adhering either to the
premises of the argument or to its conclusion but
not both. For the conditions under which the proce-
dure of construction could operate, so as to hold the
governing considerations and the end irrevocably
together, would not obtain here.
It is no longer possible, in such a situation, to
coordinate a mutual benefit-egalitarian solution with
the maximin one. To realise the one, we would have
to give up on the other.
This incompatibility in available strategies would
be a serious practical problem for any justice-aspiring
society to find itself in. There is no theoretical solu-
tion for the problem available to it, given Rawls's
argument or Rawls's text. And yet it is a problem that
requires to be resolved in a principled way.
This is the dilemma, the deep incoherence, that
Rawlsian theory necessarily leads to, or could lead to.
It is a situation that Rawlsian theory has not con-
templated, but one that we, as friends of justice and
perhaps of Rawlsian theory, need to think hard
about.
 REX MARTIN
A Postscript on Sources
and Acknowledgements
The present chapter draws on two main sources. First
it is based upon my book Rawls and Rights (Martin
(1985)), chapters 4, and 5 especially, and also chap-
ter 8. Chapter 5 is, in turn, a longer version of certain
sections of Martin and Shenoy (1983). I owe to
Prakash Shenoy both the idea for the efficiency-
cum-egalitarianism presentation of Rawls's differ-
ence principle and much of the formalism developed
for it in the present chapter and in Martin (1985). In
particular, I owe to him the idea that chain con-
nection (or, rather, something like it) must be
assumed in society with n classes (where n > 2).
And I owe to him the development of the egalitarian
matrix in which, in a situation of n classes, we
minimise (reduce as much as possible) the difference
between the top-most class x1 and the bottom-most
class x2. (Rawls's own analysis on this very point, we
may recall, was wholly confined to a two-class situa-
tion and did not take account of sub-optimal situa-
tions). Finally, the important appendix in Martin
(1985) is entirely his work. Secondly, this chapter is
also based upon my paper `Economic Justice: Con-
tractarianism and Rawls's Difference Principle', in
Boucher and Kelly (eds) (1994), pp. 245±66.
Let me mention as well two recent studies, both
Notes
1. See Rawls (1972), pp. 60±1 for the Rawlsian `first
statement' of the principles of justice, which I have
tried to convey in the paragraph to which this note is
attached.
2. See ibid., pp. 78±9, 319 for the distinction between
thoroughly/perfectly just. Rawls adds, in each of these
discussions, the proviso that the income expectations
of the various groups are to be understood as making a
positive functional contribution to one another, and
as doing so in some accredited setting ± in an open and
competitive market, say. Thus, in a `perfectly just'
situation the contribution is such that no further
improving `changes in the expectations of those better
off can improve the situation of those worse off' (ibid.,
p. 78). In a `thoroughly just' situation, on the other
hand, the key claim is that if the expectations of those
better off were somehow `decreased, the prospects of
the least advantaged would likewise fall' from their
present point (ibid., p. 78).
110
critical of arguments I have given in my works cited
above, which I have found helpful in the develop-
ment of the present chapter. These studies are
Lamont (1994) and Williams (1995). I cannot claim
to have answered all the criticisms made, especially
by Willliams, but these two studies did lead me to
significant reformulations and to some new lines of
thought, as sketched in the present chapter.
Finally, for their assistance on the present chapter
(or on earlier versions of it), I want to thank Prakash
Shenoy (with whom I've been discussing, on and off
since February 1994, the problem identified in sec-
tion IV) and Ann Cudd (with whom I've discussed
one of Williams's leading criticisms, that the effi-
ciency-cum-egalitarianism presentation of Rawls's
difference principle fails to satisfy the standard of
`contraction consistency' ± a criticism which I do not
address in the present chapter and one which I do
not credit, in the end). I have been helped too
(especially in developing the argument of section
IV) by students in my classes and by audiences, in
both the US and the UK, before whom I've pre-
sented earlier versions of this chapter. The present
chapter is a shortened version of one I published in a
volume of papers entitled Rights and Reason, edited by
Larry May and Marilyn Friedman for Kluwer (2000).
3. It is important to bear in mind throughout that the
difference principle is an aggregative principle; it
concerns groups (in the example given, the bottom
20 per cent of wage earners); it does not concern
individuals per se, i.e. people readily known to us by
their proper names (ibid., p. 78). The same could be
said of the other income groups (say, the top 20 per
cent and so on). Our focus is always on an ideal-type
average individual within that particular group, not on
any given, single individual who happens to be there.
4. Rawls makes clear that the account of distributive
shares that accords with the difference principle `be-
longs to strict compliance theory and so to the con-
sideration of the ideal scheme' (ibid., p. 315).
5. See, for example, ibid., pp. 60, 83, 302±3.
6. Rawls refers to this composite principle as his second
principle of justice at, for example, ibid., section 12;
the first principle, of course, is his principle of equal
basic liberties (ibid., pp. 60±1). And in his book there
 INDETERMINACY OF RAWLS'S DIFFERENCE PRINCIPLE
is a strict priority, of the first principle over the second:
ibid., pp. 61, 302±3.
7. Ibid., p. 78.
8. The argument for this final thought, that we should
manifest a continuing concern for minimising inequal-
ity, is a simple and appealing one. We apply the same
standard to end points (or outcomes) that we had
earlier applied, in the `democratic interpretation' of
the second principle, to starting-points. That is, we
begin by conscientiously reducing ± as we much as we
can (all things considered) ± the difference in starting
points that accrue to individuals from the two main
sources, from one's natural endowment and from one's
initial social position, and then we repeat the same
procedure with respect to outcomes. The rationale for
doing this is that, since we can never reduce the initial
differences to zero, it is permissible to reduce differ-
ences in outcomes as well. The idea of reducing
differences in outcomes between two or more feasible
options, available at a given time, is a natural com-
plement to the theme of equal opportunity with
respect to starting points with which we began. In-
deed, it is one and the same idea applied to both
starting points and outcomes. This thought, then, is a
fitting completion to the pattern already established in
Rawls's first two thoughts. The egalitarian motif of
`reducing' ± ideally of minimising ± differences in
income between the top-most and bottom-most group
is expressed most clearly in ibid., section 17 (esp. p.
104). See also Rawls's `Reply to Alexander and Mus-
grave', in Rawls (1999) at pp. 246±7, including n. 7;
and `Social Unity and Primary Goods' in ibid., at p.
374, n. 12.
9. Let me be a bit more precise here. Rawls calls the first
two ideas ± `democratic' equality of opportunity and
everyone's continual benefit, when taken together in
the way I have just described ± the idea of `collective
asset'. The folding of the third idea, of minimising
inequalities, into collective asset is, in my view, part of
Rawls's overall account, though less explicitly so than
the other two ideas. The development of, and main
argumentation for, the collective asset idea is found in
Rawls (1972), pp. 72±5, 101±4 (see also pp. 179, 523).
To this main argumentation should be added his paper
`The Basic Structure as Subject', in Goldman and Kim
(eds.) (1978), pp. 47±71, at sections 5, 7 and 8. (This
paper is reprinted as Lecture VII in Rawls (1996).) See
also Rawls (1999), pp. 246±8. For further discussion of
the relationship of the collective asset idea to indivi-
dual desert, see Martin (1985), chapter 2, section 2.
10. Rawls (1972), pp. 83, 302.
11. For a proof of this contention, see the appendix in
Martin (1985), pp. 197±201.
12. See Rawls (1996), Lecture III and, for the quoted
phrase, p. 93.
13. The procedure has many similarities to what Gabriel
111
Sandu called a `logic game' in his paper `Games in
Logic', presented at the conference on Logic and
Ontology at the University of the West, Timisoara,
Romania, in May 1998.
14. For example at Rawls (1972), p. 83.
15. Ibid., p. 80.
16. Ibid., pp. 81±3; see also Martin (1985), p. 199.
17. See Rawls (1972), p. 82.
18. Here I paraphrase Lamont (1994), p. 318. And La-
mont, in turn, was describing one argument, which he
thought successful, from Martin (1985) and my `Eco-
nomic Justice' in Boucher and Kelly (eds) (1994).
Rawls represents the reasoning of the parties in the
original position as `part of the theory of rational
choice' (Rawls, 1972, p. 172). The parties there are
taken to be rational (ibid., p. 25) and to lack envy
(ibid., p. 143). Thus, under the assumption that `more
is preferred over less', the parties are simply concerned
to ensure a suitable ensemble of primary goods (for
themselves individually and their descendants), in the
social world they will inhabit, under a rule that they all
agree to and each could have vetoed (ibid., pp. 263,
270).
19. Ibid., pp. 96±9 and section 33.
20. Some of Rawls's clearest statements of this line of
argument, stressing (as a consideration in the original
position) the continuing importance of the bench-
mark of equality in a condition of fundamental in-
equality in starting points, is found in his `A Kantian
Conception of Equality', Rawls (1999) at pp. 262±4;
`Reply to Alexander and Musgrave' in ibid., at pp.
246±7; `Some Reasons for the Maximin Criterion' in
ibid., at pp. 230±1; and `Distributive Justice: Some
Addenda' in ibid., at p. 165. Thus I think Pat Shaw is
wrong to suggest that the difference principle is un-
concerned with equality: Shaw (1992), p. 76.
21. Rawls (1999), p. 594 n. 55.
22. For Kymlicka, see Kymlicka (1990), pp. 61±6. For
Wolff, see Wolff (1996), pp. 168±87. See also Wil-
liams (1995), section III.
23. A further point. Rawls says that certain considerations
make the maximin argument appropriate and render it
usable. The three most important of these are given in
Rawls (1972), p. 154; the latter two could not be
convincingly stated and invoked without relying upon
background arguments, of the sort already familiar
from the present chapter. For example, one of the
considerations Rawls cites is that `the rejected alter-
natives have outcomes that one can hardly accept.'
But we could have no clear idea of what might be
acceptable (or not) without relying upon the very
arguments that would be made in favour of the two
principles in the first place. Maximin, in sum, does not
provide the background argumentation. That argu-
mentation must, then, be developed elsewhere; it is,
specifically, in the argumentation for the two princi-
 REX MARTIN
ples, and in inferences from these arguments. Maximin
presupposes these arguments and could not work
without them. Thus, a `straight' maximin argument
for the second principle, by-passing the existing argu-
ment from collective asset, simply could not be devel-
oped.
24. See Rawls (1972), pp. 60±1 and 302. Rawls signifi-
cantly recasts the first principle in his later writings ±
see Rawls (1996), p. 291 (in a formulation that dates
from 1982) and p. 5 (in a formulation that dates from
1993).
25. Rawls (1972), pp. 156, 158±9; see also section 49.
26. John Harsanyi, among others, has criticised Rawls's
argument here. His point is that the principle of
maximising average utility would fare as well as Raw-
ls's two principles under any sort of maximin test; see
Harsanyi (1975). D. W. Haslett has argued something
similar with respect to the principle of total utility;
specifically, he has argued that the principle of max-
imising total utility would fare better under any sort of
maximin test than Rawls's difference principle, at least
as regards the distribution of income and wealth ± see
Haslett (1985). I do not find Haslett's argument
convincing. Haslett fails to see that Rawls's main
focus, in the maximin test, is on liberties, not on
income and wealth. Even on the latter point, Haslett
fails to see that the judgement at each stage, as to
which to select, is made from a standpoint that
incorporates results from the previous stages; Haslett,
on the contrary, treats each sequential choice simply
on its own, without regard to where things stand as the
result of prior choices. Finally, in Haslett's own sample
matrix, if the scheme of selection were the one
advocated in the present chapter, where everybody's
cumulative benefit is constrained at each stage by
egalitarian considerations, then the worst-off class
would end with a higher index of income and wealth
than that afforded by total utility and the resultant
112
total wealth (of all classes) would be closely compar-
able with the result given by classical utility theory.
27. Rawls is quite explicit on this point; see Rawls (1999),
pp. 165, 246±7.
28. See ibid., p. 230; see also p. 226.
29. This figure is an adaptation of figures Rawls himself
uses; see Rawls (1972), pp. 76±7, 81. Rawls's justi-
fication of inequality (in income, wealth, social
position) presupposes that representative members
of the least well-off group are better off (or at least,
no worse off) than they would be under a `hypothe-
tical initial arrangement in which all the social
primary goods are equally distributed', which would
include that `income and wealth are evenly shared':
ibid., p. 62. This standard is met by all points in the
so-called available space in the figure. If more than
two classes were involved, we would have to repre-
sent their situation in a more complex way. I pre-
sume, though, that no significant difficulty is posed
on this particular score for my account of the figure
drawn here.
30. One thing we mean when we say that Rawls's theory is
ideal is that it is strictly complied with. Its theoretical
strictures are fully met. Thus, we would have strict
compliance if a strategy of everybody's benefit, as
constrained by egalitarianism, was consistently fol-
lowed. The stipulation here is not that the strategy
has been optimally followed (`in the best way' or with
the best result); rather, it is merely required that the
strategy be completely followed in all cases. My point,
then, is that even if this were done, a society could still
end up in the forbidden zone. A strict compliance
strategy would not preclude this. A society that so
acted would not be unjust. Thus, even if the principles
of justice are strictly complied with, a problem of
essential indeterminacy could still arise. The proble-
matic I am describing, then, is a real one for Rawls's
theory.

RAWLSIAN THEORY,
CONTEMPORARY MARXISM AND
THE DIFFERENCE PRINCIPLE
Rodney G. Peffer
In this chapter, my overall goals are to put the case
for a Marxist politics today and to proclaim the
utility to that case of Rawlsian theory ± which is,
of course, contemporary liberalism's most influential
contribution to the venerable debate about what
`justice' requires. In the history of modern progressive
political thought, socialists have persistently
tacked liberalism for what they perceive to be serious
deficiencies in their theorising of justice. Contrary to
what many people appear to have concluded in the
wake of Soviet-block communism's collapse, I do not
believe that the socialist challenge has been at all
discredited. Indeed, I continue to defend the overall
Marxist position I first elaborated in my 1990 book
Marxism, Morality and Social Justice. However, a
major theme of that work was that Rawls's theory
of justice need only be modified to yield the position
I wished to defend. In other words, there need be no
great moral gap between at least some versions of
liberalism and the ostensibly much more radical
politics inspired by Marxism. If valid, such a con-
clusion has at least two profound consequences. The
first is that, even when it is granted that arguments
for socialist politics carry greater weight in the justice
debate, it need not be necessary to abandon liberal
theory (at least in its Rawlsian form) to make them.
The second is that, in so far as Rawlsianism may be
thought to give voice to the liberalism that is widely
regarded as having won the `battle of ideas' in
9
politics, the full implications of that victory may
still be very far from being recognised and worked
through. If an authentically Marxist politics can be
justified from a Rawlsian perspective, the idea of
liberalism's `triumph' over Marxism is clearly pro-
blematised.
at- To illustrate this argument, I will first summarise
the key claims in my modified Rawlsian theory of
social justice. I shall then seek to rebut some of the
major reasons why critics have thought that Raw-
lsian concepts and prescriptions are inadequate for
`justice'. This section will necessarily be incomplete
in its coverage; there is simply not the space for a full
airing of the issues and we should not doubt that the
controversy over justice remains variegated and vig-
orous. But I hope at least to suggest how we might
construct a case for saying that Rawlsian theory still
stands up well in the debate about how `justice'
should be characterised; it would be vastly premature
to proclaim the need for its supersession. Finally, I
shall offer defence of the claim that my theory really
can count as Marxist. I shall attempt to situate it in
the general school of thought made up of contem-
porary liberal-egalitarians and what I shall call the
liberal-socialist egalitarians (among whom we find
certain contemporary variants of Marxism who reject
the official hostility to morality of its `orthodox'
version and with which my theory has sufficient
affinity to adopt the label itself).
113
 RODNEY G. PEFFER
Throughout this chapter, I shall leave aside ques-
tions of how to generate and justify principles of
justice; I am concerned only to apply intuitive and
analytical tests of the moral adequacy of those pro-
posed. Also, such is the enormousness of the various
issues raised here that I can only outline positions
and arguments, aiming to persuade readers to take
them seriously enough to pursue further. The chap-
ter's success in this aim may be gauged by the extent
to which it shows how Rawlsianism, the staple diet of
contemporary liberalism, can push us towards a
radical politics that many assume to be completely
anathema to the liberal project.
A Modified Rawlsian Theory
of Social Justice
Since my explicitly stated aim has been to create an
`adequate' Marxist moral and social theory, I should
first explain what I mean by that phrase. As I noted
in the first paragraphs of my 1990 book:
By a `moral and social theory' I mean one that
provides a set of moral principles or standards by
which to judge social arrangements and, by so
doing, provides criteria to decide between com-
peting sets of historically possible social arrange-
ments. Such a theory must contain enough of an
empirical, social-scientific theory to determine
which sets of social arrangements are real histor-
ical possibilities and ± of those that are possible ±
which best conform to the moral principles or
standards propounded by an adequate moral
theory . . .
By an `adequate' moral and social theory I
mean one that is based on a correct set of em-
pirical, social-scientific theories and on an ade-
quate (i.e. correct) moral theory. By an `adequate'
or `correct moral theory I mean one that is most in
wide reflective equilibrium with our considered
moral judgements . . .
By a `Marxist' moral and social theory I mean
one that [1] is informed by the spirit of Marx's
radical humanism and egalitarianism; [2] is based
on the empirical theses centrally important to the
Marxist political perspective (particularly, Marx's
theory of classes and class struggle and his analysis
114
of capitalism); and [3] attempts to defend the
Marxist's basic normative political positions.
The first of these positions is that socialism ±
that is, democratic, self-managing socialism ± is
morally preferable to any form of capitalism as
well as to any other form of (modern, mass,
pluralistic) society possible under the conditions
of moderate scarcity and moderate egoism. The
second is that social and/or political revolution, if
necessary (and sufficient) to effect the appropriate
1
transformations, is prima facie morally justified.
My strategy towards the end of creating such a theory
has been to combine a slightly modified Rawlsian
theory of social justice with a minimal set of Marxist
empirical assumptions to create a theory that ade-
quately argues for the desirability of working toward a
worldwide federation of democratic, self-managing
market socialist societies. This outcome is not the
one envisaged by Marx, of course. But I ± along with
many other thinkers who position themselves within
the Marxist tradition today ± reject the feasibility of
Marx's `higher stage' of communism; the lower,
`socialist' stage of post-capitalist society can suffice
for Marxism's practical purposes. Furthermore, the
`command economy' model for socialism in mass,
industrial societies has been sufficiently discredited
for us to prefer a suitably regulated form of market
2
economy as socialism's best vehicle.
Although I wouldn't go so far as to call my theory
a marriage of Marx and Rawls (since this would
probably have to be a `shotgun wedding'), I am
willing to call it an attempt to make Marx and Rawls
seem `comfortable bedfellows'. As a first step in this
direction, in the aforementioned work I surveyed
numerous leftist or left-leaning objections made
against Rawls's theory, arguing that most of them
were based on misunderstandings of his theory and/
or on faulty reasoning. However, I agreed that there
were four legitimate objections which, although not
requiring a complete rejection of his theory, neces-
sitated some relatively minor modifications. These
resulted in the following claims:
1. There must be a minimum floor of well-being
below which persons are not allowed to fall,
and observance of this principle must take
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
precedence over any other principle of social
justice.
2. There must be at least approximate equality in
the worth of liberty as well as strict equality of
liberty per se.
3. The difference principle must take the social
basis of self-respect ± as well as material wealth
± as a good to be maximised for the least
advantaged.
4. Democracy must not be limited to the political
realm but be implemented in the social and
economic realms as well, most especially in the
3
workplace.
One significant outcome of these arguments is that
in his most recent book, Political Liberalism, Rawls
has explicitly acknowledged three of the four mod-
ifications as being persuasive and, in effect, accepted
them. As remaining points of dispute between us,
this leaves only my suggested modification in favour
of a principle of social and economic democracy
(modification 4) and a disagreement over whether
or not the difference principle governing fair dis-
tributions of material primary goods (income, wealth
and leisure time) should be applied only within
nations (Rawls's position) or internationally (my
view as well as the view of many other writers
4
and theorists). However, part of our ongoing dis-
agreement over the former seems merely semantic in
nature. Rawls's stated reason for rejecting it is that, as
a first principle of social justice, it would automati-
cally require some form of socialism and thus illegi-
timately prejudge the practical project of selecting a
suitable regime for justice given the contingencies of
the circumstances in question. I believe that this
argument fails in so far as it is based on a non-
standard definition of socialism which obscures the
fact that a society which has some significant degree
of social and economic democracy need not be
thought of as `socialist' by definition. It is also at
least logically possible that a capitalist society could
be characterised as having a substantial amount of
5 6
social and economic democracy. Hence it is wrong
to think that the modification selects the best form of
regime in advance of the evaluation of actual cir-
cumstances. (I shall return to the latter dispute
shortly.)
115
Now let us consider my modified Rawlsian theory
of social justice in full (the principles being listed in
order of lexical priority), which I believe captures
more clearly the sense of `justice' than Rawls's own
two principles:
1. Everyone's basic security and subsistence rights
are to be met (that is, everyone's physical
integrity is to be respected), and everyone is
to be guaranteed a minimum level of material
well-being including basic needs (that is, those
needs that must be met in order to remain a
normally functioning human being and citi-
zen).
2. There is to be a fully adequate scheme of equal
basic liberties, including:
a. freedom of speech and assembly, freedom of
conscience and thought, freedom of move-
ment and free choice of occupation, the
right to hold (personal) property, freedom
from arbitrary arrest and seizure as defined
by the concepts of due process and the rule
of law;
b. the political liberties such as the right to
vote and to run for (and hold) political
office. These political liberties ± including
the rights to free political speech and as-
sembly ± are to be guaranteed their equal
worth.
3. There is to be fair equality of opportunity in
the competition for social positions and of-
fices.
4. Social and economic inequalities are justified
if and only if they benefit the least advantaged,
consistent with the just savings principle, but
are not to exceed levels that will undermine
(a) (approximately) equal worth of the liber-
ties required by due process, or (b) the good of
self-respect.
5. There is to be an equal right to participate in
decision-making processes within the social
and economic institutions of which one is a
member.
Let us call these principles (1) the Basic Rights
Principle; (2) the Equal Basic Liberties Principle;
(3) the Fair Equality of Opportunity Principle; (4)
 RODNEY G. PEFFER
the Modified Difference Principle; and (5) the Social
and Economic Democracy Principle. As with any
abstract theory, a number of caveats, clarifications
and principles of application must be mentioned and
I want to highlight a few that are important in the
consolidation of my argument. One important gen-
eral point to remember is that Rawls's theory applies
specifically to a society's `basic structure' (the poli-
tical constitution and the nature and structure of the
economy, including the provision of public goods,
taxation and redistribution policies, and environ-
mental regulation). Together with the fact that at
a micro-level, so to speak, many institutions, both
public and private, will have areas of activity ± such
as competitions for positions, awards and prizes ±
within which individual merit, effort or accomplish-
ment are legitimate criteria for awards, this point
bears out the insistence by Michael Walzer (among
others) that there are different spheres of justice
which may have different criteria for just distribu-
tions of benefits and burdens. Clearly, this thesis
could be perfectly compatible with Rawls's theory,
deliberately restricted in its scope of application as it
7
is. And even though we are to decide on more and
more specific questions and issues at each subsequent
stage of Rawls's four-stage decision-making sequence
± in which, after the choice of basic principles of
social justice in the initial original position, we have
the constitutional, legislative and judicial stages ± all
choices are to be made on the basis of these prin-
ciples of social justice and are applicable only to
8
public or publicly supported institutions.
It should also be mentioned that Rawlsian theories
and indeed all other theories of social justice devel-
oped within the contemporary liberal-egalitarian/
liberal-socialist tradition are motivated by a genuine
concern for individual liberty/autonomy as well as for
human well-being/flourishing. We should always
keep in view the point that we should not select a
conception of justice that overly compromises these
considerations. One manifestation of this is that,
when it comes to principles designed to govern
and judge the distribution of material resources,
we are concerned to develop a theory that, to a
relatively high degree, is both ambition-sensitive (cap-
able of rewarding autonomous initiative over indo-
lent passivity, say) as well as endowment-insensitive
116
(structured so as not to skew distribution according
to the unearned or otherwise undeserved possession
of certain attributes). Any theory that does not
adequately take these points into account will not
9
be judged adequate within this tradition.
Let us briefly highlight some features of my first,
second, third and fifth principles, saving the Mod-
ified Difference Principle for a more extended ana-
lysis in the next section. With respect to my first
(Basic Rights) principle, it is evident that different
conditions in the physical and mental health of
individuals will mean that the resources allocated
to meet people's basic needs will not be equal in
amount; no crude `uniformitarian' equality is sought.
Therefore, we must say that the equality in this
principle is cashed out in terms of everyone posses-
sing an equal right to have their security and sub-
sistence rights met (or an equal right to be given the
opportunity to meet their subsistence rights in the
case of able-bodied and able-minded people). More-
over, when it proves not to be possible to meet
everyone's basic needs (as when, say, there are not
enough hearts available to meet each need for a heart
transplant), it could be consistent and perhaps ne-
cessary to invoke notions from other traditional
canons of distributive justice to decide how distribu-
tion ought to proceed in such suboptimal circum-
stances. At a more general level, to operationalise
this principle social and political theorists must turn
to social-scientific theory and research to determine
how various types of societies (and governments)
under various historical conditions can be expected
to meet people's subsistence and security rights.
Moral and political theory shouldn't hubristically
10
claim to settle all such questions in advance.
Discussion of my first principle is a good point at
which to stress the ambition of my theory to have
global applicability. I may be going against the
normatively localising grain of much contemporary
thinking about justice here, but one of the virtues of
the Marxist account is that it forcefully illustrates the
direct dependence of the achievement of justice in
the developing world (surely, given the conditions
there, justice's most urgent priority) upon the beha-
viour of the developed world. Localised or relativised
accounts of justice (favoured by many communitar-
ians) are thus deeply irresponsible in failing both to
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
recognise how a society's own behaviour and internal
make-up can profoundly affect the prospects for
justice in other societies and for refusing to include
them in their theorising of justice as
Although it might be tempting to think that I
diverge sharply from Rawls in terms of the theory's
relevance to the developing world, this would be a
mistake. It may be the case that Rawls doesn't go as
far as he should in applying certain aspects of his
theory to the relationships between the developed
and developing parts of the world (governing the
distribution of goods between rich and poor nations).
Yet even an unmodified version of his theory has
some important things to say about this issue. First,
his distinction between the general and the special
conception of justice as fairness is of the utmost
importance because the general conception (which
applies within developing societies) actually requires
the difference principle to be applied to all the social
primary goods (liberties and opportunities, income
and wealth, and the social bases of self-respect) as a
package. Therefore, it allows trade-offs between
them. In contrast, the special conception (the more
familiar two principles of justice, applying only to
well-developed societies) lists the principles govern-
ing these goods in order of lexical priority such that
any amount of increase in `lower-order' goods (such
as income and wealth for the least well-off) cannot
be traded off for the least decrease in higher order
goods (such as liberties or fair access to opportu-
11
nities). The reason that Rawls makes this distinc-
tion is precisely because he doesn't want his theory to
be committed to the absurdly heartless doctrine that
liberties and fair access to opportunities can never be
limited to any degree in any society, even if this were
necessary, say, to save people from starvation. And
he is genuinely concerned with `how the poorer and
less technologically advanced societies of the world
can attain historical and social conditions that allow
them to establish just and workable institutions.'
Although often forgotten in discussions of Rawls's
work, his theory of natural duties which posits,
among others, the duties not to harm and of mutual
aid (that is, the duty to aid those whose basic needs
are not being met) has always been taken by Rawls
himself to apply between individuals no matter
whether or not they belong to the same societies.
These duties obviously require us to do what we can
± including getting our governments to do what they
can ± to try to make sure that everyone in the world
a result. is raised up out of absolute poverty and that no one
has their security rights violated. (Or rather, it
requires us to do at least our fair share to accomplish
these goals.) However, one of the reasons that I
believe my version of his theory has a greater degree
of clarity is that my Basic Rights Principle explicitly
states that both people's subsistence and security
rights (the correlates of his duties not to harm and
to aid the severely deprived) must be respected and
protected and that this is the most important or basic
principle of social justice which must thus take
precedence over all others in the design of social
14
institutions, policies and programmes.
Let us turn now to the first part of principle 2, the
Maximum Equal Basic Liberties Principle. To allay
socialist suspicions that Rawlsian theory is some kind
of bourgeois apologia, the nature of the property
right as referring specifically to personal property must
be stressed. We must keep in view the point that the
right to own large-scale productive property ± which is
here opposed to `personal property' ± is not taken as
one of the basic liberties. Whether there should be
such a right can only be answered at the second
(constitutional) stage of the Rawlsian four-step de-
cision procedure, after more empirical facts are
known about the effects that such a right would
have in terms of the realisation of the basic principles
of social justice. This allows for the possibility of
socialism on Rawls's view (although a democratic
form of market socialism), as he makes abundantly
15
clear in his first book.
Moving to the second part of this principle: as
Rawls argues, guaranteeing people (approximately)
equal worth of their political liberties would seem at
a minimum to require full public financing of elec-
toral campaigns (at least in a society such as ours in
12
which wealthy individuals, corporations and interest
groups can disproportionately influence electoral
outcomes) and perhaps other innovative electoral
16
modifications. In short, the kind of democracy that
it advocates will clearly not sanction those distor-
tions due to unequal wealth that capitalist democ-
racies (particularly the US) typically experience.
13
The primary moral thesis behind principle 3 ± the
117
 RODNEY G. PEFFER
Fair Equality of Opportunity Principle ± is that
persons of (approximately) equal talents and abilities
ought to have (approximately) equal chances of
succeeding in life (whatever their idea of success
may be, within the constraints of justice already
specified). In particular, such persons ought to have
(approximately) equal chances in the competition
17
for desirable offices, jobs and positions. In order to
accomplish this goal at least two conditions must be
met. First, there must be no discrimination based on
race, ethnicity, gender, sexual preference and so on.
Second, everyone must be given an equal start `out of
the gate', so to speak, in the race for life's favourable
offices, jobs and positions or, more broadly, success.
And to accomplish this it is reasonable to think that
the minimum which must be done is that all children
are given equal access to high-quality education from
a very young age. Assuming that there are no natural
differences in general intelligence and suchlike be-
tween racial and ethnic groups (which we have every
18
reason to assume ), there probably would be no
need for compensatory or remedial measures such as
affirmative action programmes if these conditions
were fulfilled since, presumably, there would be no
statistical discrimination against any of the groups.
However, in a society like ours in which such dis-
crimination clearly exists, there are ± I submit ±
strong moral arguments for such remedial
19
grammes in the construction of a just society.
Principle 5 ± the Social and Economic Democracy
Principle ± is, admittedly, particularly vague as to
what it requires given that regime selection is not to
be made at the level of principle. Specifically, I do
not take it in itself to be pre-emptively, definitionally
endorsing socialism over capitalism, as we earlier saw
Rawls mistakenly believing my theory to be doing. I
completely agree with his claim that a theory of
social justice should not prejudge the choice of social
system until we take into account empirical, social-
scientific information about how they actually per-
form or can be expected to perform under specified
historical conditions. But I would submit that justice
requires members of institutions to have some right
to participate in social decision-making processes
concerning matters that affect them and the institu-
tion. However, it doesn't necessarily give everyone
full voting rights with respect to every decision in
every institution in which they participate. Nor does
it claim that there cannot be administrative hierar-
20
chies in such institutions.
The essential distinction between socialism and
capitalism, I maintain, is that in a socialist society
the preponderance of large-scale productive enter-
prises (and investment capital) are socially owned
and utilised primarily for the purposes of fulfilling
human needs and for the public good. In capitalism
these enterprises (and capital) are largely owned
privately and are geared to maximise the profit of
their owners. Whether it is empirically possible or
likely that socialism or capitalism can or will better
facilitate social and economic democracy cannot be
read off from these institutional ideals alone. Many
theorists, I think, assume that socialism is more likely
to allow for social and economic democracy, even if
some of them doubt that this would be a good thing
or suspect that it may not concomitantly allow
increased political democracy. But whether or not
any society has (or will) meet the definitional criteria
I have lain down for socialism is still a matter for
debate, as is whether or not human needs can be
adequately fulfilled and the public good accommo-
dated in various capitalist societies by trickle-down
mechanisms and/or enlightened government poli-
cies. Here, at the level of practical application, would
pro- seem to be the place at which substantive disputes
between liberals and socialists remain. Certainly, it is
at this point that I would make my case for socialism.
This concludes the summary of my modifications
to Rawls's theory, with the exception of the Modified
Difference Principle (principle 4).
A Defence of
the Modified Difference Principle
The question we now need to take up is whether the
Modified Difference Principle is sufficiently robust to
take its place in a Marxist account. As space is
limited, I will concentrate on some reasons to think
that it is more ambition-sensitive and endowment-
insensitive than many critics, who have accordingly
formulated alternative accounts of justice, realise. I
claim that this could form the core of a fuller
argument that the Rawlsian approach should still
command our attention as an adequate characterisa-
118
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
tion of justice. I bolster this idea with an argument
which shows that supposedly more `orthodox' Marx-
ist thinking on justice does not yield self-evidently
superior concepts for its own purposes.
We might suspect that one Marxist reservation
about Rawlsian theory is that, because it is `liberal' in
origin, it is not sufficiently egalitarian in the relevant
way. So it is only fair to the difference principle to
begin by reminding ourselves that Rawlsian theory
demands (1) that the least well-off group be maxi-
mally benefited, not simply benefited (whatever that
would mean), and (2) that the difference principle
be applied lexically, that is after the life prospects of
the least well-off group are maximised then the life
prospects of the next least well-off group are to be
21
maximised and so on. In the light of this, we can
already sense the oddity of the standard Marxist
charge that Rawlsianism is merely a species of capi-
talist ideology; we must not underestimate its poten-
tial as a sanction for strongly egalitarian policies.
My claim that The Modified Difference Principle
is (even) more egalitarian than Rawls's original
version is generated from the point that, for mine,
the allowable economic inequalities are not to ex-
ceed levels that will (1) undermine (approximately)
equal worth of the liberties required by due process,
or (2) undermine the good of self-respect. I have
previously noted that Rawls has recently accepted
these points and I think he is able to do so because
they are arguably already implicit in his own work.
He thinks his difference principle compensates for
any lesser worth of liberty among the `less advan-
taged' given that it would be worth even less if the
difference principle had not determined distribution.
And self-respect, explicitly identified as a social
primary good, depends in part upon mutual respect
and social recognition, which may be less forthcom-
ing where there are widespread and large-scale in-
22
equalities. In this respect, my modifications do not
seem to be leading us in a direction that Rawls
himself would never be prepared to take. (Under
what economic conditions people's self-respect may
be undermined is again an issue for the social and
psychological sciences to decide.)
Conversely, this direction should not be thought
of as leading to the crudely ambition-insensitive
policy of `levelling-down' (the standard vulgar
119
charge against Marxist accounts of egalitarian justice
and one which I agree would, if valid, discredit this
general approach). For one consequence of Rawls's
demand that the difference principle be applied
lexically is that it can deflect the charge that it does
not meet the Pareto optimality requirement and
must consequently be rejected. This requirement
holds that if someone or some group can be made
materially better off without anyone else (or any
other group) being made worse off, then this must be
done (or allowed). Without this requirement, it
seems that the `only-if' clause of the difference
principle's claim (holding that material inequalities
are to be allowed ± by the background institutions,
policies and programmes ± if and only if they will
maximally benefit the least well-off group) does not
allow inequalities that would benefit a group of
people which is materially more well-off than the
least well-off (poorest) social group even when, for
whatever reason, nothing further could be done to
improve the life prospects of the latter. Now to deny
an improvement in material well-being for someone
if this won't prevent a benefit from accruing to
anyone else ± specifically, in Rawls's view, to the
least well-off ± does seem unjustifiable. And this is
precisely what the lexical application of the differ-
ence principle is designed to prevent since it `goes all
the way to the top'. So, for example, if every income
level group has had its life prospects maximised in
lexical order (as described above) and it turns out
that something can still be done to improve the
material life prospects of the wealthiest group, then
the background institutions must allow for it. Of
course, it is not easy to envisage clear-cut, concrete
examples of this, certainly not any in which wide
gaps would remain between the least advantaged and
those above them. Although it is possible that the
lexical application of the principle would result in
greater inequalities between upper and lower income
groups, in most real-world scenarios it still seems
likely that its effect will be significantly to reduce
inequalities. In any case, as a matter of principle this
part of Rawls's theory avoids the criticism that it
violates the Pareto optimality requirement.
Another charge against Rawlsian theory which
accuses it of ambition-insensitivity says that, by
counting talents and capacities as socially useful
 RODNEY G. PEFFER
resources rather than as vehicles for individual gain,
it leads to the `slavery of the talented'. This claim
holds that it is unjust to reward gifted individuals for
the usage of their abilities only in so far as that would
maximise the position of the least advantaged; their
gifts are not being rewarded in their own right. It
focuses upon the very heart of what might build the
bridge between Rawlsian and Marxist justice: Rawls's
interpretation of that often-neglected liberal value of
`fraternity' as a conception of `democratic equality',
in which people agree to `regard the distribution of
natural talents as a common asset . . . [t]hose better
circumstanced are willing to have their greater ad-
vantages only under a scheme in which this works
23
out for the benefit of the least advantaged.'
One interpretation of this position that may in-
advertently fuel the `slavery' critique holds that
Rawls rejects a crucial part of what G. A. Cohen
calls the principle of self-ownership: `Because it is a
matter of brute luck that people have the talents that
they do, their talents do not, morally speaking,
belong to them, but are, properly regarded, resources
over which society as a whole may legitimately
24
dispose.' No great imagination is required to read
into this a sanctioning of some kind of forced-labour
policy, which would be the grossest violation of the
stipulation that justice should facilitate, not crush,
autonomy. But Cohen's wording is unfortunately
misleading if this is the reading it prompts. Neither
Rawls's theory nor my own deny people the right to
choose how they use their talents. Society is not said
to `own' them in the sense of having the right to tell
25
people how and when they should be employed.
Strictly speaking, what society is primarily laying
claim to is the wealth used to compensate for and/
or reward their employment. Rawlsian theory insists
that people have no particular claim on this wealth
simply by virtue of the talents they have chosen to
exercise; a socially determined concern for a com-
mon good provides the distributive criterion instead.
This might still sound too radically egalitarian for
some liberals, which once led Cohen to think that
26
perhaps Rawls was not a `proper' liberal at all. But
reflection on how people ordinarily think about
justice can readily show that, on this score, Raw-
lsianism may not be all that out of the ordinary to
many. For people often justify high rewards to in-
120
dividuals on the grounds that they are producing
benefits for a wider society (capitalists in particular
typically cite this argument when justifying the large
pay rises they award themselves, implicitly conceding
the inadequacy of any argument that says their
talents demand such rewards regardless of the ben-
efits they supposedly bring). And many complaints
about low rewards often proceed from the claim that
the social benefit of the job in question is not being
properly recognised. Even when people think they
are judging the `intrinsic' rather than instrumental
(social benefit) value of someone's talents, the cri-
teria they use often still implicitly rely upon some
notion of those talents' social utility. How else, we
might wonder, could we ultimately judge what any
talent is worth? Further, even what counts as a
`talent' rather than a mere `capacity' may be at least
partly determined by social utility; in capitalist
ideology, after all, the idea that the market rewards
talented effort and achievement necessarily relies
upon the distinctly instrumentalist idea that what
counts as `talented' ± rewardable ± contributions is
determined by what other people in the marketplace
demand.
Of course this is not to claim that people are
therefore already thinking in terms of the difference
principle and I don't deny the fact that people also
ordinarily use rival notions of justice as well. But I am
saying that there is something in ordinary discourse
about justice upon which Rawlsianism can latch to
develop its case.
Such a case should stress that (in particular) anti-
socialist criticisms of Rawlsianism usually assume (1)
that individuals are garnering most of their material
resources (income and wealth) by working (as either
an employee or their own boss); (2) that individuals
are usually materially rewarded due to their effort or
socially useful contribution; and (3) that the level of
recompense is usually correlated with natural talents
and abilities, at least in a just or relatively just
society. But in contemporary class societies these
assumptions are highly dubious. In capitalist socie-
ties, members of the highest income/wealth socio-
economic group ± that is the capitalist ruling class
and its `hangers-on' ± often inherit the great pre-
ponderance of their wealth rather than garner it by
their labour or effort. Moreover, the investment
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
income they receive through rent, interest and di-
vidends (which normally dwarfs their salaried in-
comes, even if they are top corporate executives) is
not earned in any socially useful, morally significant
sense. As David Schweickart has argued, to allow
capital to be used is hardly a genuinely productive act
± since, at most, it requires signing some papers on
the part of those who own it ± even if the capital
owner's permission is legally required for its deploy-
ment within the socio-economic system in ques-
27
tion. And to claim that corporate executives,
bank presidents and suchlike really deserve what
they earn (because they are the only ones who
can do the job or are really the best qualified) surely
strains the limits of credibility. In an interesting ploy
to disprove such widely accepted nonsense, the
politically radical American entrepreneurs
and Jerry ± of the highly successful Ben & Jerry's
ice cream parlours ± some years ago advertised the
position of president of their corporation to the
general public, requiring only a one-page letter of
interest initially and then follow-up interviews. They
ended up selecting an African-American man in his
fifties who had only a high school education and had
been employed as a janitor most of his life, with
absolutely no ill effects for the corporation.
One might agree that rewards at the `top end' of
capitalist society (including entertainers and sports
stars as well as capitalists) far outstrip any plausible
assessment of their talent's worth, however this is
assessed. But the claim that even individuals in the
middle and lower classes usually get paid what they
deserve for their work in market societies
grounds, for example, that people normally receive
an income commensurate with the productive value
and relative scarcity of their work abilities and skills
under equilibrium market conditions) is highly sus-
picious because of the general impact that non-
market factors such as social custom, political reg-
ulation and the strength of organised labour (each of
which may be connected with the others) have on
the determination of remuneration by the `free mar-
ket'. In general, there are just too many counter-
vailing forces in the operation of the capitalist
market to make us happy with the idea that the
social value of people's talents is appropriately re-
warded by the wage or salary that their labour-power
28
is able to command.
Although Rawlsianism clearly poses a large chal-
lenge to capitalist thinking about justice, G. A.
Cohen has claimed that socialists should still not
be satisfied with the difference principle for another
29
reason. He notes its familiar presentation as an
`incentive principle', sanctioning unequally large
rewards to the talented rich on the grounds that
they will consequently have the incentive to work
hard and thereby generate wealth which will `trickle
down' to the least advantaged and benefit them more
than would any other arrangement. (This, of course,
has been the staple justification of regressive taxation
policies since the Reagan/Thatcher days.) Leaving
aside the debate as to whether the `trickle-down'
thesis actually holds (for if it does not, such a policy
Ben would clearly not be sanctioned by the difference
principle), Cohen claims that the `incentive' con-
struction of the difference principle's operation relies
upon certain assumptions that are incompatible with
the attitude towards one's talents that Rawls himself
believes is required by justice. Simply put, a talented
capitalist entrepreneur who is paid a particularly high
salary as an incentive to create jobs which would not
otherwise be created for the least advantaged is
typically not unable to do the same job for a lot less.
Rather, he is unwilling to do so and he is effectively
using his socially valuable talents as his own `bar-
gaining chips' to enhance his own personal gains at
no extra benefit to the community. In other words,
he is not treating his talents as `common assets', for if
he did he would not need incentives to work (in-
(on deed, he probably does not `need' them at all, but is
nevertheless able to claim so as he holds the rest of
the community to ransom by threatening to with-
hold his talents if he is not paid more).
It is hardly clear that Rawls would wish to depart
from Cohen's claim that the `incentive' scenario
described above involved an unjust rejection of
community and its claims on the talents of its
members. As we have seen, his own notion of
democratic equality rests upon a specific understand-
ing of `community' whose members would reject such
a selfish attitude to their own talents. Thus, it would
be misleading to think that the difference principle
sanctions such an incentive payment as being itself
just. (Cohen would agree that if incentives were the
121
 RODNEY G. PEFFER
only way to get the talented to work and that their
labours would indeed maximise the position of the
least advantaged, then it would be justified. But this
would be the application only of a principle to
ameliorate injustice in an unjust world and not what
full justice itself would mandate.) Admittedly, a
Rawlsian might sanction incentives on the grounds
that they would be necessary to attract people to
careers that are socially necessary but exceptionally
difficult or dangerous and hence unlikely to be
chosen without them. To this I have two brief
replies. One is to point out that this observation
does not serve many of today's incentive-claiming
talented rich particularly well; they would doubtless
still have picked their chosen careers even if the
material rewards were somewhat less because they are
not `dangerous' or particularly difficult. Indeed, in a
just society such incentives may be necessary to
attract people to the kind of work which is very
often exceptionally low-paid in capitalist society.
Secondly, my rendering of Rawlsianism as a socialist
doctrine consolidates within its terms the ideal of
people freely choosing to dispose of their talents in a
socially beneficial way. To be sure, some difficult and
dangerous tasks may still require incentive payments,
if other ways are not found for allocating them. But,
in general, I don't conceive of people needing huge
incentives to work in a socially beneficial rather than
selfish way in the just socialist society, certainly not
to the kind of inegalitarian degree supposedly re-
quired in capitalist society.
The `slavery of the talented' critique has probably
had most bite in the rather different criticism that
since his difference principle and minimum floor
requirement seem to assure the least well-off ± in
fact, everyone ± a minimal income or set of material
resources, Rawls's theory is not sufficiently ambition-
sensitive because it seems not to require anyone to do
anything in order to be entitled to this income or set
of resources; it encourages free-riding and hence
discourages the formation of the kind of community
just described above. There are actually two ques-
tions that must be distinguished here: (1) whether
Rawls's theory requires that this minimum income
simply be given to anyone whether or not they are
willing to do anything to earn it (assuming that they
are able-bodied and able-minded and that such
122
opportunities actually exist); (2) whether this is a
justified social policy (or principle of social justice)
regardless of what Rawls's theory says about it.
Now I claim that (1) Rawls's theory does not
require such a policy; and (2) assuming that there
are genuine opportunities for people to earn a living
under decent conditions and with decent recom-
pense, a social policy (or principle of justice) that
would allow such free-riders is not justified or at least
30
not required. Admittedly, both Rawlsians and non-
Rawlsians have split over both of these issues such
that, for example, I have been scolded by certain of
Rawls's former students for both (supposedly) mis-
interpreting Rawls and for being too much of a
`puritan' in rejecting such a policy. I have also
encountered both Rawlsians and non-Rawlsians
who have agreed with me on both points. Happily,
Rawls is now on record as not requiring such a policy,
since in his typescript `Justice as Fairness: A Restate-
ment' he writes:
Particular distributions cannot be judged at all
apart from the claims (entitlements) of indivi-
duals earned by their efforts within the fair system of
co-operation from which those distributions re-
sult. In contrast to utilitarianism, the concept of
allocative justice has no application. There is no
criterion for a just distribution apart from back-
ground institutions and the entitlements that
arise from actually working through the procedure
31
[emphasis added].
It should be noted, however, that to prevent able-
bodied people from being free-riders when there are
opportunities available for them to earn a decent
living by their own efforts is neither an endorsement
of the kind of movement to have `workfare' replace
`welfare' underway in the US (since the opportu-
nities may not be well-paying enough once one
includes childcare and transportation expenses into
the equation), nor an affirmation of the view that
people must work if they are able-bodied (which I
would regard as a genuinely puritan attitude). My
only claim is that able-bodied people should not be
supported by the state or society in general if they
refuse to do anything to earn a living (assuming,
again, that decent opportunities exist for them to
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
do so). If they can sponge off others while not doing
anything illegitimate to garner the material goods
and services they need, the state or society has no
business telling them that they can't live their lives
this way, as far as I'm concerned . . . at least if
requiring people to do socially useful work is not
absolutely necessary for society to meet the first
(Basic Rights) principle of justice.
I now want briefly to reflect upon the objection
that Rawls's notion of `social primary goods' mis-
identifies what theories of social justice should be
concerned to distribute (such as welfare or utility, or
material resources). Space does not allow a full
discussion of all of Rawls's rivals, so I can only sketch
the case for retaining his approach. I will, though, say
a little more on `exploitation', a category which
many might think Marxism ought to prefer as the
key term in its account of justice.
Richard Arneson helpfully distinguishes between
four alternative principles of justice, based around
two different types of good to be distributed ± `wel-
fare' and `resources' ± and two types of `end-state' ±
based upon whether justice is concerned with the
amount possessed of each good or merely with the
opportunity to acquire them:
1. the Equality of Welfare Principle;
2. the Equality of Resources Principle;
3. the Equal Opportunity for Welfare Principle;
4. the Equal Opportunity for Resources
32
Principle.
To this we might add a fifth, developed by Cohen
and, because of his own political pedigree, one that
may be of particular interest to socialists:
5. the Equality of Access to Advantage
33
Principle.
All five have been much discussed in order to
ascertain which one does the best job of capturing
what justice intuitively requires. Typically, their
respective proponents all claim superiority over Raw-
ls's approach in this regard. Important issues are
undoubtedly raised in this debate and I accept that
it would be premature to foreclose on it. But there
may be a case for thinking that it should not pre-
dominate in the discourse about justice in the way it
has done recently, particularly from a radical poli-
123
tical perspective that is inclined towards impatience
with regards to some types of `armchair' reflection.
The nub of Rawls's argument is that social primary
goods are all-purpose general resources in modern
societies that can usually be utilised by individuals
who have them no matter what their life plans are.
Because they constitute a specific, limited set of
goods, they should be easier to track in the process
of distribution than the wider and more nebulous
categories of (opportunities for) resources, welfare or
advantage. In other words, policies of social justice
may be easier to operationalise with a concept of
social primary goods as their currency. This claim is
reinforced by Rawls's insistence that his theory is one
of `pure procedural justice' in which an outcome is
judged to be just wholly in so far as it has arisen
through the operation of just procedures. He writes
that:
The great practical advantage of pure procedural
justice is that it is no longer necessary in meeting
the demands of justice to keep track of the endless
variety of circumstances and the changing relative
positions of particular persons. One avoids the
problem of defining principles to cope with the
enormous complexities which would arise if such
details were relevant. It is a mistake to focus
attention on the varying relative positions of
individuals and to require that every change,
considered as a single transaction viewed in iso-
lation, be in itself just. It is the arrangement of the
basic structure which is to be judged, and judged
from a general point of view . . . Thus the ac-
ceptance of the two principles constitutes an
understanding to discard as irrelevant as a matter
of social justice much of the information and
34
many of the complications of everyday life.
This is not to say that institutions should never be
scrutinised for their justice-promoting abilities once
they have been established. In his `Restatement',
Rawls talks of `pure adjusted procedural justice',
introducing the caveat that `certain rules for making
adjustments must be included in the basic structure
as a system of social co-operation so that this system
35
remains fair over time.'
These features are, I would argue, great strengths
 RODNEY G. PEFFER
of his theory since they make it more readily applic-
able to the workings of actual societies and less
intrusive into people's individual lives. On Rawls's
approach, so long as people act within the confines
of justice and so long as what they are able to garner
for themselves is accumulated within the confines of
the background institutions set up by the principles
of social justice they ± and their legitimate material
holdings ± are not to be interfered with. This may
contrast with those theories of justice which do not
have such in-built delimitations, for they may inad-
vertently sanction a political `micro-management' of
social and personal life that, apart from anything
else, would conflict with the fundamental respect for
individual autonomy and well-being identified ear-
lier as central to justice's concerns.
To consolidate this part of the case for Rawlsian
theory, it is worth reflecting upon this claim of
Arneson's, who, having surveyed numerous compet-
ing theories of justice, concludes that their practical
implications
may be hard to discern, and may not diverge
much in practice. Familiar information-gathering
and information-using problems will make us
unwilling to authorise government agencies to
determine people's distributive shares on the basis
of their preference satisfaction prospects, which
will often be unknowable for all practical pur-
poses. [So, for example], we may insist that gov-
ernments have regard to primary goods equality or
resource equality as rough proxies for the welfarist
36
equality that we are unable to calculate.
If it is indeed fair to say that these rival theories will
tend to converge upon similar concerns and recom-
mendations in practice, then the case for `social
primary goods' on the grounds of their theoretical
parsimony is indeed strong. The substantive disputes
about justice could then shift to the debate over
which form of regime may best bear the concerns of
justice (more on this in the next section).
One noteworthy critic of Rawls who also opposes
the `welfarist/resourcist' axis is Amartya Sen, who has
defended a `basic capacities' approach, arguing that
justice should seek to assure that people have the
opportunity to exercise their basic human capacities
124
37
in order to fulfil valuable human functionings. This
approach is not easily classifiable as either welfarist or
resourcist as, strictly speaking, it seeks to equalise
neither. But it has the advantage of carrying our
moral view to a very deep level of what is really
(intrinsically) valuable in human lives. It keeps us
focused on the fact that, due to differential natural
talents and abilities, people differentially convert
resources into valuable `functionings' (or capabilities
for such functionings) and that it is these with which
theorists of justice should be primarily concerned.
However, Rawlsian theory is not blind to this point.
It does not cease to be interested in how people stand
with reference to their primary goods once they have
been distributed. A crucial consideration in selecting
the best form of social and political order for Raw-
lsian justice is precisely which of the alternatives best
allow people the chance to utilise these goods in
pursuit of their own life-plans. My own case for
socialism, indeed, would partly rest upon the claim
that a socialist order is much more amenable to the
process of converting `goods' into `functionings'.
And once that is recognised, the advantages of Sen's
approach diminish in so far as it is limited to covering
only basic capabilities. Once these are achieved, it
doesn't tell us how to govern any further distribu-
tions of resources above the levels necessary for these
purposes. Indeed, `capacities' may simply cease to be
useful beyond this basic level. As G. A. Cohen puts
it, `capacities beyond the basic (Can I run a mile?
Can I impress Ukrainians with my impersonation of
Russians? Can I sew more quickly than you?) seem
quite irrelevant to measurement, deprivation, in-
equality or anything else of urgent concern from
38
the point of view of justice.' In any case, once we
see that subsistence rights are cited as something that
must be fulfilled by my Basic Rights Principle, Sen's
concerns with basic capabilities would already seem
to be accommodated. No practical advantage over
the `social primary goods' approach is gained.
Finally in this section we turn to `exploitation',
which many Marxists might wish to make central to
any discussion of justice. Now there has been much
discussion as to how `exploitation' should be con-
ceptualised, led by John Roemer in particular, and
39
there is no space here to enter this debate. Instead,
I simply want to offer one definition of the term ±
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
which I take to be much more `orthodox Marxist'
than Roemer's ± and then sketch an argument which
holds that `exploitation' could not substitute for
anything like my Rawlsian account of social justice
in a plausible political morality.
I contend that, for Marx, `exploited labour' is (1)
forced (because workers have no realistic choice but
to sell their labour-power to capitalists); (2) unpaid in
its yielding of surplus value or product (that is, the
profit the individual capitalist accrues from the
commodities produced by workers' labour); also
(3): the product of the labour is lost to the producer
40
± it comes under the control of someone else. The
question for theorists of justice here is what, exactly,
is wrong with exploitation? We can answer this using
a three-way distinction between types of exploita-
41
tion.
First, we have simple exploitation, denoting the
appropriation of surplus value, or the fact of the
direct producers not getting back the full value of
what they produce. Note that there is no reference to
`force' here, so the Marxist notion does not come
under this heading. It highlights merely the transfer
of value from producers and it cannot be considered
even prima facie wrong because this could come
about as the result of a genuinely free and fair
agreement. Exploitation proper is forced, unpaid, sur-
plus labour, the product of which is not under the
control of the direct producers. This brings in the
crucial element missing from simple exploitation and
I would say it is prima facie wrong. But it may not
always be wrong. Imagine that the apologists for
capitalism were correct in arguing that capitalism
was necessary to promote the undoubted moral goods
of meeting people's basic needs, preserving their
liberties and allowing for democracy. In this in-
stance, the moral harm suffered by exploited labour
is outweighed by the moral benefits that only capit-
alism brings. So at the very least there must be more
to the morality of justice than the category of ex-
ploitation if such judgements are to be made. This
fuller morality would therefore allow us to identify
instances of all-things-considered unjustified exploita-
tion, which is `exploitation proper' that is not justi-
fied by its promotion of some other weighty moral
concern. (Contra the capitalist apology, my Marxist
theory naturally claims that, with socialism a real
125
possibility, capitalist exploitation is all-things-con-
sidered unjustified.)
Further, I would also argue that not only does
`exploitation' need a fuller morality to supplement it,
it can also be replaced or subsumed by one such as my
Rawlsian theory. For the elements of `force', `robbery'
and `loss of control' in exploitation proper can all
also stand condemned by specific components of that
theory. `Exploitation' adds nothing new or necessary
to it.
`But Is This A Marxist Theory?'
Now I should make clear why I say that my overall
moral and social theory is a Marxist theory and what
is contained in its minimal set of Marxist empirical
assumptions. To the first question, my answer is (as it
has previously been) that it is not a specifically
Marxist moral theory. This should not be surprising.
Not even Marx's implicit moral theory per se is a
specifically Marxist theory. There is, in fact, no such
thing as a specifically Marxist moral theory. There is,
however, such a thing as a specifically Marxist moral
and social theory, i.e. a theory which combines a
moral theory with a set of empirical, social-scientific
theses in order to judge alternative sets of social
arrangements, programmes and policies. In fact, any
moral and social theory that utilises a recognisably
Marxist set of empirical, social-scientific theses and
supports a recognisably Marxist set of normative
political positions qualifies as a Marxist moral and
42
social theory.
However, since my 1990 book was published right
at the time of the so-called `fall of communism' in the
Eastern Bloc ± causing even some who thought it
succeeded in its stated goal of creating an adequate
(or, at least, more sophisticated) Marxist moral and
social theory to call it an excellent exemplar of
Hegel's famous quip about the Owl of Minerva flying
only at the end of the day ± I find myself, upon
occasion, having to defend the relevance of a the-
oretical project that still:
1. takes a minimal set of Marxist empirical as-
sumptions ± in particular about the nature of
capitalist society ± as plausible and relevant
(even though it is based on a slimmed-down
 RODNEY G. PEFFER
Marxism which jettisons some of its orthodox
baggage);
2. argues for a kind of socialism, albeit a demo-
cratic, self-managing form of socialism based
on a regulated market economy rather than a
43
command economy, which, within well-
developed societies, would be committed to
protecting civil liberties as much as any bour-
geois capitalist democracy;
3. defends socialist regimes and socialist revolu-
tions in the developing world as the only way
to overcome the absolute poverty and extreme
subservience characteristic of such societies,
and forcibly confronts us with the possibility
that `parliamentary' politics may not be en-
ough even in developed nations to overcome
capitalist resistance to the requirements of
justice.
Thus, I want to comment very briefly on the con-
tinued relevance of at least some components of
Marxist empirical theory as well as on the continued
relevance of some kind of socialism as a worthy goal
44
or ideal. My response to the charge that Marxist
theory is now passe
 is as follows.
1. So many of Marx's views have passed more
widely into the social sciences and our modern
world-views that perhaps we tend no longer to
appreciate the true significance and lasting
legacy of his thought and work.
2. As shown by continuing sophisticated devel-
opments of certain components of Marx's
economic, sociological and political theories
by the Analytical Marxist tradition and others,
it is only certain orthodox and dogmatic inter-
pretations of Marxism that have become
otiose, not Marxist theory in general.
3. By Marx's own definition, no society on this
planet has ever become even a fully socialist
society ± that is, one which has fulfilled all of
the requirements of Marx's so-called first stage
of communism, which include full democratic
rights ± let alone a full-fledged communist
society (in which, ex hypothesi, there would
be no political state as well as no social classes
and no significant social conflict). Thus, Sta-
126
linist and post-Stalinist claims about how far
Soviet-style communist regimes had actually
progressed towards socialism and communism
were complete travesties and misrepresenta-
tions of Marxist theory and should not be
allowed to distort our understanding of what
the theory sanctions in practice.
4. The struggle to establish and preserve social-
ism in various countries may be somewhat
more likely to succeed in light of the theore-
tical work and institutional changes advocat-
ing and implementing market socialism as
opposed to command economy socialism
(which, whatever its role in industrialising
the former Soviet Union and helping to defeat
fascism in The Second World War ± for which
the world owes the former Soviet Union an
eternal debt of gratitude ± simply does not
seem to be capable of adequately functioning
in consumer-oriented, modern, mass socie-
ties).
5. Even the dissolution of the Soviet Union and
the `fall of communism' in the Eastern Bloc
does not necessarily herald the end of attempts
to establish post-capitalist, or nascent socialist,
societies or to preserve them where they still
exist (such as in Cuba and China), especially
in light of the fact that under the hegemonic
domination of international capitalism, the
plight of the world's poor ± as well as the
world's natural environment ± has continued
to worsen, with world capitalism showing
neither the ability nor the inclination to solve
these problems.
Of all these claims, it is most immediately important
to substantiate the first one and I shall do so by
stating the minimal set of Marxist empirical assump-
tions which, when combined with my modified
Rawlsian theory of justice, yield the appropriate
Marxist political conclusion:
1. As a result of the logic of the maximisation of
profit, all capitalist societies ± developed or
developing ± exhibit and will continue to
exhibit certain economic and social problems
(inflation, depression, recession, unemploy-
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
ment, poverty, failure to regulate environ-
mental pollution sufficiently, and so on) that
can be solved only by the institution of a
regulated, partially planned (but not neces-
sarily command) socialist economy.
2. Even the mixed, welfare-state capitalist so-
cieties of the advanced, industrialised nations
in the West exhibit severe social inequalities
and ± if sufficiently threatened by mass work-
ing-class movements for social equality ± will
almost undoubtedly exhibit severe repression.
(The possibly irreversible decline of the Key-
nesian welfare state is evidence of how capi-
talist societies will inevitably `resolve' the
tensions between social justice and the logic
of capital in favour of the latter.)
3. The world capitalist system causes in the
Third World both extreme inequality and
suffering, on the one hand, and (often) ex-
tremely repressive regimes, on the other.
4. So long as it is dominant on an international
scale, the capitalist system will not allow the
massive transfers of capital, technology and
knowledge necessary to solve the Third
World's major social and economic problems.
5. Such conditions in the Third World make for
perpetual social instability since those who
are severely oppressed and/or deprived will
organise and, if necessary, fight to better their
condition.
6. The predictable response from the most
powerful nations at the capitalist `centre'
(primarily the United States now and for
the foreseeable future) is first to install and/
or aid those Third World regimes or military
cliques that can best suppress these mass
movements for radical social change and,
second, if that strategy fails, to intervene
either directly with its own military forces
or indirectly through proxy armies and `low-
intensity warfare'.
7. However else we may characterise contem-
porary post-capitalist societies such as Cuba,
it seems clear that they are not the primary
cause of the many indigenous revolutionary
movements in the Third World and do not
bear the primary responsibility for the nuclear
127
arms race.
8. The bureaucracies of such post-capitalist so-
cieties genuinely want to reduce or eliminate
arms expenditures in order better to satisfy
the consumer appetites of their own popula-
tions.
9. Without the economic, diplomatic, and mili-
tary pressure of the western capitalist powers
such post-capitalist societies may well
achieve significant democratisation (by
either evolutionary or revolutionary means).
10. Socialist transformations can occur in the
advanced industrialised countries of the
West, and such transformations can lead to
democratic forms of socialism; thus a world-
wide federation of democratic, self-managing,
socialist societies is a genuine historical pos-
45
sibility.
Although these theses were put forward just before
the fall of the Soviet Union, I don't think that this
development or any since then has decisively falsi-
fied any of them. The only one put into question at
all would seem to be the ninth thesis since, although
significant democratisation has occurred in most of
the former so-called communist countries, this has
been conjoined with a movement ± by no means
complete as of yet and certainly not completely
successful ± toward capitalism. On the other hand,
it will be observed that the theses concerning the
lack of ability of capitalism to solve the world's socio-
economic problems have tended to be strengthened
rather than falsified, not least in many of the former
so-called communist countries of the defunct Eastern
Bloc, where the movement towards capitalism has
resulted in substantial increases in unemployment,
poverty, social inequality, homelessness and lack of
adequate health care accompanied by skyrocketing
corruption, crime and gangsterism. Moreover, the
theses concerning the potential ability of Third
World socialist (or at least post-capitalist) societies
to solve their basic social problems have also, I think,
continued to be borne out, even though under
contemporary world circumstances, post-capitalist
societies such as Cuba and China have had to resort
to market mechanisms and even foreign capital
investment, usually in jointly owned enterprises.
 RODNEY G. PEFFER
(But this, of course, is no mark against these socie-
ties, so far as I'm concerned, since I support market
socialism to begin with.) And it is also arguable that
such societies would be able to do even better
economically if the powerful capitalist nations ±
especially the US ± were not constantly attempting
to undermine them.
Note that this set of empirical assumptions does
not include dialectical materialism, orthodox or even
reconstructed versions of historical materialism, or
such lower-level empirical theses as the labour theory
of value and the `collapse' theory of capitalism. This
is not to say that some of these theories ± at least in
their more sophisticated forms ± are not plausible or
arguably true. But it is to claim that the truth of these
components of Marxist theory is not necessary in
order for the basic normative political positions to be
justified. Most emphatically, this set of empirical
theses does not entail the view that socialism and/
or communism are historically inevitable, or that
genuine, full-fledged socialist societies can
achieved within one or a few countries (in isolation
from the rest of the world). Nor does it imply that
Marx's higher state of communism is historically
possible (as opposed to a utopian impossibility that
should not even be seriously considered when doing
social and political theory).
I recognise, of course, the prima facie oddity of
conjoining these theses with Rawlsian moral and
political theory. Yet that I should develop a theory
which attempts to synthesise Rawls's theory of justice
and the salvageable components of the Marxist
tradition should not, in fact, seem all that strange
to anyone familiar with moral, social and political
philosophy over the past thirty years or so. From the
last third of the twentieth century this realm's domi-
nant tendency in English-speaking societies, Scan-
dinavia and the Netherlands has been composed of
liberal-egalitarian and liberal-socialist egalitarian
theorists who, for the most part, take Rawls's theory
or similar theories of social justice to be basically
correct. Most of these theorists also respect much of
what Marx said and much of what he was trying to
accomplish and they accept some, although certainly
not all, of Marxist empirical theory as true.
Moreover, most contemporary Marxist theorists
involved in current debates in moral, social and
political philosophy within this general tendency
are part of the Analytical Marxist tradition that
has developed over the past twenty years or so. As
such, they have no problem operating within the
general analytic-philosophical paradigm that empha-
sises conceptual clarity and logical rigour or within
the general framework of liberal egalitarian theory
developed by Rawls. In fact, Analytical Marxists are
an integral part of the current discussions specifically
involving Rawls's theory and similar theories of
social justice. There is little (if anything) to distin-
guish them from the non-socialist liberal-egalitarians
who participate in these debates, save their greater
commitments to certain parts of Marxist empirical,
social-scientific theory and thus to the idea that
morally defensible forms of socialism are genuine
historical possibilities which should be advocated
or at least discussed as genuine options. It should
be stressed, of course, that these theorists are all
proponents of political democracy and civil liberties
be (which they believe can thrive in socialist societies
under favourable circumstances) and, thus, critics of
the kind of Stalinist political organisation that so ill-
served the socialist cause. In addition, most are
advocates (if not always without reservation) of
market socialism and opponents of command-econ-
omy socialism.
So let me assert again that what divides contem-
porary egalitarian liberals such as Rawls from liberal-
socialist egalitarians such as myself is not moral
theory but differences in empirical beliefs and prac-
tical political convictions. It would be difficult to
isolate any normative moral thesis that was univer-
sally agreed upon by those in one category and
universally denied by those in the other. The first
category is thus composed of theorists who don't
explicitly advocate socialism, but who could admit
that a democratic-liberal form of socialism would be
justified if certain empirical, social-scientific theses
that Marxists put forward about capitalism and so-
cialism were true. This category includes not only
Rawls but such figures as Ronald Dworkin, T. M.
Scanlon, Norman Daniels, Thomas Nagel, Amartya
Sen, Martha Nussbaum, Brian Barry, Amy Gut-
mann, Bruce Ackerman, Jeremy Waldron, Henry
Shue, Philippe Van Parijs, Samuel Scheffler and
Susan Moller Okin. The second category ± the
128
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
liberal-socialist egalitarians ± includes such figures as:
G. A. Cohen, John Roemer, Jon Elster (at least until
recently), Kai Nielsen, Michael Walzer, David Mill-
er, William McBride, Roger Gottlieb, Jeffrey Reim-
an, David Schweickart, Nancy Holmstrom, Andrew
Levine, Alex Callinicos, Norman Geras, Mark Evans
46
and me. There is also a category of philosophers
within this tradition who speak more sympatheti-
cally than the liberal-egalitarians of certain aspects of
Marxism and/or socialism, but don't explicitly or at
least very strongly advocate socialism, so it is difficult
to decide in which category they should be classified.
This would include, I think, Robert Paul Wolff,
Joshua Cohen, Allen E. Buchanan, Thomas Pogge,
Richard Arneson, Will Kymlicka and Sharon Lloyd.
(Hence my categories are useful terms of reference to
guide us in the debate, but not strict and exhaustive
classifications.)
Here, we should recognise the complication that
to rely simply on the categories of capitalism and
socialism may not do justice ± so to speak ± to the
richness and complexity of the institutional arrange-
ments suggested by these two groups of theorists
when they propose how their recommended policies
might best be implemented. For example, Rawls
suggests that his theory of justice could be best
implemented in what he calls a `property-owning
democracy', in which most large-scale productive
enterprises would not be socially owned ± hence,
it is not a form of socialism ± but in which all citizens
would somehow own some productive property (pre-
sumably in terms of stocks and/or bonds), and in
which significant wealth could be accumulated by
one within one's own lifetime but which could not
be passed on to the next generation ± or anyone else,
for that matter ± due to strenuous inheritance and
47
gift taxes. Somewhat similarly, in John Roemer's
model of market socialism room would be made in
the economy for entrepreneurs to make virtually
unlimited amounts of money within their lifetimes
if they could organise a series of successful economic
Notes
1. Peffer (1990), p. 3.
2. See Peffer (1990), especially chapter 10, for substan-
tiation of these modifications.
129
enterprises (from seed capital provided by large
socialist enterprises or the banks) and then sell them
off to the public sector after they reached a certain
level of productivity. But, as in Rawls's property-
owning democracy, no one would be able to pass on
significant levels of wealth to anyone else, although
Roemer suggests they might be able to donate all or
much of their wealth to their choice of socially
48
approved charitable or philanthropic causes. An-
other important view worthy of further attention is
represented by Phillipe Van Parijs, who, even though
a close associate of the Analytical Marxists, actually
advocates an extremely egalitarian version of social-
49
democratic, welfare-state capitalism.
But despite these `second-stage' disagreements, I
think it justified to propose that the most plausible
moral, social and political philosophy/theory must lie
somewhere within the general confines that I think
Rawlsian principles adequately map ± and it is, in
general form, liberal-egalitarian in moral theory and
Marxist in political theory and action. What motivates
my Marxist convictions above all is this final claim: if
we are genuinely concerned about justice, then we
should be firstly and profoundly concerned about the
desperate plight of the vast majority of people in the
developing world in particular. And I think that
contemporary liberal debate should increasingly focus
upon the liberal-egalitarians upon whom, if they ac-
knowledge the huge gap between the reality of global
injustice and their own standards of justice, it should be
incumbent to argue against what I take to be the fact
that such injustice can be fully rectified only by anti-
capitalist revolutions, and genuinely socialist property
relations and social policies, together with genuinely
fair relations between the developed and developing
worlds. Arguably, this may only occur with the trans-
formation of the major industrial societies of the North
into socialist societies ± a goal that the Marxist em-
pirical theses would also indicate is necessary for the
50
achievement of justice within them as well.
3. Ibid, pp. 419±20.
4. Rawls (1996), pp. 7±8 n. 7, and Rawls (1999), pp.
558±9, Peffer (1994).
 RODNEY G. PEFFER
5. See Peffer (1994), pp. 261±2 for more on the defini-
tion of `socialism'.
6. Peffer (1990), pp. 418±20; Peffer (1994), pp. 264±5.
7. See Walzer (1983).
8. See Rawls (1972), pp. 195±201 for discussion of the
four-stage decision-making sequence. Of course, these
restrictions upon the domain of Rawlsian justice do
not deny that people's conduct even within private
institutions and situations is still subject to criminal
and civil law.
9. These terms are coined by Dworkin (1981a), p. 311.
See also G. A. Cohen (1989) and his `Equality of
What? On Welfare, Goods and Capabilities', in Nuss-
baum and Sen (eds) (1993), pp. 9±29, and Kymlicka
(1990), pp. 75±85.
10. Although this is far from being uncontroversial, I
believe that a strong case can be made for the view
that, in the developing world, only certain socialist
countries (such as Cuba and China) as well as the
Indian province of Kerala (which has had a socialist
government since shortly after the Second World
War) have generally succeeded in fulfilling their
populations' subsistence rights. It seems clear to me
that if one agrees that the empirical evidence shows
the only way for developing societies to ensure that
people's subsistence rights are met is to institute
socialism and if the empirical evidence shows ± as I
believe it does after having tracked Amnesty Inter-
national reports for the past twenty years or so ± that
under normal circumstances socialist regimes in the
developing world generally do a better job of protecting
their people's security rights than capitalist Third
World countries, then one should seek to support
and promote such socialist regimes and movements
as a first duty of social justice, rather than attempting
to undermine and destroy them as the government of
the United States has so often been guilty of and
which, I'm ashamed to say, it is continuing to do with
respect to socialist Cuba.
11. See Rawls (1972), pp. 150±2.
12. Rawls (1999), p. 537.
13. See Rawls (1972), pp. 114±18.
14. My continuing disappointment with Rawls's attitude
to global justice is due to the fact that in `The Law of
People', which he had actually intended to be an
appendix to Political Liberalism, he has, for reasons I
still do not understand, continued to refuse to apply
his difference principle internationally. Although he
does call for `standards of fairness for trade and other
co-operative arrangements' and for `certain provisions
for mutual assistance between peoples in times of
famine and drought' (Rawls, 1999, p. 541) there is
the question as to why this should only apply to
`reasonably developed liberal societies' and also why
the difference principle itself shouldn't be so applied.
But, in any case, it seems clear that Rawls's theory and
130
Rawlsian theories in general ± including mine ± are
relevant to the relations between the developed and
the developing world and thus provide at least a
starting point to a case for international justice.
15. See Rawls (1972), pp. 270±4, 280±2.
16. For philosophical perspectives on these issues see
Pateman (1970), Cohen and Rogers (1983), Green
(1985), Cunningham (1987).
17. Rawls (1972), pp. 72±108.
18. See Holmstrom (1994).
19. See, for example, Sher (1975).
20. See Peffer (1990), pp. 396±403.
21. Rawls (1972), p. 83.
22. See Rawls (1972), pp. 204±5, 440±6; Peffer (1990),
chapter 9; and Peffer (1994).
23. Rawls (1972), pp. 101, 105.
24. Cohen (1986), p. 79. His concern here is, in fact,
primarily with the `right-wing liberal' or libertarian
view espoused in Nozick (1974).
25. One might want to think or hope that genuinely
`fraternal' (socialistic) individuals would choose to
use their talents in whatever maximally socially ben-
eficial way they could. But it would surely be stretch-
ing credibility to think that every individual in a
socialist society would always freely surrender all their
talents to whatever task `society' allocated them.
26. Cohen (1986), p. 79. He recognises the oddity of this
thought, produced by the claim that liberals were
defined by acceptance of a full-blown principle of
self-ownership, and retracts his characterisation in
Cohen (1995), p. 116 n. 1.
27. Schweickart (1993), p. 17.
28. Even within a just society it would be dubious to
assume that there will be a watertight correlation
between people's natural talents and abilities and their
ability to garner greater income and wealth under
`unregulated' social and economic conditions. This
is because, even if all children were to be guaranteed
equal access to high-quality education, quirks of their
personalities and/or accidents of family circumstances
(such as whether their parents or other family mem-
bers or close associates made them feel loved and
secure), and the operation of blind chance in general,
will affect whether or not particular individuals are
successful. This is not to say that in presently existing
societies there is no discernible correlation between
natural talents/abilities and success, nor to deny that
the correlation should and could be made much higher
in a genuinely just society. It is only to say that the
assumption that `the talented' in even a perfectly just
society will be able to parlay their talents into success
(however they come to define it) is, to some extent,
unjustified.
29. See Cohen (1996).
30. For a defence of the opposite position, however, see
Van Parijs (1991).
 RAWLSIAN THEORY AND CONTEMPORARY MARXISM
31. Rawls (1990), p. 42. I would like to thank John Rawls
for giving me permission to quote from this typescript
in my work.
32. See Arneson (1989). Dworkin (1981a) and (1981b)
also discusses many of the ideas within these princi-
ples.
33. See Cohen (1989).
34. Rawls (1972), pp. 87±8.
35. Rawls (1990), p. 42.
36. Arneson (1988), p. 89.
37. See Sen (1984), chapter 13, (1985),(1992), and
`Capabilities and Well-Being' in Nussbaum and Sen
(eds) (1993), pp. 30±53; for commentary on Sen, see
Crocker (1992).
38. Cohen (1994), p. 119.
39. See Peffer (1990), pp. 137±65 for my analysis of this
debate.
40. Ibid., pp. 142±6. The fact that Marx often called this
an example of `theft' by capitalists is strong evidence
that ± perhaps despite himself ± he did not think
morality could only ever be an ideological phenom-
enon relative to and supportive of a specific mode of
production, a `prop' which would become increasingly
redundant in post-capitalist societies.
41. This section follows ibid., pp. 146±7.
42. Peffer (1990), pp. 433±4.
43. The notion of `regulated' market economy as an
antonym for `capitalist economy' is pivotal here, but
would obviously need to be substantiated in a fuller
account, not least because no capitalist economy is
entirely unregulated and so cannot accurately be
defined as being in opposition to regulation. In gen-
eral, what will distinguish the market socialist econ-
omy in this regard is the degree of regulation and, most
signficantly, the ends to which it is put.
44. Here, we should also confront the claim by various
postmodernists that it is no longer possible to take
Marxism to be a serious theoretical tradition or soci-
alism to be a legitimate historical goal because they are
based on so-called grand historical meta narratives.
But the brief outline of the lower-level empirical
claims still taken seriously by the Analytical Marxists
and others given in this section should suffice to show
how utterly bogus this claim is. For the parts of Marx's
theory still being utilised have nothing whatever to do
with grand quasi-Hegelian claims about inevitable
movements in history. Indeed, they are quite the
opposite: they are the more concrete empirical claims
that can be (and are) seriously discussed by social
scientists, political thinkers and ordinary people inter-
ested in politics everywhere. To think politically we
cannot avoid making certain assumptions about what
social, economic and political institutions, pro-
grammes and policies will best accomplish given goals
or support and promote certain values. So whether we
are dealing with those postmodernist thinkers who are
131
operating on the ultimately self-refuting notion that
there is no such thing as rationality, or with those who
merely claim that Marxist theory is irrelevant and/or
illegitimate because it is (supposedly) based on grand
historical meta narratives, the claim that all aspects of
Marxism and all hope for socialism must be abandoned
just doesn't hold water. To say that Marxism and
socialism are now irrelevant theoretical or practical
political traditions is absurd given the quantity and
quality of the theorising that they still inspire, which is
due to the insights they offer into humanity's most
urgent political challenges. Given the popularity and
influence of postmodernism within certain circles, it is
vital to insist upon these points.
45. Peffer (1990), pp. 458±9. In fact, this set of assumptions
is so minimal that, if they can indeed be adopted by non-
Marxists/socialists, some would say that they are not
Marxist at all. But because they originate within the
Marxist tradition and continue to animate it, I see no
good reason to withdraw the label from them ± parti-
cularly when, as in my case, they are used to energise a
particular political position opposed to capitalism for
which Marxism has long held the banner.
46. Notice that this list of socialist theorists (who are,
broadly speaking, within the analytic-linguistic tradi-
tion of philosophy) by no means exhausts the category
of socialist or quasi-socialist theorists who participate
in contemporary debates since it doesn't mention: the
Yugoslav Praxis school of Marxian philosophers and
similar socialist humanists (such as Svetozar Stojono-
Â, Mihaelo Markovic
vic Â, Gajo Petrovic
Â); critical the-
orists (such as Ju
È rgen Habermas, Karl-Otto Apel,
Albrecht Wellmer, Thomas McCarthy, Richard Bern-
stein, Nancy Fraser, Doug Kellner and Stephen Best);
Hegelian Marxists (such as Shlomo Avineri and Ber-
tell Ollman); liberation philosophers (such as Enrique
Dussel and Horacio Cerruti); Cuban political philo-
sophers (such as Thalia Fung, Olga Fernandez, and
Pablo Guadarrama); other socialist philosophers from
the developing world (such as Francisco Miro Ques-
ada, Adolfo Sanchez Vazquez and Pablo Gonzalez
Casanova); socialist feminists (such as Alison Jaggar,
Ann Ferguson, Heidi Hartmann and Nancy Hart-
sock); socialist theorists who are primarily economists
and only secondarily philosophers (such as Alec Nove,
Branko Horvat, Jaroslav Vanek, Thomas Weisskopf
and Robin Hahnel); world-systems theorists (such as
 Gunder Frank and
Immanuel Wallerstein, Andre
Samir Amin) or theorists of development ethics (such
as David Crocker, Denis Goulet and Luis Camacho);
radical sociological theorists (such as Claus Offe,
Ernesto Laclau, Alain Touraine); social-democratic
theorists (such as Michael Harrington, Robert Heilbr-
oner and Cornell West); or the so-called Marxist
amoralist theorists who argued that Marxism is in-
compatible with morality in general or at least with
 RODNEY G. PEFFER
justice in particular (such as Allen W. Wood, Richard
Miller, Anthony Skillen and Andrew Collier). This
last view is one that I spend over 100 pages critiquing
and, I believe, decisively refuting in Peffer (1990).
47. See Krouse and McPherson (1988); Peffer (1994).
48. See Roemer (1994).
49. Van Parijs (1992).
132
50. I would like to thank G. A. Cohen and David Miller
for comments on a previous version of this paper given
at Oxford University as well as Neil Harding for
comments on a version of it given at the University
of Wales Swansea. I would especially like to thank
Mark Evans for his incisive comments and his editorial
help and suggestions.
 Part Four
PROBLEMS OF LIBERAL JUSTIFICATION
This page intentionally left blank
 10
DISENCHANTMENT AND
THE LIBERALISM OF FEAR
Peter Lassman
The wars of this century with their extreme violence
and increasing destructiveness, culminating in the
manic evil of the Holocaust, raise in an acute way the
question whether political relations must be gov-
erned by power and coercion alone. If a reasonably
just society that subordinates power to its aims is not
possible and people are largely amoral, if not incur-
ably cynical and self-centred, one might ask with
Kant whether it is worthwhile for human beings to
1
live on the earth?
I
The argument of this chapter is:
1. that pluralism, in the sense of `the wide di-
versity and deep conflict among different con-
ceptions of ends prevalent in modern societies
and different patterns of life and the institu-
tions that go with them', is the distinctive
2
problem which modern politics has to face;
2. that before this topic became of central con-
cern for current political theory Max Weber
had proposed an interpretation of the modern
world in terms of value pluralism and `the
disenchantment of the world' which provides
the (usually unacknowledged) foundation for
most contemporary liberal debate. This ac-
count, however, raises serious problems for
our understanding of modern liberalism, de-
mocracy and the legitimation of political rule.
135
One consequence is that this leads to what we
might call the `Weimar problem';
3. that John Rawls's Political Liberalism, the most
systematic attempt to construct a liberal re-
sponse, fails in its attempt to provide a solution
to the problem. Contemporary liberalism
ought instead to present itself in terms of
Judith Shklar's idea of `the liberalism of fear'.
This humble, cautious liberalism is probably
the best that liberals can hope for in such a
pluralistic social world. Fortunately, it is en-
ough to save us from conceding the bleakly
negative possibility described in the opening
quote.
This chapter's epigraph, taken from the introduc-
tion to the second edition of his Political Liberalism,
can be read as a statement of the `esoteric doctrine' or
`inner citadel' at the heart of John Rawls's political
3
philosophy. It is what motivates his project and the
centrality of the discussion of Rawls's work in con-
temporary political philosophy is a reflection of the
insightful way in which it engages with the problem
which all contemporary political philosophies face in
light of the conditions of modernity first sketched by
Max Weber. This claim, of course, appears to pre-
suppose that we accept Weber's sketch as true. What
I do think is true is that the picture of the modern
world which Weber painted has been highly influ-
ential even, or especially, among those who deny its
relevance or who fail to mention it in their work. In
 PETER LASSMAN
other words, it haunts much contemporary political
thinking despite the negative assessment many thin-
4
kers make of it. Here, then, I simply want to stand
by a recent statement made by Habermas: `I feel that
Max Weber's question regarding the paradoxes of
rationalisation is still the best key to a philosophi-
cally and scientifically informed diagnosis of our
5
time.' We shall shortly see how, despite itself, the
Rawlsian viewpoint is shadowed by the Weberian
sketch. Let us begin, though, by seeing how Rawls
conceptualises his project of political philosophy as
the positive response to Kant's question.
II
In the concluding paragraphs of his introduction to
the second (paperback) edition of Political Liberalism
Rawls refers to three kinds of conflict in modern
societies, the confrontation with which is the main
preoccupation of his political philosophy. The first
derives from `citizens' conflicting comprehensive
doctrines' (the organised sets of beliefs,
and goals by which people structure their lives);
the second derives from citizens' `different status,
class position, and occupation, or from their ethni-
city, gender, and race'; while the third is a result of
what Rawls calls the `burdens of judgement' (the
inevitable indeterminacies and consequent diver-
gent results of rational and reasonable normative
deliberation in free societies). Rawls states that his
theory of political liberalism `mitigates but cannot
eliminate the first kind of conflict, since compre-
hensive doctrines are, politically speaking, unrec-
oncilable and remain inconsistent with
another'. Nevertheless, Rawls argues that the prin-
ciples of political liberalism, `the principles of justice
of a reasonably just constitutional regime', are cap-
able of reconciling the second kind of conflict.
Once we accept the principles of justice, preferably
those of his own theory, this kind of conflict `need
no longer arise, or arise so forcefully'. It appears,
then, to be this kind of conflict whose non-resolu-
tion would plunge us into the amoral world of `pure'
power described in the opening quote. And it is the
confidence that it can be transcended which stokes
Rawls's optimism. Unfortunately though, he thinks
conflicts `arising from the burdens of judgement
always, however, remain and limit the extent of
6
possible agreement.'
This limitation may explain why, in his more
recent work, there are several hints that he is
troubled by the thought that his whole project
cannot possibly succeed, that conflict will always
undermine the possibility of agreement. He accepts
that many will `regard it as obvious' that `a just and
well-ordered democratic society is not possible'. But
this is a conclusion he thinks we must resist. One of
the interesting reasons why Rawls feels that we can
resist this conclusion is bound up with his concep-
tion of the nature of political philosophy itself. In
some of his remarks on this Rawls is responding to
critics such as Michael Walzer who have charged
that his theory is too abstract, that it does not deal
with the real problems of contemporary politics, and
that its philosophical structure displaces `genuine
politics' and is therefore entirely at odds with the
7
nature of democratic pluralist society.
The thought here is that the problems of social
values and political conflict are unlikely to find their
resolution in his kind of abstractive philosophical
methodology. But Rawls himself has always expli-
citly argued the reverse. He states very clearly that in
`political philosophy the work of abstraction is set in
8
motion by deep political conflicts.' We turn to
political philosophy when our `shared political un-
derstandings' break down and `equally when we are
torn within ourselves'. Abstraction from that which
divides us is a way of finding grounds for ideas which
might unite us and that is where we find the basis for a
principled resolution of conflict.
one The emphasis on possible unity shows that Raw-
lsian political philosophy is not in the business of
uncovering actual conflict-resolving consensus upon
principles. Operating in the realm of the possible
obviously includes acknowledging the possibility of
failure, hence Rawls's disquiet. Yet we ought not to
accept the sceptical argument against the possibility
of a conflict-resolving (or at least a conflict-mana-
ging) and well-ordered liberal-democratic society, in
Rawls's view, without first being clear about what the
consequences would be `for our view of the political
9
world, and even the world as a whole.' Rawls's
conception of the nature of political philosophy
and, by implication, of the responsibility of the
136
 DISENCHANTMENT AND THE LIBERALISM OF FEAR
political philosopher means that we must be aware
that the answer we give to general and abstract
questions does have a bearing upon our `background
thoughts and attitudes about the world as a whole.'
Thus, if we are convinced on abstract philosophical
grounds that a just and stable democratic society is
not possible then this will have a great impact upon
`the underlying attitudes of the public culture and
the conduct of politics.' It is at this point that Rawls
revealingly points to the case of the Weimar Repub-
lic and to one of its most outspoken critics, Carl
Schmitt.
Rawls presents the Weimar Republic as offering us
a particularly dramatic example of exactly what he
fears. One of the reasons for the fall of the Weimar
Republic and its replacement by Nazi dictatorship
was that both the traditional elites and a large
section of the intellectuals were unwilling to support
it on the grounds that they could no longer believe
that any form of democratic and liberal society was
possible. Given what we now know of the terrible
consequences of this particular political
Rawls urges that we
must start with the assumption that a reasonably
just political society is possible, and for it to be
possible, human beings must have a moral nature,
not of course a perfect such nature, yet one that
can understand, act on, and be sufficiently moved
by a reasonable political conception of right and
justice to support a society guided by its ideals and
10
principles.
The vital moral and political need to believe in
the possibility of justice, however, should not lead
theorists to a wilful blindness over the true nature of
the pluralism which divides societies and puts the
question of justice on the political agenda. Thus the
idea of moral and value pluralism has increasingly
become a central concern for political theorists,
especially liberal theorists, working in both what
can loosely be called the Anglo-American and Con-
tinental traditions of political thought. Pluralism is
therefore widely seen as a defining characteristic of
modernity and as the most significant challenge to
political philosophy's search for conflict-resolving
agreements. How it is to be conceptualised, then,
crucially impacts upon how we might develop pos-
sibilities for the avoidance of the Weimar problem.
III
The variety of ways in which liberal theorists have
responded to this problem is striking, indicating the
variety within the liberal tradition. There are those,
such as Rawls, Larmore, Nozick and Hayek, who
have asserted the reality of moral pluralism while
arguing, at the same time, for the possibility of a
ranking of values so that one value, liberty, can take
precedence. Others, such as Williams, Berlin and
Hampshire, have stressed the ineliminable and often
tragic character of value conflicts. For example,
Hampshire has argued the case for a strong thesis
that, on the basis of an analogy between moralities
and natural languages, `nature has so designed us
that, taking humanity as a whole and the evidences
of history, we tend to have conflicting and divergent
moralities imbedded in divergent ways of life, each
failure, the product of specific historical memories and local
11
conditions.' For him
the diversity in conceptions of the good is an
irreducible diversity, not only because no suffi-
cient reasons have been given, and could ever be
given, for taking one end, such as the general
happiness or the exercise of reason, as the single
supreme end; but also because the capacity to
develop idiosyncrasies of style and of imagination,
and to form specific conceptions of the good, is
the salient and peculiar capacity of human beings
12
among other animals.
However, this second group's acceptance of the
reality (as they see it) of such conflicts has created
problems for the coherence and tenability of their
own defence of the value of liberalism. If the ex-
istence of plural and conflicting values is accepted as
an unavoidable reality then it follows that the claims
made for any particular political theory must be at
best both severely limited and local in their applica-
tion. But the first group of theorists, in accepting that
we are in some way faced with a pluralism of funda-
mental and conflicting values, are confronted by the
problem of explaining how their own theory's sup-
137
 PETER LASSMAN
port for liberal principles is able to escape from
insoluble conflicts of value.
From Rawls's earliest work it is possible to observe
a concern with how we might get out of this parti-
cular bind. In his first published paper `Outline of a
Decision Procedure for Ethics' he asked if there
existed
a reasonable decision procedure which is suffi-
ciently strong, at least in some cases, to determine
the manner in which competing interests could
be adjudicated, and, in instances of conflict, one
interest given preference over another; and,
further, can the existence of this procedure, as
well as its reasonableness, be established by ra-
13
tional methods of inquiry?
In his most recent writings Rawls is concerned to
work out his version of `political liberalism' with the
same set of problems in mind. The central problem
which he now recognises as forcing him to modify
the account of stability put forward in Part Three of
his A Theory of Justice is what he terms `the fact of
reasonable pluralism'. Pluralism is `reasonable', he
says, because it is the outcome that we should expect
of people freely using their powers of moral reasoning
under liberal-democratic institutions; it is, as we
have noted, the product of the burdens of judgement.
In other words, such is the nature of human value
and value judgements in free societies that we should
expect such conflict.
Crucially, Rawlsian `reasonable pluralism' is not
itself supposed to be a claim about the truth of
values (for that is something that can be reasonably
contested between monists and pluralists), only
about the reasonableness of the disputes that con-
14
stitute the fact that there are plural values. It is
not officially, then, conceding the truth of the
Weberian sketch of modernity (for that is just
one of potentially numerous incompatible world-
views it is reasonable to affirm). Support for the
claim that Rawls inadvertently leans in Weber's
direction has yet to be adduced, but there is already
an initial, fleeting point of convergence: both wish
to accommodate, not suppress, pluralism. Rawls
poses the challenge thus: `How is it possible that
there may exist over time a stable and just society of
138
free and equal citizens profoundly divided by rea-
sonable though incompatible religious, philosophi-
15
cal, and moral doctrines?'
There is some disagreement among advocates of
pluralism concerning its historical status. Hampshire,
as we might ascertain from his quotation above, sees
pluralism as not being either a peculiarly modern or
democratic phenomenon or an unexplained general
fact of human existence. Moral pluralism and moral
conflict are, for him, an essential product of the
imaginative powers of human beings in general. But
the idea that an irreducible plurality of values has
transformed the way people think about how a
specifically modern politics should be conducted is
increasingly influential. Rawls makes this kind of
interpretation of our moral history absolutely central
for his account of `political liberalism'. He argues that
`modern democratic political culture' itself is to be
sharply distinguished from all that preceded it so that
its `special nature' accounts for, to a large extent, `the
different problems of political philosophy in the
16
modern as compared with the ancient world.'
Rawls cites both Weber and Isaiah Berlin as
offering important and influential accounts of the
idea of the pluralism of values. These thinkers,
however, provide ways of challenging his own ac-
count of pluralism's nature and hence his ideal
solution to the Weimar problem. Let us, then, look
first at Berlin's ideas.
Initially, Berlin appears to be equivocal on the
question of the historical character of pluralism. In
tracing the roots of the doctrine of pluralism to the
Cynics and the Sceptics he seems to be saying that
there is a perennial quality in the conflict between
monism and pluralism in western political thought.
But, in pointing to the importance of Machiavelli as
the `first person to indicate a conflict of real values',
he also indicates the significance of a peculiarly
17
modern quality to the recognition of pluralism.
According to Berlin it was Machiavelli who was
the first to make it clear that the very idea of a
Christian commonwealth is a contradiction in
terms: one cannot be a Christian and an heroic
Roman citizen at the same time. Christians must
remain humble, to be trampled on at times; Ro-
mans resist this successfully. This implies an
 DISENCHANTMENT AND THE LIBERALISM OF FEAR
irreconcilable dualism. One can choose one life or
the other, but not both; and there is no over-
arching criterion to determine the right choice;
one chooses as one chooses, neither life can
18
objectively be called superior to the other.
This prompts a significant insight according to
Rawls, which goes a long way in distinguishing
modern liberal from premodern political thought.
This is the claim `there is no social world without loss
± that is, no social world that does not exclude some
ways of life that realise in special ways certain
fundamental values. By virtue of its culture and
institutions, any society will prove uncongenial to
some ways of life'. These values may be fully objec-
19
tive, but the harsh fact of life is that their full range
is too extensive to fit into any one social world; not
only are they incompatible with one another,
imposing conflicting requirements on institutions,
but there exists no family of workable institutions
with sufficient space for them all. That there is no
social world without loss is rooted in the nature of
values and the world, and much human tragedy
reflects that; a just liberal society may have far more
space than other social worlds, but still, it can never
20
accommodate all possible ways of life.
For Berlin, as for Rawls, the reality of the pluralism
of values is tied to the possibility of political philo-
sophy itself. He argues that if
we ask the Kantian question, `In what kind of
world is political philosophy ± the kind of dis-
cussion and argument in which it consists ± in
principle possible? the answer must be `Only in a
21
world where ends collide.'
But the Berlinian thesis of pluralism (the now
much-discussed `inner fortress' of his thought) warns
political philosophy that `the pursuit of a single,
final, universal solution to human problems' is a
fundamental and dangerous error. It is essential to
recognise that there are
many ideals worth pursuing, some of them are
incompatible with one another, but the idea of an
139
all-embracing solution to all human problems
which, if there is too much resistance to it, might
need force to secure it, only tends to lead to
bloodshed and the increase in human misery.
Berlin takes the idea of the pluralism of values to
be central for his understanding of both the history of
political thought and of his own liberalism, not in
the sense that he has provided a general explanation
for the existence of pluralism, but simply in the
consequent necessity that we be reconciled with
the basic fact of life that `values can clash ± that
22
is why civilisations are incompatible.' The plural-
ism of values implies the `clash of values'. If the
reality of plural and conflicting values is ineliminable
then it follows that there is no `perfect world in
which all good things can be harmonised in princi-
23
ple.' The implication of this is, for Berlin, that the
search for perfection or utopia is both incoherent and
dangerous.
This understanding of the pluralism of values
coheres with Berlin's depiction of the good society.
The good society is to be understood as one in which
maximum space is allowed for those forms of life
which can be guided by a principle that competing
values can be allowed to flourish within the terms of
a shared social space. Berlin thinks that the form of
society in which this is most likely to occur typically
happens to be governed by the principles of liberal-
ism. The virtue of a liberal society resides in the fact
that, while the impossibility of realising all relevant
values remains an impossibility, it ± more than any
other kind of society ± is one which recognises and is
strengthened by the pluralism of values. It is one of
the marks of a liberal society that it tolerates or even
encourages value pluralism. It does not attempt to
reduce the pluralism of values in the interests of any
one particular value or set of values supported by the
authority of the state.
Clearly, for Berlin, the recognition of the reality of
the pluralism of values (his `objective pluralism') is
itself to understand a truth about the nature of
values. This is the main point of departure from
Rawls, who wants his political philosophy to abstain
from such truth-statements about what he regards as
reasonably contested metaphysical and meta-ethical
claims. Plainly, too, Berlin takes our awareness of the
 PETER LASSMAN
existence of plural and competing values as being
itself a good. He seems to take this understanding as
itself constituting an argument for liberalism because
a liberal society will express this understanding more
24
than any other. As pluralism implies that `since it is
possible that no final answers can be given to moral
and political questions, or indeed any questions
about value' then
room must be made for a life in which some values
may turn out to be incompatible, so that if
destructive conflict is to be avoided compromises
have to be effected and a minimum degree of
toleration, however reluctant, becomes indispen-
25
sable.
This suggests that those societies which respect
this truth are preferable to those which do not.
Furthermore, there seems to be an assumption here
that not all values possess an equal survival value
within a society which allows maximum space for
value pluralism (those forms of life which are less
tolerant of pluralism are perhaps more likely to be
`squeezed out', yet this may be cause for less regret
26
than if other forms of life were to wither). In this
sense value pluralism itself cannot pretend to be
neutral with regard to outcomes even if its philoso-
phical outlook and political intent is neutral with
regard to the initial support, or lack of it, given to
ways of life supported by different values.
Nevertheless, the connection between liberalism
and pluralism is not as cast-iron as one might expect.
Berlin insists that pluralism and liberalism are `not
the same or even overlapping concepts. There are
liberal theories which are not pluralistic.' While
defending the value of both he is clear that there
is no logical connection between them. The con-
nection between pluralism and liberalism is one of
27
`elective affinity', grounded as much as anything in
28
historical experience. But it could be argued that
not only is there no logical connection between
liberalism and pluralism but that they are actually
incompatible. If pluralism is taken to mean that there
is no one particular value or set of values which,
when in conflict with other values, can always claim
our allegiance then it must be incompatible with
liberalism. As a distinctive doctrine, the latter must
140
be committed to the primacy of at least the value of
liberty. In other words,
the mere fact that values are `plural' . . . tells us
nothing about which of the vast range of values
known to us from human experience are the
values we ought to choose for ourselves and
our social institutions. Pluralism tells us that we
29
must choose but not what to choose.
Furthermore, it is notable that Berlin (despite his
concern with the significance of Romantic and
Counter-Enlightenment thinkers) does not discuss
those twentieth-century theorists, such as Carl
Schmitt, who having taken the argument for plur-
alism seriously saw it as the first step in the destruction
of liberalism as a substantive normative standpoint.
This may be explained by the fact that Berlin's
pluralism ultimately stops short of being absolutely
`total' in the sense that, after all he says, he still
balances it with a belief in a universal human nature.
His conception of it is thin enough to underdeter-
mine any one substantive moral doctrine but it is still
capable of yielding the basic principles of a universal
30
`minimum morality'. This theoretically prevents
the concession of a total incommensurability of
values and the relativising impossibility of critique
levelled at any moral schema. Such a morality could
further justify a liberal social order, and values/forms
of life sympathetic to its principles, as an explanation
of the liberal bias in Berlin's view. But because this
morality does not constitute liberalism in its entirety,
it is in principle compatible with other forms of
social order. The connection between liberalism
and pluralism remains historically contingent; plur-
31
alists need not be liberals.
It is, of course, a profoundly contentious point as to
exactly how such a `minimum morality' position can
be defended from the kind of objective-pluralist per-
spective with which Berlin confronts us. An indica-
tion of the difficulty is clearly signalled when he
equates the universal minimum-moral outlook with
that of the less-than-global `western world'; as Perry
Anderson says, `the universal shrinks with every step
32
back to the local.' If this outlook is supposed to be
that upon which the possibility of `justice' ± under-
stood loosely as the peaceful, principled resolution of
 DISENCHANTMENT AND THE LIBERALISM OF FEAR
conflict ± is to rest, perhaps it is one that political
philosophy has to construct rather than just assume is
ready to hand. Here, then, is Rawls's main difference
from Berlin and his distinctive characterisation of the
fact of pluralism is the key to the hope that the
Rawlsian project will not founder on the same kind
of justificatory paradox that underpins Berlin's at-
tempt to stand for a universal minimum morality.
At this point we need to ensure that we keep clear
the distinction between (1) pluralism as a doctrine
about the nature of value (Berlin's position) and (2)
the phenomenon of deep disagreement about the
nature of the good (Rawls's position, based upon `the
burdens of judgement'). This is a move urged by
Larmore, who originally coined the term `political
liberalism'. The reasoning behind the latter is that
pluralism, understood as a doctrine about the nature
of the good, is itself a controversial doctrine. If this is
so then pluralism cannot possibly stand as a support
for `political liberalism' which itself attempts to stand
clear of all controversial philosophical and religious
doctrines in search of the agreement upon principles
which it posits as necessary for political resolution of
33
the conflicts of value. The idea that reasonable
people will tend to disagree about significant ques-
tions is not to be confused with either pluralism or
scepticism. It is, according to Larmore, the fact of
reasonable disagreement rather than the `doctrine of
pluralism' which lies at the heart of modern liberal
political philosophy.
Nevertheless, Larmore's idea of reasonable dis-
agreement shares with Berlin's concept of pluralism
the belief that they are both reflections of a pecu-
liarly modern reality. Despite his criticism of Berlin's
failure to distinguish pluralism from the persistence
of reasonable disagreement he sounds much like
Berlin in his argument that we are now reversing
the belief embedded in our `intellectual tradition'
which `has been to a large and important extent
rationalist', that `reason is what brings us together.'
On this account pluralism is a
characteristically modern outlook. It is likely to
appear attractive only against the background of
metaphysical and religious disappointment.
recommends itself to those who continue to
believe that value can be objective, something
more than just the expression of our preferences,
but who refuse to believe that it has its home in a
harmonious cosmos (as Aristotle thought) or in
God's plan.
For Larmore, liberalism too is very much a product of
this intellectual milieu. It has `distinguished itself
from earlier political philosophies by its refusal, ever
more pronounced, to base the principles of associa-
tion upon a vision of God's plan or of an ordered
35
cosmos.' It is unsurprising, then, that the doctrines
of liberalism and pluralism have seemed so closely
entwined. And even though the Rawls/Larmore
project wants to distinguish the former, in its epis-
temically abstinent `political' guise, from the latter as
a doctrine about the objective nature of value,
Larmore concedes that both are `distinctively mod-
ern in that they have something to do with the
metaphysical-religious disenchantment of the
36
world.' And this is itself a particular metaphysical
standpoint on the nature of value.
The suggestion, then, is that some version of
`doctrinal pluralism' is an idea that shadows moder-
nity and hence contemporary liberalism, even when
the latter attempts to distinguish it from the separate
thesis of pluralism as reasonable disagreement. As we
saw with Berlin, it is this idea which makes the
justification of liberalism so problematic.
It is in the light of this suggestion that we can
illuminate the problem of liberalism and pluralism by
placing it into the context of Weber's thought.
IV
Weber was probably the first to take seriously as a
basic feature of modernity the idea of a pluralism of
conflicting values in a way which has a direct bearing
on politics. Most of the themes concerning the
question of incommensurability, disenchantment
34
and tragic choices which are discussed by Berlin,
Larmore, Williams, Hampshire and others were
raised in his work. Although I do not want to claim
that Weber carried out a systematic and satisfactory
analysis it is instructive to take a brief look at his
It discussion of these topics because it starkly reveals
some of the problems and dangers involved. Weber is
therefore interesting because his problems remain, in
141
 PETER LASSMAN
my view, to a large extent our problems. Although
there is some controversy concerning the question of
whether Weber can be classified as a liberal at all it
seems to me that he was, at least, `a Liberal of a
37
strange kind'.
Weber's problem was that he wanted to defend
liberal values and institutions while, at the same
time, his general philosophical outlook with its
acceptance of disenchantment and pluralism as
our fate led to a radical scepticism concerning our
ability to produce, under modern conditions, a con-
vincing theoretical justification for them. In other
words, in Weber's account we have a defence of
liberal institutions and values without the support of
38
an explicit liberal political philosophy. At the
extremes this could be interpreted, for example, as
either an early recognition of postmodern reality, the
inevitable contingency of any `defence' available to
liberalism, as Rorty would argue, or as an example of
the failure of the whole tradition, as Leo Strauss and
his followers seem to urge.
Weber's account of the fate of politics in a world of
plural and conflicting values is embedded in a `nar-
rative of modernity'. According to him we live in an
increasingly secular, rationalised and `disenchanted'
world in which we are confronted with an opposition
between the seemingly unstoppable development of
the `housing for the new serfdom' (the instrumen-
tally rationalising bureaucratisation of social institu-
tions and practices) and the need for the individual
to make choices between ultimately irreconcilable
ultimate values, `the unending struggle between the
39
gods.' On Rawls's analysis, here is a key point of
difference between Weber and Berlin. Eschewing
any notion of objective pluralism, `Weber's view
rests on a form of value scepticism and voluntarism;
political tragedy arises from the conflict of subjective
40
commitments and resolute wills.' Rather than the
apparent reification of values through the language
of `objectivity' (which also facilitates the notion of a
thin objective morality that we have argued pluralists
struggle consistently to defend), Weber's outlook
strips `reality' right back to the bare fact of people's
different, clashing subjective opinions and wills.
Furthermore, the deeper the process of disenchant-
ment the more intense the struggle between their
contending viewpoints becomes.
142
There are doubtless numerous ways of rendering
vivid the idea that Weber somehow haunts con-
temporary liberalism, including political liberalism,
and why it is his pluralism ± as opposed perhaps to
Berlin's ± that informs the most disturbing challenges
to it. But here are a few considerations that can
contribute to it. First, Larmore is right that the
peculiarly modern phenomenon of disenchantment
has historically fuelled the politics of liberalism in so
far as people have increasingly appreciated that
politics in modern, diverse societies cannot rest upon
any one `enchanted' doctrine; to avoid outright
conflict, it must accommodate a plurality of such.
Second, this concession to the existence of plural
values has a tendency (albeit not logically) to further
the subjectivist account of value which the supre-
mely disenchanted (`scientific') practices of instru-
mental rationality promote as they burgeon in social
life. Third, this renders increasingly irresistible a
frankly Nietzschean view of politics, where subjec-
tive wills battle it out for power over each other with
no grand metaphysical-moral story to be told by
which to cloak the brute reality of the struggle.
Fourth (and this is arguably where Weber himself
ultimately stumbled), the case for liberalism ± or any
other political order ± rests only upon individuals'
decisions to support it, with nothing independent to
ground or validate them, a form of political existen-
tialism in other words. And if accelerating disen-
chantment is deepening pluralism, the prospects of
conflict-managing agreements in decisions accord-
ingly recedes.
The problem with Weber's position looked at
from some liberal points of view is that he was only
able to offer two unacceptable solutions for the
predicament of the individual under these condi-
tions. One is a re-enchantment of the world, which
would mean a return to rule by what Rawls would
call a fully comprehensive moral and philosophical
doctrine. As Anderson stresses, First World War
nationalism disturbingly seems to function in this
41
role for Weber. Here, the arbitrariness of his
decisionism propels the individual into a radical
fatalism; the accident of birth within a particular
national community is our chance to escape the
gloom of disenchantment, to embrace the destiny
of the nation as our own. The remaining sphere for
 DISENCHANTMENT AND THE LIBERALISM OF FEAR
freedom would be one in which the individual is
thrown back on his or her own resources (the
meagreness of which may encourage individuals to
give up still more of their own `being' to the `nation's
cause').
The other solution perhaps lacks some of the
ominous portent of the first, but is ultimately no
more satisfactory. In his later writings on political
affairs Weber came to argue that in a modern
democratic state the only acceptable response would
be a mixture of charismatic rule backed up by a legal
order as a form of rule. The problem here is that the
legitimacy of this form of state is based entirely upon
its success in being successful. There are `no other
42
criteria for success beyond success itself.' And if
regimes resting upon such bases are not `successful' ±
if, for example, they struggle to cope with social and
political conflict ± they have nothing upon which to
fall back to argue that they are nevertheless legit-
imate. (Compare this with a regime whose citizens
widely accept its authority as being derived from
`God'. It, too, may experience functional inefficien-
cies but its lack of success doesn't exhaust its legit-
imating grounds and hence it is likely to be more
stable.) The problems and tensions in Weber's ac-
count of the problem of the legitimation of political
rule within a plural moral universe lead directly to
the `Weimar problem', the crisis of justification
which liberalism faces as soon as the implications
of pluralism become apparent and are thought
through in a particularly radical direction.
The `Weimar problem' which follows from We-
ber's diagnosis of the political crisis of our time is one
in which the central debate concerns liberalism's
ability to cope with pluralism. A common conclu-
sion reached is that it could not. According to Carl
Schmitt, for example, all political theories contain
an element of `surplus value' of meaning. It is in their
nature to claim a truth for themselves which they
cannot possibly justify by rational means. In the eyes
of many, and Schmitt is the most well-known ex-
ponent of this view, liberalism is bound to oscillate
between a self-interpretation of impartiality, or neu-
trality mixed with proceduralism and a commitment
to the more substantive values which it peculiarly
embodies. (Schmitt's particular complaint was that
liberalism's claims to be objectively valid conflicted
143
with democracy's claim to value and embody the
collective will of the people, making `liberal democ-
racy' ultimately oxymoronic and ± particularly when
popular will runs against liberal democracy ± doomed
43
to implosion. )
It is in the light of this that we can see why, for
Rawls, it might seem to be `a great puzzle to me why
political liberalism was not worked out much earlier:
it seems such a natural way to present the idea of
liberalism, given the fact of reasonable pluralism in
44
political life.' For him, liberalism has been forced to
become political in the way in which he proposes
because, under conditions of pluralism, it cannot
assume that liberal ideas are the non-controversial
45
foundation of social and political life. It is, then, a
direct development from the specifically modern
form of post-Reformation religious pluralism. `For
the moderns the good was known in their religion;
with their profound divisions, the essential condi-
46
tions of a viable and just society were not.' This
distinguishes the modern from the ancient world.
`What the ancient world did not know was the clash
between salvationist, credal, and expansionist reli-
gions.' Crucially, this
introduces into people's conceptions of their good
a transcendent element not admitting of compro-
mise. This element forces either mortal conflict
moderated only by circumstance and exhaustion,
or equal liberty of conscience and freedom of
thought. Except on the basis of these last, firmly
founded and publicly recognised, no reasonable
political conception of justice is possible. Political
liberalism starts by taking to heart the absolute
47
depth of that irreconcilable latent conflict.
Political liberalism's task is to seek self-justifica-
tion in terms of ideas which, aside from those which
reasonably divide them, all people qua citizens of
constitutional democracies can nevertheless share.
In other words, as the basis of political agreement,
liberalism must try not to take sides in the wider,
plural comprehensive doctrines of value. Hence the
great significance he attaches to the role of political
philosophy and its abstract constructivist methodol-
ogy: it is, with hindsight, the obvious as well as
morally and politically necessary strategy to ward
 PETER LASSMAN
off the Weimar problem. It helps us, too, to see that
modernity need not treat pluralism as some kind of
secular equivalent of the fall from grace. For Rawls,
the pluralism of comprehensive doctrines is `not seen
as a disaster but rather as the natural outcome of the
activities of human reason under enduring free in-
stitutions. To see reasonable pluralism as a disaster is
to see the exercise of reason under conditions of
48
freedom itself as a disaster.'
At once humble in its attempt to avoid grand
`metanarratives' and ambitious in its quest for poli-
tical agreement amidst reasonable pluralism, it is
here ± in the account of pluralism that grounds it
± that we can begin to spot political liberalism's
problem. For, although Rawls does not see it this
way, the above assessment of pluralism is itself a
deeply controversial interpretation. After all one
could equally see this version of modern history as
a story of `error and illusion, or at least as the
49
intensification of tragic conflicts of value.' Rawls,
against the strictures of political liberalism, is already
favouring one version of history and one attitude
towards pluralism. Would those who do not accept
this version of history be included among `the rea-
sonable', those who in principle may endorse the
50
principles of political liberalism?
V
The paradoxical nature of modern liberalism resides
in the fact that it, more than other political the-
ories, takes pluralism, understood both in terms of
an interpretation of the sources of value and as
reasonable disagreement, to be both a fundamental
feature of the modern world and as a problem which
it must cope with in order to establish its own
legitimacy. One of the interesting features of Rawls's
attempt to construct a `political liberalism' is that he
sees pluralism as built into the nature of modern
constitutional liberal democracy, the natural out-
come of the `burdens of judgement'. However, to
take this example, Rawls fails to notice that in
defining one of the criteria of `the reasonable' in
his account of pluralism and the constitutional-
democratic citizen's appropriate response to it as
the `willingness to accept the burdens of judgement'
he is introducing into his theory a controversial
144
doctrine of the very kind which he is attempting to
exclude.
This profoundly suggests the possibility that there
is no one way of characterising pluralism, as a
grounding notion in any case for a liberal politics,
which could command universal agreement. It has
been frequently observed how, in other respects,
Rawls's doctrine falls back upon the kind of com-
prehensive liberalism he seeks to avoid in justifica-
51
tory arguments. What this argument suggests is
that we should not be very surprised at this, because
the account of pluralism with which the whole
project starts already begins to show `comprehensive'
biases.
The worry that we are doomed never to be able to
deliver a form of theory which is not just `another
controversial and partisan vision of the good life' and
which `will cease to be a plausible solution to one of
our most pressing moral and political problems' is at
the heart of modern liberalism's anxieties. Larmore
has argued that if we are so condemned then political
liberalism will `become just another part of the
problem'. The conviction that `we can agree on a
core morality while continuing to disagree about
what makes life worth living' may `turn out to be
baseless. Liberalism may necessarily be just one more
52
partisan ideal' ± a fear which is most perturbingly
captured in the context of the Weberian picture.
The strong resemblance the problem has to that
which contributed to the intellectual and political
collapse of Weimar democracy dramatises this pro-
blem into much more than just a philosophical
curiosity.
What chance, then, do liberals have of warding off
the Weimar problem without engaging in self-delud-
ingly reassuring `reassessments' of pluralism's proble-
matic significance? I want to suggest that Judith
Shklar's idea of the `liberalism of fear' indicates
the way forward. It has some interesting similarities
to Weber's interpretation of the fate of liberal ideas
under modern conditions and offers us a way of
thinking about the predicament of modern liberalism
which, while suitably pessimistic, is not unduly so. So
if it has been valid to talk of the Weberian under-
current in contemporary liberalism, the extent to
which the liberalism of fear can speak to it is a clear
indication of its relevance. Further, Rawls approv-
 DISENCHANTMENT AND THE LIBERALISM OF FEAR
ingly notes that the idea has certain similarities with
53
political liberalism too. However, it falls well short
of the full political-liberal strategy ± but herein may
lie its strength.
Shklar argues that `the liberalism of fear' does not
rest upon a theory of moral pluralism of the kind
advocated by Berlin. It does not deny pluralism,
though; rather, the point is that the case for liberal-
ism does not need to rest upon such metaphysical
54
claims about the nature of value and suchlike.
is the big difference between Rawls and Berlin on the
one hand, whose liberalisms start with specific ac-
counts of pluralism, and the liberalism of fear on the
other: all that the latter really needs to begin with is
an account of the `summum malum, which all of us
know and would avoid if only we could. That evil is
cruelty and the fear it inspires, and the very fear of
55
fear itself.' So for Shklar the `deepest grounding' of
liberalism as a distinct doctrine is to be found in the
conviction `that cruelty is an absolute evil, an of-
fence against God or humanity. It is out of that
tradition that the political liberalism of fear arose and
continues amid the terror of our time to have
56
relevance.' The fear which is to be prevented is
the fear created by arbitrary and unnecessary force
and by acts of torture and cruelty performed by states
and their agents. Historical experience has shown us
how such fears are not groundless nightmares; that
self-same experience teaches us how liberal political
practices and institutions have so often been the best
hope people have had to remove cruelty and fear
from political life. It is thus historical memory, as
opposed to speculative metaphysics, with which the
57
liberalism of fear should launch its case.
The case for seeing how the liberalism of fear may
address the fears raised by the Weberian picture of
modernity and the Weimar problem is twofold. First,
having explicitly eschewed reliance upon a Berlinian
account of pluralism, Shklar notes that there is a `real
psychological connection' between liberalism and
the kind of value-scepticism to which Weber gives
paradigmatic expression:
scepticism is inclined toward toleration, since in
its doubts it cannot choose among the competing
beliefs that swirl around it, so often in murderous
rage. Whether the sceptic seeks personal tranquil-
lity in retreat or tries to calm the warring factions
around her, she must prefer a government that
does nothing to increase the prevailing levels of
fanaticism and dogmatism. To that extent there is
a natural affinity between the liberal and the
58
sceptic.
Those who may be swayed by sceptical arguments
of the sort that Weber prompts are nevertheless
Here availed of a definite argument for liberal politics,
precisely because it tries to derive the latter from a
mutual concern or impulse ± the avoidance of fear
and cruelty ± rather than a substantive (contested)
metaphysical/moral doctrine. But those who have
not succumbed to scepticism, who continue to hold
other `comprehensive doctrines', can be equally im-
pressed by the case for liberalism because, again, it
rests upon a concern that is shareable not in people's
philosophies but in their psychological motivations.
Hence Shklar argues that to embrace the liberalism
of fear is not to embrace any form of normative or
cultural relativism. It makes a universal and cosmo-
politan claim for liberalism while, at the same time,
recognising its local origins and the infrequency of its
59
full realisation.
A crucial element in the liberalism of fear is what
some take to be the excessive modesty of its ends.
`Liberalism has only one overriding aim: to secure
the political conditions that are necessary for the
60
exercise of personal freedom.' Prohibiting interfer-
ence with the freedom of others is how liberal
politics removes what sources of fear it can from
people's lives; beyond that, `liberalism does not have
any particular positive doctrines about how people
are to conduct their lives or what personal choices
61
they are to make.' If some might think this is
unduly cautious, Shklar has more than just historical
memory to make us see the liberalism of fear as both
judiciously prudent and no mean achievement in its
own right. Some of us may nowadays be able to take
the security of the liberal order for granted and hope
for more adventurous and `imaginative' politics. But
our foreign news should surely remind us of the
global prevalence of the most brutal illiberalisms.
There may indeed be more morally `exhilarating'
62
liberalisms, but they are surely less urgent.
There is another way of expressing reservations
145
 PETER LASSMAN
about the apparently limited nature of Shklar's lib-
eralism. Those who are troubled by the Weberian
sketch of modernity may wonder whether, in its
refusal to theorise substantive `positive' goals for
the social and political order, it tacitly grants free
rein for instrumental rationality to dominate in
modern society. But Shklar believes the liberalism
of fear has two powerful responses here. First, if
`instrumental rationality' refers purely to political
practices which pursue only `efficiency' or `means±
end calculation', without questioning the rationality
or desirability of whatever end to which they are put,
then it does not apply to the liberalism of fear. For it
has very definite, substantive ends in its desire to
avoid the institutionalisation of fear and cruelty, and
political practices are to be explicitly scrutinised with
these in view.
The second response has `instrumental rationality'
describing a politics that places all its confidence in
procedures, without adequate attention to the ra-
tionality of the conduct and discourse of those who
participate in and follow them. It trusts the mechan-
isms for creating consent and ensuring fairness, with-
out any attention to the character of the individual
63
citizens or to that of the society as a whole.
But Shklar, of course, wants the liberalism of fear
to justify a liberal regime that may be little more than
a modus vivendi, a political accommodation shared by
people of diverse natures, tastes and goals mutually
motivated by a fear of political cruelty. To invest
politics with a more substantively rational goal, to
take sides in the pluralism of possible forms of life, is
to risk precisely the kind of totalitarian oppression
which provides part of the historical memory psy-
chologically motivating the liberalism of fear. To
think that this charge of `instrumentality' has any bite
is to show `disdain for those who do not want to pay
Notes
1. Rawls (1996), p. plxii.
2. Richardson (1994), p. 238.
3. The idea of the `inner citadel' here is drawn from
Berlin: `the deepest convictions of philosophers are
seldom contained in their formal arguments: funda-
mental beliefs, comprehensive views of life, are like
citadels which must be guarded against the enemy . . .
the inner fortress itself ± the vision of life for the sake
146
the price of utopian ventures, least of all those
64
invented by other people.'
Space has allowed only a brief sketch of the liberal-
ism of fear. But perhaps enough has already been said
to suggest the attractiveness of its politics, which
admittedly may not be evident at first sight. We
may also be able to appreciate the potential fruitful-
ness of its approach to justifying liberalism when the
fact of pluralism, itself so variably interpretable, had
apparently left all such ventures toppling into self-
contradiction. It concedes that we have to work from
our own traditions and experiences in order to arrive at
a form of self-understanding for our political ideas,
even when we deliberate for universal ideals. This may
make our political thought paradoxical but it does not
65
make it incoherent.
As a response to Kant's question, with which this
chapter began, the liberalism of fear suggests some-
thing very different in content to Rawlsian political
philosophy even though it shares with Rawls the
project of explaining what `should be the public
philosophy in a reasonably just constitutional re-
66
gime.' But the difference is what can tip the scales
of argument in Shklar's favour. The exemplars of
philosophy examined here proved unable to con-
vince us that they could answer the question, that
they could avoid the Weimar problem. All that the
liberalism of fear needs, though, is the fear that
motivates the question ± which perhaps receives
its most acute expression with the Weberian sketch
of modernity as its backdrop ± to begin the political
argument for liberalism. Under the conditions of
modern pluralism, this is probably the best we can
do. But the Weimar problem, and the existence of
widespread cruelty still in the world, should leave us
with no doubt that this argument is still very much
worth pursuing.
of which the war is being waged ± will, as a rule, turn
out to be relatively simple and unsophisticated': Berlin
(1969), pp. 200±1.
4. Weber's picture is one way of substantiating the
notion of the `default' position in political theory,
an idea suggested by Mark Evans in Chapter 11 of this
volume.
5. Habermas (1998a), p. 60.
 DISENCHANTMENT AND THE LIBERALISM OF FEAR

6. Rawls (1996), p. lx. 37. Hennis (1988). For the idea of a `strange liberal', see
7. See Walzer (1983); cf. Griffin (1994). Boesche (1987).
8. Rawls (1996), p. 44. 38. See, for example, Eden (1982). Indeed, the uncer-
9. Ibid., p. lxi. tainty about the classification merely reinforces the
10. Ibid., p. lxii. claim that Weber poses problems for liberals.
11. Hampshire (1983), p. 162. 39. Weber (1994) p. 68; Gerth and Wright Mills
12. Hampshire (1989), p. 118. (eds.)(1991), p. 153.
13. Rawls (1999), p. 1. 40. Rawls (1999), pp. 463 n. 24.
14. Rawls (1996), p. 55. 41. Anderson (1992), pp. 193±7.
15. Rawls (1996), p. xx. 42. See, on this point, Dyzenhaus (1997), p. 12.
16. Ibid., p. xxiii. 43. See Schmitt (1985); cf. Larmore (1996), chapter 8 for
17. Jahanbegloo (1993), p. 56. critical discussion.
18. Ibid., p. 45. 44. Rawls (1996), p. 374.
19. For the interpretation of Berlin as an objective plur- 45. Dyzenhaus (1997), p. 16.
alist, which sharply distinguishes him from Weber, see 46. Rawls (1996), p. xxvii.
Gray (1995b). 47. Ibid., p. xxviii.
20. Rawls (1999), pp. 463 n. 24; cf. ibid., pp. 462±3. 48. Ibid., pp. xxvi±vii.
21. Berlin (1978), p. 148. 49. Wenar (1995), p. 48.
22. Berlin (1991), p. 12. John Gray refers to Berlin's 50. Ibid.
`phenomenological certainties' in Gray (1995b), p. 62. 51. For an example of such that Rawls himself concedes
23. Ibid. but which may be far more likely than he would hope,
24. See B. Williams, `Introduction', in Berlin (1978). see Rawls (1996), pp. 152; cf. Evans (1999).
25. Jahanbegloo (1993), p. 44. 52. Larmore (1996), p. 151.
26. For a version of this argument, see Raz (1986), chapter 53. Rawls (1996), p. 374.
14. 54. Shklar (1989), p. 29.
27. This is a term used by Max Weber, taken from Goethe, 55. Ibid.
to describe such relationships; see Gerth and Wright 56. Shklar (1989), p. 23.
Mills (eds) (1991), pp. 62±3. 57. Ibid., p. 27.
28. Berlin and Williams (1994), p. 309. 58. Ibid., p. 25.
29. Crowder (1994), p. 303. 59. Ibid., pp. 29, 34.
30. Berlin (1991), pp. 80, 203, 205. 60. Ibid., p. 21.
31. Kekes (1993), p. 199. 61. Ibid.
32. Anderson (1992), p. 245. 62. Ibid., p. 38.
33. Larmore (1996), chapter 7. 63. Ibid., p. 33.
34. Ibid., p. 174. 64. Ibid., p. 33.
35. Ibid., pp. 166±7. 65. See Michelman (1995) for discussion of this claim.
36. Ibid., p. 167. 66. Rawls (1999), p. 622.

147
 11
PRAGMATIST LIBERALISM
AND THE EVASION OF POLITICS
Mark Evans
I
Despite the hostility that much of it bears to its
object, the extent of the response to Richard Rorty's
intervention in liberal political theory is testament
to the significance that so many commentators
attach to it. Rorty believes that liberal convictions
are utterly contingent upon the preferences of those
in liberal cultures, that philosophy adds nothing
useful to the holding and promotion of such con-
victions and is probably best abandoned, and that
these considerations need not disrupt the way in
which we `handle' such convictions. He wants us to
treat these claims as profoundly liberating, in that
they say liberals can be freed from thinking they need
(perpetually elusive) theoretical justifications ± in
particular the `foundationalist' claims that purport-
edly establish their `truth' ± in order to entertain
their principles. But, despite his constant reassurance
to the contrary, the idea of `contingency' and its
rejection of `foundationalism' strikes many liberals as
being fraught with danger, leaving us teetering on
the brink of nihilism, sidestepping all manner of
issues in complacently claiming that `our' liberalism
can survive the embrace of contingency and the
disengagement from philosophy intact. Some critics
have felt that Rorty is so insouciant that his ideas are
not, ultimately, worth taking seriously. Yet, unless
(implausibly) the continued critical interest in him is
merely the result of a popular desire among theorists
to engage with what they regard as an easy target, it is
148
clear that the prominence of the `Rorty debate'
reflects a widespread belief that he has thrown down
an important challenge. After all, if he is right then
the very need for liberal moral and political theory
(or, indeed, any other kind) is clearly called into
question.
Thus, in a book devoted to mapping out new
research agendas for liberalism, Rorty's challenge can
hardly be avoided. But as this chapter will contend,
he has arguably not posed it in anything like the
terms in which it could and should be presented. The
full consequences for liberal politics in particular are
not teased out and so he does not give us the
opportunity to appraise the whole effect of `contin-
gency' upon our thinking and our practice. One
purpose of this chapter, then, is to develop a `Ror-
tyean view', a characterisation of aspects of his
thought that is fuller than he himself provides but
which does not depart in essence from his pragma-
tism. This view will allow us to depict and apprise
more thoroughly the Rortyean interpretation of
liberal moral and political commitments, which
Rorty believes can survive the abandonment of
the philosophical search for their foundations.
The particular aspect of liberal commitment to
which Rorty is obviously passionately attached is
what he calls the `human rights culture' (hereafter:
1
`HRC'). Now if we grant the point that we should
be very wary of overly homogenising our conception
of `the' HRC, the chapter proceeds from the plau-
sible assumption that many who hold its commit-
 PRAGMATIST LIBERALISM AND THE EVASION OF POLITICS
ments do so in a foundationalist manner. That is to
say that they believe human rights claims are `true',
for they think such rights are objectively valid
despite whatever certain people may think or prac-
tise to the contrary. Paradigmatic of this approach to
rights is the American Declaration of Independen-
2
ce's assertion that they are `self-evident' truths.
more recent documents, the notion of `objective
validity' is conveyed by the idea that codifications
of rights (such as the Universal Declaration of Hu-
man Rights) are designed to `recognise' and `observe'
the rights, rather than to create them. In other words,
such documents are posited as an acknowledgement
of the independently grounded moral validity of
3
rights. Such ideas, I suggest, are reflected in the
way rights are popularly regarded by most of those
4
within the HRC.
If, as I will hereafter assume, the HRC is pro-
foundly foundationalist in outlook, it is hardly clear
that Rorty's pragmatist approach can allow it con-
sistently to stand in anything like its familiar form.
The modified Rortyean view, then, will be designed
to test just how conservative of the moral culture's
self-understandings Rortyean pragmatism might al-
low us to be.
The centrepiece of this enquiry concerns not only
what morality looks like from a Rortyean perspective
about those we call `rights-violators'. It also focuses
upon what many who share the HRC believe is
politically sanctioned by the `truth of human rights',
namely that violations of human rights constitute
objective moral wrongs which, in at least some
circumstances, may be sufficiently bad to warrant,
for example, the `humanitarian intervention' by
certain states in the affairs of others. The founda-
tionalist nature of human rights on the HRC view as
interpreted here gives rights a seemingly authorita-
tive basis which is sufficiently `weighty' to legitimise
the use of force in their support. Whether Rorty's
anti-foundationalism can provide a suitable substi-
tute for the foundationalist rights-enforcing sanc-
tion, or whether it should not be perturbed if it
cannot, are surely questions of considerable signifi-
cance. But here is where a charge of complacency
against Rorty has real bite, for such questions of
political justification ± seeking reasons to justify the
exercise of political power ± do not particularly vex
him. If they did, he might well have obtained a
deeper sense of what moral and political problems
may open up, rather than disappear, with the aban-
donment of foundationalism.
What may strike one as paradoxical about this
indifference is that the pragmatist approach to hu-
In man rights is motivated by the idea that, because the
philosophical search for the `foundations' of rights
has proved to be so inconclusive, it is actually of little
use in spreading acceptance of human rights around
the world and should no longer detain us. Rorty's
concern, then, is explicitly practical: he wants to do
what will best promote the HRC globally. Yet there
is no real engagement with those questions about the
justifiability of the coercive rights-enforcing actions
which often accompany the actual political pursuit
of the human rights agenda. He seems to rest content
with an idea that a liberal-pragmatist society `will get
accustomed to the thought that social policy needs
no more authority than successful accommodation
among individuals . . . It will be a society . . . that
takes reflective equilibrium as the only method
5
needed in discussing social policy.' This recommen-
dation gives us no insight into how the genuinely
deep and persistent conflicts over human rights,
either within a state or internationally, are to be
tackled when `accommodation' is impossible or un-
desirable. Two commentators, Guignon and Hiley,
have written that what is most admirable in Rorty's
work `is the courage, integrity and clear-sightedness
with which he bites the bullet and draws out the
inevitable consequences of anti-foundationalism for
6
moral and social thought.' Unfortunately, this cour-
age abruptly evaporates once the really tough ques-
tions of political justification are reached. In this
respect, we have a bullet which Rorty has barely
nibbled.
The value of suggesting what account a suitably
developed Rortyeanism may give of this problem is
probably best grasped by thinking about the follow-
ing suggestion as to why his work has attracted so
much attention. I propose that the Rortyean view
articulates what might be thought of as a `default'
position in moral and political theory. That is to say,
it represents the kind of philosophical outlook which
many contemporary thinkers fear would have to be
conceded if their alternative, opposed theories are
149
 MARK EVANS
7
shown to be unsuccessful. The worrying feature of
the `default position' lies in its apparent, complete
reduction of fundamental moral values to a whimsi-
cal subjectivism that would destroy the point of
moral and political theory and, more importantly,
leave us just a `change-of-feeling' away from sanc-
tioning the embrace of completely abhorrent values.
It is a horror at having to accept that this is all there
is to morality that so much philosophical labour is
devoted to contending that this is not the case.
The default position takes on a particularly sinister
hue for many liberals in political thinking. As we
shall see, nowadays they often argue that the justi-
fication for imposing their principles with political
power rests upon some notion that the `receivers' of
such power can, in some way, consent to those
8
principles. I suggest that Rorty, given his remarks
on reflective equilibrium above, falls into this broad
camp. Now one might not be concerned about
imposing purely subjective and contingently held
principles upon people if the latter could plausibly
agree to them in some way. Yet if, as implied by the
role of coercion in spreading the HRC, we cannot
always rely upon such agreement in politics, where
does the default position leave the argument for such
political imposition?
Rorty's indifference to the questions raised thus far
may make the `default' position in his hands rela-
tively easy to dismiss. I think it poses a more per-
plexing challenge, not least because I argue that the
Rortyean view can be made to do a better job of
defending some aspects of the HRC. This allows us
to take fairer and fuller measure ± and hence to assess
the limits ± of a theory whose conclusions may be
profoundly unsettling for many liberal political the-
orists.
II
Rortyeanism asks us not to bother pursuing any
philosophical questions whose resolution has no
significant bearing on the practical pursuit of our
goals (in this instance the spreading of the HRC's
acceptance). For Rorty, meta-ethics falls into this
category. `The difference between the moral realist
and the moral anti-realist', he writes, seems to be `a
9
difference which makes no practical difference.'
150
This might immediately suggest that nothing really
hangs upon the fact that the HRC tends to employ
10
foundationalist moral concepts. Rorty might be
happy to leave these in place, simply because it
makes little difference one way or another to the
HRC's promotion.
This resolution of the possible conflict between
the commitments to pragmatist contingency and the
HRC is too quick, however. For Rorty doesn't
abstain entirely from meta-ethical claims; he actually
adopts a provisionally non-cognitivist position, and
this crucially underpins his practical recommenda-
tions. This position is evident in the way he attacks
foundationalism, which he defines as a theory of
moral knowledge holding that we draw values from
non-moral premises: `to claim such knowledge is to
claim to know something which, though not itself a
11
moral intuition, can correct moral intuitions.' He
rhetorically demands whether there is this kind of
knowledge and argues that the extent to which the
philosophical search for it contributes little or noth-
ing to the promotion of moral values is the extent to
which we have reason to think there is no such
12
knowledge. This claim is defensible by pondering
the wide diversity of, and persistent disagreement
between, the myriad accounts of moral knowledge
and the related unpersuasiveness of any one of them
to a significantly large number of rationally enquiring
minds. Obviously this is not enough in logic to
clinch Rorty's conclusion that there is no one `cor-
rect' morality at all (the phenomena of diversity and
disagreement could be explicable in various ways).
But he is only really interested in offering enough
prima facie evidence in support of non-cognitivism to
disarm those who think that continuing the search
for moral knowledge remains the best way of pro-
moting moral values. Consequently, he urges the
embrace of what is actually the classic non-cogniti-
vist approach to moral revision as the most effective
means to pursue this goal: the manipulation of
sentiments to change the way people feel about moral
13
matters.
It could be objected that a non-cognitivist ac-
count cannot be objectivist about moral value: how
can a value be objective if it does not exist inde-
pendently of people's attitudes? Given the objectivist
quality typically assigned to `human rights' in the
 PRAGMATIST LIBERALISM AND THE EVASION OF POLITICS
HRC, we may already have reached the point at
which tensions surface between Rorty's philosophi-
cal commitments and the standpoint of the HRC.
Certainly, if we adopt what can be called a trans-
cendentalist conception of objectivity, which natu-
rally attaches to moral realism in holding that there
is a realm of facts `out there', independent of our
attitudes, we could agree that `objectivity' should be
purged from a Rortyean view. But it can be recast in
conversationalist terms, in which `objectivity'
achieved in the agreed outcome of a discourse over
values which has conformed to certain ideal condi-
tions of validity (without any implication that the
agreed outcome of moral conversation thereby iden-
tifies any of the cognitivists' `independently
14
grounded' moral facts). Successful moral arguments
create, rather than discover, what is objective. This is
precisely in accord with Rorty's claim that for prag-
matists
the desire for objectivity . . . is simply the desire
for as much intersubjective agreement
possible . . . Insofar as pragmatists make a dis-
tinction between knowledge and opinion, it is
simply the distinction between topics on which
such agreement is relatively easy to get and topics
15
on which agreement is relatively hard to get.
Yet if `objectivity' in some form need not be
denied to the Rortyean view altogether, it might
still be feared from the standpoint of the HRC that
the conversationalist variant is a very pale imitation
of `real' objectivity. The non-cognitivist underpin-
ning of the Rortyean view, so the fear runs, will
always conflict with the way in which the HRC
conceptualises its moral commitments.
III
This concern arises from the common belief that
non-cognitivism is driven towards these positions:
1. the explanation of the virtually ubiquitous
realist terminology in moral discourse by an
`error' theory, which says we mis-describe the
nature of our moral commitments when we
16
express them in such a way;
2. the reduction of moral judgement to a basis
which is at least potentially inconstant, ren-
dering the content of morality unstable, adap-
table merely as our whims desire.
Rorty veers towards both in his provisional rejec-
tion of the HRC's foundationalist underpinnings and
in his belief that a properly ironist view will always be
disposed to treat our current moral vocabularies as
is revisable. But here is where we can develop a
Rortyean view sympathetically in a way which mini-
mises the discrepancy with the HRC.
Rorty asks us to think of moral values as `we-
17
intentions', summaries of a community's agree-
ments as to what should and should not be done
within that community. Hence `the core meaning of
``immoral action'' is ``the sort of thing we don't
18
do'' ', or, better, what we would rather people
didn't do. Moral understandings evolve over time
in a community's practices as people work out for
themselves commodious ways of living and, as Mar-
as garet Macdonald (another non-cognitivist defender
of rights) describes, they become `expressed in the
life of the society and constitute its ``quality''. They
are conveyed by its ``tone'' and atmosphere as well as
19
its laws.' A non-cognitivist could describe this in
terms of socialisation imparting specific moral atti-
tudes to its subjects, furnishing them with an attitu-
dinal moral perspective that reflects how those who
share it feel about desirable and admirable forms of
conduct and their obverse. Once developed, a moral
perspective can quickly acquire the status of a set of
social facts, impressed upon people by what are
experienced as the external forces of socialisation
20
and passed on as purportedly self-standing rules.
Hence people are confronted by morality as an
apparently supra-individual, `weighty' and hence at
least apparently objective phenomenon. But in this
perspectivist analysis this feature is testament not to
the actual, external existence of moral facts but to
the content and depth of people's socialisation into
the conventions of morality.
Applying this account to describe the HRC, it can
be said that, in our socialisation into its perspective,
we acquire an outlook in which rights are attributed
to all: when we conceptualise `human being' we
conceptualise `rights-bearer'. `Rights' are the summa-
151
 MARK EVANS
ries of our culture's ideals of how all humans should
be treated. Rorty calls moral judgements `ethno-
centric' to indicate their cultural particularity, draw-
ing one's attention to the cultural specificity of such
21
judgements and the beliefs that form them. As a
`final vocabulary' he thinks the discourse of rights has
22
no non-circular means of defence. Any arguments
in favour of them are going to be, in some way and at
some level, question-begging, only working by as-
suming the validity of the very values that a dis-
putant is throwing into doubt. But, in response to
concerns that this might be a deeply unsatisfactory
feature of a final vocabulary's justification, we might
wonder whether other, if any, of our deep commit-
ments (to family, friends, personal projects, organisa-
tions) can ultimately be justified in non-question-
begging ways. Is there, for example, necessarily a
non-question-begging justification to explain why I
love my wife and have committed my life to her?
Furthermore, is it necessarily appropriate to think
that I need a non-question-begging justification for
such a commitment? (Isn't the sheer fact that I love
her enough to justify my loving commitment to her?)
A Rortyean, therefore, may be able to argue that
there is no deficiency in any final vocabulary's
circular defence.
As Rorty has frequently stressed, admitting to the
epistemological relativism of the `ethnocentric' ap-
proach does not necessarily entail a normatively
relativistic configuration of the values held within
any perspective. To see this point, it should be
remembered that perspectives are unrankable with-
out adopting standards which are only available
within a particular perspective. Outside of any moral
perspective, from the morally vacuous standpoint of
the world `as it is', no moral evaluation of any kind ±
relativist or otherwise ± can be sanctioned. No
grounds therefore present themselves from the per-
spectivist outlook per se as to why we must relati-
vistically concede equal validity to all perspectives.
Any insistence upon normative relativism has to be
based upon an essentially coherentist claim that it
would be internally inconsistent for the human
rights perspective to be constructed without it.
Clearly, Rorty finds no such grounds for this posi-
tion. He thinks it proper that people should make
judgements about fundamental rules of human inter-
152
23
action; he just denies that there is any useful way of
validating them from a transcendental perspective.
IV
Thus far, we have seen how the Rortyean view could
furnish itself with non-relativised moral prescriptions
and has at its disposal a sociological explanation of
why morality appears to have an objectivist, foun-
dationalist quality. But the question remains as to
whether a non-cognitivist should still recommend
the replacement of the realist, objectivist language of
morals, for the sake of `philosophical hygiene'. If so,
then the view still threatens to destroy the form of
discourse typically employed in the HRC.
Simon Blackburn has argued that it is perfectly
legitimate to employ realist-seeming terms when
accounting for what are in fact only products of
attitudes. We can be what he calls `quasi-realist'
in our moral judgements, projecting values on to
24
the world from one's moral perspective. Blackburn
shares a broadly similar account of the purpose of
thinking about morality to Rorty's:
we should not theorise about morality and ethics
as if they are in the business of describing aspects
of the world . . . To enter a moral claim is . . . to
take up an attitude or stance, and centrally it
involves making an emotional response to con-
templated events and states of affairs . . . Moral
and ethical vocabulary . . . enables us to share our
stances with others, to shape our sensibilities in
light of opinion of others.
In short, `the essence of ethics lies in its practical
function' ± a natural ally, then, for the Rortyean
25
view. More pertinently still, Blackburn's `projecti-
vism' suggests a way of describing human rights from
within an attitudinal moral perspective in terms that
realists often think are reserved for themselves.
Projectivism is a complex philosophical position
and only some of its basic features can be outlined
here; moreover, it is reasonable to assume that Rorty
himself might part company with it at several key
points. Enough, though, can be said to illuminate the
proposal for its partial incorporation into the Ror-
tyean view.
 PRAGMATIST LIBERALISM AND THE EVASION OF POLITICS
Although Blackburn does not believe that values
are intrinsically part of their objects, his key claim is
that there is no bar to using the language of `percep-
tion of the actual' in attributing moral properties to
people because
we speak of the perception of every single cate-
gory of thing and fact that we ever communicate.
We speak of perception of numerical truths,
truths about the future, truths about the past,
possibilities, other minds, theoretical entities of
26
all kinds . . .
± mind-dependent features, in short.
Further, he contends that realist language turns
out to be a proper way of expressing commitment to
that which we deem to be valuable. One way of
grasping this claim is to see the realist-sounding
language of the projectivist as expressions of how
things indeed are from the standpoint of our perspective.
Moral statements are properly statements from the
`inside', internal to one's moral perspective. To
believe that realist language ought to be abandoned
because our moralising has its origins in attitudes is to
step outside of the moral way of looking at things and
confuses the language which is appropriate `intern-
ally', within the moral outlook, with what is appro-
priate `externally' from the moral `view from
nowhere' that we conceptualise as lying beyond all
perspectives. Blackburn's example of characterising
cruelty illustrates this. If a projectivist, P, is asked
`why is cruelty wrong?', he is asked a moral question,
which is properly understood to enquire as to what
judgements are made within a moral outlook. If he
were to answer `because it is not something I like' ±
making cruelty wrong because of the responses he has
to it ± a realist would swiftly point out the obvious
anti-realist character of the reply. But P's answer is
not the one he has to give to a moral question.
Answers in which attitudes figure are appropriate to
`external' questions about the nature of morality,
such as `why do we say cruelty is wrong?' But this
is a different kind of question from the one asked
within the moral view and we should not give
answers to questions which are raised from different
standpoints. A parallel would be to contrast an
internal mathematics question `why is 12 the square
153
root of 144?' ± which would refer to how the con-
ventions within mathematics produce this result ±
with an external question `why do we say 12 is the
square root of 144?', which would direct us towards
the origins and nature of the conventions them-
selves. So P can legitimately answer `why is cruelty
wrong?' by referring to those features of cruel acts
which make cruelty wrong from the moral perspec-
tive, rather than having to say that his attitudes to
cruelty are what make it wrong. Within his moral
perspective, it is actually the case that cruelty is
wrong and using realist language to express this is
a coherent way of stating what is the case in his
27
outlook.
Transferring this analysis to the projection of
human rights: if we ask the `outside' question `why
do we say people have human rights?', the reply
should focus upon the attitudes people have about
how they think humans should treat each other.
From within the perspective, though, rights are
projected onto people first as an act of valuation,
28
to follow a distinction made by David Wiggins,
which identifies what is felt to be sufficiently im-
portant about a human being to warrant a certain
kind of morally obligatory respect, and then as a
directive judgement (which holds that `because X is
valued, Y ought to be done'), creating the rule
embodied in the idea of people possessing a right
to expect a certain kind of treatment with respect to
29
the valued attribute.
I suggest, then, that the Rortyean view's provi-
sional non-cognitivism need not force the abandon-
ment of the realist, objectivist discourse of the HRC.
Of course, like any other philosophical position,
projectivism has its critics and some would urge that
its usage of realist terms is too hollow to be accep-
table (I expand upon some of these doubts in the
final section). This may be so ± but it requires
philosophical argument to prosecute such conten-
tions. And that is an activity which, it is now clear,
Rorty abandons too quickly. Here, we have had to
retrieve it on his behalf to tackle the problem of the
HRC's foundationalist character. If the simple graft-
ing of projectivism onto the Rortyean view is not
enough to satisfy the doubters, then this is all the
more reason to keep philosophical enquiry alive.
 MARK EVANS
V
We now turn to the heart of Rorty's own practical
concerns: the business of promoting the HRC, most
of all among those who actively reject it. At this
point, it is apposite to distinguish clearly between
three types of justificatory argument:
1. Moral justification. With respect to beliefs, this
is generally concerned to offer arguments as to
why we should support certain moral princi-
ples and not others. With respect to our ac-
tions, moral justification asks whether morality
gives us warrant for acting in a particular way.
It requires us to connect our actions to moral
beliefs and assess their moral worth accord-
ingly.
2. Political justification. This question specifically
asks for a justification for the exercise of
political power. When talking about the po-
litical imposition of a moral principle, it may
not be enough to justify that action via a moral
justification of the principle alone. The prin-
ciple may indeed be morally valid, but that
may not suffice to make its forcible imposition
justified. (It may be morally right, for example,
to condemn adultery but many would not
think that judgement sufficient to justify its
criminalisation.) One consideration in the
search for political justification may be the
third argument.
3. The justification of political authority, the estab-
lishment of the right of an institution (such as
the state) to be the one to exercise power
(separable from (2) in so far as the question of
who should exercise power is distinct from that
of which among our (moral and non-moral)
ends can be promoted with political power).
These are crucial distinctions yet Rorty's political
theory (if indeed it can be properly called that) has
little grasp of them. Once again, we must see how we
might construct his position to accommodate them.
Instantly we should reiterate a point that helps to
frame the disquieting character of the Rortyean view
as a default position: the only way for non-cogniti-
vism ultimately to preclude the acceptability of any
154
type of action to propagate human rights among
others is to show coherentistically that the action
would conflict with other principles within the hu-
man rights perspective to which a liberal ought to
accord primacy. And that perspective, of course, is
held utterly contingently, the product of nothing
other than our affirmation of the values of our
culture.
From a non-cognitivist position, the force of the
idea that action in support of human rights can
trump other moral codes can only ultimately lie in
the fact that the rights we embrace in the HRC
represent firm, settled and passionately held attitudes
as to how people should be treated. Some ± perhaps
many ± liberals may agree with the argument that
there is nothing philosophically inconsistent in
merely possessing universalistic moral attitudes
non-cognitivistically. Yet they could still be pro-
foundly perturbed at the contention that it is per-
missible politically to act upon these, to promote or
even enforce any moral perspective among others
who do not share it when it can have no indepen-
dent warrant or authority beyond the contingent
commitment they happen to have formed to it. They
may ask themselves whether it would be appropriate
for them qua liberals to do this given the ethno-
centric provenance of their morality.
It is clearly important to think hard about coher-
ent plausible conclusions which could be drawn by
liberals from the normatively anti-relativist insistence
that all should respect human rights when made
within such a contingent perspective. The liberal
schema of universal human rights can readily be
grasped as definitive of liberalism's essence, expressing
what it means to view humans as free and equal
beings for whom cruelty is the worst thing that could
be suffered. `We liberals' can say that we feel so
strongly about them that we are convinced that all
should respect rights, even if others do not see things
the same way. Thus, our primary commitment to
rights rules out the possibility or need to admit the
legitimacy and permissibility of any moral system
which systematically denies them.
Hence, liberals do not have to adopt a `live and let
live' attitude of tolerance to rights-violating systems.
Indeed, if they were to be this sensitive to the views
of other societies then they may well be abandoning
 PRAGMATIST LIBERALISM AND THE EVASION OF POLITICS
much of their own fundamental commitment to
human rights: what is the point, after all, of taking
`human' rights seriously if they are going to be
prepared to acknowledge that some societies could
quite acceptably ignore or violate them?
One might hope that much of the concern about
the promotion of the human rights perspective could
be mitigated by adopting the essentially non-coer-
cive, persuasive or `conversational' approach to mor-
al conversion that Rorty obviously prefers. Even if it
is employed by bodies (states, quangos or interna-
tional organisations) which wield political power it is
arguable that no question of physical enforcement on
their part necessarily arises if they adopt this ap-
proach alone in HRC-supporting policies ± and of
course the persuasive strategy need not be under-
taken only by political institutions (or at all ± it
could be left to private individuals and organisations
such as churches and charities). If this is what Rorty
has in mind, the lack of a specifically political edge to
his theory might be more understandable.
Certainly, there are potentially wide resources to
pursue the persuasive-conversational strategy. We
could start by noting that many of the world's
governments pay lip-service to international under-
standings of rights and can be charged with promise-
breaking, hypocrisy and illegality in international
law to the extent that they actually violate rights
(urging coherence, or consistency, between pro-
claimed and actual standards of behaviour). The
tendency of human rights to gain support in the
international community as agreed standards of mor-
al behaviour is a further crucial point which could be
cited in favour of this perspective (`since so many
millions of people sincerely and fervently support
human rights, isn't this at least a good pro tanto
reason for others to do so as well?').
Another part of the persuasive strategy for the
liberal could be, in Rorty's rather playful phrase, the
telling of `long, sad, sentimental stories' designed to
shape the feelings and evaluative dispositions of
30
those who do not (yet) share the rights perspective.
As Rorty has observed, the perpetrators of such deeds
typically assume not that human beings are not the
sorts of creature who have human rights, but that
31
their victims are not really humans at all. Ob-
viously, then, a decisive part of what would in effect
155
be a project of sentimental education would be the
effort to break down such discriminating divisions of
humanity to forge an increasingly shared way of
viewing and treating human beings. We can hope
to encourage non-rights-respecters to make an ima-
ginative, sympathetic or empathetic identification
with the plight of those who, from our perspective,
are having their fundamental rights brutally vio-
32
lated. We can try to get others to see the awful
consequences of rights-violations and to care that
such should henceforth be avoided. That, as much as
anything, is a powerful reason for keeping alive
through historical narrative the memory of those
terrible atrocities which have made us now so con-
cerned with human rights.
Despite its status as a final vocabulary, the human
rights perspective need not shut out all opportunity
for critical reflection on the HRC's present content.
This may further placate liberal concerns that some
kind of cultural imperialism is being sanctioned here,
for this point encourages an appreciation of the
potential for refinement, improvement or transfor-
mation of our commitments even when we currently
hold fast to them. Rorty warns us to be sensitive to
how `redescription' often humiliates in undermining
the existing self-understandings of those whose views
we seek to convert. But if it is acknowledged that
redescription need not all be `one-way' ± that A
might enrich her own moral sensibility from the
dialogue she has with B even when she hopes that
B may come to share her view ± the potential for
humiliation on either part may be lessened.
However, the bottom line of the human rights
perspective must surely be that, though the moral
beliefs which supplement basic rights may be revi-
sable, there is probably little left open for debate as
far as the latter are concerned. Anyone whose views
have to be redescribed in order to begin fostering a
respect for human rights may well deserve to be
humiliated, if that is how they feel when confronted
by the viewpoint of the HRC.
VI
So far, liberals may be happy enough with the
Rortyean view's approach to the HRC. But in the
realm of conversational persuasion, of course, we
 MARK EVANS
have gone no further than the activity of moral
justification. The critical political problem in spread-
ing the HRC is, of course, that conversation is likely
to get us only so far towards this end in many cases.
And at the point at which persuasion ceases to be
effective Rorty falls silent, even though he is per-
fectly well aware that there will be those utterly
unconvinced by our redescriptions alone. If there is
nothing more that can be said to Nietzsche, Loyola
or the committed, consistent Nazi, he throws up his
33
hands and dismisses them as `mad'. But what do we
do about such people thereafter? Do we leave the
genocidal maniacs of the world alone as much as
possible, to carry out their own `ethnocentrically
determined moralities'?
Although questions of rights enforcement can
arise even within the boundaries of a liberal state
(with respect, say, to a non-liberal minority), let us
focus here upon the issue of humanitarian interven-
tion when one or more states uses force to punish
and/or halt rights violations by another state against
its own citizens. Since the end of the Cold War in
particular there has been a growing recognition that
the `international community' has rights to act thus
(grounded in international law) when basic rights ±
usually, the specific right to life ± are being violated
on a systematic, widespread scale. This recognition,
therefore, is undoubtedly part of the HRC's self-
understandings ± but it is strikingly absent from
Rorty's own account of it.
In fact it would be unfair to single out Rorty for
avoiding questions of the political justification of
coercion. Non-cognitivist moral and political theory
has always tended to dodge such questions. Because it
treats questions of value as questions of attitude, it
typically holds that `theory' has no specially privi-
leged insight into what attitudes should be held and
has consequently left the determination of what
should be done to the actual business of political
34
life.
Moreover, many contemporary liberalisms from
other traditions are also vulnerable to the charge
of evading the `challenge of the political'. Chantal
Mouffe, for example, has argued that Kantian poli-
tical theory (typified by John Rawls) seems to reduce
the resolution of political conflicts to an ultimately
consensual process of agreement formulation
156
35
through dialogue. Questions about what to do with
those who won't peacefully concur with the dialo-
gical outcome (the issue of political justification)
and, if we are to turn to coercion, how to establish a
body's right to use power (the justification of poli-
36
tical authority) are not directly confronted. Part of
Mouffe's diagnosis of this feature identifies the ap-
plication to politics of the specifically philosophical
techniques of Kantian ethics, with their emphasis
upon the securing of agreement in public reason, as
the reason why fundamental conflict is not squarely
addressed. Rorty's parallel problems with politics may
likewise come from the application of his implicit
meta-ethical Humeanism, with its emphasis upon
sentiment manipulation through story-telling and
suchlike, to what Mouffe insists is the agonistic,
conflictual realm of politics.
Yet such political quietism should be available
least of all to Rorty, given his explicitly practical
conception of theory. One would surely have ex-
pected a pragmatist to have been dismissive of
traditional non-cognitivism's attitude to political
questions and have been similarly critical of the
neo-Kantian `evasion of politics' decried by Mouffe.
So once again `the Rortyean view' has to be
developed out of Rorty's own position to ascertain
the character of a Rortyean politics that aims to cope
with political conflicts beyond the illusion that
persuasion and consensus is all we need. To focus
upon the most disturbing aspect of this as a default
position, I propose to pass fairly briefly over the
question of how political authority is constituted.
Let us assume that, qua liberal, Rortyeans would
argue that consent is the key: institutions that right-
fully exercise power must in some meaningful way
have gained the consent of their subjects to do so.
This idea may already do considerable work in
justifying humanitarian intervention. It can be
pointed out in many instances that rights-violating
states have freely joined the United Nations and
officially support the documents of international law
that require observance of human rights. Hence, it
could be said that their free adherence to the inter-
national community implies acceptance of its right
to force them to act according to its standards, or at
37
least to punish them when they fail to do so.
But let us concentrate upon the idea of a `rogue
 PRAGMATIST LIBERALISM AND THE EVASION OF POLITICS
state' which has consciously broken with the stan-
dards of international law and has effectively `with-
drawn its consent' from supra-national political
authorities. This state has embarked upon systematic
violations of human rights (a policy of genocide, say,
against a certain group of the population). What are
the pragmatically minded Rortyeans to do and what
account can be given of what they do?
The bald fact of the matter is that the only reason
available to them for not advocating the use of force
even in the absence of what they judge to be full
political authority to do so (that is, the consent of
the `rogue state' to suffer the consequences of breach-
ing international law) would seem to be that it is
prohibited according to the specifications of the
values within their attitudinal perspective. And
there is no compelling `internal' reason why these
values cannot be configured within this perspective
to sanction a coercive strategy.
This is the point, the crux of the default position,
at which many liberals (and others) will throw their
hands up in horror. How, they will want to know,
can a liberal argue for the acceptability of such
coercion when it is openly claimed that the princi-
ples in question are held purely contingently, of
`ethnocentric' origin and supported by nothing more
than arbitrary attitudinal conviction? At least with
the liberal account of political authority, even if we
accept the idea of `consent' as a contingently held
moral commitment, the fact that it sanctions coer-
cion only with some notion of agreement from the
coerced suggests a less troubling, less imperialistic
scenario. But how could the contingent liberal sin-
cerely, consistently make this claim about our atti-
tude to the rogue state? Surely, we are now in the
realms of `might is right', with a liberal `will-to-
power' seeking to stamp its image upon others in
a world that is, in itself, morally meaningless.
Now the point of the default position is that this
last proposition is indeed, in essence, the view with
which it seeks to reconcile us. `Outside' of the liberal
moral perspective, this is how we must conclude
things are. Yet the claim to be urged here is that this
gives liberals no necessary reason within their moral
perspectives to back away from such policies as
humanitarian intervention against rogue states. For
a start, a charge that this is purely `arbitrary' ± in the
157
sense that there is no independent consideration to
mandate such a policy ± fails because, from the
Rortyean standpoint, all moral positions ± including
that of non-intervention ± are `arbitrary'. The case
for saying that, despite the lack of full political
authority, there is sufficient political justification
for intervention rests upon the idea that, from the
Rortyean liberal perspective, rights matter so much
and express such deep-seated convictions about how
people should be treated, based upon our reactions to
the horrendous cruelty that has been inflicted upon
millions throughout history, that objections and
reservations about intervention can be outweighed.
Even if this is `cultural imperialism', no alternative
rights-violating culture necessarily has to be toler-
ated by liberals on their own terms.
It needs to be stressed that a belief that it is
politically justifiable to impose certain human rights
upon societies failing to respect them will not ne-
cessarily lead those of a liberal sensibility straight to
the conclusion that they must do so. The claim that
force is permissible where persuasion fails (and, qua
liberal, we can assume that Rortyeans will exhaust
the possibility of persuasion before turning to force)
need not be configured to yield a conviction that
force is obligatory. It might be that the moral costs of a
justifiable enforcement would outweigh the moral
gains. Nor does this claim entail that such interven-
tion should necessarily be a regular feature of a liberal
state's behaviour. We may admit to valid, non-moral
reasons for not enforcing rights; it may be too
difficult or otherwise impractical to force a rogue
38
state into respecting them. Further, a quintessen-
tially liberal sensibility will prevent liberals from
blithely neglecting the concerns of imperialism al-
together. Their sensitivity to diversity can still lead
them to tolerate some forms of social life even when
they find them unfitting for humans as they con-
ceptualise them, and they will not treat all aspects of
their own culture as fit to be embraced by others.
All this notwithstanding, we still have the crucial
point that Rortyean liberals are not necessarily
forced to rest content in principle with a persuasive
strategy. This is just as well because, whatever re-
servations they may have about particular instances
of possible intervention, most who share the HRC
are probably not prepared to rule it out in principle
 MARK EVANS
always and everywhere. The moral mathematics or
pragmatic costs may not always come out against
coercion and ± to take up another familiar reserva-
tion about intervention ± it is hardly clear that a
state cannot act against another to prevent, for
example, `ethnic cleansing' or genocide just because
of its own dubious moral past. It may be more
vulnerable to charges of hypocrisy if it fails to live
up to its own moral standards, but this vulnerability
need not constitute a reason for arguing against the
appropriateness of an interventionist strategy. One
does not have to deny that the United States (to take
one example of a potential `rights-enforcing' state) is
very deeply flawed on such counts and that it is
desirable to be vigilant over, and suspicious of, its
actions (for liberals should be aware of disingenuous
uses of the human rights banner) in order to agree
that these factors do not constitute decisive bars
against its ever rightfully using its power genuinely
to prevent genocidal violations of human rights.
This is particularly so if it turns out to be the only
39
power capable of doing so.
What, though, does `justification' really signify
here? From the Rortyean viewpoint it is clear that
it can only ever be a question of the consistency of a
principle and/or its practical consequences with the
overall attitudinal moral perspective. (The notion of
`wide reflective equilibrium', in which the perspec-
tive's convictions, their consequences and relevant
background considerations are brought into a coher-
ent balance, illuminates how `justification' might
40
proceed. ) In the absence of any independent cri-
teria (objective moral truths), the liberalism of the
view may always turn to persuasion first, hoping that
its moral and political cases can be based upon
reasons that the others in question can accept.
But this commitment to giving reasons acceptable
to others is now not, as it appears with Rorty himself,
something with which we should rest content. Nor is
it, as it appears to be in Kantian `public-justificatory'
liberalisms, a necessary precondition of all successful
41
or legitimate political justifications. It is at best an
aspiration, which may be consistently pushed aside
when certain moral considerations pertain. In in-
stances such as humanitarian intervention, even
liberals of the Rortyean hue will continue to give
reasons for their actions. But without the assumption
158
that they could be acceptable to the targets of their
policy, such reason-giving `justification' is in fact
`self-explication'. Liberals can think it admirable
and probably necessary on their own terms to render
their considerations clear and transparent to them-
selves and others, sincerely and rigorously. This,
though, is as far as `justification' can extend in this
case.
VIII
Thus far, it has not been the concern of the chapter
to uphold this view against its rivals, merely to
elaborate it as sympathetically as possible. But it is
appropriate to conclude by reflecting on whether
there is much point in taking this view seriously. If it
is indeed a `default' position, how urgent is the need
to try to escape it?
One important reservation about it is this: because
it cannot ultimately help but explain human rights in
terms of being reified, ethnocentric attitudes that
one (albeit very large) group of people wishes to
impose on others, it is just too brazenly advocatory of
the imposition of the powerful's preferred standards
upon others for liberals ever to feel comfortable with
it. The projectivist account of the HRC's moral
discourse, according to this reservation, is ultimately
too flimsy to reconcile liberals to this point; the
`conversationalist' idea of objectivity, for example,
patently fails to accommodate those who haven't
agreed with the results of, or even participated in, the
relevant `conversation'. It provides an all-too-fragile
drape, easily blown away to reveal the realities of
unadorned, shameless power.
A deeper worry is that liberals may thus lose
confidence in their own moral rectitude if they posit
their belief-system as just one among many of the
`warring gods' in a disenchanted world. Incapable in
principle of giving reasons why their morality is
superior to others, liberals shrink from the battle
to impose their views ± and perhaps cede the ground
to, say, a fascist doctrine whose very point is to
impose order and leadership in a fragmented social
42
milieu. This picture may be gleefully painted by
foundationalists, who would urge that foundational-
ism is the only internally tenable form liberalism
(and perhaps any other doctrine) could take.
 PRAGMATIST LIBERALISM AND THE EVASION OF POLITICS
I believe that liberalism is indeed in trouble if it
makes `public acceptability' a precondition of its
political legitimacy in such a divided and hostile
world (as with Rawlsians and more authentic neo-
Kantians), or if it does not push beyond Rorty's own
apparent evasion of politics. But the Rortyean view
has deliberately abandoned these approaches.
Hence, its ability to avoid the deliquescence of
liberalism in the tides of contingency rests upon
its propensity to maintain its convictions that respect
for principles such as rights would make the world a
much better place, even if there is nothing outside of
such convictions to `ground' them.
To conclude, I suggest that there may be a smaller
gap in this regard between the foundationalist and
the Rortyean than the former might presume. For it
can be argued that a foundationalist has great diffi-
culty in supporting principles in a way which avoids
the kind of non-foundationalist commitment held by
anti-foundationalists. A foundationalist who affirms
the existence of human rights and the rectitude of
whatever action may be appropriate in their pursuit
must be convinced that he has indeed discovered the
appropriate truths to justify the conviction. Some
foundationalists are this confident, of course. But
there are pitifully few who agree with any one of the
particular foundationalist philosophies on offer; de-
bate rages on, apparently interminably. One could
use this observation to urge that the foundationalist's
moral commitments persist without the compelling
evidence that the facts indeed support them. (`Given
the criticisms of any one foundationalist position,' so
this critique runs, `does one still not need a good deal
of faith to hold fast to one's values?')
In the absence of secure proof, then, a founda-
tionalist might say that the search for `the truth' has
simply yet to be completed. But this again embodies
a faith that there is such moral truth to be discov-
ered, which is hardly obvious given the wreckage of
previous unsuccessful philosophical searches. More-
over, few ± if any ± foundationalists who continue
the search believe that they must suspend all their
moral commitments prior to the moral discovery
they seek to verify. Yet, given that they typically
do not regard themselves as morally impotent in the
absence of secure knowledge, they must end up
159
making a pragmatic or provisional prior commitment
to their values. Put bluntly, they are committed to
them in the first instance because they are desirable;
that they correspond to some independent moral
reality is accepted on faith and, as such, the Rortyean
can say, is incidental to the commitment ± and
43
therefore dispensable.
So the Rortyean view beguilingly contends that
the kinds of moral commitment we are all entitled to
make depend rather less upon the nature of one's
meta-ethical beliefs than foundationalists in parti-
cular often urge. It can say to the foundationalist
human-rights supporter `given that we share the
same goals, the same language of rights, what profit
is to be gained by continuing the dispute about the
nature of values? We should be allies and not
enemies.' Thus, a case can be made to take the
Rortyean view as seriously as other intellectual ap-
proaches to the spreading of the HRC around the
world. As a `default' position, its increased theore-
tical strength may make it either a tougher target for
critics or, perhaps, a less troubling prospect for those
who fear that those critics may not be able to argue
convincingly against it. Of course it remains afflicted
by serious doubts and we might eventually decide
that the Rortyean view would be insufficiently cap-
able of handling its problems for us to persevere with
it. But then, which of the rival views can credibly
claim to be free of similarly extensive problems?
Moral and political theory could well have a
dangerous potential to frustrate or preclude moral
and political action if we believe we must disen-
tangle all such problems before our morality's call to
arms can be answered. If it does nothing else, the
Rortyean view suggests that this fear is, mercifully,
misplaced: we can be foundationalist or anti-foun-
dationalist and still affirm human rights in the way
they are held in the HRC. Who needs philosophy
to ground the conviction that a world where there is
no genocide is better than one where there is, and
that we ought to use whatever means we can to
secure the former? What matter most of all, we
might finally be able to agree, are the urgent moral
and political demands of those who suffer so dread-
fully when the idea of human rights is contemp-
44
tuously cast aside.
 MARK EVANS
Notes
1. Rorty (1998), p. 170.
2. See Peterson (ed.) (1977), p. 235.
3. See, for example, Brownlie (ed.) (1981), p. 143.
4. Space does not permit a full substantiation of this
claim and I do not wish to deny that different accounts
may be given of how at least some people regard the
notion of human rights. But I am more interested here
in the specific question of how well Rorty's thought
can support this particular general account, rather
than the separate issue of how widely accepted the
latter is. (I think it is sufficiently familiar not to deny
its presence in the HRC.)
5. R. Rorty, `The Priority of Democracy to Philosophy',
in Malachowski (ed.) (1990), pp. 279±302, at p. 286.
6. C. Guignon and D. Hiley, `Biting the Bullet: Rorty on
Public and Private Morality', in Malachowski (ed.)
(1990), pp. 339±64, at p. 343.
7. I am not claiming that all theorists regard this as the
only `default' position lurking in the wings and waiting
for our philosophies to fail. I suspect, though, that this
specific position frequently haunts much philosophi-
cal debate.
8. For example Rawls (1996), p. 136: `Our exercise of
political power is fully proper only when it is ex-
ercised in accordance with a constitution the essen-
tials of which all citizens as free and equal may
reasonably be expected to endorse in the light of
principles and ideals acceptable to their common
human reason.'
9. Rorty (1998), p. 172.
10. At ibid., p. 171, Rorty defines `foundationalism' as a
type of moral doctrine which tries to find indepen-
dently verifiable truths (usually about human nature)
from which our moral intuitions may be directly
inferred. Although constructivism affirms the idea
of `moral knowledge as well, it is clear that moral
realism is really what is signified here; cf. the reference
at n. 11.
11. Ibid., p. 172.
12. Ibid., pp. 171±3.
13. Ibid., p. 176 and passim. It is possible to argue that
Rorty need not have gone as far as he does in
provisionally adopting non-cognitivism because his
sentimentalist approach to moral conversion is com-
patible with the form of realism, defended by John
McDowell, which holds that, although moral proper-
ties inhere in their objects independently of our
attitudes, they can only be perceived when one is
suitably disposed attitudinally. See, for example, his
`Values and Secondary Qualities' and `Projection and
Truth in Ethics', in McDowell (1998), pp. 131±50 and
pp. 151±66, respectively.
160
14. D'Agostino (1993). The `conditions of acceptability'
typically include requirements that the agreement be
reached freely and fairly, and be based upon `consid-
ered and critical reflection', not instantaneous opinio-
natedness. Of course, these must themselves be
regarded as purely contingent criteria from the Ror-
tyean perspective.
15. Rorty (1985), p. 5.
16. A view famously adopted on behalf of non-cogniti-
vism by Mackie (1977), chapter 1, especially pp. 48±9.
17. Rorty (1989), chapter 3, especially pp. 59±60.
18. Ibid., p. 59.
19. M. Macdonald, `Natural Rights', in Waldron (ed.)
(1984), pp. 21±40, at p. 36.
20. For the concept of `social fact', see Durkheim (1982),
especially chapter 1.
21. Rorty (1985), p. 13.
22. The notion of `final vocabulary' is discussed at Rorty
(1989), p. 73.
23. See Rorty (1998), p. 170 for his conviction that a
`human rights' perspective constitutes a morally super-
ior way of viewing and treating people.
24. Blackburn (1984), chapter 6, and (1993), especially
chapters 8±9.
25. Blackburn (1996), pp. 82±100.
26. Blackburn (1993), p. 161.
27. Ibid., chapter 9, especially pp. 172±5.
28. Wiggins (1988), pp. 133±4.
29. One potential problem with this interpretation of
projectivism is that Blackburn has recently contended
that `non-cognitivism' is actually an inappropriate
term to characterise it (Blackburn, 1996, p. 82). He
claims it is compatible with a notion of `best moral
sensibility', derivable thus: conceptualise one's current
moral perspective and all possible rival perspectives as
akin to the structure of a tree in the following way.
The `trunk' contains all of those features that we
would think any acceptable moral perspective would
have to hold. It is this that provides the content of
what he says we can legitimately claim to know as
moral truth. Thereafter, the trunk branches out, in so
far as we conceptualise the perspectives diverging in
their attitudes from this shared core. Blackburn thinks
it is possible to remove the possibility of branching: if
my perspective branches out because I believe that A
is better than B while I acknowledge the possibility of
a perspective that says B is better than A, I have the
following choices: I can stick to my current belief, in
which case the alternative disappears from my `tree' as
a candidate for a suitable sensibility; or I concede its
validity, in which case my branch should disappear.
Alternatively, I concede the merits of both cases, in
 PRAGMATIST LIBERALISM AND THE EVASION OF POLITICS
which case I should improve upon my current belief
that A is better than B: again, this branch should
disappear. Now it must be a matter of considerable
doubt as to whether this could plausibly yield a single
set of moral truths, or whether the label of `truth' is
simply being used rhetorically to affirm a suitably
refined set of attitudes. For this reason, we leave this
issue aside in the present analysis.
30. Rorty (1998), p. 185.
31. Ibid., pp. 167±9.
32. Martha Nussbaum is arguably the most prominent
exponent of literature's value in shaping people's
moral sentiments; see Nussbaum (1995).
33. Rorty, `The Priority of Democracy to Philosophy' in
Malachowski (ed.) (1990), p. 288.
34. In practice non-cognitivism has usually been aligned
with some version of the liberal view that the consent
of those subject to power legitimises its exercise.
35. Mouffe (1993), chapters 3, 9.
36. In a much-debated passage, Rawls concedes that im-
position of fundamental liberal values upon a steadfast
anti-liberal (a violent religious fundamentalist, for
example) may be necessary if the latter is incapable
of accepting the principles (Rawls, 1996, p. 152).
Although it is significant that Rawls's theory does
not recoil from the advocacy of coercion, this case
appears to be sufficiently marginal and exceptional in
Rawls's mind not to need systematic analysis and
justification (or so he thinks). My suggestion here is
that politics may be a lot more agonistic and antag-
onistic than he presumes, and the coercive (as opposed
to consensual) resolution of political conflict is far
more central to everyday politics than Rawls believes.
See, in this regard, Rawls (1999), pp. 529±64, where
he sketches the possibility of an internationalist `law of
peoples', including a scheme of basic human rights.
Although applicable to all humans, Rawls recognises
that perhaps not all societies could agree to affirm such
a law; in their case, he says, there is in principle no
peaceful solution except a modus vivendi, `the domina-
tion of one side or the peace of exhaustion' (ibid., p.
564). At this point, perhaps, we have nothing other
than Rorty-type contingent attitudes by which to
orientate political responses to such societies.
37. This, of course, simplifies and glosses over a lot of
issues. I merely sketch the outline of a possible argu-
ment, to allow swift passage to the particular problem I
wish to address.
38. See Parekh (1997) and Mu
È llerson (1997), pp. 164±5
for some possible conditions for justifiable interven-
tion.
39. A familiar criticism of interventionist actions by wes-
tern powers is that realpolitik produces hypocritical
selectivity: for example, why attack Iraq for its inva-
161
sion of Kuwait and not China for its occupation of
Tibet? We need hardly be ignorant of the cynicism
that doubtless enters into the politics of intervention
to query which moral direction this point is supposed
to take us. If the argument is that we should never
sanction humanitarian intervention unless we consis-
tently apply the same moral considerations in every
other like instance, one must wonder whether the
proper object of moral concern hasn't been entirely
misplaced here. Would the Kosovar Albanians or the
East Timorese thank us for leaving them entirely alone
to their fate, appreciating that we are primarily con-
cerned that our own precious moral consistency might
be impugned if we intervened on their behalf while
continuing to tolerate the rights violations of the
Chinese state?
40. See Daniels (1996), chapter 2 for the concept of wide
reflective equilibrium.
41. I criticise the concept of public-justificatory liberalism
in Evans (1999).
42. Carl Schmitt famously offers a version of this critique,
claiming that liberalism's claims to embody objective
moral truths conflicts with democracy's claim that
legitimacy is based upon the collective will of the
people, thus making `liberal democracy' philosophi-
cally oxymoronic and, in a fractured social world,
practically unworkable. For analysis, see Larmore
(1996), chapter 8.
43. Isaiah Berlin observes that `the very desire for
guarantees that our values are eternal and secure
in some objective heaven is perhaps only a craving
for the certainties of childhood, or the absolute
values of our primitive past': Berlin (1969), p.
172. If this is so, then the need for foundationalist
belief may not be dispensable, though some anti-
foundationalists would put this down to an inability
to live without holding values in an erroneous
manner. Certainly, a purely evaluative conviction
may be vulnerable to dissolution through the waning
of attitudes, but yet again there is not quite the
unfavourable contrast with realist convictions that
one might think. Crises of faith (sometimes term-
inal) in examples of the latter are, after all, familiar
enough to challenge the foundationalists' argument
that their form of moral conviction is that much
more insulated from doubt.
44. I would like to thank John Torrance and Roger Crisp
for comments on a distant ancestor of this chapter
which helped to shape its current form. Thanks are
owed as well to the participants at the 1999 University
of Wales Swansea political theory conference at Gre-
gynog, and in particular Bruce Haddock, for valuable
discussion.
 12
LIBERALISM AND CONTINGENCY
Bruce Haddock
Liberals, or at least theorists concerned to defend
liberal political positions, have in the last decade
seemed peculiarly diffident about the foundational
arguments which were once the stock-in-trade of
liberal theory. Rawls's defection from foundational-
ism in Political Liberalism is the most celebrated (and
arresting) case of a phenomenon which is wider and
deeper, going to the heart of what we might mean by
1
normative political philosophy. Rawls continues to
espouse a comprehensive liberal doctrine, but he is
anxious to show that terms of political cooperation
can be defended independently of the particular
comprehensive doctrine he happens to endorse.
Other theorists (Rorty and Gray have been pro-
minent here) defend the institutions and values of
the liberal communities they happen to have grown
2
up in, but for purely contingent reasons. Rorty's
liberal ironist can delight in the cultural diversity
which confronts him, enriching the heritage through
complex description rather than analysis. Sentimen-
tal stories may be said to enhance possibilities for
self-understanding. They do not furnish rules to live
by. Rules which purport to be theoretically defen-
sible are in fact simply arbitrary conventions which
facilitate mutual enjoyment. But they do not facil-
itate everybody's enjoyment. The pretence that ar-
bitrary conventions are necessary conditions for a
just society is one of the principal foundational
illusions that the liberal ironist is concerned to
dispel.
Gray's position (at least in his more recent writ-
ings) is more straightforward. He simply asserts that
attachment to liberalism is a function of loyalty to a
162
(purely contingent) national tradition. Liberals may
express their loyalty in terms of doctrinal commit-
ments. But once they recognise the fragility of the
assumptions they have made (regarding progress,
rationality, personal autonomy and so on), they
are bound to accept that the sources of their attach-
ment are cultural rather than philosophical. There
may well be decisive practical reasons for endorsing
liberal institutions and practices as the best means of
preserving and developing the advantages which
accrue from flourishing civil societies. But such a
defence could never be more than conditional and
certainly could not assume the guise of a deontolo-
gical theory.
Closer inspection, however, reveals that these
retreats from foundational argument are far from
clear-cut. Rawls is the most interesting case, since
he accepts that the resources of comprehensive
doctrines can be used to defend a liberal political
consensus. His concern is to ensure that a particular
political consensus should not be identified with any
particular comprehensive doctrine. Clearly he is
trying to walk a fine line here. As Leif Wenar has
recently pointed out, a (supposedly) non-conten-
tious notion of reasonableness is being asked to do
3
a great deal of work. It is extremely difficult to
distinguish theoretically between reasonable and
unreasonable pluralisms, though in practice crude
political judgements about what should or should
not be tolerated are made all the time. Similar
problems arise when we try to distinguish reasonable
from unreasonable comprehensive doctrines. Is it
simply the case that any comprehensive doctrine
 LIBERALISM AND CONTINGENCY
which endorses a reasonable political consensus (for
whatever reasons) is to be adjudged reasonable? Or is
there something about the comprehensive doctrines
themselves which distinguishes them as reasonable?
Wenar argues that the conception of political rea-
sonableness Rawls invokes is comprehensive (Kan-
tian) rather than limited, excluding not only
fundamentalists and fanatics from a reasonable prin-
cipled consensus but also Catholics, Hobbesians and
4
Humeans. Wenar's solution is to make the concep-
tion of political reasonableness even more limited,
thus making it non-contentious from the perspective
of comprehensive doctrines. But in that situation it is
unlikely that comprehensive doctrines would gen-
erate the kind of principled support for a political
consensus which would transform a modus vivendi
into a principled agreement. It would also make
Rawls even more vulnerable to critics who argue
that political liberalism requires more rather than less
5
foundational support. What is clear is that Rawls
himself insists that some sort of bridge between a
political consensus and reasonable comprehensive
doctrines is indispensable to his position, citing
Hart's notion of `the minimum content of natural
law' as an example of the kind of argument he has in
6
mind.
Gray's position is ambivalent in precisely the same
way. He feels free to equate `a commitment to the
preservation of civil society' with `a commitment to
7
the maintenance of civilisation.' And though he
accepts that association on these terms `may be only
one of the diverse forms of flourishing our species has
achieved', he nevertheless defends a form of civil
society as not only `the best in our cultural inheri-
tance' but also `the best that the species can presently
8
reasonably hope for.' Having withdrawn from lib-
eral universalism, then, he continues to defend a
position which is far broader than the discrete
traditions which have sustained modern western
political culture. Certainly civil society can be en-
dorsed from within these traditions but cannot be
identified with them. And to specify even in the
broadest formal terms what human beings as a species
might `reasonably hope for' surely takes us beyond
the contingent theoretical resources of particular
cultures.
Like Rawls, Gray turns to Hart's `minimum con-
163
9
tent of natural law' for support. What he cannot
allow, however, is that this `universal minimum
content of morality' (whatever it might comprise)
should be specifiable in theoretical terms which
could serve as a normative foundation for a variety
10
of political settlements. Following Berlin, he insists
that value incommensurability is an irreducible fea-
11
ture of pluralist societies. In this scheme of things,
what we confront in the last resort is existential
choice unaided by practical reason. The appeal to a
`minimum content of natural law' or a `universal
minimum content of morality' does no normative
work here.
Rorty makes fewer concessions to even a residual
foundationalism. But it is worth noting in passing
that when he claims that (for liberals) `cruelty is the
worst thing we do', he is invoking an evaluative
criterion which is far from clear in its meaning and
12
scope. Mill scholars have always recognised that
the relevant sense of `harm' for liberal theory needs
13
to be specified carefully. And more recently John
Kekes has argued that priority can only be accorded
to the claims of cruelty at the cost of undermining
14
the specifically liberal character of a state. The
point to insist on here is that assertion of the
significance of cruelty in liberal doctrine needs to
be supported by argument. It is vacuous as a simple
description of liberal sensibilities.
What we see emerging here is a series of more or
less reluctant adjustments to a minimal foundation-
alism, couched often in such broad terms that the
idea of a foundation serves no normative function.
This in itself is philosophically interesting, even if it
impinges hardly at all on the more immediate prac-
tical concerns of political theorists. It is surely sig-
nificant that allusions to Hart's `minimum content of
natural law' are often made in passing, as if discussion
at the most general level could have no implications
for the claim that liberal values are contingent. And
it must be said that the way Hart presents his case
rather encourages that view. The `minimum content
of natural law' is not a contentious conclusion to an
elaborate argument, but a series of `simple truisms'
which afford reasons for adopting legal and moral
15
practices. Hart's concern in this section is with the
viability of any legal system, rather than with the
desirability of a liberal system or any other variant. In
 BRUCE HADDOCK
The Concept of Law Hart treats (almost exclusively)
liberal discussions of law within the broader context
of legal order as such. My concern in this chapter is
to reverse the direction of the argument in order to
develop the normative implications of a formal legal
framework. My claim is that reflection on the idea of
a formal legal framework furnishes grounds for pre-
ferring certain forms of political or legal order to
others. I do not claim that only liberal political
orders are defensible, or that all forms of liberal order
are desirable, only that there are philosophical
grounds for preferring to manage public affairs in a
context of liberal institutions and practices.
What Hart focuses on in the brief passage on natural
law which has been cited so frequently is the con-
tingent predicament in which all societies find them-
selves. Adapting arguments from Hobbes and Hume,
he depicts a context which must inform the moral and
legal reflection of all communities, anywhere. Indi-
vidual vulnerability to personal attack, relative equal-
ity (such that even the weak can threaten the strong
through stealth), limited altruism, limited resources,
limited understanding and suchlike put clear con-
straints on the kinds of institutions and practices
which could (in principle) be devised to sustain social
life. We can imagine situations in which these restric-
tions would not apply. We can fantasise about heaven
on earth, where both resources and goodwill are
superabundant. But whatever arrangements people
in that happy situation might devise for their delight,
we can assume that distributive and coercive proce-
dures would not be among them. Politics as we know it
would simply not be necessary. At the other extreme,
as Hume argued, absolute scarcity renders conven-
tional institutions and practices redundant. Where
starvation is a likely prospect, we cannot assume that
the law of property would be greatly respected. As
Hume expressed the point, in a dire emergency `the
strict laws of justice are suspended . . . and give place
to the stronger motives of necessity and self-preserva-
16
tion.' In the ordinary way of things, of course, we are
not faced with such stark dilemmas, except in cases
where regimes actually break down. What Hart asks us
to consider is the normative impact of truths that are
`contingent on human beings and the world they live
in retaining the salient characteristics which they
17
have.'
164
We may say, at the very least, that recognition of
mutual vulnerability commits us to the adoption of
some rules of social cooperation. These rules may be
many and varied and may in substance be wholly
arbitrary. But that there should be some rules may be
said to be a defining characteristic of a society. Given
that societies with (necessarily) limited understand-
ing and scarce resources will have to make hard
choices, we must assume that they will have author-
itative procedures for allocating resources. Govern-
ment of some kind will have to be carried on, even if
a formal specification of institutional roles is not
developed. But, of course, to recognise the necessity
of government in certain circumstances is not to
endorse any particular kind of government. Hart is
concerned to do no more in his discussion of the
minimum content of natural law than to set before
his readers the general predicament to which all legal
and moral schemes must respond; he is interested in
questions of viability rather than desirability. And
even Hume, whose discussion of justice has the
conveniences of commercial society in mind, offers
generic principles which all viable societies will have
to endorse. On the face of it, little follows from the
`three fundamental laws of nature' which constitute
18
justice for Hume. We can grant that all societies
must sustain `stability of possession' without specify-
ing precise terms on which property might be en-
19
joyed. Similarly, to move from possession to
`transference by consent' is simply to insist that there
should be some procedural rules, not that any specific
20
rules should be deemed to be appropriate. The
third dimension of justice (`the performance of pro-
21
mises') is more fruitful. Hume himself may want to
stress the utility of promise-keeping, but reflection
on the constitutive role of truthfulness in any con-
ception of agency extends the discussion to the
necessary requirements of intelligible agency.
The key issue here is not how particular practices
(like promising) might be justified but what makes
practices possible. Following Wittgenstein and
Winch, we have come to accept a capacity for rule
following as a constitutive feature of human con-
22
duct. We know that things happen to people in the
course of their lives and that we can make reasonable
predictions about what people might do from a basis
of crude generalisation. But doing things involves
 LIBERALISM AND CONTINGENCY
applying criteria, and these criteria consist of (more
or less formally acknowledged) social rules. It is
important to stress that we do not need to be able
to specify the rules we are following in order to be
said to be following rules. Wittgenstein classically
focuses on the grammar of a language as a framework
of rules which enable us to speak. Children can use a
language before they can analyse sentences. And we
can imagine languages which have never been ana-
lysed by anyone. The users of such languages can
nevertheless be said to be following rules even if the
rules have not been formally specified. If they juxta-
pose sounds in a wholly arbitrary way they will not be
understood. Children acquiring the language will be
corrected by adults, though the adults may not be
able to specify the linguistic rule which had been
wrongly applied. We may expect the range of prac-
tical usage to be wide in a language which is wholly
informal. But informal rules remain rules, specifiable
in principle even if they have not been specified in
practice. We simply cannot conceive of a language
which is not (in some sense) rule-bound. It is surely
significant that we assume that anyone could (in
principle) learn any language. They may not be
taught rules but they know how to carry on making
meaningful sounds.
The framework of rules implicit in any language
may appear to be entirely neutral. Yet, as Winch has
argued, a presupposition of truthfulness is a basic
23
requirement of any language. People tell lies and
deceive one another because it can be assumed that
(on the whole) they mean what they say. He stresses,
too, that using a language is a practical engagement
in which commitments are made to other people.
When children are taught a language they are at the
same time inducted into the ways of a form of life.
The rules which constitute the form of life could not
be specified in abstract fashion. What we have,
instead, is a series of overlapping practices, not all
of them mutually supportive, which yield a sense of
how to carry on, how to respond to other people,
what to expect of oneself.
Language is instructive to us here as an illustration
of a natural practice. We do not choose to learn our
first language, we may have little or no understand-
ing of the grammatical rules implicit in that lan-
guage, and yet we learn to correct ourselves as we
165
speak. We also acquire a rudimentary understanding
of how we should respond to other speakers. The use
of a language is thus a social engagement which
commits us to much more than a basic set of
grammatical rules.
Lyotard, among others, has stressed the insepar-
24
ability of speaking and listening. We know that
societies devise complex ways (ritually and profes-
sionally) of limiting rights to speak about certain
matters. If someone were systematically denied the
right to speak on any occasion, however, they would
effectively have been reduced to the status of a non-
person, irrespective of whatever else might happen to
them. In this sense, some basic rights accompany
language use. Rights can, of course, be forfeited
through wrongdoing. But recognition of a prima facie
case to be heard can be justified simply by focusing
on the logic of conversation. It would be absurd to
claim that everyone had a right to take part in any
conversation, even in principle. I simply do not
know enough about brain surgery or football to insist
that my voice should be heard. Yet to be denied the
right to participate in any conversation would be
tantamount to being excluded from all human com-
munity. Such things may be thought desirable in
exceptional circumstances, but they would always
have to be justified in terms of some conception of
human flourishing.
Just how far this sort of reflection can take us is a
much-disputed question. Few modern philosophers
would want to follow Hegel's attempt in the Philo-
sophy of Right to defend the necessity of a specific
institutional framework from a series of reflections on
the character of human agency. Yet it may well be
that what we should be worried about in Hegel is not
so much the ambition of the project as the detailed
development of the argument. Modern readers tend
not to find the elaborate deductive structure of the
Philosophy of Right helpful. It should be clear, how-
ever, that when he asks us in the compressed in-
troduction to reflect on what we might mean by
agency, what we presuppose about human conduct
when we make choices, how we should regard
ourselves in relation to the cultures that sustain
us, he is raising questions which any political phi-
25
losophy must respond to. It does not follow, of
course, that these issues are always treated self-con-
 BRUCE HADDOCK
sciously. For the most part, many basic presupposi-
tions will simply be taken for granted in order to
proceed with more specific and practically compel-
ling questions. But it remains the case that presup-
positions are made which have different implications
for practice.
Our concern, then, before we consider the institu-
tions and policies we might happen to desire, is to
explore the necessary requirements which enable
practices to flourish. We must assume that these
requirements will be very general indeed since we
know from ordinary experience that different cul-
tures pursue all manner of ends in a variety of ways.
Our focus should be on what people need to ac-
knowledge in order to attain any satisfaction at all
rather than on the things they happen to want. We
have accepted (following Hobbes, Hume and Hart)
that recognition of mutual vulnerability commits us
to establishing terms of social cooperation. If you do
not accept this point, you can have no interest in
normative political philosophy.
The point may be put more strongly. If you regard
social cooperation as optional or irrelevant, you will
have fundamentally misconceived the idea of human
agency. When we do things we apply and adapt rules
and procedures, and these rules and procedures have
a public (social) life even if they are not formally
acknowledged. To say or do things deliberately is to
apply criteria such that we might wonder whether or
not we had said or done the right thing. Following a
practice necessarily involves the possibility of mak-
ing a mistake. But it would make no sense to say that
it was entirely arbitrary whether or not a mistake had
been made.
Practices are thus to be understood in terms of
rules and conventions which are more or less ade-
quately observed. In relation to any specific engage-
ment, the rules and conventions are regarded as
authoritative. Sometimes (as in a game) the engage-
ment is constituted by specific endorsement of a
formal set of rules. You cannot play chess and
repudiate the rules at the same time. The overlap-
ping (and largely informal) rules of social life are
much more difficult to discern. In this case it makes
perfect sense to suspend judgement about certain
conventions while continuing to engage in coopera-
tive projects more or less normally. But we could not
166
repudiate all rules and conventions without ceasing
26
to make sense, even to ourselves.
The intelligibility of our own conduct thus de-
pends upon the social dimension of the practices we
are inducted into. We can envisage an esoteric
practice, but it is clearly parasitic on the fact that
we have been inducted into other practices naturally,
much as our ability to construct private languages is
parasitic on the fact that we already know what
27
languages are. It is only because practices are held
to be authoritative in some sense that we can be said
to engage in intelligible activity. And this applies
even if our concern is to revise radically the way a
practice is carried on. The focus again is on how
things are done, not on the specific goals that are
pursued.
The conception of rules defended here does not
depend upon the self-conscious adoption of rules of
conduct in particular communities. Our concern is
rather with the necessary conditions of intelligible
action. If we are to conceive of societies at all we
must think in terms of rules which constitute prac-
tices. Of course, societies cannot be conceived in
purely procedural terms. Cultures and subcultures
will endorse different conceptions of a worthwhile
life, often pursuing ends which are strictly incom-
mensurable. And if we focus on these different ends,
it may be that we will have to accept that certain
ways of life are incompatible. We may not be able to
find any common ground between deeply held be-
liefs. No matter what we might say, however, about
the constitutive values of particular communities, we
still have to regard the practices within these com-
munities as instances of rule-following, even if it has
never occurred to agents to describe their conduct in
this light. They may think of themselves as marion-
ettes in a display of divine virtuosity, and we must
refer to these beliefs when we reconstruct their
conduct. But what they are actually doing is making
decisions, appraising one another's conduct in terms
of assumptions and standards, correcting each other
when they make mistakes, arguing, reproaching or
punishing one another for failure to observe appro-
priate rules and procedures. The things they do can
be done well or ill, with more or less good faith. But
simply to say that they are doing something commits
us to certain ways of conceiving of conduct.
 LIBERALISM AND CONTINGENCY
Oakeshott focuses on the quality of active engage-
ment in his characterisation of a practice as `a
prudential or an authoritative adverbial qualification
of choices and preferences, more or less complicated,
in which conduct is understood in terms of a pro-
28
cedure.' The terms of reference here are so broad
that it may be thought that nothing is ruled out. The
conceptual clarification of what it means to carry on
may seem to have no implications whatever for what
we should actually do. But this formal distinction
only holds while a practice is conducted within the
recognisable terms of reference of rule following. To
behave impulsively or wholly erratically is not to
engage in a practice at all, even though observation
of such behaviour may generate reasonably accurate
generalisations. And Oakeshott himself, notably in
Rationalism in Politics, highlights the ruinous conse-
quences which follow from misconceptions of hu-
29
man agency. To suppose that all right-thinking
people will necessarily endorse the same collective
goals, or that we can somehow stand outside all
practices in order to assess the rationality or desir-
ability of a particular practice, or that liberation from
the constraints of all conventions or traditions will
somehow set us free, is to conceive of human life as it
could not possibly be lived. Philosophy, in this
scheme of things, can specify the conceptual limits
of practical life, without telling us how we should
actually live.
The stress on procedural rules should not be
misunderstood. The claim is not that a minimal
framework is sufficient to sustain a polity, rather
that the framework provides a basis for different
modes of political flourishing. Following Oakeshott,
the framework may be described as `the minimum
condition of any settled association among indivi-
30
duals.' But no polity actually flourishes on the basis
of minimal conditions alone. What is built upon this
minimal (or `weak') foundation will depend upon
the specific (and contingent) features of a given
political culture. The point, however, is that the
contingent values of particular cultures may distort
or constrain practices which can be shown to be
constitutive of a flourishing political life. The terms
of reference here are still very broad indeed. We are
dealing with the possibility of human flourishing
rather than preferences as these have often been
167
understood within liberal theory. The point to focus
on is that philosophical reflection on the necessity of
a minimal foundation may furnish grounds for pre-
ferring certain political options to others.
Hart himself offers us some help along these lines
in the opening arguments of The Concept of Law. His
modification of the received version of legal positi-
vism depended largely on his adoption of a rule-
following model of law as a practice. Following
Wittgenstein and Winch, he sought to highlight
the rules (formal or tacit) which enable agents to
set themselves standards in the conduct of their
business. An agent can only ask herself what she
ought to do (whether as a judge, a lawyer or a
litigant) within the context of a practice constituted
31
by procedural rules. This is not to say that the older
view (endorsed variously by Hobbes, Bentham, Aus-
tin and others) of law as an authoritative command
backed by coercive power is wholly redundant. This
may well be how law is regarded by people who do
not acknowledge the authority of a legal system or
who see themselves as free-riders prepared to exploit
the passive obedience of others, but it certainly does
not help us to explain what makes conduct within a
legal practice intelligible.
Hart clearly expects his distinction between inter-
nal and external perspectives (or rule-following and
habitual behaviour) to do a great deal of work for
him, but he is very cautious about drawing out
normative implications. This is perfectly intelligible
given the overall thrust of his argument. He is
interested in law as such, and not any particular
variety of law which may have emerged in a given
context. A theory of law that simply reflected the
culture and values of a particular community would
(in his terms) be no theory at all. His modification of
legal positivism thus involved only a minimal retreat
from the moral neutrality which was central to that
theory. His conception of a `minimum content of
natural law' is conceived principally as a response to
the contingent circumstances in which human life is
lived. He avoids reference to substantive concep-
tions of the good life which had been the stock-in-
trade of the traditional theory of natural law. He
accepts that a formally viable legal system may be
morally reprehensible. The fact that legal rules are
always grafted on to a moral context has (for Hart)
 BRUCE HADDOCK
no implications for the specification of the formal
requirements of a legal system.
Hart makes the point emphatically in his insis-
tence that `the question what the criteria of legal
validity in any legal system are is a question of fact',
stemming from the `rule of recognition' which hap-
32
pens to be endorsed in the particular system. A rule
of recognition may be complex (as in an elaborate
juridical culture such as the United States) or for-
mally straightforward (whatever the Queen in Par-
liament enacts will be recognised as law in the
United Kingdom) or simplistic (the tyrant's will,
Rex I in Hart's example, will be taken to be
33
law). Hart's point is that we can make sense of
each of these legal systems in terms of criteria
furnished by a rule of recognition, though in each
case the rule of recognition may be arbitrary. `Its
34
existence', Hart insists, `is a matter of fact.' What I
want to suggest, however, is that it makes the world
of difference that what we are talking about here is a
rule. To be sure, the whim of Rex I constitutes a rule
only in the most primitive sense. But recognising
that it is the existence of a rule which makes public
cooperation possible gives us grounds, all other
things being equal, to prefer a rule which is more
predictable in its operation rather than less.
Hart's description of his work in the preface of The
Concept of Law as `an essay in descriptive sociology' is
35
thus seriously misleading. It is better seen as an
exercise in normative theory extending beyond the
normal confines of ideological dispute. Indeed the
tension which some commentators have seen in
Hart's work between his conceptions of critical
morality and analytical jurisprudence can be easily
resolved if the normative dimension of analytical
jurisprudence is acknowledged. The point to insist
on is simply that the minimal criteria which enable
us to identify a legal system will be more or less
effectively embodied in actual legal systems. Quite
how far this could take us in the critical amendment
of legal or constitutional practices is an open (and
contentious) question. But if it is accepted that
predictability and transparency are essential to the
furtherance of cooperative pursuits among autono-
mous strangers, then it is clear that we do have
grounds for preferring certain institutional proce-
dures to others.
168
The crucial distinction at work here is between
36
`primary' and `secondary' rules of obligation. It is
perfectly possible to envisage a community in which
dealings between people would be regulated exclu-
sively in terms of customary rules. Social ostracism
clearly constitutes a powerful coercive constraint in
contexts where alternative lifestyles are not avail-
able. In a purely customary society one is not faced
with choices between options but with more or less
reluctant compliance. But, as Hart makes clear, such
arrangements of unofficial rules can only work in `a
small community closely knit by ties of kinship,
common sentiment, and belief, and placed in a stable
37
environment.' Customary rules are simply given.
As soon as we start wondering about their desir-
ability, we introduce hypothetical considerations
which may be contentious. The need to accommo-
date change of any kind obliges us to make judge-
ments. Our concern here is with the most
rudimentary level of social organisation. The point
to stress is simply that a (hypothetical) customary
society ceases to function effectively when alterna-
tive possibilities are made available. Disputes about
ends have to be settled somehow (through fighting,
ritual manipulation, social pressure or whatever). But
it soon becomes evident that ordinary business re-
quires authoritative procedures. Practical judgements
cannot be made in chronically uncertain situations.
The introduction of these authoritative procedures
(or `secondary rules') constitutes, for Hart, `a step
38
from the pre-legal into the legal world'.
Note that the transition to a `legal world' is not a
choice in any ordinary sense. We may regard dif-
ferent legal rules as more or less useful or desirable,
but we cannot reject the idea of legal rules without
undermining our capacity to function as agents in
complex contexts. The notion of returning to a
`pre-legal' situation is no more than a nostalgic
fantasy. In an unfashionable speculative sense, we
may be said to have crossed a divide between kinds
of civilisation. The assumptions which enable cul-
tures to function at this level are not choices or
preferences. They are, rather, the `presuppositions'
or `hinges' without which we simply could not carry
39
on.
The claim that philosophical reflection on the
necessity of a legal framework is at the same time a
 LIBERALISM AND CONTINGENCY
justification of legal order should not itself be con-
tentious. We can easily appreciate that a pure theory
of legal order would not loom large in a society that
happened to display an overwhelming consensus
around a substantive conception of the good. But
it clearly would not follow that legal order depended
upon such a substantive consensus. In practical
terms, we simply have to accept a plurality of de-
fensible values as a starting point for our reflections.
Recognition of the contingency of values at this
level, however, does not rule out the possibility that
more binding philosophical reasons might be ad-
vanced in defence of a political order which made
diversity manageable.
Examples of the kind of argument I have in mind
will be familiar to everyone from the canon of
`classics' of political philosophy. To be sure, the
usual suspects among the classics constitute a motley
crew. We find arguments ranging from the most
abstract accounts of conditions of theoretical intel-
ligibility to detailed moral and political prescriptions
for proper conduct in everyday life, informed by
personal obsessions which are (in some cases) laugh-
able or contemptible. But this is just to recognise that
classics of political philosophy operate at many
different levels, and that matters of philosophical
interest are sometimes concealed in the most un-
likely places. Whatever else might be going on in
such texts, we nevertheless find attempts to specify
what would need to be true about human beings
living in communities if the ordinary things they
take for granted are to make any sense at all. This, of
course, is not necessarily what political philosophers
always suppose they are doing, especially in cultures
40
untroubled by the `burdens of judgement'. But we
go back to them as philosophers precisely because
certain sorts of problems necessarily arise when we
take the justification of terms of social cooperation
seriously. Oakeshott put the point nicely in a cele-
brated discussion of Hobbes. He defined philosophy
41
as `the establishment by reasoning of true fictions.'
Given the things we accept, there are certain notions
which simply have to be held to be true. The point is
not that we can give a list of these indubitable truths,
rather that critical thinking depends upon presup-
positions which can and (at times) must be exam-
ined.
169
Nor is there any suggestion that `true fictions' are
unchanging. What needs to be presupposed in order
to make sense of ordinary experience will change
with the variety of circumstances, but not as a matter
of taste or fashion and some very basic assumptions
may be more or less invariable. The point, however,
is that we cannot specify what will be assumed to be
invariable before critical enquiry begins. It would be
very odd if someone (other than a mischievous
philosopher) were to deny that things happen at
some time in some place. And while we can conceive
of a human world in which moral choices were
somehow unreal, we would nevertheless be surprised
(as a matter of fact) to learn that languages existed
without an imperative mood. At some level we
picture human association as a matter of agents
making choices which affect others, even if a capa-
city for making autonomous choices is not greatly
prized within a particular culture.
The significance of agency for an understanding of
morality can be easily granted. What this concession
may commit us to at the political or institutional
level is much less clear. At the very least, we must
accept that intelligible agency depends upon stability
of expectations. Stability, of course, can be attained
in all sorts of ways. There are stable expectations in
prisons and monasteries, yet it would be odd to
regard either as models of social and political life.
Leviathan also secures stability, but at a price modern
liberals have been reluctant to pay. If we supposed,
however, with Hobbes, that an anarchic and un-
predictable state of nature was the only alternative to
authoritarian government, then we may accept that
authorising a sovereign to lay down binding rules was
a small price to pay for a measure of `commodious
living'.
In the ordinary way of things, we are not con-
fronting Hobbes's dilemma. We are faced with more
or less desirable options in viable political circum-
stances. What may be lost from sight is that stress on
what is chosen may obscure the significance of an
institutional context which makes choice possible. It
is a paradox, as Oakeshott has pointed out, that
`Hobbes, without being himself a liberal, had in him
more of the philosophy of liberalism than most of its
42
professed defenders.' Recognition of the overriding
significance of stability in contexts in which agree-

ment about ends cannot be expected does not
commit us to Hobbesian measures. Stability of ex-
pectations may also be achieved in legal orders which
43
guarantee basic rights. Philosophical interest fo-
cuses on the fact that we should have rights, not
simply on the specific rights we happen to value.
It remains the case, of course, that urgent and
contentious theoretical issues are raised when we
consider the sorts of rights that should be enshrined
in a modern legal system. And at the policy level,
too, it is by no means clear how a foundational
commitment to agency should be translated into
concrete proposals for educational and health provi-
sion, taxation and so on. Even theorists who share a
broadly neo-Kantian approach to these questions
will diverge widely in their views on the role of
44
the state in specific areas. Philosophical agreement
is thus much more likely on the priority of liberty
than (say) the difference principle.
What I am defending here is the idea of a
spectrum which runs from political
through ideology to policy questions, rather than
a categorial distinction between philosophy and
other modes of theoretical concern. Relations of
entailment do not obtain along the
rather, we are dealing with compatibility in relation
45
to what Rawls terms `lexical order'.
cance of political philosophy at the level of policy is
unlikely to be intrusive in consensual
cultures. But contentious issues (such as abortion
in the United States or welfare in Britain) cannot
be adequately handled without confronting (what
still might properly be called) foundational ques-
tions. If we ask ourselves how the citizen should
stand in relation to the state in a modern polity, we
are inviting reflection at a number of different
levels. We could be concerned with the obligation
of parents to ensure that their children are educated
and law-abiding or the obligation of the state to
provide support for the destitute or wider questions
of shared responsibility for the conduct of public
life. It would be too much to expect a seamless web
of doctrine to run through these questions. But the
claim that philosophical discussion is irrelevant,
BRUCE HADDOCK
inappropriate or meaningless would surely cripple
our critical endeavours.
The implications of this view are wide-ranging
and cannot be developed in any detail here. Even
if the radical contingency of liberal values and
institutions is granted, we are still left with the
problem of explaining what it means to make hard
choices in contexts where binding decisions have
to be made. For the most part, we can accept that
reasons which are regarded as compelling within a
tradition may have little purchase beyond it. But
traditions are no more self-contained than lan-
guage-games or forms of life. It is precisely when
terms of reference are challenged that more basic
forms of justification are called for. In these cases,
we have to give reasons for our choices which
extend beyond the values we happen to have. We
do not have to presuppose that stark moral or
theoretical dilemmas can always be resolved satis-
factorily, only that we have to appeal to (some-
philosophy thing like) practical reason whenever we are asked
to give a public justification of our preferences.
What makes liberalism distinctive is not so much
the rationalism that critics have often focused on as
spectrum; the institutional procedures which have been de-
vised to facilitate agreement in pluralist contexts.
The signifi- My claim, then, is that a logic of practical reason is
at work whenever hard choices are made. Where
political liberal theory goes astray is in treating the strong
version of practical reason that has emerged in
pluralist societies as the ideal form for the resolution
of public disputes in any context. If my argument is
correct a weak foundationalism, drawn from reflec-
tion on the necessary formal requirements for viable
agency, can be seen to be at work in the positions of
sceptical liberals even when they are reluctant to
acknowledge the fact. To be sure, the idea of a
philosophical foundation invoked here is very gen-
eral indeed. Most regularian polities could be de-
fended along these lines. And, of course, no polity is
merely regularian. My contention, however, is that
awareness of the necessary conditions for any polity
to flourish does provide grounds for tempering more
narrowly ideological preferences.
170
 LIBERALISM AND CONTINGENCY

Notes

1. Rawls (1996). 17. Hart (1961), p. 195.

2. See Rorty (1989); Gray (1995a) and (1997). 18. Hume (1969), p. 578.
3. See Wenar (1995). 19. Ibid.
4. See ibid., pp. 60±1. 20. Ibid.
5. See Bellamy and Hollis (1995). 21. Ibid.
6. See Rawls (1996), pp. 109, 161, and Hart (1961), pp. 22. See Wittgenstein (1968) and Winch (1958).
189±195. 23. See Winch (1972), especially pp. 61±3.
7. Gray (1993), p. 328. 24. See Lyotard (1993), pp. 135±47.
8. Ibid., p. 328. 25. See Hegel (1991). For further discussion see Haddock
9. Gray (1995a), p. 81. (1993) and my `Hegel's Critique of the Theory of
10. Ibid., p. 81. Social Contract', in Boucher and Kelly (eds) (1994),
11. See Gray (1995b). Whether Gray is right to portray pp. 147±63.
Berlin precisely as he does is, of course, another ques- 26. See Wittgenstein (1969).
tion. He draws attention (ibid., p. 14) to Berlin's claim 27. See `Can There Be a Private Language?', in Rhees
that `our conscious idea of man . . . involves the use of (1970), pp. 55±70.
some among the basic categories in terms of which we 28. Oakeshott (1975a), p. 55.
perceive and order and interpret data.' And he is clear 29. See Oakeshott (1962) and, for discussion, Haddock
that these `basic categories' are neither empirical nor (1996).
formal in any straightforward sense. Gray is happy to 30. Oakeshott (1975b), p. 62.
describe this as `a sort of quasi-Kantian transcendental 31. See Hart (1961), pp. 54±60.
deduction' (ibid., p. 14), without dwelling on the 32. Ibid., pp. 107, 245.
implications this might have for radical value pluralism. 33. See ibid., pp. 97±8.
In `Does Political Theory Still Exist?', however, Berlin 34. Ibid., p. 107.
also describes these `basic categories' as `universal ± or 35. Ibid., p. vii.
almost universal ± values'. See Berlin (1978), p. 166. 36. See ibid., pp. 77±96.
And in all his later writings he was careful to distinguish 37. Ibid., p. 89.
pluralism from relativism. See his `Alleged Relativism 38. Ibid., p. 91.
in Eighteenth-Century European Thought', in Berlin 39. The terminology here is drawn from Collingwood
(1991), pp. 70±90. Berlin's sympathy for the variety of (1940), pp. 3±77, and Wittgenstein (1969), p. 44.
modes of human flourishing was always coupled with 40. See Rawls (1996), pp. 54±8.
recognition of the appalling cost of political folly, 41. Oakeshott (1975b), p. 25.
especially in the twentieth century. And he never 42. Ibid., p. 63.
wavered from the view that it was the responsibility 43. See Martin (1993).
of the political philosopher to specify the normative 44. Consider, for example, the contrasting positions of
limits of a defensible pluralism. Rawls (1996), O'Neill (1996), Nozick (1974) and
12. Rorty (1989), p. 146. Hayek (1960).
13. See Rees (1985). 45. Rawls (1972), pp. 42±8. A useful distinction between
14. See Kekes (1996). the philosophy and ideology of liberalism is also
15. Hart (1961), p. 189. mooted but not defended in detail in Croce (1973),
16. Hume (1966), p. 186. pp. 235±43 and 263±7.

171
This page intentionally left blank
 Part Five
LIBERALISM VERSUS REPUBLICANISM?
This page intentionally left blank
 13
BACK TO THE FUTURE:
PLURALISM AND THE
REPUBLICAN ALTERNATIVE
TO LIBERALISM
Richard Bellamy
1
Contemporary societies are irreducibly pluralist.
Their growing differentiation and complexity gen-
erate a variety of interests, ideals, cultures and values.
This pluralism manifests itself in the wide range of
ethnic, religious, ideological, occupational and other
allegiances of citizens. The resulting diversity creates
potential conflicts between groups of people holding
different views and within single individuals. For
differing ends, outlooks and evaluations may prove
logically or practically incompatible and involve
2
incommensurable types of moral claim. These ten-
sions are the lifeblood of current democratic politics
and its greatest challenge. Clashes of opinion and
interest provide the rationale for democracy, yet
taken to extremes can make it totally unworkable.
Liberalism presents itself as the appropriate poli-
tical form for such conditions. After all, its origins lie
in the religious conflicts of early modern Europe and
the class struggles of the centuries that followed.
Liberalism tries to accommodate difference by pro-
tecting each person's capacity to pursue his own good
in his own way to the extent that is compatible with
the similar pursuits of others. According to the
standard liberal formula, this proposal involves seek-
ing the maximal set of equal liberty rights capable of
3
being held by all. These rights are usually to be
secured by a constitutional framework of universal,
175
general laws which constrains and provides the basis
for the operations of democratic institutions and the
economic market. However, if different liberties em-
body incommensurable views of human flourishing,
as pluralists believe, then a clash of liberties will place
this doctrine in trouble. No metric will be available to
adjudicate between different sets of liberties and show
4
which combination most maximises freedom. Earlier
liberals avoided this problem by holding to a histori-
cist faith in the progress of society. A view still held
implicitly and occasionally explicitly by many liberals
today, this thesis assumes pluralism fosters autonomy
and that an invisible hand reconciles the various
heterogeneous ends pursued by individuals in mu-
5
tually supportive ways. Today's complex and diverse
societies render such solutions sociologically unten-
6
able as well as theoretically debatable. Different
spheres of life and world-views generate competing
and conflicting ethical codes, goals and commit-
ments. Worse, the very principles of liberalism can
collide and are themselves subject to a plurality of
incommensurable evaluations and interpretations.
Witness the debates over issues such as abortion,
freedom of information and social rights between
rival factions within the liberal camp. Consequently,
liberalism cannot hold the ring for such disputes. It is
within, not above, the fray.
 RICHARD BELLAMY
To accommodate plurality requires an alternative
approach, one I dub democratic liberalism. Of re-
publican inspiration, it eschews a pre-political con-
sensus on basic liberties for political negotiation
aimed at a fair compromise. Instead of relegating
the differences that accompany pluralism to the
private and civil spheres, it gives them public re-
cognition. Republicans view liberty as a civic rather
than a natural condition. Freedom only exists within
a republican polity where self-government removes
the possibility of domination by others through
allowing citizens to ensure the laws respond to their
7
various concerns and values. Within pluralist socie-
ties, these deliberations entail negotiating fair com-
promises between the different ideals and interests in
play. Such reciprocal decision-making produces laws
that recognise and reconcile the members of a
pluralist polity, enabling them to live on an equitable
basis with others. Though republicanism and liberal-
ism are historically entwined, with elements of both
present in the political systems of most western
8
democracies, coexistence should not be taken for
complementarity or overlap. As I shall argue below,
republican justifications and conceptions of liberty,
rights and the rule of law differ from the liberal's in
important respects ± most especially in relation to
9
the role and nature of democracy. Republicanism
issues in a democratic liberalism, where liberty and
rights are intrinsic to and only achievable through
democracy, not preconditions for the political pro-
cess as liberal democrats contend.
The argument proceeds as follows. The next sec-
tion analyses the weaknesses of liberal democracy
within complex and plural societies, while the suc-
ceeding section presents the democratic liberal alter-
native. I then present a defence of this proposed
return to the liberties of the ancients against the
charges that it is now both impractical and undesir-
able.
Liberal Democracy
Liberal democracy rests on a distinction between the
state and civil society. Liberals standardly see con-
stitutionalism as a normative framework that sets
limits on and goals for the exercise of state power.
Traditionally, its principles are grounded in a social
176
contract designed to legitimate the state's monopoly
of violence. According to this argument, free and
equal citizens would only consensually submit to a
polity that removed the uncertainties of the state of
nature while preserving the most extensive set of
equal natural liberties. Interference by the state or
law is only justified to reduce the mutual interfer-
ences attendant upon social life so as to produce a
greater liberty over all. The separation of powers
supposedly fosters this aim by preventing any one
from being judge in his or her own cause, thereby
constraining the arbitrary and partial framing and
interpretation of legislation. The rule of men is
replaced by the rule of universal and equally applic-
10
able general laws.
Two features of these arrangements are worth
highlighting. First, as James Tully has observed,
the normative consensus assumed by the `modern'
liberal conception of constitutionalism hypothesises
a degree of uniformity among the constitutive peo-
11
ple. It assumes that behind different beliefs and
customs lies a common human nature, a natural
equality of status, and shared forms of reasoning
sufficient to generate agreement on constitutional
essentials. What divergences remain are supposedly
eroded as historical progress leads to more homo-
geneous and less stratified societies that conform to a
similar pattern of social and political organisation,
and stand in contrast to the ranked societies and
cultural particularisms of the past. Nation-building
further strengthens this process. As co-nationals, the
people share a corporate identity as equal citizens of
12
the polity.
Second, the rights-based approach goes together
with a conception of freedom as non-interference
and of the state as a neutral ringmaster, unconcerned
13
with upholding any particular set of values. This
understanding of the constitution encourages in its
turn a purely preference-based picture of the econ-
omy and an interest-based account of democracy. In
each case, what matters is the degree outcomes
correspond to the uncoerced choices and express
desires of those concerned. The conditions of pro-
duction and the protection of public goods enter
with difficulty into this view of the economy. The
first are assumed to be the result of voluntary con-
tracts, the latter left up to an invisible hand. Like-
 BACK TO THE FUTURE
wise, politics becomes a competitive market within
which rival interest groups bargain with each other,
and involves no or little attempt to evaluate the
interests concerned. Its purpose is largely instrumen-
tal: to protect against incompetent or tyrannous
rulers by allowing their removal, and to aggregate
individual preferences through majoritarian or con-
sensus voting and encourage politicians to pursue
14
policies that conform to them.
Liberals accept that economy and democracy need
regulating when they threaten the constitutional
structure. However, identifying when such threats
occur and who possesses the authority to remedy
them proves problematic. Because the economy
forms part of the private sphere, there are difficulties
about whether the requisite interference is either
legitimate or perpetrates an even greater intrusion
into people's lives than those it prevents. Such
decisions cannot necessarily be left up to democratic
governments, since interest groups may use the
state's coercive power to further their personal goals.
This dilemma raises a further source of tension
between the hypothetical consent underlying the
constitution and the express will of the people.
Liberals try to avoid this crux by treating the con-
stitution as a `higher' law that provides the precondi-
tions for the `normal' legislation arising out
democratic politics. They see judicial review by a
court buttressed by a bill of rights as the best bulwark
against the democratic subversion of the constitu-
15
tion.
Pluralism erodes this liberal settlement. The social
and economic complexity of advanced societies, and
the consequent multiplicity of interests and values
within them, render majoritarian decision-making
more problematic, increase the difficulty of regulat-
ing the unaccountable power located in civil society
and subvert the rights consensus upon which liberal-
16
ism rests. Increased functional differentiation re-
sults in a proliferation of autonomous centres of
17
power. These centres are capable of making deci-
sions according to a variety of criteria specific to their
respective domains, with unpredictable knock-on
effects for other parts of the social and economic
system. Citizens find themselves locked into a variety
of these spheres, and get pulled in opposite directions
by the inner logic of each. Problems become more
technical, less amenable to general regulations, and
hence harder to control through centralised demo-
cratic mechanisms. The range and scale of decisions
handled by unaccountable specialised bureaucracies,
and involving considerable technocratic discretion,
expands.
As the various areas of social life operate with
increasingly distinct and largely self-validating cri-
teria, so they become more conflictual. They become
ever more taken up with their own concerns and
tend to interpret the world from their own perspec-
tive, generating incommensurable and incompatible
claims. Reconciling such clashes by democratic
horse-trading proves highly problematic. Clashes
of interests appear more zero-sum, and their aggrega-
tion harder to legitimate and enforce because their
relation to any given collectivity is unclear. Making
and sustaining collective decisions is further com-
plicated by the spread of multiculturalism as im-
proved social mobility renders states more
pluriethnic as well. Differences of beliefs and iden-
tities prove even less amenable to democratic bar-
gaining and the formation of stable and fair
majorities than economic and social interests. As
a result, the likelihood of conflict or the oppression
of minorities rises.
of Liberals have responded to these failings of de-
mocracy as an instrument for interest aggregation
and accountability by advocating the enhanced jur-
18
idification of politics. Worried by the destabilising
effects of pluralism, they look to a legal constitution
grounded in a consensus on rights and principles of
justice to guarantee the polity stays just. Contem-
porary liberals suggest citizens can trim their diver-
gent substantive beliefs on the good to arrive at
19
agreement on what is politically right. But this
distinction proves elusive. How and when rights to
privacy and to freedom of speech clash, for example,
involve invoking notions of the presence and ab-
sence of constraints which may be normatively and
empirically evaluated from a range of reasonably
different perspectives, giving divergent answers in
each case. In pluralist societies, the basic liberties and
their interpretation become contested matters. In
practice, therefore, liberal trimming entails the im-
position of a particular reading of liberalism and the
judicial exclusion of dissident voices. Moreover,
177
 RICHARD BELLAMY
without democratic support the constitution and the
courts risk appearing self-validating and being either
practically impotent or obliged to adopt coercive
measures. It remains to be seen if a more inclusive
and political strategy is available.
Democratic Liberalism
Democratic liberalism harks back to a pre-liberal
conception of constitutionalism that identified the
constitution with the social composition and form of
20
government of the polity. Much as we associate a
person's physical health with his or her bodily con-
stitution and regard a fit individual as someone with
a balanced diet and regimen, so a healthy body
politic was attributed to a political system capable
of bringing its various constituent social groups into
equilibrium with each other. The aim was to disperse
power so as to encourage a process of controlled
political conflict and deliberation that ensured the
various social classes both checked and ultimately
cooperated with each other, moving them thereby to
construct and pursue the public good rather than
narrow sectional interests.
As Quentin Skinner and Philip Pettit
21
shown, the heart of the republican approach lies
in a different conception of freedom to the liberal's.
As I noted, liberty is a civic achievement rather than
a natural attribute. It results from preventing arbi-
trary domination rather than an absence of inter-
ference tout court. Domination denotes a capacity to
intentionally control and diminish an agent's realm
of choice, either overtly through various explicit
forms of restraint or obstruction, or covertly by more
subtle forms of manipulation and influence. Arbi-
trariness rests in the power to exert domination at
whim, and without reference to the interests or ideas
of those over whom it is exercised. Pettit notes that
an absence of interference can be consistent with the
presence of domination. Those with such power may
simply choose not to wield it. Social relations will be
adversely affected nonetheless. A good king may
leave his subjects alone but they remain his subjects
and will treat him with deference for that reason
alone, regardless of whatever merit or demerit he may
possess. Likewise, seeking to reduce interference may
in given contexts be compatible with leaving certain
agents or agencies with considerable power over
others. For example, attempts to reduce the arbitrary
hold men have traditionally exerted over women in
marriage have been challenged on the grounds that
they are too intrusive and themselves involve a
greater degree of interference. Similar arguments
have been used against laws to protect employees
from unscrupulous employers. Even social liberals,
such as L. T. Hobhouse, accept that the onus of proof
rests on the proponents of state intervention to show
22
that less interference is thereby created overall.
Republicans by contrast, see debate about the legiti-
macy of interference per se as misconceived. They
concentrate on providing a non-dominating envir-
onment where citizens can lead secure lives, plan
ahead and live on a basis of mutual respect ± con-
23
ditions which may require intervention.
This view of liberty shapes the republicans' dis-
tinctive linkage of the rule of law with the distribu-
tion of power and democracy. Instead of the
constitution being a precondition for politics, poli-
tical debate becomes the medium through which a
polity constitutes itself. This occurs not just in
exceptional, founding constitutional moments, as
24
have some liberals grant, but continuously as part of
an evolving process of mutual recognition. Domina-
tion and arbitrary power involve more than an
infringement of the formal rule of law espoused by
liberals. It is entirely possible to promote general
rules based on whim or self-interest and that entail a
gross curtailment of people's freedom of action. The
generality and universality requirements can also
seem themselves arbitrary if employed to disqualify
special rules that refer to properties that apply to only
some groups ± as when maternity leave for women or
affirmative action policies are accused of being dis-
criminatory, or when such considerations are used to
block any form of regulation which might seek to
25
focus on particular contexts or outcomes. Such
formal criteria appear particularly inadequate at
tackling structural forms of domination, where dis-
crimination and selective blindness have been built
into the institutions, norms, social and economic
relations, and procedures within which the rules are
framed.
Contemporary liberal jurists try to get around
these difficulties by adopting a more substantive
178
 BACK TO THE FUTURE
view of the rule of law, identifying it with the
26
upholding of rights by an independent judiciary.
As I noted, this approach proves problematic. A
political constitutionalism takes a different tack.
Justice becomes identified with the process of poli-
tics. Political mechanisms not only ensure all are
subject to the laws and that no one can be judge in
their own case ± the traditional tasks of the separa-
tion of powers ± but also that the laws connect with
the understandings and activities of those to whom
they are to apply ± the side benefit of dispersing
power so that more people have a say in its enact-
ment. Audi alteram partem forms the watchword of
legal fairness, not the formal or substantive properties
27
liberals associate with the law.
`Hearing the other side' within a pluralist polity
implies respecting that people can be reasonably led
to incommensurable and incompatible understand-
ings of values and interests, and seeing the need to
engage with them in terms they can accept. This
criterion places constraints on both the procedures
28
and the outcomes of the political process. It obliges
people to drop purely self-referential or self-inter-
ested reasoning and to look for considerations others
can find compelling, thereby ruling out arguments
that fail to treat all as of equal moral worth. Political
actors must strive for common ground through mu-
tually acceptable modifications. Because the clashes
of principle and preferences associated with plural-
ism preclude substantive consensual agreement, po-
litical compromise takes the place of a pre-political
29
consensus.
The type of compromise, and the style of politics
needed to achieve it, depends on the issue and the
character of the groups involved. Compromise is
most familiar to us in `more or less' contexts where
we can `split the difference' so as to converge on an
agreed decision through incremental mutual conces-
30
sions. Haggling over the price of a commodity such
as a house or wage bargaining are typical examples.
Such conflicts, however, are usually along a single
dimension and involve a common denominator. As
such, this procedure is unsuited to the conflicts
associated with pluralism since it assumes commen-
suration. Compromises in `either-or' situations in-
volving plural and competing values and interests
prove more exacting but not impossible. They have
179
to take the form of constructing a distinctive position
through argument that seeks to accommodate at
some level or another the various claims, values
and interests at stake.
How such compromises are to be achieved will
vary according to the nature of the case. Thus, a
compromise might take the form of a composite
agreement in which each party gets some but not
all of what it wants. This strategy usually proves
easier when more than one policy is involved.
Because groups often have rather different priorities
with regard to a range of policies, it frequently proves
possible for them to concede on issues they do not
feel strongly about but others do in order to obtain
similar concessions from them or other allies on the
matters they regard as most important. Such pro-
cesses are crucial to the log-rolling and coalition-
building practised by political parties. This practice
does not entail that all views are reducible to inter-
ests, as is sometimes assumed. In fact, these sorts of
deals are more like bartering than normal bargaining.
Indeed, they generally occur precisely because people
hold a plurality of moral positions that lead them to
evaluate goods and even what constitutes a matter of
morality differently ± much as beads can be ex-
changed for gold when one group's worthless shiny
substance is another's valuable commodity. The
exchange entails incommensurability, since it might
well not have taken place had it been conceived in
standard `more or less' terms, with all goods trans-
lated into a single common currency. It also proves
more responsive to intensity of feeling than a simple
bargaining model. However, it will only work if those
involved can avoid judging particular issues on their
merits.
When a compromise has to be reached on a single
issue, such trade-offs cannot be achieved. Moreover,
a mutual adaptation of views may be highly unsa-
tisfactory, producing a result that nobody wants
rather than something for everyone ± precisely the
situation critics who accuse compromise of incoher-
ence most abhor. One way out of this dilemma,
recently proposed by Robert Goodin, is an agree-
31
ment on a second best. A notion adapted from
economics, the basic idea is that modifications to
one's preferred option may be less desirable than
obtaining one's next best or an even lower ranked
 RICHARD BELLAMY
choice. A cheap sports car that pretends to be an
expensive one may have less appeal than an unpre-
tentious family estate, for example. Indeed, indivi-
duals or groups with conflicting first preferences may
have shared second preferences. A compromise on
an agreed second best may prove a more coherent
and acceptable decision, therefore, than an attempt
to combine first preferences in ways that transform
them out of all recognition or involve inconsisten-
cies of various kinds.
When a group makes a claim that involves special
treatment or imposing burdens on others, they may
have to compromise more unilaterally by appealing
either to some norm of equity that the others also
share, or to ideals or interests that their interlocutors
can identify with through appeals to precedent, use
32
of analogy or a fortiori arguments. Thus, a religious
group seeking public recognition for its beliefs, say in
state support for its schools, will be unlikely to get
very far if it desires to set up a theocratic state. But
they may be able to invoke a norm of equal treat-
ment by pointing to the existence of other publicly
funded religious schools. Or, where state education
has hitherto been secular, the claim might involve
drawing a parallel with state support for some hu-
manist analogue to religious belief, such as the arts,
in which the cultural aspect of religion is empha-
sised.
Nonetheless, sometimes no compromise on sub-
stance proves possible. This dilemma often arises for
purely practical reasons to do with the nature of the
decision rather than because of any lack of good will
among those making it ± such as a choice between
two somewhat different but equally good candidates
for a post, for each of whom a sound case might be
made. Only one can be chosen, and a compromise
candidate does not always make sense. In such
circumstances the compromise might have to be
on accepting the outcome of a fair procedure, such
as tossing a coin. Within purely procedural versions
of democracy, the majoritarian principle can be seen
33
as a compromise of this sort.
Constitutional design has the task of shaping the
political system so as to foster the form of democratic
debate appropriate to the requisite kind of fair
compromise. These considerations guide where de-
cisions are to be made, how groups should be re-
180
presented and the degree of autonomy particular
bodies or sections of the community may claim.
They are integral to a political constitutionalism,
with its intimate linking of justice, the rule of law
and the democratic dispersal and division of power.
In the ancient ideal of mixed government, the
favoured mechanism was to assign particular govern-
mental functions to different social classes. In con-
temporary societies, the answer lies in multiplying
the forms of representation and the sites of decision-
making power.
As I noted, the key lies in ensuring all sides get a
hearing so that decisions can be justified to and
accepted by all citizens. A notion linked to an ideal
of equality of concern and respect, the aim is not
only to render inputs more equitable, but also out-
34
comes. On the whole, this goal is more demanding
than simply guaranteeing that everyone has their say
or gets a piece of the action, important though these
may be in attaining it. When the desired end is
simply to `split the difference', then the proportion-
ate weighting of interests or preferences may be all
that is required. But the character of compromise is
different in matters of principle, and equalising
representation in this way may be either too much
or not enough. Here the object will be to ensure
equal consideration of the content and intrinsic
importance of different values for particular groups
of people, so that they seek solutions that are ac-
ceptable to a variety of different points of view.
Instead of bargaining, participants in this sort of
dispute must negotiate and argue. In the case of
bargained compromises, preferences can be taken
as exogenous to the system and democracy seen in
largely instrumental terms. A negotiated compromise
involves a more deliberative model of democracy
that leads to preferences being shaped and ranked
endogenously through the democratic process itself
as otherwise inaccessible information regarding the
range and intensity of the moral and material claims
involved comes to light. Achieving this result re-
quires that groups reach a sufficient threshold to
have a voice that people take seriously. With very
small groups, that may involve more than propor-
tionate voting power; with others somewhat less will
35
suffice.
Within a more complex and differentiated social
 BACK TO THE FUTURE
context centralised and hierarchical ways of distri-
buting power will be inadequate. Territorially based
representation has to be supplemented by functional
and cultural forms within particular sectors. Social
and cultural interests are often territorially dispersed
or located below any specific territorial unit. Em-
powering certain groups may require their represen-
tation within a specific location, or across a given
sector, or in the case of vertical cleavages according
to segment. Workplace democracy and parent gov-
ernors at schools are examples of the first, corporatist
representation of unions, employer organisations and
professional associations of the second, consocia-
tional representation for given ethnic, linguistic,
36
religious and cultural groups of the third. Such
mechanisms allow minority opinions to have both a
degree of autonomy within their own sphere com-
bined with a say in collective decision-making. On
the one hand, all groups (those asking for special
consideration included) are obliged to consult the
broader interests and concerns of society as a whole.
On the other, these same mechanisms operate as
checks and balances on the purely self-interested or
partial exercise of power.
Democracy plays a central role in this system,
protecting against arbitrary rule and enabling edu-
37
cative engagement with others. Decisions may be
contested and the rationale behind them tried.
Interests are not simply advanced and aggregated,
as in liberal accounts of the democratic process. They
get related and subjected to the criticism of reasons,
38
transforming politics into a forum of principle.
Positions can shift and common views emerge. De-
mocracy also operates within civil society as well as
the state, acknowledging the power exercised by
private bodies and the need to publicise them ± a
point liberal democrats tend to overlook by making
too sharp a public/private distinction. Power is not
simply devolved down in a hierarchical manner to
lesser levels of the state, as in a standard federal
system. It is dispersed among semi-autonomous yet
publicised private bodies. In this way, politics shapes
rather than being simply shaped by social demands.
Paradoxically from a liberal perspective, a demo-
cratic liberalism contends that the rule of law de-
pends upon the rule of men. The only guard against
arbitrary and dominating rule is if the law-making
181
power remains with the people or their representa-
39
tives. That position remains consistent with the
acknowledgement of the standard liberal set of in-
dividual rights to freedom of expression, association,
bodily integrity and the like. Indeed, it compels
40
recognition of them. Yet it also allows the demo-
cratic balancing of their relative weight in relation to
both each other and to additional values and inter-
ests according to the issue and the people involved.
Diversity in the ethical reasoning of agents and of the
moral relevance of given factors is naturally to be
expected within pluralist and complex societies.
Dispersing power helps both the appropriate mix
of voices to be heard and the peculiar circumstances
of particular contexts to be taken into account. Not
only can general rules be tailored to a wide variety of
objects and concerns, and their implementation and
monitoring enacted to meet the special requirements
of a given situation and constituency, but also ± and
often more importantly ± specific norms can be
established to meet special circumstances and rele-
vant differences. In consequence, the need di-
minishes for a judicially monitored principled
constitution to frame democracy. Judicial review
can track whether reasoned debate occurs, but need
not substitute for an absence of such deliberation. As
in a trial, the judge may advise on points of existing
law and ensure debate is conducted in a procedurally
correct matter, but like a jury the demos or its
representatives will make the decision for themselves
and may even overturn established precedent if the
41
circumstances warrant it and make new law.
This politicised account of justice suits a concern
with domination and the fact of pluralism. Domina-
tion most commonly manifests itself through inhi-
biting or preventing groups or individuals from
having a voice in the decisions governing their lives.
When this occurs they are far more likely to find
themselves oppressed by social structures that stunt
or prevent their capacity to employ and develop
their capacities or lead lives expressive of their beliefs
42
and culture. For if political institutions constrain
certain groups' ability to influence both the issues
that get discussed and how those that do get decided,
they will lose status and standing, and become
marginalised and ultimately exploited. Typically,
oppression has manifested itself indirectly through
 RICHARD BELLAMY
the mechanisms of hegemony. At the extreme,
however, exclusion from politics may eventually lead
oppressed groups to resort to extra-political means to
put their case ± a move that often lends legitimacy,
albeit of a spurious nature, to the employment of
outright force against them. Such has been the fate of
indigenous peoples in colonial societies, of many
ghettoised immigrant communities, of workers de-
nied the vote either in general elections or in the
workplace, to name but a few well-known examples
of this phenomenon.
As these cases indicate, such domination involves
not just an oppression of human beings but of a
diversity of ways of life, values, interests and alle-
giances. In other words, it is linked to the denial of
pluralism. Since these diverse attitudes and concerns
are often related in their turn to different concep-
tions of the right and the good, the injustice their
suppression entails is not fully captured by reference
to a particular conception of justice that may ±
perhaps unwittingly ± itself embody aspects of the
dominant position. Justice can only be done, there-
fore, by being seen to be done within a non-dom-
inating political structure that allows the various
conceptions to confront each other so that their
conflicts can be resolved through free negotiation.
Two Criticisms
Two criticisms are often ranged against schemes of
the sort proposed. Modern complex societies are said
to render a more participatory and deliberative pol-
itics together with any notion of agreement on
common policies not only impractical but also un-
desirable. A political constitutionalism is said to be
suited only to the small-scale and homogeneous
communities that originally gave birth to the idea.
Such criticisms are historically and substantively
mistaken. They gain credibility from certain com-
munitarian views of republicanism. These latter tend
to stress the supposed ethical and social unity of the
43
ancient polity. However, the key issue for repub-
lican theorists from Aristotle to Machiavelli was
class conflict and the difficulties of balancing the
44
various interests of different social groups. Their
view of politics was consequently less idealised and
more realist than many communitarians appreciate,
182
their aim not so much a unified as a mixed polity. It
was the insistence on the need for politics to con-
front the social reality of class and ideological strug-
gles that attracted the writers of the Federalist Papers,
who adapted republicanism to the needs of a modern
large-scale commercial society. As Machiavelli's
modern heir Max Weber appreciated, this feature
also makes it suited to tackling the problems of
45
pluralism. When republicanism is so conceived,
the two standard criticisms of it prove misdirected. I
shall take each in turn.
Critics of republicanism's practicality assume that
the complexity of large-scale societies can only be
handled by the market or a centralised bureaucratic
state or ± more usually ± some combination of the
two, with both social democrats and moderate lib-
ertarians falling into this last camp. However, de-
volved and more deliberative decision-making, far
from being obsolete, has many advantages over these
alternatives. Pluralism and complexity usually go
together in the sense that what makes a problem
complex is not simply its scale and the sheer number
of factors involved, but the fact that it can be
conceived and evaluated in numerous different ways.
As such, complex problems resist decomposition into
their component elements so that priorities can be
hierarchically ordered. Different actors and observers
may not agree on what these are let alone on what, or
whether, they can be ranked on any given scale.
It is precisely this kind of complexity that markets
and bureaucracies often handle badly. Take the
example of environmental issues, which typically
46
display these characteristics. When discussing
how best to handle acid rain, for instance, experts
dispute both its causes and consequences. It has been
traced to a number of natural and humanly produced
emissions and associated with various sorts of damage
to a wide range of objects and entities. Yet the extent
to which particular damage can be attributed to any
given cause is far from clear. Additionally, the hu-
man actors involved as both potential contributors
to and victims of pollution are similarly varied ± from
private motorists and consumers, to a whole host of
industrial and agricultural producers of one sort or
another. Their concerns will be similarly diverse, and
it is highly likely that each may be to some degree
both polluter and polluted. Since they will almost
 BACK TO THE FUTURE
certainly hold differing beliefs and values, how even
members of a given group appraise their interests is
going to vary greatly.
We have here two basic aspects of political dis-
agreement: namely cognitive indeterminacy and
47
conceptual essential contestability. Bureaucratic
management will be highly inefficient and reductive
in such instances. For there will be no clearly
demarcated ends or interests to be served, and the
relationships between the multifarious actors in-
volved cannot be tracked. As Hayek noted, admin-
istrators would have to be omniscient, omnipotent
and unwaveringly angelic to carry out such a task.
Markets sometimes fare better in offering a system of
dispersed knowledge, but not when public goods or
bads are concerned for the well known reason that
the externalities involved are non-excludable. The
price mechanism and the view that all values are
mere subjective preferences are also insensitive to
conviction-based or cultural difference. Moreover,
market exchanges are rarely solely among individual
consumers and entrepreneurs, but between private
institutions that are themselves bureaucratically or-
ganised and suffer from the self-same managerial
defects libertarians level at the state.
By contrast, a political constitutionalism attempts
to confront this complexity head on by bringing
together the actors concerned in an effort at mutual
understanding and accommodation. Devolved and
deliberative decision-making enables appropriate
ends and means to be fixed on by fostering coopera-
tion and coordination by multiple actors across a
host of domains. Both the norms to be applied and
their enforcement arise out of discussion, and could
not be fixed in advance without introducing biases
and oversimplifications of the problem with poten-
tially disastrous knock-on effects. Dryzek cites the
resolution of a dispute over the construction of a dam
and water supply system near Denver, Colorado as a
48
successful instance of this approach. A plethora of
federal and local agencies and groups were involved
and produced an appropriate fair compromise in
which the dam was built but substantial mitigating
measures were instituted. The success of this `Foot-
hills' decision led to the institution of a Metropolitan
Water Roundtable to mediate future cases.
It is of paramount importance that those involved
183
in such decision-making negotiate a collective policy
rather than engage in pluralist interest-group bar-
gaining if they are to arrive at a full perspective on
the dimensions of the problem and prevent stalemate
or distorted solutions, such as prisoners' dilemmas.
The way institutions operate will be crucial to
achieving this, as we have seen. However, this
feature gives rise to the second set of criticisms of
republicanism. Some critics worry that this sort of
institutional design amounts to social engineering
and is driven by a misguided pursuit of an unobtain-
able agreement on an elusive common good that can
only be attained through manipulation and covert
coercion. The deliberative ideal is charged with
assuming unity to be a prior condition of reasonable
49
discussion, and with having consensus as its goal.
With regard to the first part of the charge, certain
theorists have argued that a deliberative democracy
assumes a demos bound by a common fate and shared
50
understandings. They note how constitutional set-
tlements motivated by this ideal ± notably the Uni-
ted States Constitution ± were simultaneously
exercises in nation-building. Critics, however, point
out the often exclusionary consequences of these
projects, and in particular the marginalisation of
those deemed not capable of joining in the conver-
sation ± such as women workers, and members of
non-western cultures. If a people are to subject
themselves to often onerous burdens to generate
collective benefits for others, then they will have
to feel bound to them in some significant way. How
far such bonds must rely on a common culture as
opposed to mere functional interdependence and
territorial contiguity is difficult to say. States display
a bafflingly wide variety of mixes of these and other
elements. Historical contingency probably plays the
major factor. While starting from the status quo is
inevitable, a political constitutionalism is not tied to
any particular definition of the demos or the polity.
On the contrary, both are explicitly seen as political
artefacts fashioned by the people themselves. More-
over, it operates with a dynamic of inclusiveness that
places the burden of proof on those wanting to
exclude others or, via secession, themselves. They
must show such choices to be consistent with a norm
of equal worth and mutual acceptance, and that they
do not cause even more damage to others than
 RICHARD BELLAMY
benefits to themselves. But that leaves ample scope
for the renegotiation of the terms by which citizen-
ship is defined and political structures operate. In-
deed, once it is understood that the reconstitution of
a polity is an ongoing process, then there can be
considerable openness and flexibility in how the
constituent groups might relate to each other.
With regard to the second part of the charge, I
argued above that consensus need not and often
could not be the goal of discussion, as certain
theorists of deliberative democracy claim. They look
for a `republic of reasons' in which the most compel-
51
ling arguments prevail. However, pluralism renders
such reasoning problematic, since more than one
rationally compelling argument may be in play. That
makes compromise necessary. Achieving a fair com-
promise also entails a change in the character of
politics, but of a slightly different nature to one
oriented towards consensus. It requires a move from
a purely individualistic and instrumental politics to a
more interactive and problem-solving model. Rather
than viewing other people's interests and values as
mere constraints on getting one's own way to which
minimal concessions should be made, this approach
leads to the search for solutions that attempt to
integrate the various concerns of the parties in-
volved. This possibility need not rely on transcen-
dent criteria, however, merely a reciprocal
understanding of the frameworks of other actors,
the ability to engage with them and to seek agree-
ment on what is desirable even if this is based on
differing views of why it might be.
At its heart, this conception of politics has an
attachment to civic liberty, the guiding principles of
which are non-domination, mutual acceptance and
accommodation. What these conditions all point to
is a vision of society in which all enjoy equal status.
This implies universality and impartiality, in the
sense that all persons must be treated as of equal
moral worth and claims based on self-serving bias or
prejudice should be ruled out. But that does not
mean that all citizens must be regarded as having
identical needs or the same values ± quite the
reverse. This supposition appears to arise from the
belief that only general rules can meet these criteria,
and that these will be blind to people's particular
52
concerns or convictions. A claim for respect for
184
one's particular practices, however, is perfectly con-
sistent with a universal principle of equal moral
worth, and particular rules may be applied in as
consistent and unbiased a manner as general ones.
Universality and impartiality in these senses protect
particularity by insisting that any case for special
consideration must avoid subjecting others to dom-
ination. Arguments, such as those of sexists or racists
that assert that certain people's views or interests
count for less than those of others will be treated as
prima facie unacceptable, for example. A claim by a
religious minority for public funding of its own
schools will probably not be, since it asks for equal
recognition rather than denying it.
A political constitutionalism also allows for dif-
ferent styles of argumentation and suggests that as
much attention should be paid to the different
cultures of negotiation as to differences in the sub-
53
stance of what is negotiated. This need has long
been recognised by students of diplomacy looking at
54
international negotiations across cultures. Infrana-
tional pluralism means that domestic politics must
also take such factors into account. Here too a
problem-solving approach seems the most appropri-
ate, with symbolism and honour counting for as
much as argument and strategic gains.
Nevertheless, it is doubtful that any process of
negotiation would be possible unless people saw the
political norms framing the discussion as generating
some form of common good. Civic freedom operates
in this context as what Joseph Raz has called an
55
`inherent public good'. That is a good the benefits
of which are under the sole control of each potential
beneficiary and which by their nature could not be
voluntarily controlled and distributed by any single
agency. For one cannot create such an environment
except through the active collaboration with others,
nor control the beneficial externalities that it gen-
erates so as to channel them only to certain others,
though one could cut oneself off from them through
one's own anti-social and intolerant actions. Put
another way, the condition of living as equals has
to be desired in and for itself ± as an intrinsic aspect
of a certain kind of society ± rather than instrumen-
tally, since that would allow selective domination to
acquire personal advantage. To the extent pluralist
conflicts are rightly characterised as struggles for
 BACK TO THE FUTURE
recognition, then this goal should appeal to the
parties concerned.
Though this argument goes beyond the purely
formal proceduralism advocated by thinkers like
Habermas, who offer an entirely circular case for
constitutional norms via the self-validating character
56
of democratic discourse, it does not imply agree-
ment on substantive ends. Indeed at times it may be
possible to do no more than agree to disagree and
accept the authority of the democratic procedure
itself. In these cases the majority principle acts as a
means for resolving conflict in an authoritative
manner when a compromise on substance cannot
be achieved. Authority here rests on neither claims
to superior reason nor coercion but the simple
acceptance of the procedure as authoritative (in
the sense of being `in' authority) for the disputing
parties. Here the authority of law rests on the
legitimacy of the political system which generates
it. Parties acknowledge that in some cases there may
not be any `correct' or `most just' way of resolving a
clash between incommensurable plural values, but
that the ways of ending the dispute are acceptable.
The procedural fairness of the process of justice can
be more important than the favourability of, or
57
consensus about, the outcome.
A more political constitutionalism does not turn
on the existence of a homogeneous community,
therefore, as some communitarian theorists maintain
and certain critics complain. The political system
can operate as a public good for a plurality of social
groups, without assuming they share other values
(indeed, perhaps for the very reason that they do
Notes
1. Much of this chapter's argument emerged from an
ESRC research project on `Sovereignty and Citizen-
ship in a Mixed Polity' (R000222446) undertaken
with Dario Castiglione of Exeter University. I am
grateful for his comments, those of the editor and
the participants at the Swansea `Liberalism at the
Millennium II' conference, to Richard Dagger and
others at a seminar at All Souls College, Oxford, and
to Cecile Laborde. I have also drawn on my Liberalism
and Pluralism (Bellamy, (1999), especially chapter 5.
2. See Bellamy (1999), `Introduction', and Kekes (1993)
for a discussion of this problem.
3. For example, Locke (1991), Second Treatise, chapter
185
not). Secession or conscientious objection may still
have to be options for groups or individuals whose
values and convictions prove totally incompatible
with those of the majority. Their reasons could not
be, however, claims to superiority ± that their values
and interests are worth more than those of others ± as
was effectively argued by advocates of apartheid, for
example. Rather, the case must be that such drastic
measures are necessary to ensure equal worth and
that otherwise their cultures and concerns might be
totally eroded, as indigenous peoples seeking protec-
tion against a dominant post-colonial community
have contended. Even here alternative solutions,
such as greater autonomy or special rights, might
be available to keep them within the polity.
Conclusion
I have argued that to meet the challenges of the new
millennium, liberalism needs to take a step back to
the future and draw on the resources of the repub-
lican tradition. The resulting democratic liberalism
employs the notion of a political constitution based
around the dispersal of power and the mixing and
balancing of social classes. It incorporates the liberal
concern with freedom and justice into the demo-
crat's desire to ensure that citizens have an equal say
in influencing and holding to account the rules and
rulers governing them. As such, it offers a means of
both recognising and reconciling differences through
negotiation of fair compromises that embody mutual
acceptance and accommodation.
VI, para. 57; the French Declaration of the Rights of
Man and the Citizen of 1789, especially Articles 1, 2,
4, 6, 14 and 16; I. Kant, `On the Common Saying:
``This May be True in Theory, But It Does Not Apply
in Practice'', in Kant (1991), p. 73; J.S. Mill, `On
Liberty', in Mill (1972), p. 75; and Rawls (1972), p.
60.
4. On this problem, see O'Neill (1979) and Gray (1989),
especially chapter 9.
5. The locus classicus of this view is of course Adam
Smith: Smith (1976), p. 456. But the thesis is far from
being confined to economic liberals. See, for example,
Mill (1972), pp. 73±4; Raz (1986), pp. 369±70, 394.
 RICHARD BELLAMY
6. I explore the breakdown of this social thesis and the
related philosophical thesis in late nineteenth- and
early twentieth-century thought in Bellamy (1992).
7. In employing this formula, I am referring to the neo-
Roman version of republicanism identified by Skinner
(1998) and Pettit (1997), rather than the more loosely
defined republicanism espoused by communitarians
such as Michael Sandel, for example in Sandel (1996).
8. See Isaac (1988).
9. My main dissent from Dagger (1997) arises from his
underplaying these differences.
10. See the references cited in note 3.
11. Tully (1995), chapter 3.
12. For example, see Locke (1991), Second Treatise,
chapters II and VIII; Mill (1972), pp. 188±98; and
R. Dworkin's `Liberalism', in Dworkin (1985), pp.
181±204.
13. Once again the examples cited in note 3 provide
exemplary instances of this mode of thinking.
14. For an analysis of historical and contemporary models
of liberal democracy, see Held (1996), chapters 3, 5, 6
and 7.
15. For examples from the left and right of the liberal
spectrum, see Dworkin (1985), p. 197 and Hayek
(1976), pp. 62±3.
16. These processes are naturally exacerbated by globali-
sation. Many liberal theorists believe this challenge
can be met by extending liberal democracy in a
cosmopolitan direction, for example Held (1995).
However, the forces supposed to underpin this exten-
sion of liberalism have in reality given rise to the very
kinds of diversity that make liberal democracy hard to
sustain. For the same processes that drive globalisation
have augmented functional differentiation in the
economy and society and fostered multiculturalism.
I have criticised the cosmopolitan liberal thesis, and
provided an indication of how democratic liberalism
can be extended to transnational political commu-
nities in Bellamy and Castiglione (1998). Despite
global pressures, however, the state remains the pri-
mary locus of political authority and this chapter
addresses that context.
17. The analysis that follows draws on Luhmann (1981)
and Zolo (1992).
18. For example Dworkin (1995), pp. 2±11 and Rawls's
identification of public reason with judicial reasoning
in Rawls (1996), Lecture VI.
19. Rawls (1996) offers the most sophisticated example of
this approach. See Bellamy (1999), chapter 2 for a
detailed critique.
20. For a brief history, see my `The Political Form of the
Constitution: The Separation of Powers, Rights and
Representative Democracy', in Bellamy and Casti-
glione (eds) (1996), pp. 436±56.
21. Skinner (1998), Pettit (1997).
22. Hobhouse (1964), p. 71.
186
23. See Sunstein (1993), pp. 3±4.
24. For example, Ackerman (1991).
25. See Hayek (1960), pp. 153±4.
26. For example, Dworkin (1985), p. 2.
27. Pettit (1997), p. 189. See too Hampshire (1991), pp.
20±1.
28. See Gutmann and Thompson (1996), p. 57 and J.
Cohen, `Procedure and Substance in Deliberative
Democracy', in Benhabib (ed.) (1996), pp. 95±119,
at pp. 100±1.
29. What follows draws on R. Bellamy and M. Hollis,
`Consensus, Neutrality and Compromise', in Bellamy
and Hollis (eds) (1999), chapter 5, and Bellamy
(1999), chapter 4.
30. I owe the distinction between `either/or' and `more or
less' conflicts to Hirschman (1994), pp. 203±18.
31. Goodin (1995).
32. See Manin (1987), p. 353.
33. Singer (1974), p. 32 sees it as `a paradigm of a fair
compromise'.
34. Beitz (1989), p. 155.
35. See Kymlicka (1995), chapter 7.
36. For examples of each of these mechanisms see, respec-
tively, Hirst (1994), Dahl (1985) and Lijphart (1968).
I review the merits and weaknesses of these forms of
pluralist politics in Bellamy (2000).
37. Pettit (1997), p. 30 and Skinner (1998), p. 74 n. 38
stress the first benefit but regard the second as a civic
humanist rather than a neo-Roman concern, which
smacks dangerously of `positive' liberty. Putting history
to one side, substantively I do not believe a weak
`positive' appreciation of the virtues of participation
can be totally excised from republicanism.
38. See Sunstein (1991), pp. 3±34 for discussion of these
points.
39. See Skinner (1998), pp. 74±6 for the history of this
insight.
40. See J. Cohen, in Benhabib (ed.) (1996), pp. 102±5.
41. As Jeremy Waldron has recently insisted, we are sadly
lacking a model of jurisprudence which takes legislation
as seriously as judicial reasoning. Republicanism sup-
ports his plea for such an account. See Waldron (1999).
42. I'm here following Young's distinction between dom-
ination and oppression, and the relation between the
two; see Young (1990), pp. 37±8 and chapter 2.
43. For example Taylor (1995), pp. 141, 191±2, 302 n. 15.
44. See Bellamy, `The Political Form of the Constitution',
in Bellamy and Castiglione (eds) (1996).
45. Bellamy (1992), chapter 4.
46. I owe this example and much of the argument that
follows to Dryzek (1990), chapter 3. I have also drawn
on Cohen and Rogers (1995).
47. See Mason (1992), chapters 1 and 2.
48. Dryzek (1990), pp. 73±4.
49. See, for example Young (1990), chapter 4.
50. For example, Miller (1995), pp. 150±2.
 BACK TO THE FUTURE

51. Sunstein (1993), chapter 1. 54. For example, Cohen (1991).

52. See, for example, Young (1990), chapter 4. 55. Raz (1986), pp. 198±9.
53. Young, `Communication and the Other: Beyond De- 56. Habermas (1979), p. 90.
liberative Democracy', in Benhabib (ed.) (1996), pp. 57. See Lind and Tyler (1988) for empirical evidence to
120±35, at pp. 128±32. this effect.

187
 14
ACCOMMODATING
REPUBLICANISM
David Rasmussen
Presently, there is a debate raging in political theory
over two approaches to theorising justice: the deonto-
logical, which holds that justice can be achieved by the
establishment of specific principles and procedures as
opposed to the achievement of a substantive common
good as a goal or `end-state', and the teleological, which
conceptualises the achievement of justice as end ± or
goal-oriented, justifying whatever principles and pro-
cedures may be necessary as means to the realisation of
that specific end. No doubt, both Ju
È rgen Habermas
and John Rawls would want to see themselves on the
deontological side of this debate. In their own ways,
they present their theories as proceeding from the
acceptance of certain assumptions about the need to
respect wide value-pluralism. These assumptions are
held to be incompatible with the teleological ap-
proach, whose specified end ± however characterised
± is in some way reasonable to contest and disavow in
the light of them. However, each accuses the other of a
kind of latent Aristotelianism, which is one way of
1
characterising `teleology'. The source of that accusa-
tion is clearly to be found in both thinkers' attempts to
accommodate the influential critique of liberalism by
republicanism. They both want to acknowledge the
connections between public political life and `private'
liberty upon which the critique insists, but without
sacrificing an essential respect for individual liberty
and diversity that liberalism aspires to typify. The
accusation by one implies that the other has somehow
failed in the attempted synthesis of liberalism and
republicanism.
188
In the following I want to turn to that portion of
the 1995 debate between Rawls and Habermas
which focuses on the Ancients and the Moderns
in order to clarify what is at stake in what I have
called `accommodating republicanism'. Such is the
significance of these two thinkers in modern political
theory that it is useful to set out the `liberal ±
republican' debate at least partly in the terms sug-
gested by their exchange. We will see how `law' turns
out to be at the centre of this controversy, whether
one wants to see it in the form of Rawls's constitu-
tionalism or Habermas's co-originality thesis. It will
become clear that the question of the distinctiveness
between the two may rest upon whether Rawlsian
constitutionalism provides a genuine alternative to
the co-originality thesis. This will be a crucial test of
Rawlsian liberalism's conceptual strength, if we grant
that accommodating the republican insight is a
necessary move for it to make. We will then be in
a position to identify an important issue that should
be placed on the liberal ± republican research agen-
da.
Rawls and Habermas:
`Ancient versus Modern'?
The battle between Rawls's `interpretative construc-
tivism' and Habermas's `argumentative constructi-
vism' comes to the fore over Benjamin Constant's
distinction between the liberties of the moderns and
the liberties of the ancients, and the claim that these
 ACCOMMODATING REPUBLICANISM
should be nourished by the same root. Constant
claims that `modern' liberty ± the liberties of private
individuals ± is best directed towards self-develop-
ment which, in his view, is most effectively achieved
through free, collective political participation. (This
is the exercise of `ancient' liberty, the ideal which
particularly attracts contemporary republicans.)
Hence Constant calls for an efficacious combination
2
of the two types of liberty.
Habermas's point connects ancient and modern
liberty much more tightly than what, in Constant's
approach, is still an instrumental linkage of the two.
He argues simply that what he calls `public' and
`private' autonomy (analogues for `ancient' and
`modern' respectively) are `co-original'; neither is
prior to the other in any plausible conceptual sense.
He thinks that Rawls finds himself defending the
quintessentially liberal position of prejudicing the
liberties of the moderns, the private rights of private
individuals, over those of the ancients, the political
rights of public individuals, or `citizens'. This is
because the Rawlsian `political' sphere, from which
are drawn the shared ideas used to justify liberal
politics (and which for Rawls, of course, prioritise
individual liberties in the lexically primary first
principle of justice), is already `given' in his theory.
It just appears, so to speak: a ready-made domain
demarcating the sphere in which constitutional de-
mocratic politics thereafter takes place and from
which reasons to provide public justifications for
3
such politics are drawn.
Now the republican critique says about liberalism
the very thing that Habermas is here saying about
Rawls, namely that liberalism functions to justify
private rights above all else and, indeed, tends just to
assume them, as being simply `ready-to-hand.'
Among other complaints, republicanism believes
that this leaves liberalism fatally inattentive to
how a society can establish and maintain the public
conditions necessary for the kind of liberty-promot-
ing politics it favours. Given that republicans do not
want to reject wholesale liberalism's general ideals of
just governance, we can conceptualise the heart of
the liberal ± republican debate as posing this ques-
tion: how does one retain the basic framework of
liberalism while at the same time accommodating
the republican critique of liberalism's prioritisation of
189
private rights? Given that, as we shall see, Haber-
mas's response involves a radical conceptual and
practical extension of democracy, we may go on
to ask how one might realise the substantive mor-
al/political goal of justice as fairness while assuring
the liberal goal of impartiality in a political process
that is comprehensively democratic.
Habermas's solution is based on a notion of re-
ciprocity derived from the procedure of argumenta-
tion. He stresses how language, with which people
reason and communicate with each other, is essen-
tially intersubjective. He urges that we must there-
fore think of `justified' norms as being those which
would emerge from free and equal discourse. The
Habermasian thesis is that justice as fairness can only
be achieved through an adequate notion of recipro-
city. In turn, such a notion can only be gained from
the procedure of argumentation in which the actors
in the political process mutually recognise the claims
and counter-claims of one another. Here, we en-
counter the centrality to Habermasian political the-
ory of public will-formation ± by which is meant
democratic procedure ± which can be reconstructed
as a public and intersubjective process derived from
the procedure of argumentation. The emerging thesis
postulates the assumption that the norms of private,
individual rights can only be derived from the pro-
cess of public will-formation; they are not prior to it.
It follows from this that private and public autonomy
are co-original, the consequence of which being that
rights are democratically grounded from the outset.
The point of this argument is the claim that for
Rawls rights are not democratically grounded. Ha-
bermas believes the upshot is as follows: if Rawls had
accepted his critique, which epistemically associates
the reasonable and the true with the consequence
that consensus and validity are linked from the
4
outset, then he wouldn't have had to use the
original position as a device of representation. This
is because he would have found that `the moral point
of view' is implicit in the socio-ontological constitu-
5
tion of the public practice of argumentation.'
Further, Rawls would not have had to make the
second step in his theory, that is the step to over-
lapping consensus, because the public practice of
argumentation requires the `complex relations of
mutual recognition that participants in rational dis-
 DAVID RASMUSSEN
course ``must'' accept (in the weak sense of trans-
6
cendental necessity).' In other words, Rawls should
have built consensus fully into the first stage of the
theory (the original position in which rationally
choosing agents choose under reasonable procedural
constraints) rather than adding it on in the second
stage (the step to overlapping consensus as the
achievement by justice as fairness ± the outcome
of original-position deliberation ± of the right kind of
stable public support). This would have enabled his
political liberalism to escape both the arbitrariness of
the second stage (for here `the philosopher can at
most attempt to anticipate in reflection the direction
of real discourses as they would probably unfold
7
under conditions of a pluralistic society' ) as well
as the implicit monological perspective of the first
stage (for Rawls `imposes a common perspective on
the parties in the original position through informa-
tional constraints and thereby neutralises the multi-
plicity of particular interpretive perspectives from
8
the outset' ). In the Habermasian alternative, with
its `speech-act' view based on the nature of perfor-
matives from which one derives a certain intersub-
jective necessity, intersubjectivity ± and its notion of
reciprocity to guarantee impartiality ± is written into
the theory from the outset. In this sense, practical
reason is actualised through mutual recognition: `we
could say that precisely those principles are valid
which meet with uncoerced intersubjective recogni-
9
tion under conditions of rational discourse.' `Con-
sensus' need not be an afterthought, following in the
wake of the construction of principles of justice.
From Habermas's point of view Rawls's distinction
between the private and the public, which underlies
his distinction between the political and the com-
prehensive, is not only unnecessary but leads to
undesired consequences, particularly when one fo-
cuses on liberties and rights. It runs against the
historical insight that the `boundary' demarcating
the relative spheres is always shifting, due to the
basic shifts in political will-formation (Habermas's
example is that of the welfare state's development,
which turned formerly private matters of welfare into
10
public political concerns). From this perspective,
Habermas thinks that Rawls would be better off
subsuming this distinction under that of `legal reg-
ulation' which would in turn be determined by the
190
democratic life of a political community. In other
words, given such a scenario legal regulation would
determine the distinction between the public and
private in a democratic way in the sense that those
under the law would be its own authors. The reigning
question would then be: `which rights must free and
equal persons mutually accord one another if they
wish to regulate their existence by the legitimate
11
means of positive and coercive law?' If one were to
put the question that way it would follow that
Rawls's characterisation of a private domain, which
contains the realm of the comprehensive as distin-
guished from a public realm that characterises `the
political', would be a lot less necessary than his
political liberalism would have us think. Habermas
would therefore shift the Rawlsian distinction be-
tween the private and the public, or the compre-
hensive and the political, to the domain of the
procedure of legal regulation which would accord
the right to equal `actionable subjective liberties' as
well as the public autonomy of its citizens.
In brief, Habermas holds that legal procedure
presupposes the subjective autonomy of its citizens,
what Rawls would call the domain of private right.
At the same time legal procedure in a democratic
society cannot be legitimate without the public
exercise of democratic law-making. So private and
public autonomy presuppose each other. Hence, one
would infer from the argument that Rawls should
have concentrated on the procedure of democratic
12
law-making. He could have done this, it seems,
without moving beyond the bounds of the procedur-
al towards teleology because Habermas's democratic
law-making entrusts even more responsibility in
formulating norms than Rawls does to that political
13
procedure.
In his `Reply', Rawls sets out a rigorously con-
structed rebuttal of Habermas's characterisation and
critique of his project. He contends that if one pays
careful attention to the actual practice of designing a
constitution in relationship to the process of govern-
ment it can be seen that private autonomy cannot be
regarded as pre-political at all. The public political
culture which grounds it is not just `given', for there is
a real, founding constitutional convention that se-
lects a constitution with its Bill of Rights and this
`restricts majority legislation in how it may burden
 ACCOMMODATING REPUBLICANISM
such basic liberties as liberty of conscience and
14
freedom of speech and thought.' To be sure, then,
rights `trump' popular sovereignty, but only as pop-
ular sovereignty is expressed in the legislature in-
asmuch as those very rights are derived from a
democratic process of constitutional foundation.
Hence, basic liberties are `incorporated into the
constitution and protected as constitutional rights
on the basis of citizens' deliberations and judgements over
15
time' (emphasis added). In other words, and this is
a distinctly American view, popular sovereignty is to
be distinguished from `parliamentary supremacy'.
Following Locke's dualist idea of a constitutional
democracy Rawls contrasts ordinary law-making by a
legislature and the `people's constituent power to
form, ratify and amend a constitution' in order to
clarify the distinction. A parallel distinction is then
drawn between `higher law of the people' and the
16
`ordinary law of legislative bodies'. It is important
that the example which illuminates Rawls's argu-
ment is historical, not abstractly conceptual. Follow-
17
ing Bruce Ackerman, he cites the three most
significant, decisive innovations in American con-
stitutional history: the founding and initial develop-
ment of the constitution during the period 1787±91,
the post-Civil War Reconstruction, and the New
Deal under President Franklin D. Roosevelt in the
mid±1930s. Rawls observes that in these periods it
was the electorate which `confirmed or motivated
the constitutional changes that were proposed and
18
finally accepted.'
Of course, Rawls's case, derived from historical
interpretation and not purely from philosophical
argumentation (hence the contrast between inter-
pretative and argumentative constructivism), is pre-
sented as reliant upon the development of the
constitutional history of the United States. As such,
he argues this reference proves that justice as fairness
is a political notion rather than one derived from
pre-political natural law. He insists that for him `the
liberties of the moderns do not impose the prior
19
restrictions on the people's constituent will' ± and
to that extent he, too, thinks he can accommodate
the republican critique.
But does Rawls's argument prove anything beyond
something which is peculiar to the constitutional
history of the United States? Or, to rephrase the issue
191
at stake, we may ask if the question of the co-relation
of the liberty of the ancients and the liberty of the
moderns, of private right and popular sovereignty,
can be resolved independently of a contingent con-
stitutional history. In Rawls's view it could be that
`Habermas may have no objection to justice as fair-
ness but may reject the constitution to which he
thinks it leads, and which I think may secure both
20
the ancient and modern liberties.' Fair enough.
Rawls is willing to concede that `he and I are not,
however, debating the justice of the United States'
constitution as it is; but whether justice as fairness
allows and is consistent with the popular sovereignty
21
he cherishes.'
However, Rawls is clearly sensitive to the inter-
pretation, and thus the critique, that Habermas gives
of the liberalism he articulates. Hence, in support of
his argument that individual rights are not pre-
political, and to some extent in support of his
own more interpretative view, he cites an argument
which Habermas uses against Charles Larmore. Lar-
more has argued that one right has to proceed and
constrain democratic will-formation: `no one should
be forcibly compelled to submit to norms whose
22
validity cannot be made evident to reason.' Ha-
bermas interprets this as a denial of co-originality,
meaning the establishment of a pre-political right
against the state which is perceived under the cate-
gory of violence. Against this view Habermas pre-
sents a two-stage model which begins with the
`horizontal sociation of citizens who, recognising
one another as equals, mutually accord rights to
one another. Only then does one advance to the
constitutional taming of power presupposed with the
23
medium of law.'
Now Rawls believes that his own construction can
be viewed in a similar way. Hence, he arrives at two
conclusions, namely that Habermas's position is not
24
that different from his own, and that Habermas
mistakenly unleashes a critique in his interpretation
25
of liberalism. Habermas thinks that liberalism
founders on the dilemma generated by the conflict
between these two propositions: `no moral law can
be externally imposed on a sovereign democratic
people' and `the sovereign people may not justly
26
. . . enact any law violating these rights.' But for
Rawls there is no true dilemma here: both proposi-
 DAVID RASMUSSEN
tions are correct and `simply express the risk for
political justice of all government, democratic or
otherwise; for there is no human institution ± poli-
tical or social, judicial or ecclesiastical ± that can
guarantee that legitimate (or just) laws are always
27
enacted and just rights always respected.' It is quite
conceivable that a single person could stand alone
and be right to say that law and government are
unjust. `No special doctrine of the co-originality and
equal weight of the two forms of autonomy is needed
to explain this fact,' Rawls writes. `It is hard to
believe that all major liberal and civic republican
writers did not understand this. It bears on the age-
old question of how best to unite power with law and
28
justice.'
At this point, we see Rawls making no attempt to
do anything other than defend his position against
the presumption of a liberal bias for private rights.
But clearly he doubts that Habermas's theoretical
construction, such as the co-originality thesis, will
guarantee the proper exercise of law and justice in
practice. Equally, he questions the emphasis on the
political in Habermas's apparent claim that `the
internal relation between the two forms of autonomy
depends upon ``the normative content of the mode of
29
exercising political autonomy.'' ' Rawls wants to
know how we might justify the idea that it is
incumbent specifically upon the political realm to
provide the basis for private autonomy, given that
activity in the institutions and practices of civil
society could also provide a sufficient basis for it.
He speculates that the origin of Habermas's prior-
itisation of political activity is found in civic human-
ism (by which it is reasonable to assume that he
means Aristotelianism). The crucially relevant fea-
ture here is that for civic humanism the greatest good
is conceived of as active participation in political life,
a paradigm of the teleological approach. For Rawls,
this notion is really derived from a comprehensive
doctrine. I take his critique of Habermas to be based
upon the idea that it is reasonable that not everybody
pursues the good defined specifically in terms of
political life and activity. Non-political public pro-
jects and private interests and pursuits are equally
reasonable. From the point of view of political
liberalism, then, it would be inappropriate to assert
one comprehensive view over others. Hence, for
192
Rawls, not everybody need pursue the good embo-
died in the politics of old as an intrinsic element of
the good life today ± but Habermas's theory seems
implicitly to curtail this possibility.
The Persistence
of The `Teleological Question'
So Rawls's reply to Habermas raises a very interesting
point to which, to my knowledge, Habermas has yet
to reply. I would contend that it can be used to
formulate an urgent issue for the liberal/republican
research agenda. His critique can be formulated as
follows: if one asserts the co-originality thesis in the
way that Habermas does, with its implicit assumption
about the primacy of the political in the ancient as
opposed to the modern sense, then one buys into a
teleology which can only be justified through a
30
comprehensive philosophical framework. If valid,
it suggests that Habermas may have gone too far in
his attempt to accommodate the claims of repub-
licanism; a crucial modern insight about the plur-
alistic nature of the good has gone unrespected.
Rawls claims to get by his interpretative constructi-
vism what he contends Habermas does not get by his
alternative argumentative constructivism. As I un-
derstand it, Rawls claims to be able to assert a merely
political and non-comprehensive notion of demo-
cratic procedure through his use of the claims of a
certain form of constitutional legal theory. It would
be equally interesting to know if Habermas believes
that Rawls, by relying on American constitutional-
ism, could successfully avoid a similar dilemma ±
namely, the assertion of the teleological primacy of
the public on grounds which depend too specifically
(`comprehensively') on the particularities of Amer-
31
ican historical experience.
Either way, the republican insight into the rela-
tionship between the public and the private may
seem to be one that must be coherently adopted by
any theory sharing the broad moral and political
purposes of Rawls and Habermas. But, as their ex-
change shows, it is as yet far from clear that it can be
accommodated without reliance upon the kind of
teleological assumptions from which many contem-
porary political theorists ± Rawls and Habermas
included ± emphatically wish to distance themselves.
 ACCOMMODATING REPUBLICANISM

Notes

1. The appropriateness of the term `Aristotelianism' may concept of law has been clarified in this way, it
be questionable. I presume that its spirit is what becomes clear that the normative substance of basic
Habermas invokes when he associates Rawls with a liberal rights is already contained in the indispensable
classical theory which legitimates its theory of justice medium for the legal institutionalisation of the public
by assuming the actuality of just institutions in chapter use of reason of sovereign citizens'; ibid., p. 130.
2 of Habermas (1996). He also thinks that the fact 13. Ibid., p. 131.
that `the issue of principles of justice can arise only in 14. Rawls (1996), pp. 404±5.
the guise of the question of the just distribution of the 15. Ibid., p. 405.
primary goods' is more consistent with `Aristotelian or 16. Ibid., pp. 405±6; cf. ibid., pp. 231±40.
utilitarian approaches than with a theory of rights, 17. Ackerman (1991), chapters 3±6.
such as [Rawls's] own, that proceeds from the concept 18. Rawls (1996), p. 406; cf. ibid., pp. 233±5.
of autonomy' (Habermas, 1995, p. 114). Further, I 19. Ibid., pp. 406±7.
assume that the term captures what Rawls means when 20. Ibid., p. 407.
he associates Habermas with civic humanism, which 21. Ibid.
he does at Rawls (1996), p. 420. 22. Cited in Habermas (1996), p. 456.
2. See B. Constant, `The Liberty of the Ancients Com- 23. Ibid.
pared with that of the Moderns', in Constant (1988), 24. Rawls (1996), p. 407.
pp. 308±28, at p. 327; cf. Habermas (1995), p. 129. 25. Ibid., p. 412.
3. Habermas (1995), esp. pp. 126±9. 26. Ibid., pp. 411, 416.
4. Ibid., pp. 119±26, especially p. 122. 27. Ibid., p. 416.
5. Ibid., p. 127. 28. Ibid.
6. Ibid. 29. Ibid., p. 419.
7. Ibid., p. 121. 30. Ibid., p. 420; cf. pp. 376±80.
8. Ibid., p. 117. 31. Habermas has recently had more to say about the
9. Ibid., p. 127. distinction between the reasonable and the true in
10. Ibid., p. 129; cf. Benhabib (1992), pp. 89±120. Rawls's work. However, at the time of writing there
11. Habermas (1995), p. 130. remains much more mileage in their exchange; see
12. Ibid., pp. 130±1. Habermas believes that `once the Habermas (1998b), pp. 75±101.

193
This page intentionally left blank
 Part Six
THE `AUTONOMOUS
INDIVIDUAL': FEMINIST
CRITIQUES AND LIBERAL REPLIES
This page intentionally left blank
 15
LIBERALISM, FEMINISM,
ENLIGHTENMENT
Kate Soper
I
Like many other intimate relationships, that of
feminism and liberalism has from the outset been
troubled by contention, but it has also proved to be
very resistant to any final severance. By common
consent, the modern feminist movement has its roots
in the philosophical liberalism of the Enlightenment,
but it has also since its origins constituted itself at
least in part through an immanent critique of the
partialities and hypocrisies of its humanist and ega-
litarian claims. Yet it has belonged within, even as it
has conducted a continuous polemic against, the
framework of liberal-Enlightenment thinking. This
is so in two senses: firstly, in the sense that although
they were highly critical of the patriarchal culture of
the liberal Enlightenment, most of those arguing for
female emancipation from the end of the eighteenth
century until relatively recently have viewed that
project as not only consistent with liberal principles
but as requisite to their full implementation. For
Wollstonecraft, it seemed only rational that a Lock-
ean-liberal conception of rights and citizenship
should extend to women on the same grounds and
in equal measure as it did to men, even though in
demanding that it should be, she was asking for an
extension of rights and franchise that was totally
unacceptable to the political culture of her time.
From the perspective of the form of liberalism de-
fended by John Stuart Mill and Harriet Taylor in the
nineteenth century, the continued subjection of
197
women appeared a barbaric anachronism. For them,
parity of treatment for the sexes was viewed as
essential to bringing political practice into confor-
mity with liberal philosophy ± and this view has been
shared by all those who have subsequently carried
forward the `equal opportunities' campaign for an
end to female subordination.
What is essentially at issue for all such critics is the
arbitrary and unjustifiable exclusion of women as
proper candidates for the freedoms and rights
claimed to be due to humanity at large: a line of
objection which has in recent times been focused on
the extent to which liberal humanist discourse has
functioned historically as a cover under which both
gender and racial difference can be safely ignored and
the particular interests of a white, male and bourgeois
elite be the better secured.
But feminism has also been viewed as rooted in
liberal-Enlightenment values in the further and re-
lated sense that in developing this immanent cri-
tique, it was doing no more than observing the self-
critical, constantly revisionist spirit of those values
1
themselves. To point this out, however, is clearly to
beg the question as to how self-critical this spirit can
become before it turns heretical. At what point does
feminist criticism of liberalism cease to be immanent
to it and become outright dissidence?
One possible answer might be at the point when it
casts doubt on the capacity of liberal-democratic
institutions and political process to deliver the `free-
dom and equality' for all individuals that liberalism
 KATE SOPER
abstractly professes to uphold and seek to advance. It
is always debatable, however, at what point exactly
this may be said to occur in any feminist critique of
liberalism. Virginia Sapiro, to take but one example,
has presented Wollstonecraft as challenging not just
the masculinity of the liberal conception of citizen-
ship but also ± and even more radically for her times
± its rigid distinction between `public' and private'
spheres. But Sapiro nonetheless seems to view such
demands as capable of accommodation, at least in
principle, within the existing framework of a liberal
politics. Thus she writes that Wollstonecraft's argu-
ment on democracy offered a `means of stretching
the liberal temperament to incorporate into political
thinking explicit concern for the quality of personal
relations and day-to-day conditions of the lives of
2
citizens.' On the other hand, one might be inclined
to argue in the converse sense, that if Wollstonecraft
can indeed be interpreted as launching an early
`feminist' assault on the private-public division, then
she is not so much `stretching' as directly contesting
3
liberal sentiment. In other words, what seems raised
here is how far liberalism, not only in Wollstone-
craft's time but also since then, is consistent with the
dissolving or transcending of the public-private di-
vide that is called for in this form of feminist critique,
or tolerant of its insistence on such a `politicising' of
the `personal'.
But whether one reads Wollstonecraft as pushing
the limits of liberal politics or as more radically
breaking through these, there is little doubt that
in the later argument of the socialist feminists, it is
liberal political institutions themselves that are un-
der attack. In arguing that without the overthrow of
capitalist exploitations and its class-based forms of
oppression, female emancipation will always be par-
tial, more formal than real and tend always to be
restricted to the more affluent and educationally
privileged, socialist feminism would seem clearly at
odds with liberal reformism. Even so, given that it
clearly retains a commitment to some of the key
emancipatory goals of the liberal Enlightenment (to
those of freedom, parity of treatment, self-realisation
and so on), one may still question whether socialist
feminism really constitutes a decisive break with the
immanent critique of liberalism's essential mode of
reasoning. After all, much of the dispute (if admit-
198
tedly not all of it) between liberal and socialist
feminists would seem to be over means rather than
ends: what is at issue is the capacity of liberal forms of
polity to deliver on the emancipatory agenda rather
4
than the agenda itself. In this sense socialist-fem-
inist critique remains ultimately consistent with the
commitment to liberal goals and values even as it
rejects the liberal conception of the procedures
through which they might be realised.
If this is right, then one might claim that the point
of final rupture with the immanent feminist critique
of liberalism has come only more recently in the form
of the postmodernist rejection of the `foundational-
ist' metaphysics and universal humanist commit-
ments of liberalism. For we are talking here of an
attack on liberalism not for its political shortcomings
in realising its professed Enlightenment commit-
ments, but of an attack directed against liberalism's
Enlightenment pedigree itself. Nor is it any accident
that much of this critique has been formulated in
terms of liberalism's unfortunate inheritance of, and
continued reliance upon, a so-called `Enlightenment
subject'.
In effect, there are two aspects of this critique
corresponding to two related, but distinguishable,
constructions to be placed upon the notion of the
`Enlightenment subject'. In a first and more histor-
ical understanding, the idea refers us to the ideals of
human subjectivity first promoted in the late seven-
teenth century and more fully elaborated in the
historical period that we know of as `Enlighten-
5
ment.' Most prominent among these is the idea
of `autonomy': the liberated or Enlightened indivi-
dual is one who, in Kant's phrase, `dares to know',
who is no longer in thrall to received ideas and
externally imposed rules of comportment but thinks
and acts in accordance with self-ordained principles
± who prescribes the law unto the self. Self-mastery,
independence, freedom from superstition and un-
thinking subservience to religious dictates, the ex-
ercise of reason in pursuit of self-chosen ends: these
are the recommended qualities of the subject as first
promoted in the writings of the Enlighteners. The
subject in this conception is central to a normative
theory about what it is to be a fully realised, auton-
omous human person, and in an important sense it is
a historical category, referring us to one who comes to
 LIBERALISM, FEMINISM, ENLIGHTENMENT
be as a consequence of the new rationality and
scientific understanding illuminating the period of
the Enlightenment. One might note, in this con-
nection, the contrasts which scholars have drawn
between conceptions of reason in the seventeenth
and eighteenth centuries. Cassirer, for example,
argues that, for the later century, `it is no longer
the sum total of ``innate ideas'' given prior to all
experience which reveals the absolute essence of
things. Reason is now looked upon as rather an
6
acquisition than a heritage.'
In a second and more abstract conception, on the
other hand, the `Enlightenment subject' refers us to a
rather different theoretical entity, namely that ab-
stracted individual who underpins the cosmopolitan
and humanist aspirations of the liberal-Enlighten-
ment project. This is the subject stripped of the
accretion of culture, the bare-forked creature, the
human being conceived (in Wollstonecraft's phrase)
as `naked human nature': the subject who is invoked,
either implicitly or explicitly, as the ground of the
project to unify the various tribes and nations of
man. It invokes that `common humanity' through
which, or on the basis of which, it alone makes sense
to aspire to liberal forms of solidarity and the realisa-
tion of universal values within the human commu-
nity.
Even if these two conceptions of subjectivity are
linked in the sense that it was thought that everyone,
by virtue of possessing those attributes which are
definitive of human nature, could in principle realise
the ideal of autonomy, they are clearly distinguish-
able inflections, and corresponding to them, as sug-
gested, we find differing emphases in the feminist
polemic.
In their critical engagement with the `historical'
sense of Enlightenment subjectivity, feminist the-
ories have delivered a series of powerful arguments
against the idealisation of autonomy, objecting that
it makes a fetish of masculine attributes (or at any
rate of those associated with male performance in
patriarchal culture). They have relatedly claimed
that this devalues other ± `feminine' or culturally
coded `feminine' ± attributes which can contribute as
much if not more to civilised culture and are there-
fore just as worthy of advancement. Most damningly
of all, they have dismissed the very idea of the
199
`autonomous' individual as a delusion of the bour-
geois epoch itself. The objection here is not so much
that this conception of the self as `sovereign being'
validated forms of independence whose enjoyment
was systematically denied to women, but that it
promoted a misguided ideology of personal autono-
my through which all individuals were encouraged to
overlook their actual ties of dependency on others,
and men, in particular, were enabled to justify their
detachment from the sphere of domesticity and
reproduction. Individuals are not only culturally
embedded and embodied to an extent and depth
that liberal humanist ideals of selfhood utterly fail to
appreciate; they are, in virtue of being so, fragmented
selves: the unified subject of the liberal-Enlighten-
ment is a fiction as well as a masculine fetish.
Postmodernist criticism of the Enlightenment sub-
ject in the `abstract' sense, on the other hand, is
directed not at the discounting or downgrading of
specifically feminine values and attributes, but
against what is said to be a mistakenly universalist
and essentialist conception of humanity which lies at
the heart of the Enlightenment and liberal-humanist
projects. The problem is not (as a feminist such as
Wollstonecraft supposed) that women have been
irrationally and hypocritically excluded from the
claims that are made about the human subject,
but rather that this notion is itself a dangerous
fiction: to invoke the idea of a universal humanity,
so it is said, is to be guilty of a certain imperialism, to
lend oneself to an anti-democratic impulse to over-
look cultural and personal difference.
Going together with this feminist critique of the
Enlightenment conceptual legacy there has also been
a tendency for feminist historians of the eighteenth
century to portray the ferments of the French Re-
volution in a very negative light, and to deny they
had any positive role to play in the advance of female
emancipation. In the argument of Joan Landes, the
period is associated with the opening-up of the
exclusively `masculine' space of the `public' and
the dominion of a `reason' which is again viewed
7
as the exclusive attribute of the male. Women, we
are led to believe, had little or nothing to gain from
the process of liberalisation, and those of them who
were calling for women to be included within the
remit of the so-called `rights of man' tend to be
 KATE SOPER
dismissed as colluding in the maintenance of patri-
8
archal culture. Modernist feminists from de Beau-
voir onwards have been similarly charged by the
postmodernists with perpetuating an `Enlighten-
ment' ± and hence, so it is implied, inherently
9
`masculine' ± cultural perspective. As Pauline John-
son has pointed out, such feminists have presumed
that the critique of the claims of transcendent reason
has `established modern feminism on the path of
counter-Enlightenment' and that feminism now re-
quires a `fundamental break from the Enlightenment
cause of reason and truth, which is exposed as
nothing more than a distorted and disguised will-
10
to-power.' In some cases, in fact, the Enlighten-
ment and liberal humanism by association have been
treated as if they were more or less synonymous with
everything white, male, middle class and heterosex-
ual, and we have been invited to revile them accord-
ingly.
Now, for the most part, where feminists have
countered the postmodernist positions on the En-
lightenment subject, they have focused on the in-
consistencies of their anti-essentialist argument.
They have sought to reveal the incoherence of
denying any existential unity and autonomy at the
level of subjectivity while seeking to expose cultural
norms and `constructions' as having negative con-
sequences for female emancipation and self-fulfil-
11
ment. They have pointed out that it makes little
sense to advocate the politics of `identity' while
theoretically denying subjects any powers over their
self-formation. They have insisted on the incompat-
ibility of a democratic politics of a kind purportedly
being defended in postmodernist argument with
respect for an indefinite plurality of cultural values
and aspirations. They have, relatedly, taken issue
with the extremely reactionary forms of traditional-
ism and fundamentalism that they see as licensed by
the postmodernist refusal to recognise any essential-
12
ist or universalist position on human nature. Even
Derrida, Spivak and other figures associated with the
deconstructionist position have, in the light of this
ethical problem, made some gestural moves towards
acknowledging the necessity of a minimalist essenti-
13
alism. Although I shall have a word more to say on
these arguments in my concluding section, I believe
the case against the incoherence of postmodernist
200
relativism within feminism has been stated with
sufficient potency for it to be rather pointless for
me to add much further to it here. Those who have
not been so convinced to date are unlikely to have
their thinking on these issues altered by further
restatements of the realist argument.
But there is a somewhat different line of response,
which has received less attention, and which I
propose to develop in more detail here. In essence,
what I shall be arguing is that critics of liberalism
who see this as shamed or deeply flawed by virtue of
its Enlightenment origins and conceptual sources are
basing their argument on an extremely partial con-
ception of this theoretical and cultural legacy. In
other words, the postmodernist feminist critique of
liberalism is itself flawed to the extent that it is
reliant on a distorted and caricaturised version of
14
liberal-Enlightenment thinking on gender issues.
II
In this context any attempt to offer a corrective to
this version will necessarily have to be rather brief
and synoptic. But a first point to be made is that the
feminist dismissal of the liberal-Enlightenment is
often based on an overly homogeneous conception
of its thought and culture. As recent scholars of the
period have been insisting, the scope and quality of
Enlightenment thinking even within a single nation
± let alone across Europe (and into America) ± is a
good deal more varied, complex and uneven in its
development than is allowed for by this feminist
rhetoric. In recognition of this, Dorinda Outram
proposes in her recent study what she terms a
`capsule' conception of Enlightenment as containing
sets of debates, stresses and concerns rather than
15
representing a single, unified value system. The
Enlightenment, according to Outram, presented it-
self as a series of processes and problems to those, like
Kant and Mendelssohn, who attempted to define it.
It was not so much a `modern paganism' as a quest for
16
a `rational Christianity'. It was a period obsessed
with cultural difference, of continuous questioning
and relativisation of European `civilisation', and of
critique of colonialism. Most contemporary com-
mentators also emphasise the disparate and uneven
development of the `pockets' of Enlightenment con-
 LIBERALISM, FEMINISM, ENLIGHTENMENT
cern in the different countries of Europe and their
emergence in non-European locations during the
eighteenth century, most notably in
Where once the Enlightenment was viewed as con-
fined to the argument of a few French philosophes
towards the close of the century, the contemporary
assessment favours a very long and geographically
extensive Enlightenment, by no means local to
Europe, and stretching temporally from the mid-
seventeenth right up to our own times ± and even
beyond. As Habermas has famously pronounced, the
task of Enlightenment has still to be completed if we
are to be enabled to act as human beings.
It is also important that those relying on the anti-
Enlightenment rhetoric recognise that there is ex-
tensive debate today about the role that women
played in Enlightenment and the gains they derived
from it. Margaret Jacob and other recent historians of
Enlightenment have challenged accounts of its un-
remittingly patriarchal character, insisting on its
overall positive impact for the feminist cause, and
arguing that it represented a very significant advance
on the oppression of the Church-dominated and
Absolutist political orders which it supplanted.
To deny that, Jacob suggests, is implicitly to align
oneself with the position of the eighteenth-century
enemies of the revolution and with a long tradition
18
of right-wing rejection of its legacy. She acknowl-
edges that participation in the `public sphere' re-
quired one to be both literate and in possession of
adequate funds for club membership, theatre atten-
dance and so forth, and that this restricted it to the
19
better-off among women. But she denies that there
was any increasing `masculinisation' of the `public
sphere' in any of the main centres of Enlighten-
20
ment, and rejects the currently fashionable view of
the French Revolution as a defeat for women.
Rather than imagining the French movement
towards egalitarianism, leading to the contestations
of the 1790s, as the signal of women's first defeat in
modern democratic politics, the evidence, when seen
comparatively, suggests quite the opposite.
emergence of the public sphere put women's rights
on the western agenda just as it augured all the other
struggles about the meaning and truthfulness of
democratic politics with which we still live.
Jacob's Habermasian defence of the Enlighten-
ment's gains is open to the charge that it too readily
abstracts from gender-blind or patriarchalist dimen-
America. sions of Habermas's social theory. Indeed, my own
position would be that you can endorse a broadly
Habermasian position on Enlightenment while
being a good deal more critical of Habermas's general
neglect of gender issues along lines discussed by
Nancy Fraser and others. (Fraser has suggested that
Habermas is guilty of a certain gender-blindness in
failing to see how close a critique of the commodi-
fication of the cherished `symbolic' realm of child-
rearing and nursing can come to rationalising female
22
domesticity. ) All the same, I certainly think Jacob
is right to remind the postmodernist feminist histor-
ians of the reactionary quality of the politics they risk
endorsing in their dismissive position on the Enlight-
enment legacy. I think they should acknowledge,
too, how problematic their own position looks in
relation to the arguments of those seeking to advance
female emancipation in the age of Enlightenment
itself.
One might briefly illustrate this point by reference
to one of its better-known disputes ± that between
17
Wollstonecraft and Burke. From Burke's perspective,
Wollstonecraft is indeed an abstractor from cultural
difference bent on denuding society of those distinc-
tions which are of its essence and which endow it
with cultural richness. Thus, he professes, he cannot
relate to `human actions and concerns, in a simple
view of the object, as it stands stripped of every
relation, in all the nakedness and solitude of meta-
23
physical abstraction.' For Burke, it is what distin-
guished man from beast, culture from nature, that
humanity is differentiated, its culture functioning as
the sanctum of such difference:
All the pleasing illusions, which made power
gentle, and obedience liberal, which harmonised
the different shades of life, and which, by a bland
assimilation, incorporated into politics the senti-
ments which beautify and soften private society,
The are to be dissolved by this new conquering of light
and reason. All the decent drapery of life is to be
rudely torn off. All the super-added ideas, furn-
ished from the wardrobe of moral imagination,
21
which the heart owns, and the understanding
ratifies, as necessary to cover the defects of our
201
 KATE SOPER
naked shivering nature, and to raise it to dignity
in our own estimation, are to be exploded as
ridiculous, absurd, and antiquated fashion. On
this scheme of things, a king is but a man; a
queen is but a woman; a woman is but an animal;
24
and an animal not of the highest order.
For her part, Wollstonecraft was happy to plead
guilty to these charges in a polemic against the
`gorgeous drapery' in which Burke had `enwrapped
his tyrannic principles': a polemic through which she
seeks both to expose the principles themselves and to
denude them of their rhetorical garb. For Wollstone-
craft, Burke's eloquence masks a profound irration-
alism, a defence of bigotry in the name of reverence
for cultural tradition, whose logical implication
would be that slavery ought never to have been
abolished. The Burkean doctrine would have us
submit to this `inhuman custom' simpy because it
was the creation of `our ignorant forefathers', who,
`not understanding the native dignity of man, sanc-
tioned a traffic that outrages every suggestion of
25
reason and religion.'
No doubt some will object to this association of
the conservative Burke with contemporary defen-
dants of cultural relativism, given the difference in
their motivations: Burke's to preserve aristocratic
privilege, the latter's to check the normalising power
of western discourse. But the comparison certainly
serves to dramatise the tensions, and potentially
reactionary implications, of these positions which
we have been invited to view as a wholly positive
rupture with humanist thinking. Moreover, the fact
that certain parallels can be drawn here should give
pause for thought to all those who would sponta-
neously acclaim Wollstonecraft's contribution to
feminist and democratic theory while subscribing
to theoretical positions very much at odds with hers.
In this connection, one might note how paradox-
ical it is, in view of the current emphasis on `differ-
ence', that it tended to be the more progressive, and,
if one could call them so, `feminist' elements in the
eighteenth century who were most resistant to re-
cognising any major natural, as opposed to culturally
induced, divergences between the sexes. The con-
verse side of this paradox is that so far from obscuring
or abstracting from the specific conditions of women,
202
the Enlightenment, (to the distress of the more
feminist-minded) was in reality a period obsessed
with the topic of sexual difference: its origin, nature
and social implications. Indeed, it has been main-
tained that it is only in the period of Enlightenment
that the issue of feminine difference ± both corporeal
and mental ± begins in any serious sense to be
addressed, and hence only in the Enlightenment,
when women are recognised as possessed of a quite
separate and sui generis sexuality, that the sex-gender
distinction, with its acknowledgement of the differ-
ence between what is given by nature and what is
socially cultivated, is seen as having application. `No
longer would those who think about such matters',
writes Thomas Laquer of the eighteenth century,
`regard women as a lesser version of man along a
vertical axis of infinite gradations, but rather as an
altogether different creature along a horizontal axis
26
whose middle ground was largely empty.' Correla-
tively he argues that it was only then that sex was
invented, that `science fleshed out, in terms presen-
table to the new epistemology, the categories ``male''
and ``female'' as opposite and incommensurable bio-
logical sexes', and hence only then that the speci-
ficity of the female body was fully recognised with all
27
its various discursive implications. In this connec-
tion he notes the influence on the eighteenth cen-
tury of Poulain de La Barre, who had challenged the
notion of the inferiority of women on the ground
that since sex difference was solely a matter of the
body, there could be no social grounds to maintain
28
it. Among the philosophes themselves, of course,
there were many voices ± most famously Diderot and
Condorcet ± who argued for political and legal
equality for both sexes, and deplored the neglect
of the education of women. Condorcet even went so
far as to deny that natural differences between the
sexes had any bearing on their rights of citizenship:
why deny to women on the grounds of their suscept-
ibility to pregnancy what no one dreamt of denying
to men cooped up for half the year with the gout?
Of course, as Joan Scott has pointed out, Con-
dorcet was in something of a minority in pursuing
this line on the issue of equality and natural differ-
29
ence, and I am not disputing the existence of the
numerous other discourses on female difference
which were prejudicial and in many cases deeply
 LIBERALISM, FEMINISM, ENLIGHTENMENT
offensive in their suggestions about female limita-
tions and deficiencies. The point is, rather, that there
was a much wider and more varied spectrum of
arguments about gender and subjectivity than is
recognised by the contemporary dismissive refer-
ences to the Enlightenment subject. One might cite
as one example here the extremely crenellated de-
bate on the extent to which women had gained from
30
the process of civilisation.
There was some general agreement even among
those critical of the contemporary treatment of
women that the latter had only benefited in the
progress out of savagery. For Catherine Macaulay,
woman `in the barbarous stages of mankind' had
been reduced to a state of abject slavery; for Diderot,
`the unhappy woman in the cities is far unhappier
31
still in the midst of the forests'; and he, William
Alexander and John Millar all thought that the
degree of civility gained could be judged by the
manner in which women were treated. Alexander
additionally presented women as having ± through
their cultivation of beauty ± also stimulated men to
experience a romantic affection for a particular in-
dividual, a personalisation of attachments through
which they had spared themselves the misery and
barbarism of anonymous sexual relations. Such con-
ceptions, however, did not sit easily alongside the
recognition by these thinkers of the continued en-
slavement of women in their society ± especially
where this went together (as it did very notably in
the case of Diderot) with a critique of the corruption
and decadence of civilisation and a call `to return to
nature'. Diderot tends to forget (or is necessarily
forced to abstract from) his negative assessment of
the condition of women in primitive society, in his
praise of primitivism. `Ask civilised man whether he
is happy. Ask the savage whether he is unhappy,' he
writes. `If both reply ``No'' the dispute is settled.' But,
as Tomaselli points out, `to have asked a woman the
same set of questions would have yielded, judging by
32
Diderot's own account, both a ``no'' and a ``yes''.'
Rousseau, on the other hand, viewed the turn to
`Romantic' love not as a source of commendable
civility and liberation from barbarism, but as the
enslavement of men. Though agreeing to the role of
women in cultivating luxury and beauty, he saw the
advance of culture as a progression in a deplorable
203
effeminacy, and regarded women as a threat to a
33
`natural' masculinity. In this connection, one
might note, Tomaselli has questioned the tendency
of much feminist theory to represent women as
having been thought closer to nature. At least in
the period of Enlightenment, she suggests, culture ±
the move away from nature ± is represented as a
(variously benign or malignant) emasculation.
The more conventional Enlightenment emphasis
on the difference of the sexes went together, of
course, with the genderisation of mental faculties.
Reason, presented in Kantian philosophy as the
distinctive attribute of personhood, was viewed as
a pre-eminently masculine possession, and in the
writings of a number of eighteenth-century thinkers
± notably Rousseau and Kant ± women are depicted
as by nature more capricious and emotional creatures
incapable of attaining to a fully rational autonomy.
But even Kant implies this is a deficit brought about
by lack of education and inappropriate patronage.
Kant has often been rebuked by feminists for his
condescension. But when he remarks ironically in his
essay on Enlightenment that `those guardians who
have so kindly assumed superintendence over them'
will soon see to it `that the step to competence is held
to be very dangerous by the far greater portion of
34
mankind (and the entire fair sex)' his point is
perfectly consistent with the sort of claims made
by Wollstonecraft and others concerned in the per-
iod with improving female autonomy.
But more importantly, perhaps, one must note
how much there is in the culture and argument of
the period which runs counter to any simplistic
presentation of the liberal-Enlightenment as the site
or time of the privileging of the type of reason which
modern critics refer to as `masculine', meaning by
that a narrowly instrumental, overly logical and
abstracted mode of cognition. We have only to
consider the attention paid by theorists from the
late seventeenth century onwards to the role of
sensory experience, to the imagination, to the pos-
session and development of what was termed `sen-
sibility' to the analysis of taste and aesthetic
judgement, to the aesthetic as mediator between
reason and the sense, to realise how much more
there was to the Enlightenment discourses of the
35
self. The Age of Reason `distrusted Reason,' argues
 KATE SOPER
E. L. Tuveson, `as it had been understood for many
centuries ± far more deeply than did any preceding
36
period.' Another commentator goes on to claim:
It would be mistaken to think of reason as the
rallying cry of the Enlightenment thinkers except
in so far as it was opposed to faith, and the Age of
Reason as opposed to the Age of Superstition. If
one's gaze shifts away from the battles with l'In-
à me, then the `Age of Sentiments', `Sentimen-
fa
tality', `Feelings', `Passions', `Pleasure', `Love' or
`Imagination' are apter titles for the movement of
37
ideas in the eighteenth century.
This is surely to overstate the case against reason,
which remained the sovereign court of appeal, and
was viewed by Condorcet, Voltaire and Diderot and
most key thinkers of the age as of critical importance
to controlling the flights of fancy and excesses of the
imagination. But we certainly need to recognise the
force of the point made earlier regarding the im-
portance of distinguishing between a typically se-
venteenth-century view of reason as `sum total of
innate ideas' and the more characteristically eight-
eenth-century Kantian ± and liberal ± view of it as
committed to an unprejudicial `critique' of `pure'
reason, and to sceptical investigation of the knowl-
edge claimed on the authority of `innate ideas'. We
should recognise, too, that the tendency of some
contemporary feminist criticism to equate Enlight-
enment with the exercise of a narrowly conceived
`masculine reason' is to abstract from the many
nuanced discourses that went into its intellectual
ferment. It is also to ignore the voices of those (such
as Hume, Addison, Hutcheson, Shaftesbury, Burke,
Smith) who in the period itself did challenge the
exclusive endorsement of reason. Among the women
who did was Elizabeth Carter, acclaimed by Johnson
as the best Greek scholar in the kingdom, who
followed Addison in her scepticism about science
and argued (as did Addison and Adam Smith) for
the indispensability of the imagination as a source of
empathy with others and the need to strengthen
what Hutcheson called the `sympathetick sense'. It is
not reason, she argued, but imagination which dis-
tinguishes us from the beast.
Admittedly, it may be argued that a voice such as
204
Addison's or Carter's is better interpreted as belong-
ing to a proto-Romantic reaction to the liberal-
Enlightenment rather than as contributing to its
stock of ideas. Yet if we follow the contemporary
disposition to think in terms of a `long' Enlight-
enment, then we must view it as encompassing much
pre-Romantic and Romantic argument on the self,
directly opposed though such argument was to scien-
tific positivism, mechanistic conceptions of nature
and the dogmatic pursuit of reason. Or, to put the
point otherwise, Romanticism, which retains the
core Enlightenment idea of a unitary subject pos-
sessed of autonomous agency ± and which, after all,
owes a great deal to the influence of two figures,
Rousseau and Kant, usually deemed to belong un-
controvertibly to the Enlightenment ± can be viewed
itself as part of the unfolding process of Enlight-
enment self-criticism. One might argue, too, that
there is something of this same Romantically-
mediated Enlightenment rationality still at work
in John Stuart Mill's liberalism and that this gives
a particular inflection to his argument on a range of
topics, including that of female subjection.
III
I have tried to show here that Enlightenment culture
was a good deal more complex, and in a number of
ways more sensitive to the limitations of its own
universalising and rationalising claims, than glib
postmodernist dismissals often allow. I have tried
to show, too, that one rejects the period of Enlight-
enment and its legacy as radically unfriendly to the
advancement of feminist demands only at the risk of
appearing to opt for a theoretical perspective that
shares more with Burke's patrician defence of culture
and custom than with Wollstonecraft's attempts to
strip away the veils of patriarchal privilege. It would,
of course, be quite mistaken to present either Woll-
stonecraft in particular, or the voices of liberal-En-
lightenment in general, as offering the last word on
female emancipation. My plea here is only that they
should be acknowledged as an important first step in
the right direction, and that when feminists de-
nounce liberalism simply in virtue of its Enlight-
enment `rationalism' or `humanism' or conceptions
of subjectivity, they are making use of an easy
 LIBERALISM, FEMINISM, ENLIGHTENMENT
rhetoric rather than engaging in serious and discri-
minating critique.
They are also, as suggested earlier in my brief
remarks on the inconsistencies of the postmodernist
defence of `difference', frequently arguing in ways
which are covertly reliant on liberal-democratic
values themselves. What sense, for example, can
we ultimately make of postmodern feminist critiques
of the conflationist and Eurocentric tendencies of
liberal humanism if not in terms of its failure to
advance a more genuinely universal conception of
the human community? How do we understand the
charge that liberal approaches to emancipation are
`ethnocentric' or less than `democratic' in the respect
extended to `otherness' or `difference', if we do not
presuppose a commitment to some minimally essen-
tialist and universalist conception of human nature:
unless, that is, we hypothesise a common suscept-
ibility to pain and humiliation in the absence of
certain forms of respect and recognition? Liberalism
in general, and liberal feminism in particular, are
indeed open to the charge of having taken class-
biased and culturally partial views of what is suppo-
sedly common to everyone. But it is one thing to
expose this bias and partiality; it is quite another to
condemn liberal thinking for its invocation of a
shared structure of cognition, need and affectivity,
since little sense can be made of the request for
greater tolerance of other viewpoints and more
recognition of specific identities and constituencies,
unless some commonality of cognitive and emotional
response to the world is presupposed. Indeed, it is the
Notes
1. As Peter Hulme and Ludmilla Jordanova claim of the
Enlightenment period, since this was `in many of its
aspects, highly critical, it was inevitable that, even as
its identity emerged, it was subverted.' Hulme and
Jordanova (eds) (1990), p. 2.
2. Sapiro (1992), p. xiv.
3. Cf. my own discussion of Sapiro's position in Soper
(1997).
4. Obviously, this is a rather bald claim as it stands and
much could be said in qualification of it. I would not
want to deny that liberal and socialist feminists will
differ on the `order' of the agenda and the relative
priority to be given to the items on it ± especially those
of freedoms, on the one hand, and equality, on the
205
denial of this that has licensed some of the most
callous political neglect of the miseries and oppres-
sions of others.
Postmodernist feminists should therefore accept
that in the end their critiques of liberal humanism
make better sense and retain more internal coher-
ence if presented as a continuation of the immanent
critique of Enlightenment first sounded in the dis-
course of Wollstonecraft and other proto-feminist
liberal critics. Or if they do not agree to this, they
should at least be ready to accept the full implica-
tions of their value-relativism and anti-realism, and
thus accept that there are ultimately no objective
criteria of female emancipation, and no grounds
therefore on which to defend any projected vision
of this as more or less `progressive'. To deny that
there is any link between the discourses of `power'
and `oppression' and specific extra-discursive condi-
tions of existence and forms of human experience, is
to disown the political teleology associated with the
feminist movement, whether this be given a more
liberal or more socialist interpretation. A postmo-
dernist position can, in other words, escape the
charges of being inconsistent in its treatment of
liberal-democratic values, but only at the cost of
licensing an indefinite and uncontrollable play of
meaning such that any and every treatment of
women might count, under some discursive regime
or within a particular political context, as a further-
ance of feminist goals. The question then is: can this
still count as a distinctively `feminist' theoretical
offering at all?
other. But I think there is enough shared commitment
to these formal principles for the point to have some
validity.
5. I put the point in these terms in order to leave open
the question of the periodisation of Enlightenment,
which it is not my intention to address in this chapter.
Suffice it to note that there is still considerable dispute
on both the temporality and the definition of Enlight-
enment. As Hulme and Jordanova point out, there are
basically two approaches to the issue, the first con-
fining it more narrowly to the ideas of a group of
(mainly French) intellectuals during the period dating
from the death of Louis XIV (1715) to the events of
the French Revolution (1789±94); the second ap-
 KATE SOPER
proach, which is altogether broader, would emphasise
the origins of Enlightenment in the work of Bacon,
Newton and Locke in the seventeenth century and
argue for a much longer Enlightenment ± at least into
the early decades of the nineteenth century, and even
beyond: Hulme and Jordanova (eds) (1990), pp. 1±15.
Although the recent Blackwell Companion to the En-
lightenment focuses on the years 1720±80 (Yolton, et
al. (eds), 1991), historians of the period today increas-
ingly favour the idea of a `long' Enlightenment. For a
brief guide to the recent history of Enlightenment
interpretation, see Outram (1995), chapter 1.
6. Cassirer (1951), p. 13; cf. Pauline Johnson's discussion
(and illustration of this difference in respect of the
concepts of reason employed respectively by Mary
Anstell and Mary Wollstonecraft) in Johnson
(1993), pp. 3±12.
7. Landes (1988).
8. See Jardine (1985); cf. also the dispute between
Timothy Reiss and Frances Ferguson in Kauffman
(ed.) (1989) (where Ferguson defends Wollstonecraft
against Reiss's claim that her assertion of the rights of
women was bound to prove ineffective because it was
so immersed in the dominant discourse of Enlight-
enment). See also Johnson (1993).
9. See, for example, Evans (1985); Lloyd (1984), chapter
6. Toril Moi discusses and defends de Beauvoir against
similar charges in Moi (1992). For a brief survey, see
again Johnson (1993), pp. 3±4.
10. Johnson (1993), p. 3.
11. See Fuss (1989); Kate Soper, `Feminism, Humanism,
Postmodernism' and `Postmodernism, Subjectivity
and the Question of Value', in Soper (1990), pp.
228±45; and Squires (ed.) (1993), pp. 17±30 respec-
tively; also Lovibond (1989) and (1994).
12. Soper (1990), and `Postmodernism. Subjectivity and
the Question of Value', in Squires (ed.) (1993); see
also Nussbaum (1992).
13. See, for example, Derrida (1989), esp. p. 56f.; Spivak
(1991) and the essays in Spivak (1993).
14. In this respect I am pursuing the general line of
response already opened up by Johnson (1993). I
see my own survey here as offering further support
for her claim that contemporary feminism is working
within a distorted and partial version of Enlighten-
ment and its commitment to reason.
15. Outram (1995), p. 12.
16. See ibid., chapter 3; cf. Margaret Jacob's argument that
the `two extremes of Spinozism, pantheism and ma-
terialism, lurked in the underground on both Protes-
tant sides of the Channel, but from Samuel Clarke to
Johannes Stinstra, Betje Wolff, and Aagje Deken,
godly Enlightenment ± born out of their reading of
the new science ± seemed a sufficient antidote to
dogmatism, enthusiasm, or absolutism'. Jacob
(1994), p. 98.
206
17. Ibid., pp. 95±113; see also the studies by Jacob, Phyllis
Mack, Ruth Perry and Margaret Hunt in Women in the
Enlightenment, in Women and History, 9 (1984) (rep-
rinted in book form by the Institute for Historical
Research and the Haworth Press, 1984).
18. Jacob (1994), p. 98 f.
19. Ibid., p. 100f.
20. As Jacob points out, most historians now accept the
importance for understanding of the eighteenth cen-
tury of the category of the `public sphere' (by which is
intended the universe of `public opinion formation,
informal sociability, intellectual interaction and net-
working beyond the family and outside the formal
institutions of the state), as they do the role of
Habermas's work in drawing attention to it, notably
Habermas (1992).
21. Jacob (1994), pp. 106±7.
22. The Habermasian framework tends both to obscure
the very material character of much child-rearing
activity and the extensive permeation of family life
by the systemic forces of power and money. Feminists
have also accused him of failing to appreciate the
dualised and gendered nature of welfare systems
(which may reduce female dependency on the male
breadwinner, but only at the cost of exposing women
to a patriarchal and androcentric state bureaucracy).
See Fraser's `What's Critical About Critical Theory?',
in Benhabib and Cornell (eds) (1987); cf. also the
article by Linda Nicholson in the same volume. (Of
course this type of feminist criticism is itself caught in a
certain tension since it is reliant on a privileging of the
`symbolic', feminine-associated values and forms of
human interaction in its overall critique of patriarchy,
even as it is critical of Habermas's failure to draw out
the implications of his normative schema for feminist
theory and practice.)
23. Burke (1961), p. 19.
24. Ibid., p. 90.
25. `A Vindication of the Rights of Men, in a Letter to the
Right Honourable Edmund Burke', in Wollstonecraft
(1989), p. 14.
26. Laquer (1990), p. 148. Cf. Outram (1995), chapter 5,
which also stresses the concern with female difference
± and its contradictory implications in view of the
Enlightenment emphasis on a universal human nature
and human history, both validated by the possession of
a single human form of rationality. See also the
discussion of the concept of the `individual' in Scott
(1996), pp. 5±14.
27. Laquer (1990), pp. 149±50.
28. Ibid., p. 155±6.
29. Scott (1996), pp. 7±8.
30. As discussed in Tomaselli (1985).
31. Both cited in ibid., p. 109.
32. Ibid., p. 117.
33. Ibid., pp. 118±20. Cf. Outram (1995), pp. 88±9.
 LIBERALISM, FEMINISM, ENLIGHTENMENT
34. Kant (1963a), p. 3 (a more cutting translation of Kant
(1991), p. 54). Kant continues: `After the guardians
have first made their domestic animals dumb and have
made sure that these placid creatures will not dare take
a single step without the harness of the cart to which
they are tethered, the guardians show them the danger
207
which threatens them if they try to go alone'. Kant
(1963a), p. 3.
35. See, for example, Tuveson (1960); Engell (1982); Taylor
(1989), parts III and IV; Eagleton (1990), chapters 2±4.
36. Tuveson (1960), p. 24.
37. Yolton, et al. (eds) (1991), p. 446.
 16
FEMINISM AND WOMEN'S
AUTONOMY: THE CHALLENGE
OF FEMALE GENITAL CUTTING
Diana Tietjens Meyers
Terminology for the multiform phenomenon I shall
discuss is itself a matter of vexed controversy. Some
scholars prefer the expression `female circumcision'
on the grounds that this is the expression preferred
by women who are now subject to the practice. To
adopt any other language, they maintain, would be
to show disrespect for these women and their cultures
and perhaps to engage in cultural imperialism. Other
scholars opt for medicalised terminology ± either
`female genital surgery' or `female genital operations'.
Still other scholars prefer the expression `female
genital mutilation' on the grounds that no other
language expresses the suffering inflicted by this
practice and that other labels implicitly condone a
cruel practice that sustains male dominance by tor-
menting women.
I have adopted the expression `female genital
cutting' in order to address several concerns. On
the one hand, I see the medicalised terminology and
the culturally relative terminology as euphemistic.
Although the medicalised expressions accurately
represent the hygienic conditions of the hospitals
in which some Euro-American and African female
genital cutting is performed, they give the false
impression that female genital cutting is usually a
sanitary procedure performed with anaesthesia. Un-
fortunately, the procedure is seldom performed under
safe conditions, and it never improves women's
health. The culturally relative terminology is also
208
euphemistic because the analogy with male circum-
cision suggests a relatively risk-free, minor procedure
that does not interfere with sexual pleasure.
Although one form of the female procedure is aptly
analogised to male circumcision, most are not.
On the other hand, the morally condemnatory
language, `female genital mutilation', prejudices the
question of women's autonomy vis-a
Á -vis this practice.
Since there is a presumption that a person who
chooses to be mutilated does not choose autono-
mously, this language is question-begging in the
context of my present topic. Moreover, I suspect
that Euro-Americans are attracted to this language
largely because they do not imagine that women like
themselves have in the past consented to such
`mutilation' and that mothers in their own society
today continue to authorise comparable procedures
on their daughters. Consequently they fail to em-
pathise with contemporary women in other societies
who make similar choices. Thus, I have sought
terminology designed to avoid denying the pain
and impairment often associated with female genital
cutting but also to avoid presuming that women
involved in the practice have no autonomy.
Many Euro-Americans might doubt that there is
any basis for ascribing autonomy to women whose
cultures mandate female genital cutting. Yet, the
feminist literature on female genital cutting provides
ample evidence that many women exercise effective
 FEMINISM AND WOMEN'S AUTONOMY

agency with respect to this practice. One striking programmes may not be the most effective way to
finding is that autonomy is to be found among its eradicate female genital cutting, they have proved
accommodators as well as resisters, and that neither to be the most effective way to promote women's
group can be presumed to enjoy greater autonomy autonomous control over their sexuality and health.
than the other. If this is so, one cannot identify Finally, I consider how the cultural imperative of
autonomy or lack of it simply by looking at what self-perpetuation entails both support for and lim-
people choose or refuse. Autonomy must dwell in the itation of autonomy. If cultures are to meet the need
process of deciding, not in the nature of the action for change in response to historical contingencies,
decided upon. Moreover, students of female genital they must ensure that cultural initiates have auton-
cutting regard it as essential to avoid the trap of omy skills that enable them to adjust their tradi-
conceding that resisters are western dupes and, as tions. Yet, if cultures are to survive, they must
cultural outsiders, have no right to press for change, channel exercise of these autonomy skills into
whereas accommodators are more authentically re- constructive critique framed by cultural allegiance,
presentative of their culture of origin and, as female and they must discourage the development of
insiders, confirm the legitimacy of the practice. To autonomy skills which heighten discontent and
overcome this dilemma, one must resist the tempta- precipitate defection. I argue that autonomy-aug-
tion to explain greater autonomy as a function of the menting educational programmes succeed by devel-
reduced influence of cultures that mandate female oping autonomy skills that cultures have
genital cutting. Culture cannot be cast as the villain systematically, but needlessly, suppressed. Con-
that paralyses some women's autonomy, nor can it be ceived as the exercise of self-discovery, self-defini-
cast as the hero that frees women into autonomous tion and self-direction skills, autonomy-within-
lives. Still, in acknowledging that some social con- culture is not only intelligible, it is a morally
texts are more conducive to autonomy than others, defensible and politically viable conception of
students of female genital cutting spotlight the am- autonomy for an era of global feminism.
biguous role of culture with respect to the question of
women's autonomy.
Cultural Contexts
My project is to explore the implications of these
and Female Genital Cutting
understandings of the relation between culture,
female genital cutting and women's agency for Among Euro-Americans, misapprehensions about
autonomy theory. I begin by surveying the reasons culture and female genital cutting abound. For ex-
different cultures give for female genital cutting and ample, many people associate female genital cutting
the different types of female genital cutting. My with a single region, usually sub-Saharan Africa. But
purpose in exhibiting the enormous variation in the practice is also found indigenously in North
this practice and in women's responses to it is to America, Asia and the Middle East, and immigrants
1
undercut simplistic dismissals of women's autonomy have spread their practices far and wide. A related
with respect to female genital cutting that rely on misconception is that female genital cutting is rooted
attention-grabbing horror stories and generalised in a single culture, but this is far from the case. A
theories of patriarchal domination. I then examine highly diverse array of cultures furnish a variety of
two kinds of autonomy theory ± latitudinarian, rationales for female genital cutting.
value-neutral accounts and restrictive, value-satu- Many people are astonished to learn that, among
rated accounts ± and I argue that neither approach middle-class white Americans in the late nineteenth
adequately addresses the phenomena of women's century and continuing into the early twentieth
autonomy regarding female genital cutting. To century, clitoridectomy was a popular and medically
bring these phenomena into focus, it is useful to certified `cure' for hysteria, nymphomania, lesbian-
2
review a number of strategies for augmenting wo- ism and `excessive' masturbation. Curiously,
men's autonomy regarding female genital cutting. I though, there are cultural groups that practice female
urge that, although culturally anchored educational genital cutting today for the opposite reason: they

209
 DIANA TIETJENS MEYERS
believe that female genital cutting enhances wo-
3
men's sexual desire. More typical, though, are ra-
tionales similar to those formerly current in the US:
many cultural groups maintain that female genital
cutting reduces women's sexual appetite, enforces
norms of chastity and thereby protects family hon-
4 11
our.
Repressing women's sexuality is by no means the
only, nor even the most frequent, reason given for
female genital cutting. Gender regimentation is a
prevalent rationale, as well. From this perspective,
the purpose of female genital cutting is to ensure that
the individuals assigned to dichotomous gender ca-
tegories have anatomically suitable genitals.
In the USA today, female genital cutting is a
routine, though increasingly contested, medical in-
tervention when a condition called `ambiguous gen-
5
italia' is diagnosed. Physicians and parents generally
presume that left `untreated' an elongated clitoris
will have deleterious effects on a girl's psychological
6
development and will prove socially disabling. To
the extent that this negative prognosis is warranted,
it is because contemporary US culture insists that
there are only two kinds of people ± males and
females ± and because this particular culture pre-
scribes genitalia conforming to restrictive norms for
7
each category. Without `corrective' demasculinising
surgery, it is thought that a genetic female endowed
with `ambiguous' genitalia will be unable to attain a
feminine identity. The theme of demasculinisation is
also found in some non-western cultures. Anticipat-
ing Freud, those cultures view the clitoris as a mascu-
line organ and therefore regard excision as necessary
8
to achieving a feminine gender identity. World-
wide, it seems, people believe that babies can be born
with `unnatural', though not sexually or reproduc-
tively dysfunctional, genitalia.
Whereas American culture currently mandates
female genital cutting as a corrective for what it
deems `abnormal' genitalia, members of many other
cultural groups regard female genital cutting as an
indispensable rite of passage through which all girls
must pass in order to attain the identity and the
status of women. Some see female genital cutting as a
test of courage and endurance that prepares the
9
individual for labour pains and birthing. Others
contend that female genital cutting physically puri-
210
fies the individual and enhances her fertility, for they
believe that the clitoris can kill a man if it touches
his penis or that it will kill the baby during child-
10
birth. Female genital cutting may be thought ne-
cessary to morally appropriate fertility, and morally
sanctioned fertility may be thought necessary to
producing socially and morally viable children.
In one Sudanese community, `son of an uncircum-
cised mother' is among the most derogatory epithets
12
one can utter. Gender norms are associated with
aesthetic standards as well as moral standards, and
aesthetic considerations sometimes dictate female
genital cutting. A woman who belongs to a culture
that considers gaping orifices, including the mouth,
distasteful asks, `which is better, an ugly opening or a
13
dignified closure?' In some cases, religious doctrine
is pivotal. Some cultural groups believe that Islam
mandates female genital cutting and that Allah will
14
not hear a woman's prayers if she has not been cut.
Another cluster of reasons concerns intercultural
political relations. Within a local framework, cultur-
al identification is sometimes cited: whether wo-
men's genitals are cut or not and, if they are cut,
how they are cut may be an important way in which
cultural groups inhabiting the same geographical
15
region are differentiated. From a global perspective,
loyalty to one's cultural roots and repudiation of
western influence are sometimes cited: female genital
cutting was often seen as testifying to opposition to
colonialism and more recently as affirming indigen-
ous tradition in the face of western contempt for
non-western cultures and western economic and
16
cultural penetration.
Although cultural rationales for female genital
cutting can be classified according to the broad
themes of sexual repression, gender identity, and
group cohesion, there is no uniformity whatsoever
in the specifics. This heterogeneity is echoed in the
variety of forms that female genital cutting takes.
Many Euro-Americans believe that female genital
cutting is a single procedure, but this is not true. In
addition to western cosmetic procedures designed to
`feminise' ostensibly male genitalia, practices range
from sunna ± that is removal of the clitoral prepuce ±
to infibulation ± that is excision of the clitoris, the
labia minora and the labia majora and suturing the
remaining tissue together to create a minuscule
 FEMINISM AND WOMEN'S AUTONOMY
17
orifice. Correlated with this spectrum of outcomes
is a spectrum of health risks in the immediate after-
math of the procedure and a spectrum of long-term
consequences for women's sexuality, physical health,
and psychological well-being. Moreover, different
groups perform the cutting procedure at different
ages ± some on infants or very young girls, some on
adolescents. Adult women participate in this prac-
tice not only by authorising and/or carrying out the
procedure but also, in groups where infibulation is
the rule, by agreeing to reinfibulation after child-
birth. It is a mistake, then, to suppose that female
genital cutting has the same impact on all affected
women's lives.
Finally, it is necessary to dispel some prevalent
misconceptions about culture. The misconception
that most incenses non-western women is the `other-
ing' of the very idea of culture ± that is the supposi-
tion that culture itself is a non-western phenomenon
and that westerners are not themselves enculturated.
Alternatively, westerners may acknowledge that
everyone is enculturated yet assume that, unlike
western cultures in which dissent is tolerated and
cultural change is ongoing, non-western cultures are
18
homogeneous and static. Nothing could be farther
from the case. Members of a cultural group do not
think and live in lockstep. Within cultural groups,
interpretations of cultural beliefs vary, and so do
interpretations of cultural practices. Drawing on the
countless beliefs and practices that constitute a
culture ± beliefs and practices that are in some
respects in tension with one another ± and accenting
and linking some of these beliefs and practices while
de-emphasising or isolating others, people individua-
lise their cultural heritage. As a result, traditionalists
often coexist, albeit uneasily, with nonconformists.
Moreover, the survival of cultures depends on their
capacity to modify their beliefs and practices to meet
new contingencies, and their capacity to adjust
depends in part on the internal variegation I have
19
described. A seamless, timeless culture is a dead
culture.
This brief survey of the anthropological back-
ground of female genital cutting is sufficient to
demonstrate the daunting complexity of analysing
women's agentic position with respect to this prac-
tice. Evidently, judgements about women's autono-
211
my vis-a
Á -vis female genital cutting cannot rest on
generalisations about the severity of the outcome, for
the medical and psychosexual consequences vary
substantially. Women's autonomy with respect to
female genital cutting cannot be ruled out on the
grounds that no one could freely choose the endan-
germent and impairment resulting from these pro-
cedures, for the consequences are not always grave.
Nor can one assume that the reasons women have for
either accommodating or resisting female genital
cutting are consistent across cultural groups. So a
theory of women's autonomy that is applicable to
societies in which female genital cutting is practiced
cannot rely on ascriptions of particular reasons to the
women who are subject to this practise. The content
of women's deliberations cannot settle the question
of their autonomy, for the reasons they invoke are
not uniform.
It is clear that some cultural groups pre-empt girls'
autonomy by cutting their genitals when they are
young and helpless. However, female genital cutting
is part of an adolescent initiation rite in many
cultural groups. Although it is troubling that these
girls may have little experience of their own genital
eroticism and presumably have no experience of the
interpersonal intimacies that this sexual responsive-
ness makes possible, it is doubtful that girls of this age
20
altogether lack autonomy. Likewise, adult women
participate in this practice as midwife practitioners,
as mothers who arrange female genital cutting for
their daughters and as postpartum mothers who
request reinfibulation. It is important to bear in
mind, as well, that during the nineteenth century
many adult Euro-American women anxiously sought
relief from psychopathology or sexual deviance and
willingly underwent clitoridectomy. Whatever one
may think of juvenile agency, these women's agency
must be analysed on its own merits.
It is wrong, moreover, to assume that all female
members of cultural groups that practice female
genital cutting collaborate with this practice. Wo-
men are not in the grip of totalising, internalised
cultures which drive them to comply with a female
genital cutting imperative. There are many instances
of individual women persuading their families not to
uphold the practice, and instances of women orga-
nising against female genital cutting are on the rise.
 DIANA TIETJENS MEYERS
Although data about the prevalence of female geni-
tal cutting is disputable, it is quite possible that
adherence to the practice is waning. Accordingly,
to declare women who belong to cultural groups that
practice female genital cutting devoid of autonomy
would be to deny existing opportunities for choice
and to erase the real, sometimes courageous, choices
women have actually made.
Women exercise agency, then, both by complying
with and by resisting female genital cutting. A theory
of women's autonomy must take these realities into
account, but it must also ask whether women have as
much autonomy as they should have with respect to
female genital cutting.
Female Genital Cutting's
Challenge to Autonomy Theory
The phenomenon of female genital cutting presents
two opposed temptations for autonomy theory. On
the one hand, latitudinarian, value-neutral accounts
of autonomy ± whether rational choice models or
hierarchical identification models ± are attractive
because they do not withhold respect from women
who accommodate female genital cutting by impugn-
ing these women's ability to make their own judge-
ments and choices. On the other hand, restrictive,
value-saturated accounts of autonomy that deny that
people can be both oppressed and autonomous are
attractive because, in claiming that false conscious-
ness blocks the self-determination of women in
cultures that mandate female genital cutting, they
highlight the harsh personal cost of living under such
oppressive social regimes. Neither of these
proaches is ultimately convincing, however.
Preliminary to characterising the weaknesses of
these approaches, it is useful to review what an
account of autonomy should accomplish. An ac-
count of autonomy aims principally to explicate
an especially valuable mode of living. That mode
of living is captured in a number of familiar expres-
sions: `she lives by her own lights'; `he is his own
man'; `she's always been true to herself'; and the like.
Autonomous individuals are not mere conformists.
They need not be eccentric, but they do rely on their
own judgement. They know who they are ± what
really matters to them, who they deeply care about,
what their capacities and limitations actually are,
and so forth ± and they enact this introspective
understanding of their `true' selves in their lives.
There is a good fit between their identity, their
attitudes toward themselves and their conduct. Col-
loquial expressions, such as `I feel at one with myself'
and `I feel right in my skin' voice this sense of
integration. As these idioms suggest, living autono-
mously is satisfying, sometimes exhilarating. Subjec-
tively, then, the value of autonomy stems from the
fascination of self-discovery and the gratification of
self-determination. Objectively, it rests on the dig-
nity of the distinctive individual and the wondrous
diversity of the lives individuals may fashion for
themselves.
A second aim an account of autonomy should
fulfil is to explicate the nature of the personal and
social costs of suppressing autonomy. Individuals
experience lack of autonomy as a sense of being
out of control or being under the control of others ±
whether other identifiable individuals or anonymous
societal powers. At odds with themselves, at odds
with their behaviour, or both, non-autonomous in-
dividuals often feel anxious about their choices,
contemptuous of themselves and disappointed with
their lives. Alternatively, they may simply feel hol-
low, for they may feel they have been made into
vehicles for projects that they do not disavow but
that are not their own. In one way or another, non-
autonomous individuals suffer from alienation from
self. I would add, moreover, that societies that are
not conducive to autonomy incur a moral loss since
they thwart (or try to thwart) insightful social cri-
ap- tique. When a society discourages self-exploration
and self-expression, it discourages attention to symp-
toms of discontent and shields social ideologies and
institutions from probing examination and opposi-
tional activism. Autonomy exposes the need for
social change and equips people to pursue it.
I shall not linger long over latitudinarian, value-
neutral views for their weaknesses have been diag-
nosed and elaborated elsewhere. Briefly, rational
choice views take people's desires, values and goals
for granted and identify autonomy with devising
21
plans that maximise satisfaction. Hierarchical
identification views subject first-order desires to
scrutiny in light of second-order volitions and link
212
 FEMINISM AND WOMEN'S AUTONOMY
autonomy to reconciling the two levels to achieve a
22
harmonious whole. Both types of theory neglect
the possibility that an oppressive social context
could subvert people's autonomy by imparting detri-
mental desires, values and goals or warping people's
second-order volitions. Such theories make no ade-
quate provision for `authenticating' the concepts and
commitments that structure one's interpretations
and propel one's deliberations and choices. In con-
trast, restrictive, value-saturated accounts of auton-
omy, such as Susan Babbitt's and Paul Benson's,
insist on the need to distinguish real from apparent
desires and authentic values from spurious ones, and
they draw these distinctions by placing constraints
on what people can autonomously choose.
According to Susan Babbitt, ideological oppres-
sion instils preferences and desires that do not ade-
quately reflect an interest in one's own flourishing
and that prevent one from pursuing one's `objective
interests' even when one is aware that one has an
23
option to pursue them. The problem, claims Bab-
bitt, is the individual's `not possessing a sense of self
that would support a full sense of flourishing' ± that is
one has been deprived of a precondition for wanting
24
to pursue one's objective interests. Although the
oppressed have non-propositional knowledge ± that
is, knowledge in the form of intuitions, attitudes,
ways of behaving and so on ± that adumbrates their
objective interests, this knowledge is inexpressible
within the existing ideological regime and is not
25
translatable into autonomous action. Mute and
subjugated, the agency of these individuals can only
be salvaged through `transformative experiences'
26
that, as it were, upgrade their selfhood.
I doubt that oppression renders people's non-
propositional knowledge inexpressible. In fact, I
think one of Babbitt's examples amply demonstrates
my point. Commenting on Alice Walker's novel
about domestic violence, The Color Purple, Babbitt
claims that Celie's knowledge that she is a morally
worthy person is non-propositional and inexpressi-
ble. I would argue that Celie's knowledge indeed
stems from non-propositional sources, her feelings,
attitudes and perhaps her intuitions. But, as Babbitt
reports, when taunted by Mister ± `You nothing at
all' ± Celie trenchantly replies, `I'm pore, I'm black, I
27
may be ugly and can't cook . . . but I'm here.'
213
Babbitt is correct to say that the categories of Mister's
ideology provide no direct, authoritative way for
Celie to assert her moral status, but it is evident
that Celie is able to give her knowledge a proposi-
tional form and to encode her knowledge in intel-
ligible speech.
Oppression deprives Celie of conventions ± read-
ily available, generally accepted discursive formulae
± through which she can articulate her convictions,
28
protests and aspirations. To articulate their self-
knowledge, oppressed people must resort to circum-
locution, devise figures of speech and work to re-
define the terms that frame intrapersonal
understanding, interpersonal relations and moral
reflection. Thus, they must summon extraordinary
imaginative and linguistic powers if they are to gain a
rich understanding of who they are and why their
needs and desires deserve respect. Nevertheless, in
light of her defiant self-recognition and her poignant
self-assertion, it seems doubtful that Celie lacks an
`adequate' sense of self. I see no reason, then, not to
trust her to sort out her own values and goals and
conduct her own life.
Still, it is undeniable that Celie's social context is
doing everything possible to stifle her autonomy and
to defeat her. Likewise, women who demand reinfi-
bulation because it is beautiful, because it is a pre-
condition for womanly virtue, because it is `our way'
or because the World Health Organisation is trying
to force us to stop doing it hardly enjoy cultural
environments that encourage them to do what they
think is best for themselves. For this reason, I expect,
Paul Benson would argue that attributing autonomy
to these women betokens a `Pollyanaish' confidence
in their agentic capabilities.
According to Benson, `certain forms of socialisa-
29
tion are oppressive and clearly lessen autonomy.'
There are two forms that oppressive socialisation
takes: (1) coercive socialisation that inflicts penalties
for noncompliance with unjustifiable norms; and (2)
socialisation that instils false beliefs which prevent
30
people from discerning genuine reasons for acting.
Autonomous people are `competent criticisers' who
can `detect and appreciate the reasons there are to
31
act in various ways.' But oppressive socialisation
systematically obviates and obfuscates victims' rea-
sons for acting. As Benson puts it, oppressive socia-
 DIANA TIETJENS MEYERS
lisation limits in `well-organised ways what sorts of
32
reasons to act people are able to recognise.'
over, bi-directional integration ± that is, mutual
adjustment and reconciliation of first-order desires
and second-order volitions about first-order desires ±
is no guarantee of autonomy, for oppressive sociali-
sation could implant such deficient values that a
person could have degraded first-order desires that
would nonetheless be endorsed by second-order
33
volitions. Internalised oppressive socialisation, it
seems, can `shanghai' a person's entire life.
Presumably, Benson would conclude that, because
of their socialisation, the many women who do not
repudiate female genital cutting have defective rea-
son-detection faculties ± defective at least in so far as
they are oblivious to the `decisive' force of the
reasons against this practice ± and thus that they
have been deprived of autonomy at least with respect
34
to this choice. But I would urge that Benson's grim
assessment of the sinister potency of oppressive
socialisation exaggerates the impact of socialisation
generally. It just isn't true that oppressive socialisa-
tion always decreases autonomy. Some people be-
come oppositional activists, and some of
flourish in that role. In cases of firebrand, adven-
ture-loving resisters, one suspects they would have
had a hard time fitting in and living autonomously if
they had been born into a just society during peace-
time. Other people carve out lives that enact `in-
appropriate' values in the interstices of
constraints that society imposes. They find pockets
of lapsed surveillance or permissiveness within op-
pressive regimes and devise ways to express their
unorthodox values and commitments in
spaces. Still others endorse at least some of the values
upon which oppressive constraints are based and on
balance accept the constraints and conform their
lives to them. We find all of these possibilities
represented in connection with female genital cut-
ting ± women fleeing their homeland and seeking
asylum to avoid having their genitals cut or to
protect their daughters from cutting, women orga-
nising locally and nationally to modify or eradicate
female genital cutting, women concluding that cul-
tural tradition and cohesion or getting married and
bearing children are more important than bodily
integrity. I would venture that there are women in
each of these groups who have experienced compar-
More- ably oppressive socialisation. Undeniably, they have
more to overcome in order to attain autonomy than
some other women, for they must mobilise extra-
ordinary introspective and volitional powers to figure
out what they really want and to act on their own
desires. But it is hard to believe that none of them
chooses autonomously.
The fact is that we are all immersed in a culture at
a historical moment. How do we know that some of
us have attained adequate selfhood and thus have
the epistemic perspective needed to grasp what full
flourishing is like? How do we know that some of us
have highly developed, acutely sensitive reason-de-
tection faculties and thus have the epistemic skills
needed to determine what cannot be a good reason
to act or what is a dispositive reason to act? It seems
to me that we would need far more consensus than
we presently have (or are likely to get) about human
nature and social justice before we could conclude
that women who opt for compliance with female
genital cutting norms never do so autonomously. We
would have to be persuaded, in other words, that all
them women's interests are such that this decision could
not accord with any woman's authentic values and
desires under any circumstances. If we are prepared to
acknowledge that a woman who has undergone
oppressive socialisation but who rebels against its
dictates may be accessing her authentic values and
the desires and acting autonomously, it seems to me that
we cannot rule out a priori the possibility that a
similarly socialised woman who chooses otherwise
may be autonomous, too.
those In sum, restrictive, value-saturated accounts of
autonomy are troubling because they homogenise
authentic selves and autonomous lives. The para-
doxical effect of ahistorically, acontextually foreor-
daining what individuals can and cannot
autonomously choose is to deindividualise autonomy.
Yet, latitudinarian, value-neutral accounts of auton-
omy are troubling, too. On this view, failures of
autonomy are failures to obtain and take into ac-
count relevant information, or they are failures to
integrate one's values, desires and the like into a
coherent outlook and a feasible course of action. The
paradoxical effect of neglecting the possibility that a
well integrated, smoothly functioning self could be in
214
 FEMINISM AND WOMEN'S AUTONOMY
need of rigorous scrutiny and drastic overhaul is to
abandon the individual to the influence of a culture's
prevailing beliefs and practices. Such a view over-
simplifies self-alienation and blunts autonomy's po-
tential to spur social critique. Neither of these
approaches offers a compelling theory of individua-
lised autonomous living.
Autonomy-Augmenting Programmes
and Female Genital Cutting
Feminist students of female genital cutting have
analysed a number of programmes that have en-
hanced women's autonomy in regard to this practice.
Moreover, they have proposed legal reforms, and
they have documented the effects of social trends
on female genital cutting. I shall briefly survey this
literature with the aim of using it as a springboard for
developing an account of women's autonomy-with-
in-culture.
My reading of this literature suggests that educa-
tion is the least controversial and most effective
approach to augmenting women's autonomy regard-
ing female genital cutting, but the success of this
approach depends on the nature of the educational
programme. Again and again, scholars stress the need
for grassroots participation in conceiving the aims of
educational initiatives, in formulating educational
agendas and in devising educational presentations
and materials. Phrases like `education for critical
consciousness' and `emancipatory education' imply
the need for institutionalising self-determination at
the founding and throughout the implementation of
35
education programmes. This imperative has been
put in practice through dialogic programme planning
and dialogic `classrooms' and also by adopting cul-
36
turally familiar media for `instruction'. Songs, plays
and puppet shows followed by exploratory discussion
sessions often replace lectures and rote learning.
Experience shows that many women are receptive
to information about the health risks associated with
37
female genital cutting. Successful educational pro-
grammes often capitalise on women's dissonant ex-
perience Á -vis
vis-a female genital cutting. For
example, health education can juxtapose women's
commitment to female genital cutting with their
awareness of traumatic outcomes (sometimes deaths)
215
and their awareness of their own sexual impairment.
Likewise, health education can kindle cognitive
dissonance by highlighting potential conflicts be-
tween women's desire to bear children and their
commitment to female genital cutting, which can
38
reduce fertility. Programmes designed to elicit en-
quiry into the symbolic meanings that one's culture
attaches to female genital cutting have proved ef-
39
fective, too. An international organisation called
Women Living Under Muslim Laws has developed
educational strategies that encourage women to
analyse and reconstruct their own identities within
their cultures. Through education, this organisation
seeks to debunk the myth of one universal Muslim
identity and to offer women the opportunity to read
40
and interpret Islam's sacred texts for themselves. In
discovering how differently women live under dif-
ferent Islamic regimes and how weak the tie between
Islamic scripture and current religious doctrines re-
quiring female genital cutting actually is, many
women realise that they could modify their beliefs
regarding female genital cutting without sacrificing
their religious beliefs. Intellectual and emotional
space hospitable to envisaging a different conception
of the ethics and aesthetics of womanhood is thus
opened.
Sometimes women become convinced that they
can flourish as women with their genitals intact.
Kenyan women and their families have created
`circumcision through words' rituals in which cultur-
al teachings about womanhood are transmitted and
girls' entrance into womanhood is celebrated, but the
41
traditional cutting is eliminated. Sometimes wo-
men remain convinced that female genital cutting is
necessary, but they alter the practice either by re-
ducing it to a symbolic pricking or by medicalising
42
it. Sometimes they insist on upholding tradition.
Hofriyati women whose husbands have returned
from working in Saudi Arabia convinced that in-
fibulation is contrary to Islam and that sunna is the
religiously sanctioned form of female genital cutting
have resisted the ensuing pressure to adopt the less
43
extreme practice. In some cultural groups, the
ritual surrounding female genital cutting not only
confers the status of womanhood on the initiate but
also creates lifelong social bonds among the girls who
undergo the ritual together. Thus another important
 DIANA TIETJENS MEYERS
task of educational programmes has been to create
substitutes for the `age groups' established by female
genital cutting. In central Kenya, the social benefits
formerly delivered by age groups ± solidarity among
women and consolidation of women's social author-
ity ± are now secured by women's self-help economic
44
organisations.
In addition to such culturally attuned, participa-
tory education, we also find more coercive legal and
social forces at work. Female genital cutting is pro-
hibited in a number of countries where the practice
has been common. However, the history of crim-
inalising female genital cutting and disseminating
information about women's legal rights does not
inspire confidence in this approach. Either these
measures have had little impact or they have in-
45
tensified reactionary traditionalism. But legal sanc-
tions could take a different form. One scholar, Asma
Mohammed A'Haleem, proposes that laws be passed
giving adult women the right to sue the practitioners
46
who cut their genitals when they were children.
a precedent for her proposal, she points to the
passage of parallel laws protecting women from
compulsory marriage. Women who were adequately
informed of these laws frequently went to court and
47
sued to enforce their rights. Perhaps women would
also avail themselves of rights to seek compensation
from female genital cutting practitioners, which
would discourage women from entering this trade
and encourage practitioners to find other sources of
income. It seems clear, however, that whereas legal
measures produce mixed results, urbanisation and
industrialisation predictably curtail female genital
cutting. In Nigeria, for example, one report shows
how the rise of western-style education has led to a
marked decline in female genital cutting. Busy with
book-learning and preoccupied by extracurricular
activities, girls have no time at puberty for lengthy
48
seclusion periods and ceremonies. As education,
employment, and marriage increasingly take place
away from girls' communities of origin, the power of
cultural tradition wanes and the incidence of female
49
genital cutting plummets.
I do not object to social and economic mobility
provided that individuals welcome it and benefit
from it, nor do I object to using law to bring about
ameliorative social change. Still, it is important to
point out that these two strategies are not necessarily
compatible with cultural perpetuation, and they do
not necessarily ensure increased autonomy for wo-
men. Endowing women with the right to penalise
female genital cutting practitioners pits women from
the same cultural group against one another and pre-
empts the possibility of their working together to
transform cultural norms and obtain economically
viable employment. Some women may gain auton-
omy. But their gains may be made at the expense of
other women's autonomy, and relying on the na-
tional judiciary to make these gains may short-circuit
cultural processes through which womanhood might
be redefined without shattering all continuity with
traditional conceptions of womanhood. While some
women's autonomy may be enhanced, their auton-
omy may be dissociated from their culture of origin.
More obviously, social and economic mobility
gained through geographic mobility often dissolves
networks of mutual recognition and shared beliefs
As and practices that are necessary to sustain cultures.
Migration often leads to assimilating prevalent
norms in one's new place of residence and work.
Assimilation weakens, if it does not sever, one's ties
to one's culture of origin. Some women who move to
urban areas and marry into families with different
cultural backgrounds gain greater autonomy by aban-
doning their own culture and becoming integrated
into a more congenial cultural milieu. Others suffer a
loss of autonomy. They find their cherished beliefs
denigrated or derided; they experience a sense of
dislocation and disorientation, and they are blocked
from enacting values they consider authentically
their own.
What these observations show, I would urge, is
how important it is to bear in mind the difference
between eradicating the practice of female genital
cutting and augmenting women's autonomy with
50
respect to female genital cutting. Ceasing to be
subject to a cultural imperative mandating female
genital cutting does not in itself constitute increased
autonomy. The educational programmes I described
may well try the patience of feminists who are (in my
opinion, justifiably) appalled at the harm inflicted by
female genital cutting. Yet, these programmes have a
distinct advantage from the standpoint of autonomy,
for they are designed to engage women's autonomy
216
 FEMINISM AND WOMEN'S AUTONOMY
skills without exposing them to autonomy-disabling
cultural alienation. In short, these programmes pro-
mote autonomy-within-culture.
Autonomy-within-Culture
Michelle Moody-Adams points out that successful
cultures must preserve people's capacities for the ex-
51
ercise of judgement and discretion. `Any culture that
worked to impair these capacities', she adds, `would be
52
creating the conditions for its own demise.' I agree
that a viable culture cannot turn its adherents into
indoctrinated automatons who cannot question cul-
tural beliefs and practices and who cannot instigate
53
cultural change. Still, I find Moody-Adams' view of
autonomy-within-culture overly sanguine.
A thriving culture must evolve, but it must persist
as well. If cultures are self-perpetuating systems, they
must have in-built mechanisms that shield their
beliefs and practices from criticism so zealous and
damning that it triggers cultural decline or foments
mass defection. The slightest acquaintance with hu-
man history confirms that cultures need not depend
on the justice of their beliefs and practices to secure
the loyalty of their adherents. Indeed, cultures, in-
cluding the most unjust ones, may have a willing co-
conspirator in human psychology. People commonly
exhibit a conservative bent ± preferring the known
over the unknown, even preferring the security of
having more or less mastered coping with a known
evil over the risk of being thrown off balance by
whatever might succeed it. Still, this conservative
disposition is to a significant degree culturally
abetted. Cultures ward off the perils of internal
dissension and disruption by circumscribing adher-
ents' exercise of autonomy skills. Adroitly channel-
ling exercise of these skills where they are needed
while limiting the scope of their application enables
cultures to evolve and endure.
Moody-Adams might reply that what I have de-
scribed is not impairment of capacities for judgement
and discretion. On the contrary, she might say,
cultures guide and modulate the exercise of auton-
omy skills, and rightly so. To some extent, I would
not take exception to this rejoinder. No doubt,
cultures exist because people cannot lead fulfilling
lives without stable systems of shared beliefs and
217
practices. Thus, serviceable cultures nurture those
autonomy skills that people need to get along within
their existing cultural framework and those that they
need to improvise adjustments when circumstances
render long-standing cultural beliefs and practices
untenable. However, in so far as cultures perpetuate
injustice by selectively suppressing autonomy skills, I
would insist that they are not merely regulating
people's capacity for judgement and discretion, but
rather that they are impairing it.
It seems to me that cultures do this in two main
ways. First, cultures furnish repertories of concepts
and interpretive schemas that focus perception and
shape reflection. Thus, cultures lead people to notice
some phenomena and overlook others, and they lead
people to ascribe certain meanings to their experi-
ence and to disregard other possible meanings. Sec-
ond, cultures valorise some autonomy skills over
others. Hence they commend child-rearing practices
that nurture these skills and establish social struc-
tures that reward individuals who have them, and
they let other autonomy skills languish. The fa-
voured autonomy skills are skills that enable people
to function well in their cultural context, and the
repertories of concepts and interpretive schemas
provide input for these deliberation skills that is
pre-selected and pre-processed in culturally conge-
nial ways. The disfavoured autonomy skills are skills
that might lead people to question the adequacy of
the culturally approved repertory of concepts and
interpretive schemas and, perhaps, to condemn their
cultural heritage.
For example, middle-class Euro-American culture
prizes means/ends rationality and vigorously culti-
vates the skills needed to pick goals with high
satisfaction yields and to plot successful goal-directed
campaigns. In this culture, however, valorised auton-
omy skills are gendered. Although child-rearing
practices and reward structures do not extinguish
means/ends rationality in middle-class Euro-Amer-
ican girls, their interpersonal skills are accentuated. I
confess I do not know which autonomy skills are
valorised in the innumerable cultures that practise
female genital cutting. But I believe it is possible to
infer which autonomy skills they devalue in women
by examining the education programmes that aug-
ment women's autonomy.
 DIANA TIETJENS MEYERS
Successful educational programmes mobilise wo-
men's introspection, empathy and imagination skills.
One programme invited women to explore their
feelings about their sexuality ± their sufferings, their
frustrations, and their disappointments, as well as
their pleasures ± and to sift through the possible
54
meanings of these stirrings. In effect, this pro-
gramme encourages women to acknowledge the
complexity of their emotional lives and to take their
own subjectivity seriously. Another programme ca-
pitalised on situations in which local girls and wo-
men had become infected or died and invited
women to empathise with the suffering of the in-
fected individuals and the grief of the families of girls
and women who died. Instead of rationalising the
pain and dismissing the sorrow as passing, women
were freed to regard such suffering as significant and
55
avoidable. Some programmes invite women to
imagine the lives of women whose cultures are
different but whose religion is the same as their
56
own. Sharing a religion builds a bridge to empathy
with these cultural strangers and provides a link that
legitimates using empathic understanding of these
other women's lives to reflect critically on one's own
experience and culture.
I suspect that these educational programmes owe
their success to synergy among introspection, em-
pathy and imagination. Empathy with others can
embolden one to confront and enrich one's inter-
pretation of one's own experience, and insightful
introspection can facilitate empathy with other peo-
ple's experience and equip one to interpret their
experience sympathetically and respectfully. As evi-
dence, let me offer a sequence I have observed.
Most middle-class Euro-Americans initially find it
virtually impossible to empathise with an African
mother who consents to and actively participates in
the infibulation of her daughter. Yet these indivi-
duals typically find it easy to empathise with an
American mother who consents to reconstructive
surgery on a daughter whose genitals are considered
ambiguous. Not only does one assume that surgery in
the US will be performed under sterile conditions
and that anaesthesia will be used, but also one is
oneself imbued with the same image of an appro-
priately gendered female body that is guiding the
American mother's decision. Thus one can readily
218
and vividly apprehend the American mother's dis-
tress about her daughter's purportedly unnatural
condition and her worries about the problems she
expects her daughter will encounter if her `patholo-
gical' genitals are left untreated. It is easy, then, to
understand her decision to authorise surgery. Indeed,
many middle-class Euro-Americans might find with-
holding permission to operate incomprehensible or
irresponsible.
Sometimes when I start a conversation with a
middle-class Euro-American acquaintance by elicit-
ing empathy along the lines I have just sketched and
by eliciting reflection on the reasons for the ease of
this empathy, it proves possible for my acquaintance
to transfer this understanding to the larger issue of
female genital cutting. Sensitised to the role that
western gender norms are playing in one's empathy
for the American mother, one now appreciates how
potent culturally specific feminine bodily norms are,
and one can sympathetically reconstruct how a
vastly different set of norms could figure in an
African mother's feelings and decision about infibu-
lation. Thus empathy with a fellow cultural initiate
and introspective reflection on how culture facil-
itates one's empathy for this individual make it
possible to empathise with a woman from a different
culture and to interpret her subjectivity and choices
charitably.
Narrow categories and prejudicial values often
distort or block introspection and empathy. Yet,
opportunities to examine one's subjective experience
with supportive associates, to discover hidden simi-
larities between others' experience and one's own,
and to talk about unorthodox ideas in a safe envir-
onment can correct and deepen both self-knowledge
and insight into others. Introspection and empathy
can expand one's repertoire of concepts and inter-
pretive schemas beyond the culturally furnished
stock. The educational programmes I have described
suggest that this struggle for introspective and em-
pathic access and for language that adequately ex-
presses one's new-found knowledge is best
undertaken as an open-ended, collective endea-
57
vour. Not only does interaction with others expose
individuals to novel viewpoints that may never have
occurred to each of them separately, but also the
sharing of conclusions and the process of reaching
 FEMINISM AND WOMEN'S AUTONOMY
them infuses these dissident views with an authority
they might otherwise lack. It seems, in sum, that
women who participate in these educational pro-
grammes gain intellectual tools and self-confidence,
and that they license themselves to recirculate their
discoveries and use them to refurbish or jettison
traditional beliefs and practices. Introspection and
empathy thus emancipate their imaginations. Beliefs
and practices that once seemed degraded and abhor-
rent can now be pictured as constituting an honour-
able and fulfilling way of life.
The capacity for imagination is complex, for it
recruits subsidiary skills and orchestrates them. In-
trospection and empathy can spark imagination, as
can dialogic and consensus-building skills. More-
over, imagination depends on the concepts one
has at one's disposal and the meanings one associates
with phenomena. Prescriptive norms shape imagina-
tion. Thus, cultures can limit imagination by furnish-
ing an impoverished repertoire of concepts and
interpretive schemas, by stunting the development
of skills that contribute to imagination, or by deva-
luing imagination itself. My hypothesis, then, is that
cultures that mandate female genital cutting either
altogether devalue imagination in women or else
value imagination in women only when it is serving
specific approved purposes. In the latter case, cul-
tures may foster the subsidiary skills that imagination
enlists only in so far as these skills promote the
condoned forms of imaginative activity.
A culture's ultimate defensive weapon is to make
alternative ways of life unimaginable or imaginable
only as bizarre or loathsome specimens. There is no
more effective way to galvanise cultural allegiance
than to suppress people's ability to imagine a pro-
foundly different life as a life they might gladly lead.
There is no better evidence, I would add, that
cultures typically perpetuate themselves by impairing
empathy with cultural outsiders and thwarting ima-
gination of cultural alternatives than the fact that
ethnocentricity and xenophobia are as rampant and
virulent as they are.
Whereas I am convinced that Moody-Adams
overestimates the autonomy proficiency that viable
cultures must secure, I am convinced that the two
restrictive, value-saturated theories of autonomy I
discussed earlier underestimate the damage to auton-
219
omy that alienation from one's culture and isolation
from other members of one's cultural group can
inflict. Still, I would be remiss if I did not acknowl-
edge that Babbitt and Benson bring out important
dimensions of women's autonomy with respect to
female genital cutting, and these insights into wo-
men's autonomy lend support to a skills-based, pro-
cessual account of autonomy.
Benson's chief contribution to a theory of wo-
men's autonomy over female genital cutting rests on
his observation that oppressive socialisation com-
partmentalises critical skills and receptivity to rea-
58
sons. His example is a woman who exercises critical
competence with respect to her education but not
59
with respect to her appearance. Similarly, we have
seen that cultures place female genital cutting be-
yond the purview of women's introspection, empathy
and imagination skills, yet they rely on women to
exercise these very skills in other areas of life such as
child-rearing. Although the importance of introspec-
tion, empathy and imagination for gaining autonomy
over female genital cutting suggests that it would be a
mistake to confine autonomy skills to critical reason-
ing and the need for cultures to deflect exercise of
autonomy skills away from culturally destructive uses
suggests that it would also be a mistake to suppose
that only oppressive enculturation compartmenta-
lises autonomy skills, it is true that cultures' self-
protective devices include compartmentalisation
and that these devices can have deleterious effects
on autonomy. Indeed, it seems to me that the success
of the educational programmes I discussed is due in
part to their effectiveness in decompartmentalising
and redirecting autonomy skills that women already
have.
Babbitt claims that altered settings can prompt
transformative experiences that enhance autonomy,
and this view is borne out to some extent by the line
of thought I have presented. I would take issue with
her, though, when she allows that foisting a new set
of opportunities on people is an ethical and effica-
60
cious way to increase their autonomy. That legal
reforms and social or economic upheaval produce
mixed results for women's autonomy over female
genital cutting raises doubts about the defensibility
of manipulating the context of women's choices.
Since autonomy-augmenting educational pro-
 DIANA TIETJENS MEYERS
grammes are typically developed by cultural initiates
who rely on traditional modes of expression and
appeal to traditional values, it is reasonable to sur-
mise that autonomy is best extended by avoiding
cultural alienation and by building on women's
61
existing autonomy skills. Still, I agree with Babbitt
that women's participation in these educational
programmes can be transformative, for breaking
through the barriers erected by a culturally trans-
mitted cognitive system can radically reconfigure
perception and open previously unthinkable vistas
of reflection and choice.
Successful educational programmes embody four
assumptions about autonomy:
1. Autonomy is best understood as socially situ-
ated ± that is, as autonomy-within-culture.
2. Autonomy couples self-discovery and avowal
with choice and self-redefinition.
3. Gaining autonomy consists of exercising a
complex set of skills.
4. One's authentic values and real desires are
those that emerge as one exercises autonomy
skills.
From their inception, autonomy-augmenting pro-
grammes regard women in cultures that practise
female genital cutting as self-determining indivi-
duals. Since it is obvious that these women have
some degree of proficiency with respect to auton-
omy skills, such as introspection and empathy,
before these programmes ever become available, it
is clear that they have some understanding of what
matters to them and how best to proceed in light of
Notes
1. Toubia (1995), p. 21; Obiora (1997), p. 298.
2. Gifford (1994), pp. 334, 361; James (1998), p. 1037.
3. Ogbu (1997), p. 414.
4. Obiora (1997), p. 298; Kouba and Muasher (1985), p.
104; Ogbu (1997), p. 414.
5. Diamond and Kipnis (1998); Fausto-Sterling (2000),
chapters 3 and 4.
6. For documentation and critique of these attitudes, see
Diamond and Kipnis (1998).
7. In the US today, we find gender parity where ambiguous
genitalia are concerned. Cosmetic surgery is also routi-
nely performed on genetically male infants whose pe-
220
their values and commitments before they partici-
pate in such programmes. Understanding autonomy
as I have proposed, then, makes sense of the claim
that women who are subject to female genital
cutting are not without autonomy. But this under-
standing of autonomy entails neither endorsement
of female genital cutting nor resignation to its
persistence. Since cultures that practise female gen-
ital cutting may have selectively nurtured and stifled
women's autonomy skills, and since these cultures
may shield the issue of female genital cutting from
exercise of autonomy skills, developing and coordi-
nating women's autonomy skills and extending the
range of application of these skills are key to
augmenting their autonomy. Individuals may iden-
tify points of self-estrangement ± doubts about
values or tensions between private feelings and
conduct ± and decide to revise their convictions
and openly oppose tradition. They may find them-
selves unconflicted about their cultural heritage and
renew their traditionalist convictions from a broader
perspective. Or they may pursue some nuanced,
intermediate path, embarking upon a process of
renegotiating and revitalising cultural traditions.
Although no uniform outcome is guaranteed, the
trend when the scope of women's autonomy is
expanded has been towards erosion of the practice
of female genital cutting. This skills-based, proces-
sual view of autonomy insists on respecting people
as self-determining agents and refuses to prejudge
what values and practices autonomous people can
endorse. Nevertheless, this view does not collapse
into indifference to values or cynicism about social
62
reform.
nises are considered too small or who lack testicles. It is
worth noting, though, that gender symmetry is not
necessarily the rule. Just as some cultures practise female
genital cutting but not male genital cutting, others cut
boys' genitals to bring about a masculine identity (some-
times the cutting is as radical as infibulation) but do not
cut girls' genitals; see James (1998), pp. 1041±3.
8. Kouba and Muasher (1985), p. 103; Obiora (1997), p.
297.
9. Obiora (1997), p. 296.
10. Ogbu (1997), p. 415; Kouba and Muasher (1985), p.
103.
 FEMINISM AND WOMEN'S AUTONOMY

11. Boddy (1989), pp. 57, 60. custom' and as keeping girls' sexuality under control
12. Sussman (1998), p. 21. until marriage, but they also asserted that it was `bad'
13. Boddy (1989), p. 52. because it was illegal and unchristian (Walley, 1997,
14. Kouba and Muasher (1985), p. 104. p. 411). One student who had `enthusiastically in-
15. Obiora (1997), p. 297; Ogbu (1997), p. 415; James vited' Walley to her sister's genital cutting neverthe-
(1998), pp. 1041±3. less declared that it was another way of `destroying'
16. Toubia (1995), p. 37. women's bodies (ibid.). As Walley observes, the stu-
17. Obiora (1997), pp. 288±9; Kouba and Muasher dents' arguments in favour of female genital cutting
(1985), p. 96. parrot local authorities' pronouncements on cultural
18. Walley (1997), pp. 408, 420; Narayan (1997), pp. 14± propriety, while their critiques parrot doctrines taught
17. in the official Social Education curriculum (ibid.). I
19. Meyers (1993), pp. 16±19; Moody-Adams (1994), pp. suspect that they saw the writing assignment as an
305±7. academic exercise which called upon them to recite
20. An interesting complication concerns the potential what they had learned, rather than as an opportunity
for dramatic development of the individual's erotic to reflect and reveal what they personally thought. If
responsiveness over the course of time. In this con- such exercises contribute to their autonomy, it is only
nection, it is worth noting that throughout our lives in virtue of maintaining their consciousness of dis-
we are obliged to make choices that, for better or parate views of female genital cutting. Perhaps they
worse, cut off other possibilities. Alas, we cannot appropriate and individualise some of the ideas they
always foresee what options we are eliminating, nor have been exposed to in their private conversations
can we always appreciate just how desirable foresee- and kvetching sessions.
able, forgone options really are. Rational choice the- 37. Obiora (1997), p. 361.
ory's requirement that choosers have full knowledge of 38. Ibid., p. 362.
the consequences of their choices is an ideal that 39. Ibid., p. 361.
usually cannot be approximated very well. Thus, a 40. Shaheed (1994), pp. 1009, 1011; Shaheed (1995), pp.
realistic theory of autonomy must work within the 80, 86.
limits of human prescience. 41. James (1998), p. 1046. This cultural change highlights
21. For development of this critique, see Babbit (1993) the importance of bringing men into the educational
and Meyers (1989), pp. 77±8. and cultural reconstruction processes. When and how
22. For critique, see Meyers (1989), pp. 33±41 and (2000), women choose to do this varies greatly from cultural
pp. 169±72; Benson (1991), pp. 391±4. group to cultural group.
23. Babbit (1993), pp. 246±7. 42. Obiora (1997), pp. 368, 371±3; Gunning (1997), pp.
24. Ibid., p. 248. 455±6.
25. Ibid., pp. 252±4. 43. Boddy (1989), pp. 52, 319.
26. Ibid., pp. 252±3. 44. Robertson (1996), pp. 616, 631.
27. Ibid., p. 253; emphasis added. 45. Obiora (1997), p. 357; Kouba and Muasher (1985),
28. Walker (1998), pp. 125±8. pp. 104±5.
29. Benson (1991), p. 385. 46. A'Haleem (1992), p. 155.
30. Ibid., pp. 388±9. 47. Ibid.
31. Ibid., pp. 396±7. 48. Ogbu (1997), p. 420.
32. Ibid., p. 396. 49. Ibid., pp. 420±1.
33. Ibid., p. 395. 50. It might be argued that eradicating female genital
34. I want to acknowledge that Benson realises that cutting is objectively in the best interests of the
oppressive socialisation does not necessarily rule out cultural groups that practise it, for repudiating the
autonomy in all aspects of the victim's life. He dis- practice would decrease disease and increase fertility.
cusses the possibility that a person's critical compe- Although this may be so, I believe that social change
tence can be compartmentalised ± that is, one can that is disconnected from individual autonomy is, as
exercise critical capabilities in one arena but be unable often as not, counterproductive. It often proves cul-
to exercise these capabilities in another; ibid., p. 397. turally destructive, for it alienates members of the
35. Kouba and Muasher (1985), p. 108; Obiora (1997), p. group. But when it is culturally advantageous, it often
361; Mugo (1997), p. 467. sacrifices some members of the group and their inter-
36. Mugo (1997), p. 467. Conventional educational ests to the interests of the larger social collectivity.
methods do not seem particularly fruitful. In her role Individual autonomy may not always be the most
as English teacher, for example, Kathleen Walley gave efficient way to promote cultural interests, but I am
her pupils the option of writing an essay about female convinced that it is the most humane way to advance
genital cutting. Many students defended it as `our this goal.

221
 DIANA TIETJENS MEYERS
51. Moody-Adams (1994), p. 307.
52. Ibid.
53. See Meyers (1993), p. 17.
54. For discussion of the need to pay attention to sub-
jectivity, see Gifford (1994), p. 338, and Meyers
(1989), pp. 79±89. For related discussion of `outlaw'
emotions, see Jaggar (1997).
55. For related discussion of empowerment, see Okin
(1998), p. 47; for related discussion of empathy, see
S. Bartky, `Sympathy and Solidarity: On a Tightrope
with Scheler', in Meyers (ed.) (1997) and Meyers
(1994). In an intriguing aside, Walley wonders whether
incidents in which girls reveal `cowardice' ± in one
cultural parlance, they `cry the knife' ± could be enlisted
to reconstruct values (Walley, 1997, pp. 418±19).
Might this ostensible cowardice be reinterpreted as
resistance?
56. Shaheed (1994), pp. 1009±10.
57. For relevant discussion, see Scheman (1993), chapter
222
3; `Outliving Oneself: Trauma, Memory and Personal
Identity', in Meyers (ed.) (1997).
58. Benson (1991), p. 397.
59. Ibid.
60. Babbit (1993), p. 256±7.
61. More generally, I would urge that transformative
experiences that do not engage autonomy skills are
extremely hazardous. Cults specialise in transformative
experiences, and some people who undergo transfor-
mative experiences get converted into neo-Nazis in-
stead of feminists. Although there is no way to
immunise people against such ill-fated transforma-
tions, autonomy skills are the best protection there is.
62. I would like to thank Ken Kipnis for a pre-publication
copy of the Diamond and Kipnis paper I have cited
and for his helpful comments on an earlier draft of this
chapter. I am also grateful to Berel Lang and Sara
Ruddick for their suggestions.
 Part Seven
LIBERALISM BEYOND
THE NATION-STATE
This page intentionally left blank
 17
CIVIL ASSOCIATION:
THE EUROPEAN UNION
AS A SUPRANATIONAL
LIBERAL LEGAL ORDER
Robert Bideleux
It is widely recognised that the integration theories
which have hitherto dominated studies of the Eur-
opean Union have failed to shed much light on the
nature of the nascent supranational polity, as distinct
from the processes that brought it about. Such
theorising has latterly been going round in circles
and repeating itself, without getting us much closer
to a satisfactory conceptualisation of the order that is
emerging in Europe. In order to gain a surer under-
standing of the new political order, it is necessary to
draw upon and creatively adapt elements of main-
stream political theory. This chapter attempts to
develop a different way of discussing and thinking
about the emerging European order, using a language
which owes rather more to the writings of two
eminent liberal thinkers, Friedrich Hayek and Mi-
1
chael Oakeshott, than to integration theory. They
would perhaps have been surprised, but I hope not
too displeased, at the ways in which their ideas and
concepts have been deployed.
Democracy and the Liberal Legal Order
It has always been imprudent to treat the `liberal' and
the `democratic' components of liberal democracy as
if they were two halves of the same walnut. In most
western states, as is well known, the advent of liberal
225
democracy was preceded and made possible by sev-
eral decades of limited government, liberal consti-
tutionalism, the rule of law, and the separation of
public and private as well as political and economic
spheres. The liberal institutions, procedures, prac-
tices and patterns of conduct which paved the way
for liberal democracy have subsequently helped to
safeguard most western democracies against the perils
of unconstrained majoritarianism. The essential
point is that liberty is safeguarded not so much by
democracy as such as by liberalism, limited govern-
ment and the rule of law, which long antedated and
did not depend upon the existence of democracy. It
might be objected that, when democrats and theor-
ists of democracy and democratisation talk about
`democracy', they are merely using that word as a
form of shorthand and take it for granted that
democracy must be buttressed by liberalism, limited
government and the rule of law, without which a
system based upon freely contested multi-party elec-
tions and a broad suffrage is unlikely to function well
or survive for long. However, this shorthand is
misleading, in that it appears to suggest that the
crucial safeguards of liberty are the specifically de-
mocratic components ± procedural democracy (pri-
marily the holding of regular elections), the political
mobilisation of the demos and some form of repre-
 ROBERT BIDELEUX
sentative and/or participatory government ± rather
than the liberal elements. In fact, inasmuch as
democracy has an in-built tendency towards unlim-
ited government, because majorities (however tem-
porary) generally hope and expect to get their own
way and often resent any obstruction of their will by
law courts, second chambers, heads of state and other
checks and balances, democracy as such not infre-
quently poses major threats to the liberties which are
in practice upheld by liberalism, limited government
2
and the rule of law. `Liberalism' here refers not only
to the classic liberal political and economic doc-
trines, but also to the belief that societies ± in the
broad sense ± should primarily be based upon and
held together by impersonal law, interests, calcula-
tion and contractual relations of a Lockean kind,
rather than by `primordial' ties of religion, race,
ethnicity and kinship. It will be contended that
religious and ethnic collectivism are fundamentally
at odds with the cosmopolitan, supranational liberal
legal order which is being constructed by the Eur-
opean Union.
These distinctions are of crucial importance. They
enable us to see more clearly that the political
performance and EU membership prospects of states
such as Serbia, Croatia, Albania, Romania and
Bulgaria are being impaired not so much by a lack
of democracy (they actually have executives and
legislatures which have been more or less freely
and fairly elected by universal suffrage in multi-party
contests) as by insufficient respect for minorities and
for the rule of law, by inadequate constraints on the
exercise of power by the state and its officers, and by
the weakness or shallow roots of liberal doctrine and
sentiment. Conversely, what has placed countries
such as Hungary, the Czech Republic and Poland at
the front of the queue for membership of the EU is
not a greater commitment to democracy as such, but
rather their stronger commitment to the rule of law,
their greater respect for the liberties of individuals,
their stronger constraints on the power of the state,
and their fuller understanding of the requirements of
liberal governance. These same qualities also explain
the greater success of their transitions to market
economies.
226
The European Union as a Liberal
Rather Than a Democratic Project
The European Union, crucially, is a liberal project
rather than a democratic one. It rests upon a
supranational system of rules, regulations and laws
engendered by judicial and quasi-judicial processes
as well as by highly consensual negotiation-cum-
horse-trading between liberal democratic states.
Liberalism, rather than democratic theory, is the
key to understanding what the European Union is
and how it works. Admittedly, European federalists
in the mould of Altiero Spinelli and Mario Alber-
tini wanted European unification to assume the form
of a supranational European democracy structured
3
as a federal European state, but such a path has
always been rejected by both public opinion and the
powers that be. The constitutive treaties of the
European Communities and the European Union
have been ratified by the parliaments (and in some
cases the electorates) of each of the member states,
while successive treaties have acquired the status of
a constitution or fundamental legal charter. Never-
theless, the Union does not rest upon direct demo-
cratic legitimation at a supranational all-Union
level. The Union has a telos, but there is no nascent
European demos standing above the separate demoi
of the constituent member states, nor is there a
transnational EU `public opinion'. There is very
little articulation and aggregation of interests at a
supranational level ± only business interests are well
4
organised in this respect. There are no truly `Un-
ion-wide' political parties fighting EU elections on
transnational `European' platforms and issues. Elec-
tions to the European Parliament are characterised
by low and steadily declining turnouts and minimal
public engagement with `European' as distinct
from `national' programmes, policies, personalities
and issues. Moreover, the Union's institutional
design
does not conform to the traditional separation of
powers between legislative, executive and judicial
branches of government. Rather, decisions evolve
from intense bargaining within and across the
policy-making institutions . . . The Union is
not a traditional hierarchy with a clearly defined
 CIVIL ASSOCIATION AND EUROPE
centre of political authority . . . As a result,
Europe's citizens have difficulty in identifying
`who governs' in the Union and cannot . . .
5 7
dismiss them at elections.
Increased `openness' and `transparency' may be de-
sirable in their own rights but they do not of
themselves reduce the so-called `democratic deficit'
of the European Union. They skirt round rather than
6
tackle the problem head on.
As a polity, the European Union and its policy-
making processes have been based upon mutual
accommodation of elites, rather than on specifically
democratic procedures. It has been constructed, not
by democratic mobilisation of peoples in support of
`ever closer union' and the `idea of Europe', but
rather by consensual negotiation and horse-trading
between European governments and officials. There
have been no great waves of popular support for a
united Europe ± certainly none that are even remo-
tely comparable to the periodic populist backlashes
against the idea. From the outset, the construction of
the European Communities and the EU has been the
work of cosmopolitan elites pursuing their own elite
agendas, rather than a response to popular demands
for European integration. However, this elitism has
been not a weakness, as many critics complain, but
one of the Union's strengths. If it had been necessary
to wait for its peoples to voice demands for `ever
closer union', Western European integration would
not have advanced nearly so far as it has done in so
short a time. The crucial point is that European
states would have had to devise ways and means of
coexisting more peacefully and harmoniously, no
matter what the subjective preferences, prejudices,
bugbears or grievances of their citizens, in order to
avert any potential return to the destructive inter-
and intra-state conflicts and beggar-my-neighbour
protectionism of Europe's `Second Thirty Years
War' (1914±45). These matters were too important
to be submitted to the vagaries of public opinion. In
any case, as argued by John Gray, the current shrill
and single-minded insistence on democratic legiti-
macy is somewhat over the top. In practice, forms of
governance `are legitimate in so far as they meet the
needs of their citizens. Those that fail in this will be
judged by their citizens to be illegitimate whether or
227
not they are democracies.' Thus it is not whether a
polity `is a liberal democracy that most fundamen-
tally determines its legitimacy; it is how well it
secures its citizens against the worst evils.'
Post facto popular approval of what has been
achieved has been a bonus rather than a precondi-
tion for the success and the legitimacy of an elite-
driven integration project which has contributed so
strongly to the peace and prosperity of Western
Europe since the 1950s. The latter stands in marked
contrast to the devastating and debilitating conflicts
which resulted at least in part from the democratic
mobilisation of ideological mass movements during
the first half of the twentieth century. West Eur-
opean elites have in effect succeeded in putting the
destructive genies of mass politics back into their
bottles, thereby making Western Europe a much
freer, safer and more prosperous place than it had
been earlier in the twentieth century, and than
south-eastern Europe and the former Soviet Repub-
lics still were at the end of it.
The frameworks of EU law and regulation which
increasingly govern the Single Market, environmen-
tal policies, health and safety provisions, the terms
and conditions of employment, monetary policy,
industrial policies, regional policies and agricultural
policies have transformed the EU into a multi-tiered
`regulatory federation' which increasingly defines
8
and upholds the rights and obligations of its citizens.
The state-like institutional skeleton of the EU has
fostered growing expectations that it should behave
and grow `like a state'. Relative to other multi-tiered
federal systems of governance, however, `the Eur-
opean framework has strikingly modest resources of
its own. Consequently, there are major constraints
upon the nature of the activity that can be pursued.'
In particular, there is very limited scope for redis-
tributive measures, which in turn virtually imposes
9
the emphasis on regulatory governance. The EU is
emerging as a legal federation, as an economic
federation, and as a regulatory federation, but it lacks
either the design or the wherewithal to discharge the
classic Hobbesian and Weberian functions of a state
± especially the monopoly of organised violence ±
and it can and should be prevented from acquiring
them.
Within the European Union, political and civil
 ROBERT BIDELEUX
rights and liberties are being ever more deeply em-
bedded in a strong supranational legal and constitu-
tional framework, albeit one which in the last resort
can only be upheld by national law courts and national
police forces ± as the EU has no national-level courts
or police of its own. This supranational legal order
helps to safeguard rights and liberties against poten-
tial violation by various powerful actors, including
the governments of member states which still control
national armed forces, police forces, law-courts, na-
tionwide bureaucracies and other legitimate instru-
ments of enforcement and surveillance at that level.
Contrary to the fears and alarmism of the Euro-
phobes, the EU does not itself control or possess
any such forces with which it could conceivably
coerce those citizens, enterprises and member states
who occasionally violate its rules and laws, although
it has generated ever-growing political and economic
incentives for compliance as well as penalties for
non-compliance. The only conceivable direct
threats to liberty within the European Union come
from states, large companies, organised crime, terror-
ists and private security forces rather than from the
European Union as such. The European nation-state
has lost some functions while gaining others `in what
had previously been more private and local spheres',
but overall `the bars of the cage have not changed
very much. Citizens still need to deploy most of their
10
vigilance at the national level.' So long as the
British and the Scandinavians continue to hold out
against the aspirations of some member states to
create a strong supranational European army or a
supranational European police force with a strong
capacity for electronic surveillance of (and data
banks on) EU citizens, the possibility of the EU
becoming an Orwellian `big brother' superstate will
remain nothing more than a figment of the `black
propaganda' put about by prominent British `Euro-
phobes' such as Margaret Thatcher, Norman La-
mont, William Hague and John Redwood.
Even the external relations of the European Un-
ion have become increasingly enmeshed in a supra-
national framework of rules and regulation. The
Common Commercial Policy, the Common Exter-
nal Tariff, the Common Agricultural Policy, the
Common Fisheries Policy, the successive Â
Lome
Agreements with successive African, Caribbean
228
and Pacific countries, the Europe Agreements with
former Communist states, the Euro-Med Partnership,
international cooperation on environmental regula-
tion and the inchoate Common Foreign and Secur-
ity Policy: all have involved the creation of
mechanisms and procedures through which the ex-
ternal economic and (increasingly) non-economic
relations of EU member states can be continuously
and intensively coordinated. Indeed, on most mat-
ters pertaining to external economic relations (espe-
cially foreign access to the Single Market), the EU
Commission is empowered to act and to negotiate
agreements on behalf of the member states. More-
over, as Michael Smith has emphasised, `the see-
mingly ``domestic'' concerns of market regulation,
standards setting and competition policy' have in-
creasingly become `the stuff of international poli-
11
tics'. The mounting importance of economic and
regulatory matters in international relations and in
regional security, together with the launching of the
Single Market programme, has steadily increased the
number and importance of the areas within which
EU institutions have supranational competence to
act on behalf of the member states. Even before the
end of the Cold War, the security of the EU and its
member states was becoming less and less concerned
with traditional questions of war and peace and
military capabilities, and increasingly to do with
economic security, market access, fishing rights, en-
vironmental regulation, human rights, minority
rights, immigration issues, drug trafficking and inter-
national crime. As argued by Michael Mann, during
the 1990s Western Europe gained `the geopolitical
security it has always wanted, as well as client states
and even supplicant states between itself and any
threat coming out of the East.' So long as such
conditions continued to pertain (and here both
Mann and the EU were in danger of complacency),
the EU would have `little need for efficient geopo-
litical decision-making'. The EU represents a `Lock-
eian heartland' of civil association and contractual
relations, although not far beyond its southern and
eastern borders (for example, in the Maghreb, Egypt,
Bosnia, Kosovo and south-eastern Turkey) one
12
comes into more `Hobbesian' terrain.
The mere existence of the EU has helped to
transform the very nature of European external
 CIVIL ASSOCIATION AND EUROPE
relations. Much of what used to be regarded as foreign
policy or international relations (especially the often
bellicose relations between European states in the
past) has been `domesticated' and `civilianised' ±
becoming part of the civil order and internal relations
of the EU, dealt with by institutions and mechanisms
formerly associated with `domestic' matters. The EU
comprises a complex mesh of institutions, rules and
laws which structure both the internal and external
dealings of the EU as a whole and its individual
member states, as a result of which there is `a
constant rule-governed process of negotiation be-
tween actors which produces policy positions and
13
international policy outcomes.' The scathing pub-
licity given to the Union's inability to project im-
perialistic military might in the manner of the
United States has diverted public attention from
the large areas of consensus and joint activity in
the Union's external relations, and from the fact that
the EU as a vast and economically powerful `civil
association' exerts enormous influence and `magnetic
pull' on the states and peoples lying to the south and
east of its current borders.
From `Democratic Deficits'
to `Cosmopolitan Democracy'?
Many critics of the European Union regard the fact
that it is a supranational liberal legal order, rather
than a form of democracy, as a woeful deficiency. Yet
in reality that is its greatest virtue. Critics decry the
EU's so-called `democratic deficit', as if either its
present constitution or even the very existence of the
European Union were the intrinsic source of the said
deficit. In fact, it is the development of mighty global
market forces and global information and commu-
nications networks, rather than the increasingly
federal character of the European Union, that pose
the greatest threats to `sovereignty' and democratic
control and accountability. This is so, not only at the
national level, but also indirectly at the EU level.
Globalisation and the diminishing capacity of either
national or EU parliamentary and executive institu-
tions effectively to control and hold to account the
forces of global capitalism and global information
and communications networks have created demo-
cratic deficits at all levels. This is not to deny that
229
transnational capitalism and information/communi-
cations technologies can in many ways help to
promote the geographical diffusion of the rule of
law, liberal values and liberal democratic forms of
governance. But it should be pointed out that they
also give rise to market forces and concentrations of
power which are increasingly beyond the control of
the individual states or groups of states which remain
the principal sites of democratic accountability. This
growing democratic deficit afflicts states which are
not members of the EU every bit as much as (and
often more than) those within it, and therefore should
not be simplistically or demagogically `blamed' upon
the EU. Indeed, during the 1990s the economies of
many non-EU states were even more severely buf-
feted than those of EU states by global market forces
over which their political and monetary authorities
have little control. Public allegiance to the European
nation-state persists, but this is increasingly out of
kilter with the actual locus of economic and military
power and decision-making, which has to a large
extent shifted to higher levels (not just to the EU,
but to NATO, the WTO, the IMF, other interna-
tional agencies and giant transnational corpora-
tions). Moreover, inasmuch as the `national'
governments of the member states of the EU con-
trolled 52 per cent of GDP (on average) in 1993,
whereas the institutions of the EU controlled a mere
1.3 per cent of their combined GDP, `democratic
deficits' actually pose much more of a problem at the
national than at the EU level.
Even though there has been a lot of influential
wishful thinking about supranational forms of `cos-
mopolitan democracy', much of it by eminent poli-
14
tical theorists, no one has convincingly
demonstrated any way in which it might be possible
to establish direct democratic accountability and
control over governance at the supranational level.
An important body of supranational `cosmopolitan
law' has already emerged (most notably in the form
of European Union law), but `cosmopolitan democ-
racy' remains a mirage. It is likely that any thorough-
going attempt to subordinate all supranational
decisions, policies and regulations directly to detailed
scrutiny and control by `sovereign' national parlia-
ments would induce political paralysis or deadlock,
while attempts to subordinate them to `sovereign'
 ROBERT BIDELEUX
supranational elected bodies could lead to excessive
centralisation of power and to a real possibility of the
Orwellian `superstate' so feared and berated by Brit-
ish `Europhobes'. Although the overriding aim of
European federalists in the tradition of Altiero Spi-
nelli and Mario Albertini was to transcend the perils
and defects of the European states system, the crea-
tion of a European superstate would merely repro-
duce many of those self-same defects (and others
besides) on an even larger scale than before. That
would be to jump out of the frying-pan into the fire.
In a European `superstate', whether unitary or fed-
eral, the police forces, the bureaucracies and the
military forces holding sway over Europe would be
even larger and even harder to subordinate at all
times to democratic control and accountability than
are their present-day `national' counterparts. Thus,
although the European Union and its member states
do indeed suffer from democratic deficit, these would
not be significantly reduced either by creating a
European `superstate' or by reasserting full national
sovereignty and autonomy ± even if either of these
options were to be politically feasible.
The European Union as presently constituted, far
from being the intrinsic source of Western Europe's
democratic deficit, is the most promising means of
limiting and mitigating the potential negative conse-
quences of the democratic deficits caused by the
expansion of transnational capitalism, communica-
tions networks, bureaucratic `big government' and
transnational military bureaucracies beyond demo-
cratic control. As the doyen of American democratic
theorists has emphasised:
The boundaries of a country, even one as large as
the United States, have become much smaller
than the boundaries of the decisions that signifi-
cantly affect the fundamental interests of its
citizens . . . Thus the citizens of a country cannot
employ their national government, and much less
their local governments, to exercise direct control
over external actors whose decisions bear criti-
cally upon their lives.
The European Union, in Robert Dahl's view, affords
its citizens some mechanisms of indirect political and
legal monitoring, regulation and control over the forces
230
that increasingly govern their lives. The citizens of
small and medium-sized states in other parts of the
15
world have no such means at their disposal. Thus:
except for the European Union, the prospects for
even moderately `democratic' governments of
transnational political associations are poor. Even
if transnational political systems are greatly
strengthened, for a long time to come decisions
are likely to be made by the delegates appointed
by national governments. Thus the link between
the delegates and the citizens will remain weak;
and the democratic process will be even more
attenuated.
Dahl suggests that disgusted democrats should seek
solace in enhanced participation and deliberation in
the national and subnational democratic processes
and institutions which could indeed exercise `what-
ever democratic control may be possible over the
authority delegated to transnational decision makers'
and thus help to sustain a `healthy democratic
political life within the large sphere of relative
autonomy that democratic countries will still pos-
sess', including increased local control over such
matters as education, town and city planning, public
16
amenities, and social and health services. Half a
loaf is better than none. It is mainly at these lower
levels that more `radical' conceptions of democratic
participation and control, including `electronic de-
mocracy' or `cyberdemocracy', are most relevant and
17
realisable. It is sometimes claimed that the devel-
opment of electronic democracy offers a means of
overcoming the EU's so-called democratic deficit by
creating a `wired-up Europe' (or, through the next
generation of information technology, a `wireless
Europe'), by connecting the European Union's citi-
zens directly to the EU institutions and one another,
by-passing national governments. However, it is
more likely that the new technologies will merely
reinforce elite dominance and magnify the inequal-
ities of power and influence even further, because it
is Europe's elites that have the greatest resources for
doing this and that are most adept and assiduous at
deploying them to their own advantage.
 CIVIL ASSOCIATION AND EUROPE
The EU Legal Order
and Limited Government
The French legal scholar L. Cohen-Tanugi has
argued that two conceptions of democracy currently
coexist in Western Europe: the Jacobin tradition of
`elective democracy founded solely on universal
suffrage', and that of the constitutional state
founded on higher principles of law, equity and
procedure . . . Like other organisations of a fed-
eral type, the Community is building itself by
law . . . European integration has promoted the
`return of law' and the promotion of a democratic
ideal which does not draw its legitimacy solely
from the law of the majority, but also from
fundamental principles ± those of the European
Convention of Human Rights, for example ± and
from a system of judicial recourse largely open to
18
citizens against national states.
In the words of J. Temple-Lang, each member of the EU
now has two constitutions: its national constitu-
tion and the Community constitution, which it
shares with other member states. Member states'
legislative, executive and judicial powers under
their national constitutions can be exercised only
within the framework and limits imposed by
19
Community law.
Ian Loveland, another legal scholar, contends that
EC Treaties stand in marked contrast to our own
domestic law-making process. They are not the
product of majoritarian or even super-majoritar-
ian law-making. Their every provision has been
arrived at through a consensual negotiatory
process . . . The prospect of EC law being nar-
rowly majoritarian, and hence oppressive or irra-
tional, has been reduced almost to vanishing
point. In that sense, it is a `higher' form of law
than can be produced by any of the EC's member
20
states within their own legal systems.
The European Union thus represents an original
and constructive solution to what Friedrich Hayek
231
described as `the most crucial problem of political
organisation, namely how to limit the ``popular will''
21
without placing another ``will'' above it'. The
exigencies of accommodating the ever-increasing
political, economic and cultural diversity of the
membership of a steadily expanding EU further
ensure that it will have to remain a polity based
on limited government, consensual negotiation and
the rule of law ± indeed, unless it is held together by
an iron frame of law, it could burst apart at the seams.
Of course, even this might not be enough. Never-
theless, the national and/or religious ties on which
nation-states (whether liberal democratic or other-
wise) have hitherto implicitly relied to foster the
active commitment and loyalty of citizens to their
polities are being steadily loosened and eroded by
globalisation, regional integration, multiculturalism,
acceptance of difference and the growing fragmenta-
tion of political and cultural identities.
Squeezed between a globalised economy and
aggressively introverted cultures, the political
space is fragmenting and democracy is being
debased ± reduced, at best, to a relatively open
political marketplace, which no one troubles to
defend because it is not the object of any intel-
lectual or affective investment . . . The danger
here is that [multi]culturalism, in the name of
respect for differences, will encourage the forma-
tion of localised power that will impose an anti-
democratic authority within particular
communities. Should that happen, political so-
ciety will become no more than a marketplace
where communities locked into an obsession with
their identities and internal homogeneity enter
22
into loosely regulated transactions.
So far, however, global capitalism and regional in-
tegration have seemed incapable of fostering feelings
of belonging and allegiance to replace those which
they are helping to dilute and disempower at the
level of the nation-state. In such circumstances,
political order can only be maintained within the
broader framework of a large and consensual polity
whose laws are impartial in the specific sense of being
incapable of being merely the expression of the
interests and/or wishes of a dominant majority group
 ROBERT BIDELEUX
and whose function is not to promote agreement on
substantive goals and projects but rather to allow a
great diversity of persons and states the maximum
scope for defining and pursuing their diverse goals
and projects while minimising the scope for friction
and mutual interference.
Significantly, in an article published in 1939,
Hayek argued that `the abrogation of national sover-
eignties and the creation of an effective international
order of law is a necessary complement and the
logical consummation of the liberal programme', if
only because `an essentially liberal economic regime
is a necessary condition for the success of any inter-
23
state federation.' Indeed, since the citizens of such
a federation `will be reluctant to submit to any
interference in their daily affairs when the govern-
ment is composed of people of different nationalities
and different traditions', it follows that `the govern-
ment in a federation composed of many different
peoples will have to be restricted in scope.' Since the
full realisation of the potential economic benefits of
such an entity would in Hayek's view entail eco-
nomic and monetary union, there would have to be
`less government all round' if such a federation was to
be practicable. Economic policy would then `have to
take the form of providing a rational permanent
framework within which individual initiative will
24
have the largest possible scope.' Hayek's linking of
inter-state federation to limited government and
projects for economic and monetary union pres-
ciently anticipated what were to become the most
distinctive features of European Union governance.
Of course, the European Union is far from being a
perfect embodiment of liberal governance. Its ten-
dencies towards corporatism, clientelism, corruption
and a `fortress Europe' increasingly concerned with
keeping out unwanted `aliens' compromise its liberal-
ism. Nevertheless, its central emphases on the rule of
law, the uniform adoption and enforcement of stable
and predictable rules and the creation of an extended
spontaneous legal order are bedrocks of liberalism on
which a more completely liberal and cosmopolitan
order is gradually being built. The horse-trading,
deals and manoeuvres which take place at or prior
to meetings of the European Council and the Coun-
cil of Ministers often appear unseemly and cynical.
Yet they are intrinsic to the consensual adoption or
232
extension of uniform rules, regulations and laws
which all the participants can live with and the
placing of the EU's policies, functions and activities
on an increasingly law-governed footing.
The European Union
as an Oakeshottian `Civil Association'
According to the British philosopher Michael Oa-
keshott, the proper business of government is `to try
to keep its subjects at peace with one another in the
activities in which they have chosen to seek their
happiness.' Therefore:
the only appropriate manner of ruling is by mak-
ing and enforcing rules of conduct . . . which, in
the circumstances, are least likely to result in a
collision of interests . . . Thus governing is re-
cognised as a specific and limited activity; not the
management of an enterprise, but the rule of
those engaged in a great diversity of self-chosen
25
enterprises.'
The rule of law is, in Oakeshott's view, `a method of
government remarkably economical in its use of
power' and, `while losing nothing in strength, is
itself the emblem of that diffusion of power which
it exists to promote'. It is `the greatest single condi-
26
tion of our freedom'.
Oakeshott's conception of civil association was
put forward in contradistinction to `enterprise asso-
ciation', in which the state is perceived as an orga-
nisation for the attainment of prescribed ends which
are incumbent upon its citizens (as in the Thatch-
erite project to transform the UK into UK plc, of
which we were all to have become compliant em-
ployees, as well as in the more overtly teleological
`planned economies'). John Gray has incisively
summed up Oakeshott's conception of a civil asso-
ciation as `an association of persons who, having no
ends or purposes held necessarily in common, never-
theless coexist in peace under the rule of law. On this
account, the office of law is not typically to impose
any particular duty or goal on men, but instead seeks
principally to facilitate their dealings with one an-
27
other'. The distinctive feature of the Oakeshottian
concept of civil association is that `the rules which
 CIVIL ASSOCIATION AND EUROPE
constitute it are non-instrumental: they do not em-
body any specific project, but instead act as condi-
tions whereby individuals and groups pursue their
own several and diverse projects. In its moral aspect,
civil association is
that mode of association which exemplifies in-
dividuality ± the condition in which human
beings accept and celebrate their autonomy . . .
In its political dimension, civil association is
characterised by the diffusion of power through-
out society ± by a complex structure of counter-
vailing institutions of precisely the sort that is
threatened by contemporary collectivist projects
(of the Right and Left) of transforming govern-
28
ment into an enterprise association.
Terry Nardin has drawn upon Oakeshott's distinc-
tion between civil and enterprise associations to
fashion a similarly useful distinction between `prac-
tical' and `purposive' conceptions of international
society and of international associations:
Practical association is a relationship among those
who are engaged in the pursuit of different and
possibly incompatible purposes, and who are as-
sociated with one another . . . only in respecting
certain restrictions on how each may pursue his
own purposes . . . Purposive association is a
relationship among those who co-operate for
the purpose of securing certain shared beliefs,
values and interests, who adopt certain practices
as a means to that end, and who regard such
practices as worthy of respect only to the extent
that they are useful instruments of the common
29
purpose'.
Admittedly, any association is created to serve a
purpose and is in that sense purposive but:
The purposes of practical association, however,
are of an altogether different order from those
served by purposive association . . . The values of
practical association . . . are those appropriate to
the relations among persons who are not neces-
sarily engaged in any common pursuit but who
nevertheless have to get along with one another.
233
Hence values such as legality, morality and justice
are `best regarded not as ends but as values embodied
in the constraints governing all action . . . Even
peace is better regarded as a constraint than as a
purpose', since it requires not the avoidance of all
uses of force but rather an acceptance that `force is to
30
be used only in authorised ways'. Thus purposive
associations are constituted by the ends they serve or
pursue, whereas practical associations are based upon
the forms of conduct which they either authorise or
prohibit. In these terms, international society and by
implication most forms of international association
cannot be `purposive' (even though they are often
seen as such), on account of the great diversity of
31
goals and aspirations that exists among states.
`International society' is best conceived as an
inclusive `practical association' geared to the main-
tenance and observance of certain practices and rules
of conduct designed to minimise the risks of armed
conflict and to maximise the scope for peaceful
coexistence between states engaged in the pursuit
of widely differing goals, values, beliefs and projects.
Some international associations are purposive;
they exist for the sake of pursuing various shared
purposes; the rules they adopt and the ethic they
foster are instrumental to the achievement of the
purposes for which they were created. But inter-
national society as such ± that inclusive society of
states, or community of communities, within
which all international association takes place
± is not a purposive association constituted by
a joint wish on the part of all states to pursue
certain ends in concert. It is, rather, an associa-
tion of independent and diverse political com-
munities, each devoted to its own conception of
the good . . . The common good of this inclusive
community resides not in the ends that some, or
at times even most, of its members may wish
collectively to pursue but in the values of justice,
peace, security and coexistence, which may be
enjoyed through participation in a common body
32
of authoritative practices.
In Nardin's view, international society has to take a
form that is capable of accommodating culturally and
ideologically heterogeneous states. This kind of in-
 ROBERT BIDELEUX
ternational society first emerged in Europe following
the sixteenth- and seventeenth-century wars of re-
ligion, when it became necessary
to work out principles of coexistence among
individuals and groups committed to different
conceptions of truth and the good . . . Thus
the kind of legal order that developed in Europe
was shaped by the gradual discovery that, in the
absence of agreement on ends, agreement on
procedures is required if destructive conflict is
33
to be avoided.'
Therefore, it is also necessary to
reject all extreme claims that shared purposes [or
values or beliefs] are a necessary condition of
international order . . . International law, and
especially customary international law, embodies
the minimum principles of mutual accommoda-
tion required for the coexistence not only of
separate states but of different kinds of states . . .
It is through the experience of international
association defined and directed by this common
law that an appropriate international culture
34
bridging other cultural differences is created.
Part of the attraction of the terms of reference
employed by Oakeshott, Hayek and Nardin is that
they eschew the impersonal pseudo-scientific abstrac-
tions involved in analyses of behaviour, forces, pro-
cesses, variables and correlations, and speak instead of
conduct, society, law and individuals as moral agents
who deliberate and make reasoned judgements and
choices. As Nardin puts it, his concern is
not with international relations understood as a
system of causally or functionally related variables
but with the arrangements of an international
society constituted by the actions of thinking
agents who must take each other into account
in making decisions . . . and whose acts are
understood as being shaped and guided by rules
35
of conduct rather than laws of behaviour.
Although Nardin focuses on relations between
states in the world at large rather than on regional
234
international associations such as the EU, most of his
insights can be applied to Europe as well. In practice,
the participants in the EU are united not so much by
engagement in common endeavours (however much
they say they are!) as by their acceptance that they
are members of an association defined by a common
set of rules and laws designed to allow each of them
to pursue their own goals and ambitions with the
minimum of friction and mutual impairment.
Given the existence . . . of significant differences
of belief, value and interest, the expectation
of . . . unity on the basis of shared purposes is
sure to be disappointed. The basis of international
association lies in deference to practices that
embody recognition of the fact that we must
coexist . . . with others with whom we some-
times share little beyond a common predica-
36
ment.
The fact that the European Union lacks a single
centre of power, decision-making and authority but
instead has a multiplicity of such centres has been
conducive to the dispersal of power, authority and
functions which Oakeshott regarded as the chief
safeguards of liberty. Further, the fact that the Eur-
opean Union's budget has not been allowed to
exceed 1.27 per cent of the combined GDP of its
member states has in effect forced it to remain an
Oakeshottian `civil association', even though some
observers consider that it harbours (in my view
unrealisable) ambitions to become an `enterprise
association' in certain sectors. The Single Market
programme, the EU's growing social and environ-
mental regulation responsibilities and the project for
Economic and Monetary Union (EMU) are major
extensions of the law-governed regulatory federation
and civil association characteristics of the EU, rather
than great leaps towards the creation of `big govern-
ment' at the supranational level. Indeed, the Eur-
opean Union is governed by impersonal rules and
laws to a far greater extent than any `really existing'
democracy. Contrary to the claims of British Euro-
phobes, EMU is governed by stringent rules and a
regulatory committee of bankers rather than by an
emergent European superstate with major redistribu-
tive capabilities. The EU corruption scandals of 1998
 CIVIL ASSOCIATION AND EUROPE
and 1999 highlighted how little (rather than how
much) power is wielded by individual commissioners
and other officials. It is not only power that corrupts,
so does powerlessness.
The main virtue of the creation of a supranational
liberal legal order which functions in the manner of
an Oakeshottian `civil association' rather than a
more purposive, redistributive and potentially coer-
cive `enterprise association' is that it erects a strong
supranational framework of rules and laws (a form of
`governance without government') which can reg-
ulate political and economic activities and power
without fostering a European `superstate' with more
obviously Hobbesian, Weberian or Orwellian fea-
tures and potential. Another of the virtues of such a
civil association is its ability to maximise the cer-
tainty, equity and transparency of the rules govern-
ing economic, social and environmental matters,
above all through the equal application of its rules
and laws to all. Rule certainty and stability of ex-
pectations lie at the heart of the Single Market
programme and Economic and Monetary Union,
from which Europe's hitherto inflation-prone and
relatively unstable less developed peripheries poten-
tially have much more to gain than do the already
stable and disciplined `core' states, which have been
enjoying low levels of inflation and correspondingly
37
low `risk premia' on their interest rates for decades.
In this situation, the case for redistribution of income
from the `rich core' to `poor peripheries' and for an
expansion of the European Union's redistributive
capabilities is correspondingly diminished.
As emphasised by Hayek, the aim of the rules
must be to facilitate that matching or tallying of
the expectations on which the plans of the in-
dividuals depend for their success . . . But all rules
can achieve in this respect is to make it easier for
people to find together and to form that match;
abstract rules cannot actually secure that this will
always happen . . . Indeed, the superiority of
certain rules will become evident largely in the
fact that they will create an effective order not
only within a closed group but also between
people who meet accidentally and do not know
each other personally. They will thus . . . create
an order even among people who do not pursue a
235
common purpose. The observance of rules by all
will be important for each because the achieve-
ment of his purposes depends on it, but the
respective purposes of different persons may be
38
wholly different.
This is the way in which the European Union has
accommodated a diversity of states and peoples,
assisting them to pursue their own very diverse
purposes within a common framework of rules and
laws which has maximised peace and prosperity and
minimised the scope for friction and conflict be-
tween them. This has much more to do with the rule
of law, consensual negotiation of mutually accepta-
ble rules and laws, negotiated accommodation of
differences and `judicious' interpretation of rules
and laws by the (European) Court of Justice than
with democracy as such. As Hayek remarked, `the
ideal of individual liberty seems to have flourished
chiefly among people where, at least for long periods,
39
judge-made law predominated'. In making their
judicial or quasi-judicial rulings, the task of the
judges and of the EU Commission is not to ascertain,
reflect or uphold the wishes of a demos or even of the
fifteen separate demoi, but to endeavour to maximise
the internal consistency and viability of the EU legal
order and to minimise the scope for friction, conflict,
malfunction and acrimony. The major political safe-
guard is not a democratic one involving the submis-
sion of the decisions reached and policies adopted
after protracted and tortuous negotiation to various
forms of popular and/or parliamentary approval or
ratification. Rather, it is the fact those decisions and
policies can rarely go forward without gaining ac-
ceptance (often reluctantly and only after a great
deal of horse-trading) from fifteen separate states.
One of the major implications of all this is that the
current `eastern' applicants for EU membership need
neither expect nor be required to undergo major
extensions of democracy as such. Indeed, procedu-
rally, in most cases their parliaments and govern-
ments are as freely and fairly elected as their Western
European counterparts. Rather, what they have to
accept and adjust to is working within a political and
economic environment which is much more exten-
sively and exactingly regulated by consensually ne-
gotiated and/or judge-made supranational rules,
 ROBERT BIDELEUX
regulations and laws ± most obviously those govern-
ing the Single Market, but also those now pervading
40
almost every aspect of the EU's operations. Eur-
opean states which do not uphold liberal democracy,
basic individual rights and liberties, market princi-
ples and the rule of law exclude themselves from
potential membership of the European Union be-
cause of their consequent incapacity to discharge in
full the essential requirements of EU membership.
This is no longer merely a matter of choice or
preference on the part of the current member states,
but rather a sine qua non for the long-term survival
and effective functioning of the European Union.
Ian Ward has rightly pointed to `the role that law
plays in determining who is ``European'' in the new
41
Europe, and who is not.'
Power and Governance
in a Metamorphosing World
The erosion of state power and autonomy, to which
the emerging western neoliberal policy consensus of
the 1980s and 1990s most powerfully contributed,
has by no means been confined to the economic
domain and the growth of transnational information
and communications networks. Mary Kaldor has
highlighted the degree to which the state has lost
its former near-monopoly of the means of coercion,
which has been `eroded from above by the transna-
tionalisation of military forces' through membership
of security organisations such as NATO and the
former Warsaw Pact,) and `eroded from below by
the privatisation of organised violence' as a result of
the often interlinked emergence of private security
services, paramilitary forces, mercenary armies, large-
scale gangsterism and arms/drug-dealing mafias, and
the widespread disintegration and/or demobbing of
regular armed forces engendered by proliferating
instances of `state failure' as well as by the end the
Cold War (for example, in parts of Africa and the
42
former Soviet and Yugoslav federations). The pri-
vatisation of organised violence is assuming less
blatant forms in Western Europe and in the USA,
where private security guards now outnumber the
police by two to one, but Kaldor warns that it is
nevertheless taking place and that the West should
43
beware of complacency in these matters. More-
236
over, government itself is `increasingly contracted
out through various forms of privatisation and semi-
44
privatisation'. These momentous changes are in
turn blurring the distinctions between what is public
and what is private, between state and non-state,
between internal and external, between what is
military and what is civil, and even between war
45
and peace. Such distinctions were crucial to the
rise of limited constitutional government, civil so-
ciety and liberal democracy in the West, and the
blurring of such distinctions is potentially a very
ominous development. Indeed, armed conflicts
now arise mainly within and across rather than
between states, as the emergence of new privatised
and transnationalised forms of organised violence
erodes and in some cases completely disintegrates
states. Thus the growing `privatisation of organised
violence' in the Balkans, Transcausasia, Africa and
parts of Asia and the Middle East is a reversal of the
process by which states emerged and, if allowed to
continue unchecked, could bring about similar re-
46
sults in the West.
However, it would be more appropriate to see this
`privatisation of organised violence' as just part of the
47
more pervasive `privatisation of power' which has
been in full flood since the neoliberal counter-re-
volution in economic policy and governance during
the 1980s. Privatisation, economic liberalisation and
the unfettered development of transnational capit-
alism and global communications has engendered
autonomous market forces and concentrations of
private economic power and influence whose com-
bined strength far exceeds that of any elected gov-
ernment. The vast daily financial flows on the
world's currency and stock markets periodically set
in motion tidal waves which could even overwhelm
the US authorities and their Federal Reserve Board
or the EU authorities and their European Central
Bank. `Over $1 trillion flows across the world's
foreign exchange markets every day; over 50 times
the size of world trade and dwarfing the collective
foreign exchange reserves of the world's richest
48
states.' The genies of transnational capitalism have
been released from their old `national power-con-
tainers', which were much more amenable to na-
tional-democratic control than are the new global
forces and regional power blocs. Yet any attempt by
 CIVIL ASSOCIATION AND EUROPE
democrats to put them back into `national power-
containers' would precipitate a collapse of global
capitalist trade, production and economic confi-
dence even worse than the one that occurred be-
tween 1929 and 1931. Quite simply, liberalised
transnational capitalism has attained levels of eco-
nomic activity, specialisation, innovation and dyna-
mism which would be unsustainable in a world
which attempted to reassert `democratic' control
over supranational economic forces, especially if that
control were to be exercised by nation-states.
Whereas it was still possible to reassert national
control of economies and thereby promote various
nationalist and etatist forms of autarkic or semi-
autarkic economic development from the early
1930s to the 1970s, these have ceased to viable
options since the late 1970s. Thatcherites rant in
vain against the erosion of `national sovereignty' and
state autonomy which they have misleadingly attrib-
uted to the development of the European Union but
which was mainly a consequence of the neoliberal
counter-revolution in economic policy and the ram-
pant globalisation which they themselves helped to
unleash. The most that Europe can hope to do for
now is to regulate, constrain and channel the liberated
global market forces and the privatisation of power,
both through the elaboration of supranational rules
regimes of the sort maintained by the UN and the
World Trade Organisation and through the suprana-
tional legal order created by the EU. The Single
Market programme and the project of Economic and
Monetary Union are therefore designed to promote
`more Europe' as well as `more market' ± more
effective regulation at the EU level to compensate
for the diminishing efficacy of regulation at the level
of the nation-state.
Ways of Coping with
Uncertainty and Friction
Europe has now entered wholly uncharted waters.
Several centuries of Western political and economic
thought and prescription premised upon the exis-
tence of polities which are capable of exercising
more-or-less effective jurisdiction over `their own'
bounded territories, economies, citizenries, societies
and cultures have very little relevance to the (more
237
anarchic) brave new world which has already ar-
rived. The study of these great bodies of thought and
policy prescription might remain a valuable intellec-
tual exercise and of interest to historians, but it will
offer precious little guidance either as to what is
happening or to what policies might be effective in
the future.
In these changed circumstances, the quest for
increased democratic control ± whether national,
supranational or `cosmopolitan' ± is the pursuit of
a mirage. Demands that supranational regulatory
bodies and rule-making organisations should be sub-
jected to democratic control and accountability, as
well as the anxieties arising from the fact that they
are not, often stem from mistaken assumptions con-
cerning the origin of most laws at the national level.
As a result the so-called democratic deficit at the EU
supranational level is misleadingly contrasted with
supposedly `democratic' processes at the national
level. In fact, the executive branches of western
governments have come to exercise inordinate con-
trol over the legislatures, with the result that parlia-
ments have become less and less effective as checks
49
on the power of the executive. Moreover, only a
small fraction of the huge volume of government
business comes under the direct control and scrutiny
of the elected parliaments in most western liberal
democracies, especially under the secretive and oli-
garchic British system of `elective dictatorship' (as
Lord Hailsham so aptly described it). Every four or
five years electorates have a fleeting opportunity to
vote in a new `dictator' for the next few years ± that is
the realistic extent of democracy in most western
countries most of the time. What has made such
systems of government more bearable and effective
than any alternative system yet devised is not their
(actually rather limited) elements of democracy. It is
the very considerable heritage of liberal constitu-
tionalism, limited government, accommodation of
cultural diversity and rule of law which has safe-
guarded western liberties whose origins long predated
democracy as such and have usually helped to pro-
tect western democracies against the worst dangers of
majoritarian tyranny.
Fundamentally, as argued by Hayek (among
others), few of the legal frameworks and other rules
that regulate human conduct have ever been con-
 ROBERT BIDELEUX
sciously and deliberately devised, laid down or or-
dained by particular (far-seeing) parliaments, judici-
aries, governments or individual rulers. Rather, they
have emerged `through a gradual process of trial and
error', to which the experiences and learning pro-
cesses of successive generations have contributed.
Many rules have come into being because, in certain
types of situation, friction is likely to arise among
individual actors about what each is entitled to do,
which can be prevented only if there are rules to tell
each clearly what their respective rights and obliga-
tions are. The gradual, piecemeal, iterative elabora-
tion of such rules is one of the means by which
mankind has learned to cope with the inherently
incomplete, fragmented and dispersed nature of all
human knowledge. Rules do not just restrict; they
also empower and facilitate by fostering increased
certainty, stability and security. For the most part
our `actually existing' rules, regulations and laws
have emerged, not as the results of conscious gov-
ernmental or parliamentary decisions, but as the
outcome of endless mutual adjustment of human be-
haviour, without the deliberate intervention of an
ultimate arbiter ± not unlike the continuous mutual
adjustment of the spontaneous activities of indivi-
50
dual actors in market economies.
In an increasingly `globalised' and `regionalised'
world, laws, rules and regulatory regimes will increas-
ingly have to deputise or substitute for the ever-
diminishing scope for democratic control and ac-
countability as the (supposedly) standard means of
safeguarding civil and political liberties and/or eco-
nomic prosperity. Hayek's emphasis on the limits of
our knowledge about the present (let alone the
future) is even more appropriate to the new circum-
stances in which we now find ourselves. The best we
can hope for is trial-and-error improvisation and
adaptation of rules that can continue to allow us
to live, work and trade together with the minimum
of friction and the maximum of mutual benefit in a
shrinking and increasingly dangerous world. These
`discovery processes' are iterative and pragmatic rather
than democratic. Indeed, one must be on one's guard
against nationalists, socialists, populists, neo-fascists
and even some well-meaning democrats who pedal
preconceived blueprints and panaceas for the future,
because it is simply not possible to identify in
238
advance what will be successful and what will not,
what will meet the paramount needs of the future
and what will not, or even to specify in advance what
those needs will be, as those who suffered several
decades of Communist economic planning learned
to their great cost. It will be safer to base the EU on
the experimental, empirical liberalism of Hayek and
Oakeshott than on the more controlling and pre-
scriptive rationalist versions of liberalism. Explicitly
encouraging the EU to develop along more Oake-
shottian lines, as a civil rather than an enterprise
association, ought to be Britain's distinctive contri-
51
bution.
The Blessings of `Cosmopolitan Law'
Supranational European Union law has a particular
advantage over the `democratic' laws of liberal de-
mocratic states with multi-ethnic, multi-racial or
multi-denominational populations (that is, all of
Europe's liberal democracies, with the possible ex-
ception of Iceland). No matter how liberal and
democratic a liberal democracy might become, its
laws are still likely to embody (or at least to be seen as
embodying) the interests and preferences of the
dominant national, ethnic or religious group(s),
and to place (or at least to be seen as placing) various
minority groups in disadvantageous positions. Spe-
cial protection of the group rights of ethnic and
religious minorities rarely solves such problems, and
more often than not entrenches divisions, conflict
and perceptions of difference in ways that are coun-
52
ter-productive. By contrast, a major advantage of
EU laws and regulations is that they do not reflect,
nor are they based upon, the interests and aspirations
of any ethnic, religious or national majority. It is
arithmetically impossible for any single ethnic, na-
tional or religious group to dominate the law-making
and governance of the European Union. The largest
single nation/ethnic group makes up no more than
one-quarter of the population and it is greatly under-
represented in the EU Commission, the Council of
Ministers and the European Parliament due to the
significant weighting in favour of small nations.
Roman Catholicism, the European Union's largest
single religious denomination, does not enjoy special
favour. Indeed, EU law-making and governance are
 CIVIL ASSOCIATION AND EUROPE
in no way directed against the national, ethnic and
religious minorities `native' to Western Europe, even
though few (if any) of the member states can wholly
escape accusations of racism and/or religious intol-
erance towards various categories of immigrants of
non-EU origin. There are indeed mounting tensions
between the European Union's liberal-cosmopolitan
aspirations and founding ideals and its increasingly
exclusionary and intolerant stance towards immi-
grants from outside the EU, laying it open to charges
of malfeasance and hypocrisy. Its main line of de-
fence must be the morally lame one that `political
communities endure because they are exclusive, and
most establish their peculiar identities by accentuat-
53
ing the differences between outsiders and aliens.'
The EU has often been likened to the Austro-
Hungarian Empire, which was similarly held together
by a supranational rule of law, and within which no
single ethnic group held an absolute majority. Never-
theless, the cosmopolitan universalism of the EU is a
vast improvement on the ethnic particularism, paro-
chial narrow-mindedness and widespread religious
bigotry and segregation which many Central Eur-
opeans have acknowledged as central characteristics
of the Habsburg domains, which did have dominant
(albeit non-majoritarian) ethnic groups with ethnic-
collectivist mindsets and a strongly collectivist domi-
nant religion. EU law, which has acquired `supre-
macy' and `direct effect' in the member states, is
probably the only `actually existing' legal framework
in the world within which `cosmopolitan law-en-
54
forcement' could be rendered fully operational.
This supranational system of `cosmopolitan law',
rather than utopian conceptions of `cosmopolitan
democracy', offers the basis on which the EU can
best help to defuse Europe's potentially explosive
`minority problems'. The essential point is that
`minority problems' can never be fully resolved with
a framework of nation-states which contain hege-
Notes
1. I was first drawn to Hayek through my earlier work on
centrally planned economies and to Oakeshott by
Samuel Brittan's columns in The Financial Times,
but my understanding of both has been enhanced
by several years of fruitful discussion with Bruce
Haddock. He and Noel O'Sullivan drew my attention
239
monic ethnic and/or religious majorities. Such pro-
blems can only be resolved within a larger framework
in which no one group can establish a privileged or
hegemonic position, because every group has become
a small minority. Thus Timothy Garton Ash is right
to characterise the EU as a non-hegemonic liberal
order, although (like so many Anglo-Saxon com-
mentators) he seriously misunderstands the political
implications of EMU and misconstrues it as a threat
to ± rather than as an extension and reinforcement of
55
± this supranational liberal legal order.
The resultant shift of emphasis from unfettered
sovereignty and democracy to a supranational liberal
legal order that drastically circumscribes national
sovereignty and `unlimited' democracy, far from sig-
nifying some huge `sacrifice' or `retrograde step' (as
many `Europhobes' and the more blinkered demo-
crats would have us believe), is actually the major
benefit that the EU can confer on its member states.
The alternatives foregone are only illusions of na-
tional sovereignty and democracy, mere fig leaves
which barely conceal the reality of inescapable de-
mocratic deficits at all levels, since national democ-
racies have succumbed to increasing executive
branch control of the legislature and the media while
Europe as a whole is falling increasingly under the
sway of global capitalism, global information and
communications networks, and an expanding trans-
national security alliance (NATO). The emerging
supranational EU legal order cannot make good
Europe's growing democratic deficits, but it can
mitigate the consequences and confer major com-
pensating benefits which would be unattainable
within a more fragmented and fractious Europe of
nation-states. My pessimism concerning the pro-
spects for liberal democracy is combined with a
considerable optimism that this particular cloud
has a very substantial silver lining.
to Terry Nardin's use of Oakeshott's distinction be-
tween civil and enterprise associations.
2. Hayek (1993), volume 1, pp. 1±3.
3. See Spinelli (1957) and Albertini (1999), passim.
4. Merkel (1994), p. 38.
5. Laffan (1999), p. 336.
 ROBERT BIDELEUX

6. Lodge (1994), pp. 364±5. 32. Ibid., pp. 18±19.

7. Gray (1998), pp. 149±50. 33. Ibid., p. 318.
8. Majone (1992), pp. 77±101. 34. Ibid., pp. 222±3.
9. Bulmer (1993), pp. 375±6. 35. Ibid., p. 32.
10. Mann (1993), p. 130. 36. Ibid., p. 324.
11. Smith (1996), p. 253. 37. Gros and Steinherr (1994), p. 76.
12. Mann (1993), pp. 131, 135. 38. Hayek (1993), volume 1, pp. 98±9.
13. Smith (1996), p. 259. 39. Ibid., p. 94.
14. For example, Held (1995) and (1996). 40. This argument is presented at greater length in Bide-
15. This is a major reason why other regional blocs, most leux (1999b).
notably in Latin America, are seeking to emulate the 41. Ward (1997), p. 79.
European Union's success. 42. Kaldor (1999), pp. 4, 5, 28, 91±6, 140.
16. Dahl (1994), pp. 26, 32±3. 43. Ibid., p. 11.
17. This is the main theme of Tsagarousianou, Tambini 44. Ibid., p. 74.
and Bryan (eds) (1998). 45. Ibid., pp. 2, 29.
18. Quoted by Ladrech (1994), pp. 78±9. 46. Ibid., pp. 4, 5, 11.
19. Temple-Lang (1996), p. 126. 47. This apt phrase may have been coined by Bruce
20. Loveland (1996), pp. 534±5. Haddock, from whom I first heard it.
21. Hayek (1993), volume 1, p. 6. 48. McGrew (1997), p. 6.
22. Touraine (1997), pp. 2, 4. 49. Hayek (1993), volume 3, chapter 13.
23. Hayek (1939), p. 146. 50. Hayek (1960), pp. 157±9.
24. Hayek (1939), pp. 141±3. 51. Bideleux (1999a), pp. 116±18.
25. Oakeshott (1962), p. 189. 52. This is argued in more detail in Bideleux (1996), pp.
26. Ibid., pp. 40±1. 288±90, and in Bideleux and Jeffries (1998), pp. 407±
27. Gray (1989), p. 207. 34 and 460±6.
28. Ibid., p. 210. 53. Linklater (1998), p. 1.
29. Nardin (1983), pp. 9, 14. 54. Kaldor (1999), pp. 124±31, uses this in a more military
30. Ibid., pp. 12±13. and intergovernmental sense.
31. Ibid., pp. 14, 24. 55. Garton Ash (1998), pp. 59±64.

240
 18
THE IDEA OF A LIBERAL-
DEMOCRATIC PEACE
Howard Williams
An impressive literature has developed in recent
years dealing with the topic of a `liberal-democratic
peace'. Indeed, a debate has raged in American
political science journals such as International Secur-
ity and Comparative Political Studies about the validity
of the hypothesis that the growth in the number of
states with liberal-democratic polities will lead to a
more stable and harmonious international order.
The debate is conducted in a most civil and fair
way, with a great deal of emphasis being given to the
empirical evidence for the truth or otherwise of the
hypothesis. Nonetheless, lines have been drawn:
some stand on the side of optimism and a whole-
hearted endorsement of the thesis, while others are
more lukewarm and await more evidence. Yet others
are entirely sceptical, doubting whether liberal in-
stitutions bring peace and whether or not interna-
tional peace is possible at all. It looks as though the
debate is set to run and run as more evidence is
gathered and greater subtlety is shown in the inter-
pretation of regimes and liberal theory itself.
My object is neither to add to nor query the
relevance of this interesting literature. I think the
debate is most welcome, particularly as it demon-
strates the relevance of political theory to issues of
international politics and security. My object is
rather to examine the philosophical basis of this
discussion. I want to travel down the path of demo-
cratic peace, not with the present preoccupation in
mind of deciding whether or not such a democratic
peace is on its way, but rather with the intention of
241
deciding what the argument looks like from Kant's,
or a Kantian, perspective. In the literature on demo-
cratic peace, Kant has been much referred to: it is
true to say that he has become an emblem of the
thesis. This is so to such an extent that authors need
only cite Kant in their title to show that they are
engaging with the debate. Others make play with
Kant's name to demonstrate their topicality, such as
Christopher Layne's title to his article on `Kant or
1
Cant: The Myth of the Democratic Peace'.
The treatment of Kant in such articles tends to be
quite cursory and the emphasis generally turns very
quickly to the evidence. Here I want to alter this
emphasis by attempting to measure the democratic
peace thesis by Kant's standards. We will find that
there are intriguing differences of emphasis between
current international peace theorists and Kant's
original argument, which tells us a lot about the
priorities of liberal internationalism now, and Kant's
contrasting objectives for practical philosophy and
humanity in general.
Michael Doyle
The debate on the democratic peace has its origins in
a resurgence of interest in Kant's political philosophy
and international relations which occurred in the
early 1980s. One of the writers who most effectively
caught on to this trend was Michael Doyle, who in
two 1983 articles in Philosophy and Public Affairs
highlighted Kant's legacy to liberalism and interna-
 HOWARD WILLIAMS
tionalism; these were distilled into the one key 1986
article `Liberalism and World Politics' in the Amer-
2
ican Political Science Review. Although in the latter
Kant is cited along with Machiavelli and Schump-
eter it is clear that Kant forms the centrepiece of
Doyle's argument. For Doyle Kant's views on politics
not only have a valuable policy component in that
they show how American foreign relations ought
best to be framed but they also have a significant
empirical relevance in that Kant is `a liberal repub-
lican whose theory of internationalism best accounts
for what we are.' In contrast Machiavelli is `a classi-
cal republican whose glory is an imperialism we often
3
practice.' Doyle's object, then, is to put forward
Kant's political theory as a model of good practice for
liberal-democratic states to follow.
Doyle's thesis has a moral and an empirical side to
it. The moral side is naturally that policy-makers
should be informed by Kant's liberal international-
ism and the empirical side shows that where policy-
makers have been informed by Kant's principles they
have, in terms of promoting peace, turned out to be
quite successful. In this liberal-democratic thesis,
then, the behaviour of states that conform as near
as possible to Kant's model of a republic comes under
close scrutiny. The number of wars they have been
engaged in, their nature and their length are docu-
mented accurately and the results weighed very
carefully. This meticulous research is a feature of
Doyle's work. The 1986 article has a three-page
appendix where all the liberal regimes which have
existed from 1700 to 1982 are listed and all the
international wars since 1817 are given chronologi-
cally. Doyle rests some of his conclusions upon this
evidence. As he states in the article's abstract, `liberal
states are different. They are indeed peaceful.' But
the evidence is that they are not always inclined to
peace: `they are also prone to make war. Liberal states
have created a separate peace, as Kant argued they
4
would.' The evidence is therefore mixed but in no
sense refutes the position Doyle believes Kant and
other liberals to hold. Living by the rules outlined by
Kant in his Definitive Articles to Perpetual Peace can
pay in terms of the avoidance of war. Indeed Doyle
seems to believe that even closer attention to Kant's
arguments might have paid off because he fears that
in some respects the rulers of liberal-democratic
242
states may have been prone to be `trigger-happy',
contrary to Kant's pacific recommendations. Where
liberal-democratic states have become involved in
wars to defend liberal causes this has often turned out
badly. In so far as western democratic states have
found liberal grounds for aggression Doyle thinks
they have strayed from Kant's conceptions.
Francis Fukuyama
Further credence has been given to the democratic
peace thesis by Francis Fukuyama's writing on the
End of History. Fukuyama is perhaps better known for
the Hegelian connections of his thinking, especially
the striking claim that there are no major progressive
political developments yet to come in human his-
tory, but nonetheless he does draw very heavily upon
Kant and the idea of a pacific union. However,
Fukuyama gets to democratic peace via a very dif-
ferent route from Doyle. He is particularly struck by
the notion of a struggle for recognition that occurs in
human history, borrowing this from Hegel's philo-
sophy (the idea of such a struggle figures quite
prominently in the Phenomenology of Spirit and the
earlier unpublished work from his Jena period
5
(1804±6)). Fukuyama uses the notion of the strug-
gle for recognition ± a struggle which affects all
human individuals ± as a means of dismissing all
but the democratic republic as a legitimate form of
government. According to him, the difficulty with
all non-democratic forms of government is that they
prevent the struggle for recognition, which always
occurs between free human individuals, from work-
ing itself out in a non-destructive form. The beauty
about the democratic mode of government is that it
allows everyone to follow their own paths in a
competitive way without preventing others from
doing the same. Earlier, more authoritarian forms
of government stand in the way of such an equal
form of recognition. They are based upon forms of
recognition which elevate one individual or a group
of individuals above the rest. Fukuyama believes that
societies which are properly democratic are non-
antagonistic within themselves and therefore are
also likely to be non-antagonistic towards each
other.
Like Doyle, Fukuyama provides his own list of
 THE IDEA OF A LIBERAL-DEMOCRATIC PEACE
6
democratic states in his writings. He sees the spread
of democracies as part of the universal history of
mankind which is leading to the worldwide recogni-
tion of all individuals as equals. Fukuyama is highly
optimistic about this course of events. In his view,
`the growth of liberal democracy, together with its
companion, economic liberalism, has been the most
remarkable macropolitical phenomenon of the last
7
four hundred years.' In liberal democratic states no
adults are in general excluded from the political
process. They all equally are given the right to vote
and can themselves usually stand for office if they
wish to do so. This widening of participation in the
political process takes the edge off the natural com-
petition for preference among individuals. The com-
petition does not disappear but is channelled in a
civil direction. States that allow full democratic civil
competition fall outside history and those who fail to
allow it are regarded as still within history. They
have yet to make the step to the condition wholly
appropriate to human life. `The post-historical world
is one in which the desire for comfortable self-pre-
servation has been elevated over the desire to risk
one's life in battle for pure prestige, and in which
universal and rational recognition has replaced the
8
struggle for domination.'
Fukuyama believes that Kant foresaw such a con-
dition of universal peace in his political philosophy
and therefore provided the basis for the development
of a convincing liberal interpretation of this phe-
nomenon. So although it is Hegel who inspires his
vision of the universal homogeneous state Fukuyama
gains the idea of a democratic world peace from
Kant. He is well aware that many of the liberal
internationalist movements inspired by Kant, such
as the League of Nations, have ended in failure. But
he argues that `what many people have not under-
stood, however, is that the actual incarnations of the
Kantian idea have been seriously flawed from the
9
start by not following Kant's own precepts.'
Broadly speaking, Fukuyama believes the move-
ment towards peace and democracy represents an
irreversible process. This does not mean that he
thinks it is an inevitable outcome at any point in
time. The leaders of states may make poor choices or
fail properly to see the road ahead. Their desire for
power and prestige may momentarily stand in the
243
way of their seeking mutual recognition in the world.
But Fukuyama believes that the growth of modern
science and the technological improvement which
follows from it means that human desire and human
reason point in the same direction. Economic effi-
ciency flows from the adoption of liberal-democratic
practices and rules. No state can afford to go it alone
in the modern world economy. To flourish, states
have to open their markets and encourage both trade
and investment. In this respect, Fukuyama envisages
a kind of economic determinacy which forces people
towards greater democracy and the opportunity of
peace. This argument seems very similar to the
`guarantee' of eternal peace which nature provides
that Kant puts forward in the first supplement to
10
Perpetual Peace. As evidence of Fukuyama's opti-
mism about the applicability of his own peace thesis
we can take the following: `In fact, such a Kantian
liberal international order had come into being
willy-nilly during the Cold War under the protective
umbrella of organisations like NATO, the European
Community, the OECD, the Group of Seven,
11
GATT.' With Fukuyama the democratic peace
theory ceases to be a hypothesis and becomes instead
a reality.
`Liberalism and World Politics'
Clearly, the empirical tone of the debate is most
marked in both Fukuyama and Doyle, focusing upon
the factual reality or otherwise of Kant's outline for
perpetual peace. This is very flattering for Kant's
argument but I am not sure at all that it is the kind of
reception he had in mind. At the general philoso-
phical level it should be borne in mind that Kant is
more often regarded as a transcendental idealist
rather than a straightforward defender of empiricism.
Kant was very impressed by the empiricist philoso-
phies of Locke and Hume but ultimately wished to
differ with them about the importance of observation
in the construction of human knowledge. Kant
wanted to emphasise in his critical philosophy the
active human part played in the production of
knowledge as well as the passive part played by
the absorption of data. This brings us to the most
striking overall difference between the Doyle/Fu-
kuyama approach to the growth of a peaceful world
 HOWARD WILLIAMS
order and Kant's. In many respects Doyle and Fu-
kuyama play the role of passive spectators in obser-
ving the role of liberal republican regimes in bringing
about peace whereas Kant's main emphasis is on the
role that the leaders of states and an enlightened
public can play in realising the goal. Kant's Perpetual
Peace represents the most explicit kind of moral
advocacy in which facts are cited to demonstrate
that the author's goals are not wholly unrealistic,
whereas Doyle and Fukuyama's work is directed more
to showing the empirical feasibility of the goal of a
democratic peace and thereafter its possible moral
desirability. For Doyle and Fukuyama the lure of the
democratic peace thesis lies mainly in its apparent
present empirical validity whereas for Kant the allure
of his thesis stands independently of particular pre-
sent facts. For him, we ought to aim for a world made
up of free, republican states because this is the right
thing to do even if present circumstances do not
indicate with total clarity that this will bring lasting
peace. Kant thinks that we have to work on the
assumption that the thesis can be shown both mo-
rally as well as empirically to be correct.
Doyle's position is therefore not at odds with
Kant's argument for peace. The criticism that might
legitimately be levelled at Doyle and Fukuyama is
that they tell only part of the story. The part they tell
is the one which is empirically reassuring. As Doyle
eloquently puts it:
Beginning in the eighteenth century and slowly
growing since then, a zone of peace, which Kant
called the `pacific federation' or `pacific union',
has begun to be established among liberal socie-
ties. More than 40 liberal states currently make up
the union. Most are in Europe and North Amer-
12
ica, but they can be found on every continent.
It is interesting to note that Doyle draws his argu-
ment primarily from one section only of Perpetual
Peace: this is the section in which Kant outlines the
three definitive articles for a lasting peace. Doyle
makes little use of the earlier section containing the
three preliminary articles and hardly touches at all
upon the various supplements and appendices which
follow upon the definitive articles. As the supplements
and appendices constitute more than a third of the
244
essay by themselves, and more than half when taken
with the preliminary articles, this is a striking omission.
Doyle does have a strong justification in that the three
definitive articles represent the main positive points
that Kant wished to highlight in his essay. These
stipulate the requirements for (1) a republican form
of government; (2) a federation of free states; (3)
cosmopolitan law to be limited to the right of universal
hospitality. These three definitive laws provide the
backbone for Doyle's argument and the first two are
13
also highlighted by Fukuyama. Arguably, neither
analyses in sufficient detail the `guarantee' for perpe-
tual peace which Kant presents in the first supplement
to his essay which would seem, dealing as it does with
the course of history upon the development of peace, to
have an important bearing on their arguments. Over-
looked also are Kant's views on the role of the philo-
sopher, his discussions of morality and politics, and his
reflections on the role of publicity in politics.
Doyle's account of Kant is therefore a highly
selective one. But in those matters he chooses to
discuss he is persuasive. Indeed, in very many re-
spects both he and Fukuyama get Kant right. Doyle
in particular provides a highly accurate report of the
kind of internal social and political structure it is
important to have in order to ensure the most rapid
progress towards peace, and his discussion of Kant's
cosmopolitanism is illuminating and valuable.
Although Doyle's Kantian argument has been called
the `democratic' peace theory, he himself sensibly
stresses that Kant's main emphasis is upon the re-
publican nature of the ideal form of government. In
fact we can add that in his original account of the
kind of rule most suitable for bringing international
harmony in Perpetual Peace Kant is quite hostile to
the democratic form. As he puts it:
Democracy, in the truest sense of the word, is
necessarily a despotism, because it establishes an
executive power through which all the citizens
may make decisions about (and indeed against)
the single individual without his consent, so that
decisions are made by all the people and yet not
14
by all the people.
Kant's republican ideal is intended to prevent all
kinds of despotism, not only the despotism of one
 THE IDEA OF A LIBERAL-DEMOCRATIC PEACE
individual who rules arbitrarily from the top but also
the despotism of a group of people who might seek to
rule arbitrarily from the bottom. In a republic, as
Kant understands it, there has to be a proper separa-
tion of powers between the executive, legislature and
judiciary. Not only are the rulers not allowed to
make the law, they are also not allowed to interfere
in the adjudication of infringements to the law. Law-
makers should not also be in a position to affect the
actions of rulers in the execution of the laws, nor can
they seek to influence members of the judiciary in
the decisions they make. Equally, members of the
judiciary are not under any circumstances to interfere
in the work of the legislature and executive.
The First Definitive Article
This very strict separation of powers rules out direct
democracy as an ideal of government for Kant.
Certainly, he is very much in favour of citizens
developing political awareness and letting
views be known. He thinks that every citizen is as
valuable as the next. He does not take an elitist view
of politics, believing that political decision-making
should be confined to those `in the know'. But he has
a logical or technical objection to the idea of every-
one trying to have a hand in running a state. As he
sees it, the people as a whole cannot rule themselves.
No one should be allowed to be a judge in their own
cases. And where each branch of the polity is under
the control of the people as a whole this can easily
occur. Just as it would not be right if one individual
sought to control each branch of government so also
it has to be wrong if the people as a body seeks to
control the executive, legislature and judiciary. We
have to be ruled by our representatives. As Kant puts
it, `any form of government which is not represen-
15
tative is essentially an anomaly.' If we expect rules
to be fairly administered then those who are subject
to them should not also at the same time be respon-
sible for making them.
Although Doyle stresses this claim of Kant's,
there is a large gap in his presentation of Kant's thesis
in relation to democracy. While making no attempt
to suggest that Kant is an enthusiast for democracy,
he does not bring out the objections Kant has to it.
We cannot, for instance, glean from Doyle's discus-
sion that Kant rules out democracy as a possible form
of viable government. Kant thinks there are essen-
tially only two forms of government: the despotic or
17
the republican. It might be argued here that, in his
critique, Kant has in mind a more radical kind of
direct democracy than is now usual in western states.
Yet still, I would argue, there are elements of Kant's
objections to democracy which have a bearing on
the current practices of western democracies and on
the application of Kant's peace theory. Although
Kant's advocacy of the republican form of govern-
ment can be read as favouring the kind of represen-
tative democracy to be found in countries such as the
United States, Germany and France it can also be
read as a warning against tendencies towards a
majoritarian style of democracy which, from time,
to time, influence the government of these countries.
Like Mill and Tocqueville, Kant fears the tyranny of
the majority and believes a genuinely rational form
of government has to be properly structured to
their prevent it occurring. For example, it is difficult to
see how his republicanism might wholly embrace the
kind of party politics now prevalent in western states.
It is conceivable that Kant's guidelines are compa-
tible with parties competing for representation with-
in legislatures and so with their success putting
forward a certain legislative programme. However,
it is difficult to see how Kant's principles might be
used to endorse parties competing for sole executive
authority, particularly if that dominance were
matched by an equal dominance in the legislature.
Where a party uses executive authority to press
through its own programme, and this programme
is willed by a certain section of the population and
not all, then the government must, according to
Kant, tend towards despotism. It is true that most
parties in western states are not monolithic structures
and have their own internal representative features
which mitigate the problem. But where such parties
tend towards unity or uniformity the danger of
straying towards despotism (as Kant understands
16
it) must exist. Thus, peculiarly, from Kant's stand-
point the more diverse, open, representative and
pluralist a party, the more it will conform to his
ideal of government should it gain executive
power.
The fact that Kant's prescription for the ideal
245
 HOWARD WILLIAMS
polity requires majoritarian tendencies to be kept in
check has important implications for his peace the-
sis. Kant would regard states whose representative
systems were liable to factional control as possibly
more inclined to belligerence than other states where
the republican model he outlines is more rigidly
adhered to. Democratic states which allowed one
party to control both the legislature and executive
might be more prone to war than others, just as
democratic states which allowed one set of opinions
to predominate at the expense of others through
inadequate public debate might be seen as less peace-
able. Doyle and Fukuyama's optimistic approach to
Kant's thinking seems to overlook these possibilities
which surely would have exercised Kant. It is very
important for Kant that all citizens should regard the
rulers as acting on their behalf, and with their
authority, even if in many respects they disagree
with the policies being pursued. Should political
structures within a state systematically exclude one
section of the population (which might be seen as
occurring where some discriminatory political prin-
ciples are taken to their extreme) then they cannot
be fully regarded as co-citizens. Subjects who are
alienated in this way from their states are more liable
to be indifferent to possible aggressive foreign poli-
cies and are possibly more likely to be used as
`cannon-fodder' in any war. Thus Kant's liberal
republicanism needs emphasising if the `democratic
peace' thesis is to be true to its origins.
The Second Definitive Article
What Doyle also perhaps glosses over in his writ-
ings, and the liberal treatment of Kant in general
has to deal with carefully, is Kant's attitude to the
role of women as citizens. Essentially, Kant excludes
all women from active citizenship, as he does all
dependants including wage-workers, servants and
18
the jobless. Far from being unfortunate prejudice
which is readily dispensable in his overall account,
he makes this stipulation on the grounds that to be
in a proper position to have an enlightened and
disinterested view of the rules that bind together a
society you have to owe your existence to no one
other than yourself. Such people, he thought, could
not. This is absolutely pivotal in his account of the
246
possibility of a democratic peace. He thinks that
such independent individuals who form a republic
will be unlikely to seek war with other states, first
because through their representatives they would be
making decisions on their own behalf and, second,
if there were costs to be borne, they themselves
would have to bear them.
In this respect Doyle believes that `the apparent
absence of war between liberal states, whether ad-
jacent or not, for almost 200 years thus may have
significance.' In states where the Kantian features of
a separation of powers, representative government
and some form of adult suffrage have been absent
there seems not to have been the same restraint
shown in relation with other states. `Feudal, fascist,
communist, authoritarian forms of rule' do not,
19
according to Doyle, have such a good record. What
he finds particularly striking is the behaviour of
liberal states (which best approximate to the Kantian
model) when major wars actually occur. Not only do
such states not fight wars with each other, they also
tend, when forced, to concur in their view as to
which constitutes the `bad' side. As Doyle puts it,
`when states are forced to decide on which side of an
impending world war they will fight, liberal states all
wind up on the same side despite the complexity of
20
the paths that take them there.'
Doyle is correct to emphasise how the behaviour
of non-despotic states tends to point in the same
direction. But it is doubtful that Kant would see this
as proof of his contention. This is because even if the
leaders of a republican state were to take the `wrong
side' in a world war (which Finland is alleged to have
done at the time of the Second World War in some
of the literature calling the democratic peace theory
21
into doubt ) this would not absolve the leaders of all
states from the duty of trying to hold to the project of
perpetual peace in all their future actions. It is
interesting that in Perpetual Peace Kant does not
cite the behaviour of republican states to demon-
strate the truth of his plan; rather, he refers to such
states from the standpoint of the example they might
set to other states. He makes this clear in his com-
ments upon the Second Definitive Article where he
states: `If by good fortune one powerful and enligh-
tened nation can form a republic (which by its nature
is inclined to seek perpetual peace) this will provide
 THE IDEA OF A LIBERAL-DEMOCRATIC PEACE
a focal point for federal association among other
22
states.'
A great deal of the discussion in the democratic
peace debate has focused on the extent to which
states with Kantian-style constitutions form a pacific
union. Enthusiasts for the proposition have discov-
ered that states with liberal constitutions tend to
keep the peace in relation to each other and those
who question the peace theory suggest that there is a
great deal of evidence that states with such consti-
tutions are still liable to war. Doyle's thesis is that
`liberal republics will progressively establish peace
among themselves by means of the pacific federation
(foedus pacificium) described in Kant's Second Defi-
23
nitive Article.' In his view, `most liberal theorists
have offered inadequate guidance in understanding
the exceptional nature of liberal pacification.' Where
many fall down, he thinks, is in their inability to
explain the way in which some liberal states tend
also to get embroiled in war. In this respect, he claims
24
that `Immanuel Kant offers the best guidance.'
Kant is able to understand the tendency of liberal
states to form with each other islands of peace and
their tendency to be less at home with non-liberal
states.
Doyle understands Kant to be recommending in
his Second Definitive Article neither a formal fed-
eration nor any institutionalised form of cooperation.
Mainly, he is emphasising a disposition of states with
republican constitutions not to attack each other
militarily. But Doyle does go a bit further to suggest
that Kant `appears to have in mind a mutual non-
aggression pact, perhaps a collective security agree-
25
ment.' This seems very close to what Kant suggests
when he proposes that the `law of nations should be
26
based on a federalism of free states.' But care is
necessary here before we endorse Doyle's interpreta-
tion in its entirety. Particularly important in the way
in which Kant formulates the Second Definitive
Article is its markedly counterfactual nature. He is
not speaking directly of any empirical state of affairs,
present or future, in the Article itself. He has in mind
much more the intellectual presuppositions which
form the basis for an effective international law.
Indeed, I think the main thrust of the Article for
perpetual peace is a criticism of prevailing concep-
tions of international law which ground that law in
247
the unrestricted sovereignty of states. The recom-
mendation that a pacific federation of free states be
formed is one that flows from this initial objection.
Kant holds that very little progress can be attained in
international politics if the leaders of states behave as
rulers of entirely self-contained and wilfully inde-
pendent territories who can declare war wherever
they see fit upon any other sovereign entities. The
very conception of an international politics based
upon such wholly unrestricted sovereignty carries the
seeds of its own destruction. As Kant sees it.
the concept of a law of nations becomes mean-
ingless if interpreted as the right to go to war. For
this would make it a right to determine what is
lawful not by means of generally valid external
law, but by means of one-sided maxims backed up
27
by physical force.
Doyle is highly conscious of the federative aspect
of Kant's peace plan but does not bring out this
indictment of the traditional notions of state sover-
eignty which underlie Kant's critique of interna-
tional law. He notes that Kant's
pacific union is not a single peace treaty ending
one war, a world state, nor a state of nations. Kant
finds the first insufficient. The second and third
are impossible or potentially tyrannical. National
sovereignty precludes reliable subservience to a
state of nations; a world state destroys the civic
freedom on which the development of human
28
capacities rests.
Now although it is indeed true that Kant rules out for
the foreseeable future a world state resting upon a
worldwide civil society, this is not because the pur-
suit of such a condition is morally wrong. In his view
we ought to give up the total and arbitrary indepen-
dence of our national state and seek to approximate to
a world state. Our present condition is not one of
which we can be proud. What prevents us from
attaining the condition of world statehood which
reason prescribes is the sheer physical difficulty of
coordinating government against the entrenched
habits of the leaders of sovereign states, and over
huge tracts of territory, some of which are not
 HOWARD WILLIAMS
developed (in Kant's day this was of course a much
greater problem than it is now). In Kant's view the
former point poses the greatest difficulty. There is an
irrational immaturity in the attitude of states' leaders.
They are prepared to drive humanity to destruction
to retain their own standing as unrestricted sover-
eigns. The attainment of state sovereignty is indeed a
great step forward in the development of the human
race, but it is not an absolutely valid arrangement. As
Kant sees it, state sovereignty attains its proper
validity when it merges over into a wider interna-
tional sovereignty of free states. The splendid ma-
jesty of the absolutely unrestricted sovereign state is
an obstacle to world peace:
We look with profound contempt upon the way
in which savages cling to their lawless freedom.
They would rather engage in incessant strife than
submit to a legal constraint which they might
impose upon themselves, for they prefer the free-
dom of folly to the freedom of reason. We regard
this as barbarism, coarseness, and brutish debase-
ment of humanity. We might thus expect that
civilised peoples, each united with itself as a state,
would hasten to abandon so degrading a condi-
tion as soon as possible. But instead of doing so,
each state sees its own majesty (for it would be
absurd to speak of the majesty of a people) pre-
cisely in not having to submit to any external
legal constraint, and the glory of its ruler consists
in his power to order thousands of people to
immolate themselves for a cause which does
not truly concern them, while he need himself
29
not incur any danger whatsoever.
Fukuyama also does not realise the extent to
which Kant's Second Definitive Article implies a
critique of the traditional notions of state sovereignty
and international law. Kant wants to rewrite the
rules of international society in a rational form
whereas Fukuyama is more concerned to point out
how international society has already taken on a
rational form. He tends as a consequence to cham-
pion the changes Kant wants without fully realising
how difficult they might be to achieve. In keeping
with Kant, Fukuyama argues that `international law
30
is merely domestic law writ large,' yet there is no
248
direct recognition of the kind of mutation of state
sovereignty this might involve. In the kind of world
Kant envisages no one state can prevail over an-
other; each has always to accommodate itself to the
next as though they were neighbouring citizens
within a civil society and state. But Fukuyama and
Doyle both realise that in a peaceful condition of the
Kantian variety state sovereignty cannot play the
role it once did. Without clearly stating it both imply
that the kind of unrestricted state sovereignty that
once prevailed in international society has had its
31
day.
The Third Definitive Article
One of the great virtues of democratic peace theory is
its moral egalitarianism. Each human individual for
the proponent of a democratic peace counts for as
much as every other. Each individual, wherever they
might live, is deserving of representation and of the
right to elect that representative. With Fukuyama
this moral egalitarianism is evident in his emphasis
upon what he calls isothymia. Fukuyama defines the
term in relation to its opposite megalothymia. This is
the `desire to be recognised as superior to other
people', whereas `its opposite is isothymia, the desire
32
to be recognised as the equal of other people.' His
thesis is that the world must change, and is changing,
to a situation where each person lives under a system
of rule which fosters isothymia or equal recognition.
He sees this happening, however, mainly within
states. As what he calls `the universal homogeneous
state' becomes more widespread where such equal
recognition is found this will have a beneficial effect
upon peace and security amongst states. In drawing
this picture of extending worldwide equality Fukuya-
ma's inspiration is Hegel who, in his conception of
the state, properly develops the idea of mutual
democratic recognition.
This is not incompatible with the Kantian per-
spective. However, it is important to bear in mind
that for Kant there is a moral equality which both
historically and geographically extends beyond the
state. By definition human individuals, who are the
only rational animals, are equal. It is an offence
against our dignity to treat another human indivi-
dual or group as inherently unequal to us. In this
 THE IDEA OF A LIBERAL-DEMOCRATIC PEACE
respect Doyle captures the spirit of Kant's philosophy
well. He sees that Kant's cosmopolitanism leads to a
different style of international politics from that to
which we are now accustomed. Kant's doctrine of
right would provide no grounds for the western
imperial project of the modern world, or for any
missionary intention forcefully to `civilise' non-wes-
tern countries. Kant rules out a patriarchal approach
33
to government both internally and externally.
development of societies outside Europe must possess
a voluntary dimension. Natives have to play a part
themselves in drawing towards civilisation. The most
that European visitors can ask of the inhabitants of
territories they arrive in is that they not be treated
with hostility. This point comes out in Doyle's
interpretation of the Third Definitive Article to
Perpetual Peace. `Hospitality', says Doyle, `does not
require extending to foreigners either the right to
citizenship or the right to settlement, unless the
foreign visitors would perish if they were expelled.
Foreign conquest and plunder also find no justifica-
tion under this right.' Imperialism of any kind is ruled
out. `Hospitality does appear to include the right of
access and the obligation of maintaining the oppor-
tunity for citizens to exchange goods and ideas with-
out imposing the obligation to trade (a voluntary act
34
in all cases under liberal constitutions).'
Doyle argues that liberal states, despite their inter-
nal liberal structures, have been prone to war with
non-liberal societies. As he puts it, `the historical
liberal legacy is laden with popular wars fought to
promote freedom, to protect private property, or to
support liberal allies against non-liberal enemies.'
He suggests that Kant's position would be at odds
with this outcome. Partly, these wars have occurred
because the leaders of liberal states have demanded
more than hospitality for their citizens and subjects
in their relations with other states. In seeking to
promote the freedom and advantage of visitors to
foreign territories they often have asked not simply
for safe passage but that a certain form of life be
respected which is contrary to native traditions. Kant
believes we ought not to venture beyond peaceful
persuasion in trying to advance the condition of
other nationals in their own states. Similarly, liberal
states are in no position to guarantee the property of
their nationals abroad simply through the exertion of
their own powers. They cannot extend their sover-
eignty in such a way. Rather, they should look to the
host country to guarantee the property of their
nationals while they are resident in those territories.
The civil order of a foreign state is primarily its own
responsibility. All other states can ask is that the
liberty of their subjects should not be arbitrarily
limited.
The
The Guarantee of Perpetual Peace
In one of the most interesting supplements to Per-
petual Peace Kant outlines what he describes as a
guarantee provided by `the great artist nature her-
36
self'. Here, he argues that the apparently unguided
course of historical development tends in the direc-
tion of human improvement, often against our will.
He cites a number of mechanisms that help to bring
this about such as our competitiveness, self-inter-
ested pursuit of trade, the rivalry among nations and
even war itself. Even when against our will, these
forces can oblige us to enter into more peaceful
relations with one another.
Both Doyle and Fukuyama cite this mechanism of
nature with approval. They commend Kant's guar-
antee because it highlights an empirical process
which can help underpin the liberal-democratic
peace. For Doyle the guarantee works not because
we are politically moral beings but through our less
agreeable side. Our desire for advantage over others
makes us into `asocial-sociable beings' who are just as
much inclined to rivalry as cooperation with them.
35
Our paradoxical aim is to succeed in the eyes of
others but also at their expense. However, we learn
through sad experience that this is not the proper
way to conduct human affairs. Under the pressure of
insecurity and actual war we learn the advantages of
37
republican government. Progress can occur in his-
tory both through our good and our bad side. Doyle is
not quite sure how to handle this dialectical process.
Whereas it might be argued that Kant is quite happy
to see human short-sightedness and downright evil
inadvertently playing a progressive role and this
conflicting with the moral spirit within us, Doyle
wants us to unify the two moral forces. Those who
are entirely focused on their selfish interests are also
playing a moral role. With Doyle the `hidden plan' of
249
 HOWARD WILLIAMS
38
nature complements our moral duty. However,
with Kant the hidden plan steps in only where we
fail in our moral duty. For him, the plan is not right
in itself; rather it provides us with an incentive to do
what is right.
This is a significant point. Kant argues that we
cannot depend on the guarantee from a theoretical
standpoint. In other words, if we view the guarantee
as a scientific prediction we cannot expect it to come
true. As Kant puts it, `we cannot actually observe
such an agency in the artifices of nature, nor can we
39
even infer its existence from them.' However, the
thesis possesses sufficient reality for us to entertain it
from a practical standpoint: from the moral point of
view we can assume it to make sense, at least
sufficient sense for us to persist with a foreign policy
based upon morality.
We can rightly suppose that there are natural
mechanisms at work which promote the aim of
lasting peace. Despite the expressed aims of less
scrupulous political leaders and their subjects, we
can discern over the longer term underlying trends
which appear to be taking the human race forward.
Perversely, war itself might further these trends by
bringing home to people the doleful effects of war-
fare. Societies who engage in this aggressive beha-
viour are forced to modernise quickly in order to
employ the most modern means for engaging in war.
But this modernisation takes place with a concomi-
tant growth in the potential and demand for freedom
within the state. Self-interest is, above all, what
brings the progressive mechanisms of war into play.
The hard-nosed pursuit of economic gain may have
peaceful consequences. Although primarily moti-
vated by the pursuit of profit those professionally
engaged in trade and commerce come to dislike the
disruption caused by wars. Over the long term Kant
thinks that business prefers peace to war.
Kant realises that such projections about the
progressive mechanism of history are not subject
to clear empirical proof. There are countless events
which may tell in the opposite direction, but again
no one can offer conclusive proof that the human
race is bound not to improve. Kant's object is not to
create sufficient certainty from a scientific stand-
point but rather sufficient certainty from a moral
standpoint. He thinks there is enough evidence to
250
justify us from a moral standpoint pursuing progres-
sive policies. Nature may help us some of the way but
ultimately we have to choose to improve. Doyle and
Fukuyama tend to overlook the voluntary side of
human progress which forms the essential part of
Kant's guarantee. Both Doyle and Fukuyama are
enthusiastic about the natural mechanism, especially
Fukuyama who devotes several passages in his writ-
ings to the economic, social and political forces that
40
drive us willy-nilly towards cooperation. Successful
material acquisition under a free-market system re-
quires long-term peace and security. According to
Fukuyama, a preoccupation with acquiring more
consumer goods diverts people away from aggressive
foreign policies and war. In keeping with Doyle and
Fukuyama's interest in the determinist side of Kant's
`hidden plan' the literature on democratic peace
concentrates heavily on the empirical verifiability
41
of the perpetual peace thesis. This may not take us
entirely off the Kantian track, since the more em-
pirical evidence for the thesis which exists the fewer
excuses there can be for refusing to seek progress in
external policy. However, in the light of Kant's
priorities the emphasis is still a little one-sided.
His main purpose is to bring morality and politics
into accord; only then will perpetual peace be
achieved. The actual past course of events naturally
plays a key role in conditioning the present, but so
also must the intentions of policy-makers and those
they represent.
Conclusion
What does our examination of the liberal-demo-
cratic peace thesis advanced by Doyle and Fukuyama
from the standpoint of Kant's original peace theory
tell us? Our first conclusion might justifiably be that
there is more to Kant's original view than the
democratic peace theorists convey. There are ele-
ments in Kant's argument, such as his powerful
objections to forcible intervention in the affairs of
other states, his views on property rights and the use
of national debt to finance war, which are not fully
considered. Just as Doyle and Fukuyama's new inter-
pretations of Kant's thinking have uncovered many
valuable ideas we might legitimately claim that there
is even more to be discovered which might not only
 THE IDEA OF A LIBERAL-DEMOCRATIC PEACE
add to the democratic peace thesis but also provide
insights in other connected areas of politics. Classical
political theory, such as Kant's essay on Perpetual
Peace, provides a permanent source of new insight
into current national and international issues.
But such sources have always to be treated criti-
cally. Kant would have himself emphasised that we
have to judge what constitutes knowledge and un-
derstanding ourselves. There are no wholly reliable
external authorities. Kant's political writings should
of course be treated with respect, but not slavishly
followed. Doyle and Fukuyama are, each in their
own way, critically receptive. I am not sure they go
far enough. For a second conclusion that we might
draw from our engagement with present and Kantian
peace theory is that just as where Kant seems correct
and affords insight so also where he seems mistaken
he may equally lead us to think carefully and judge
problems afresh. Two particular issues stand out as
obviously problematic in his account: (1) the exclu-
sion of women from politically relevant roles; (2) his
apparently optimistic view of the role finance and
trade (operating through self-interest) can play in
the forging of international peace.
In terms of the first, it is clearly fair to say that,
empirically, the facts run against Kant's position. We
can say this not only in the sense that since the end
of the eighteenth century most democratic republics
of the Kantian variety have extended the franchise
to include women, but also in the more interesting
sense that the involvement of women in politics has
tended to add to the peaceful tendencies of poli-
42
ties. Evidence seems to suggest that women are less
likely than men are to condone the use of violence to
resolve international disputes. In terms of the sec-
ond, it might be argued that the extension of trade
and financial interests in the international sphere
has seemingly had as many drawbacks as advantages.
Nineteenth-century imperialism, which was some-
times violent and rapacious, seems to have ridden on
the back of financial and trade interests just as much
as it was restrained by them. The seeds of wars seem
often to lie in conflicts of economic interests. There
seems to be no shortage of interpretations of twen-
tieth-century armed conflicts which provide grounds
for hostilities in disputes over economic spheres of
influence, markets and sources of raw materials.
251
Although Doyle stresses once again the peaceful
impetus brought about by overlapping economic
43
interests, the optimistic Kantian picture has to
be weighed against the ambiguous, if not entirely
contrary, empirical evidence. It is true to say that
Kant was not depending on the world economic
system automatically to bring about peace. His whole
philosophy implies that there has to be an effort of
will to overcome our previous bad habits. His point is
that the long-term contract ± upon which trade and
commerce rest ± cannot function without trust and a
sense of community. From an intelligible or logical
standpoint war and trade are plainly at odds. But
maybe Kant was not sufficiently alert to the way in
which, in the short term, trade can in fact reinforce
our bad habits. Empirically, the case for the moral
worth of unrestricted commerce is not a strong one.
There is an irony here in that contemporary
democratic peace theory has strongly emphasised
the empirical dimension of Kant's thinking. Above
all, the literature has concentrated on analysing the
behaviour of states with constitutions approximating
to the one recommended by Kant. Their conclusion
is that in relation to each other such states behave
well, but they do not tell us it is because of the
convergence of economic interests of these states or
the peaceful intentions of their rulers. From a Kan-
tian perspective, the main gap in the contemporary
literature is its tendency to under-emphasise the
moral side of Kant's argument. Kant's allusion to
empirical developments is intended primarily to
strengthen his moral argument rather than to pro-
vide a knock-down proof. Those who require knock-
down proof of the viability of the peace project
before signing up to it are precisely those Kant
was not aiming to attract to his argument. His `pitch'
for perpetual peace is made to those who already
have moral intentions in politics. The primary pre-
mise for the success of his argument is goodwill. The
liberal-democratic peace will work only for those
who aim to live in a morally regulated peace with
others.
Not surprisingly our comparison of peace theory
then and now leads us to the conclusion that there is
more to be said ± and much more in international
politics itself that can be done ± to engender this
goodwill, the basis of peace as theory and practice.
 HOWARD WILLIAMS

Notes

1. Layne (1994). 27. Kant (1991), p. 105.

2. Doyle (1983a), (1983b), (1986).
3. Doyle (1986), p. 1151.
4. Ibid.
5. See Hegel (1969), pp. 213±17 and also Avineri
(1972), chapters 3, 4. Jonathan Seglow's chapter in
the present volume also explores the notion of `re-
cognition' as a concern for liberal political theory.
6. Fukuyama (1992), pp. 49±50.
7. Ibid., p. 48.
8. Ibid., p. 283.
9. Ibid., p. 281.
10. Kant (1991), pp. 108±14.
11. Fukuyama (1992), p. 283.
12. Doyle (1986), p. 1156.
13. Fukuyama (1992), p. 281.
14. Kant (1991), p. 101.
15. Ibid., p. 101.
16. `By republican Kant means a political society that has
solved the problem of combining moral autonomy,
individualism and social order. A private property and
market-oriented economy partially addressed that di-
lemma in the private sphere. The public, or political
sphere was more troubling. His answer was a republic
that preserved juridical freedom ± the legal equality of
citizens as subjects ± on the basis of a representative
government with a separation of powers. Juridical
freedom is preserved because the morally autonomous
individual is by means of representation a self-legis-
lator making laws that apply to all citizens equally,
including himself or herself. Tyranny is avoided be-
cause the individual is subject to laws he or she does
not also administer.' Doyle (1986), p. 1158.
17. Kant (1991), p. 101.
18. Cf. Saage (1973), pp. 83±93; Kersting (1993), p. 382.
19. Doyle (1986), p. 1156.
20. Ibid., p. 1156.
21. See Russett (1995), p. 168.
22. Kant (1991), p. 104.
23. Doyle (1986), p. 1158. Cf. Doyle (1983a), p. 226.
24. Doyle (1983a), p. 225.
25. Ibid., p. 227.
28. Doyle (1986), p. 1158.
29. Kant (1991), pp. 102±3.
30. Fukuyama (1992), p. 281.
31. This is a fairly striking example from Fukuyama
(1992), p. 283: `The United States and other liberal
democracies will have to come to grips with the fact
that, with the collapse of the communist world, the
world in which we live is less and less the old one of
geopolitics, and that the rules and methods of the
historical world are not appropriate to life in the post-
historical one. For the latter, the major issues will be
economic ones like promoting competitiveness and
innovation, managing internal and external deficits,
maintaining full employment, dealing co-operatively
with grave environmental politics, and the like.'
What might be said from a Kantian perspective is
that rulers should have attempted to overcome geo-
politics just as much before the collapse of commun-
ism as after it.
32. Fukuyama (1992), p. 182.
33. Cf. Williams (1983), pp. 129±37.
34. Doyle (1986), p. 1158.
35. Ibid., p. 1160.
36. Kant (1991), p. 108.
37. Doyle (1983a), p. 225.
38. Doyle (1986), p. 1159.
39. Kant (1991), p. 109. Contrary to Doyle's view the
supposition of progress in history is primarily a kind of
heuristic device (Doyle, 1986, p. 1159) ± but a
heuristic device with a practical rather than theore-
tical importance.
40. Fukuyama (1992), especially chapters 1, 5 and 6.
41. Cf. Oneal and Ray (1997) and Mansfield and Snyder
(1995). Note the conclusion of the second article: `In
the long run, the enlargement of the zone of stable
democracy will probably enhance the prospects for
peace. But in the short run, there is a lot to be done to
minimise the dangers of the turbulent transition' (p.
38).
42. Lisa Brandes, Public Opinion, International Security and
Gender: The United States and Britain Since 1945, Yale

26. This is my own translation; cf. Kant (1991), p. 102: University PhD dissertation, 1994. Cited in Russett
`The Right of Nations shall be based upon a Federa- (1995), p. 167f.
tion of Free States.' 43. Doyle (1986), p. 1161.

252
 19
CONSTRUCTING
INTERNATIONAL COMMUNITY:
LIBERAL THEORY,
DEVELOPMENTAL
COMMUNITARIANISM
AND INTERNATIONAL ETHICS
Peter Sutch
Introduction
This chapter offers a reading of liberalism's normative
theory of international politics. As the ideology of the
dominant powers in the recent history of interna-
tional politics and as the theoretical substructure that
underpins the claim that human rights should be the
guiding force in the future development of interna-
tional relations, liberalism undeniably forms a vital
part of the world order. In trying to determine whether
the expansion and (to use a very popular liberal
characterisation of the process) ultimate universalisa-
tion of a liberal human rights culture really is a move
towards international justice as liberals claim, an
exploration of the way in which this process is justified
in contemporary theory offers some very interesting
insights. In particular, in this chapter it becomes clear
that there are two distinct traditions in contemporary
liberal international theory: the cosmopolitan strand
which has been traditionally attributed to liberalism
and a more particularistic norm-governed approach
which is derived from liberalism's often hidden, non-
Kantian (occasionally ex-Kantian) ancestry.
253
That these currents are present in liberal interna-
tional theory is highlighted in the recent critical
1
work on John Rawls's contribution to the debate.
Rawls's international relations theory is often viewed
as falling between a rock and a hard place. He is
either a failed liberal universalist who dilutes human
rights in order to accommodate statism or a failed
communitarian who relies upon covert liberal as-
sumptions in order to make international politics
and justice compatible. One of the principal sources
of this interpretation is mainstream liberal interna-
2
tional relations theory; perhaps this explains, in
part, the hostile reaction to (or, more often, the
sheer unwillingness to engage with) Rawls's ideas or
with the ideas of those thinkers who take a similar
approach to questions of international justice.
In exploring contemporary liberal international
relations theory, this chapter falls into two parts.
First, I want to suggest that the dominant liberal
response to developments in world politics and
international relations theory does not adequately
show why international justice must be cosmopolitan
3
in character. My principal concern is that the
 PETER SUTCH
constructivist move from the requirement that prin-
ciples of ethics be universalisable to the requirement
that principles of ethics be cosmopolitan is incom-
plete. What is missing is an appropriate conception
of international community. I will then propose that
the more cautious reaction of what I will term (in
order to make a clear distinction between the types
of liberal theory on offer) the constructivist and
developmental communitarianism of thinkers such
4 5
as Frost, Rawls and Walzer offers the invaluable
link between international politics and constructi-
vist liberal theory that the liberal tradition needs if it
is to take account of the problems and possibilities
posed to it by the extension of its concerns into the
international arena. The Hegelian liberalism
Frost, the communitarian liberalism of Walzer and
the political liberalism of Rawls promise a genuine
alternative to cosmopolitanism by (as Brown writing
about the position of Frost notes) `starting from the
position that the value that people and states quite
clearly do assign to sovereignty cannot simply be
dismissed as the product of defective powers of
reasoning or some kind of persisting mass hallucina-
6
tion.'
Constructivist Cosmopolitanism:
Contemporary Liberal Internationalism
Contemporary liberal international relations theory
can best be described as `constructivist cosmopoli-
tanism'. Theoretically it is constructivist, showing
the influence of liberalism's engagement with con-
temporary debates in political philosophy. Politically
it is cosmopolitan in that it maintains that the
projects it has inherited from its Kantian origins still
represent the only practical liberal solution to the
moral problems of international politics. There is
nothing particularly surprising about this opening
statement. The extension of liberal values and prac-
tices to international relations is rarely regarded as
anything other than an absolute good, yet there is
nothing in constructivist liberal political theory that
requires or necessarily underwrites cosmopolitan in-
ternational relations theory. In this chapter I suggest
that a successful constructivist liberalism (hereafter
simply `liberalism'), if it is to produce a political
theory suited to world politics, must take proper
account of the way it constructs or portrays inter-
national community. In doing so it must relinquish
or radically reconstruct some of liberalism's most
prized goals. My reasons for making this argument
stem from several sources: firstly, a belief that the
justifications that liberals are obliged to construct for
their theory are not strong enough to ignore the facts
of pluralism in international politics; secondly, an
uneasiness with the cosmopolitan-liberal assumption
that sovereignty ± one of the political cornerstones of
international relations ± is antithetical or irrelevant
to justice; and thirdly, my belief that the assumption
that sovereignty is hostile to concepts such as human
rights which are thought by definition both universal
of and cosmopolitan in character is obscuring a very
fruitful seam in liberal international relations theory.
One of the most theoretically complete examples
of constructivist cosmopolitanism is to be found in
7
the work of Onora O'Neill. It comprises both the
latest and arguably the most sustainable form of
theoretical justification for international justice.
Her insistence that approaches such as those of
Walzer and Rawls are neither liberal nor just gives
my choice of her work as the example of liberal
international relations theory particular pertinence.
Here I intend to outline the three main claims
motivating O'Neill's argument to highlight construc-
tivist cosmopolitanism's basic position. While this
cannot do justice to its sophistication, this exercise is
useful because it shows how the cosmopolitan out-
look can be maintained within a constructivist jus-
tificatory strategy.
The claims are as follows:
1. The basis of any theory of international pol-
itics must be relevant to all parties concerned;
that is, the premises of international relations
theory must be universalisable across the re-
8
levant domain.
2. The actual circumstances of justice and injus-
tice in the modern world (understood by
looking at the assumptions that agents make
when acting) are more-or-less cosmopolitan in
9
scope.
3. Those theories that acknowledge the theore-
tical relevance of borders (sovereign, commu-
10
nal or moral) are inadequate.
254
 CONSTRUCTING INTERNATIONAL COMMUNITY
From these, O'Neill concludes that constructivist
cosmopolitanism is the only valid option. Now in
the cause of brevity `cosmopolitanism' as a political
theory can be usefully explained via Thomas Pogge's
description of its core elements. He thinks that three
elements are shared by all cosmopolitan positions:
First, individualism: the ultimate units of concern
are human beings, or persons ± rather than, say,
family lines, tribes, ethnic, cultural or religious
communities, nations or states. The latter may be
units of concern only indirectly, in virtue of their
individual members or citizens. Second, univers-
ality: the status of ultimate unit of concern at-
taches to every living human being
universally . . . Third, generality: this special
status has global force. Persons are ultimate units
of concern for everyone ± not only for their fellow
11
compatriots, fellow religionists, or such like.
This quotation is particularly interesting because of
the description of universality that Pogge offers. It is
not simply a depiction of formal universality. The
substantive liberal commitments to individualism
and generality are incorporated into the description
of moral universalism. This (sleight of hand?) is char-
acteristic of much liberal international relations the-
ory. My concern here is to ensure that this fact does not
go unnoticed and to examine the pattern of argument
that allows (or purports to allow) cosmopolitan the-
orists to link the formal criterion of moral universali-
sability with the liberal-cosmopolitan political project.
What is instructive about O'Neill's theory is that its
goal is to lay out fully the predicates of her thesis and to
show the validity of her reasoning in so doing.
So let me state my reading of O'Neill's work
boldly. My principal contention is that claims two
and three in her argument, which are not (in my
12
opinion) defensible, illicitly `construct' the con-
ception of international community which allows
O'Neill to suggest that principles of practical reason
must be universalisable and cosmopolitan. The po-
litical core of O'Neill's cosmopolitanism (and that of
liberal international relations theory more generally)
is the claim that `once the legitimacy of treating
actual state boundaries has been called into question,
an appeal to sovereignty or to assumed lack of
255
responsibilities to non-citizens cannot automatically
13
outweigh an appeal to justice.' But how does
O'Neill call the legitimacy of such boundaries into
question? Essentially, she does so by suggesting that
the requirement that practical reason be universali-
sable plus the fact that we are connected to all other
agents on a more-or-less cosmopolitan scale requires
a cosmopolitan approach to international politics.
The `either-justice-or-sovereignty' argument has
been one of the principal reasons why liberal ideal-
ism (as it is often termed) now has little support in
the study and practice of international politics. The
alternative form of liberal international relations
theory that I wish to outline is, I shall argue, appro-
priately universalist but not cosmopolitan ± and, as
such, seems to make more sense of both our actual
commitments in world politics and the liberal drive
for justice.
Universalisability and Cosmopolitanism
The flip-side of my position is that, of O'Neill's three
claims, only the first is truly stable. Its stability comes
from its constructivist methodological basis and, to
advance the overall argument, we should now grasp
the meaning of `constructivism', a vital part of con-
temporary liberalism's armour, in this context. To
construct a political theory is to avoid reliance upon
discredited metaphysics, or contested truth-claims,
or the revealed word of one culture's prophets, and
thus to avoid the criticisms of foundationalist reason-
ing that many fear threatened the paralysis of justi-
ficatory political philosophy. O'Neill's elegantly
simple case for theoretical constructivism in ethics
is that the reasons we give for selecting principles of
international ethics must be relevant to all con-
cerned parties. For her:
Constructivism in ethics . . . is not a novel
philosophical method or procedure. To construct
is only to reason with all possible solidity from
available beginnings, using available and followable
methods to reach attainable and sustainable con-
14
clusions for relevant audiences.
The key to understanding O'Neill's work is the idea
that principles of ethics must be accessible to the
 PETER SUTCH
`relevant audiences'. O'Neill's case, put simply, is
that any stretch of reason that cannot appeal to all
relevant agents and is therefore not universal in its
domain cannot be called `reasoned'. She goes on to
show that, of the various conceptions of practical
reasoning offered by many universalist and particu-
larist thinkers, most do not meet even this basic
standard. Arguments that employ `end-oriented' rea-
soning, such as that used by `platonist' thinkers who
provide a conception of a specific, independently
specified good `end' as being ultimately the purpose
of agents' actions, will appear arbitrary unless the
appropriate metaphysics is universally validated (ex-
tremely implausible in the light of the apparent
diversity of human ends). Instrumental reasoning
(again end-oriented) will appear `barefacedly' arbi-
trary to those who do not share its contestably
subjective conception of practical reason's goal.
Even `act-oriented' reasoning (theorising the re-
quirements of action without reference to any spe-
cific end towards which it may be directed) of the
particularist variety (communitarian norm-oriented)
would appear incomprehensible to `outsiders' (those
alien to the particular context) even when those
15
`outsiders' fall within the relevant sphere of action.
Remember that all O'Neill is claiming here is that
principles have to meet the requirement that they
could be universalised to cover all relevant others.
Her blanket rejection of all end-oriented and source-
derived conceptions of practical reason does not (on
the strength of arguments made thus far) have to be
accepted, but the requirement that they have to be
accessible to all concerned does. O'Neill argues that it
is vital that we develop intelligible and abstract
principles of action that can avoid the problems of
relativism and idealisation. Here, `relativism' and
`idealisation' are key notions in understanding the
failure of much international and political theory.
The main conclusions from her brief survey of con-
temporary thought are that particularistic relativism
cannot adequately provide for the modern circum-
stances of justice and that much universalist writing
appears arbitrary in that it relies upon idealised
starting points. The simple case being made here
is that, in accepting that foundational universalism
idealises (makes up) its ethical starting point, we are
not required to endorse relativism. We are not
256
allowed to pretend that all the world is a liberal
utopia but neither are we allowed to pretend that
there are no ethical and political issues that we
genuinely share and that constitute the context of
our actions. So to generate genuinely shared reason-
ing, one must operate with premises that are based
upon considerations abstracted from agents' actual
situations, accessible to all because they really do
apply to all.
The question now is whether constructivism sup-
ports cosmopolitan universalism. There are, O'Neill
claims, two senses in which principles may be for-
mally universal. The most elementary is if a principle
`applies to or holds for all rather than merely for some
cases in its domain.' The second sense is where `the
range of beings is seen as extensive: the scope of
principles is inclusive, perhaps (more-or-less) cosmo-
16
politan.'
Of these senses of universalism, the first definition
is relevant to O'Neill's case that all those agents
concerned must (potentially) be able to agree that
the principles of practical reason adopted are rele-
vant to them in their shared condition. This, it must
be noted, does not tell us which agents are concerned
by what problem in any particular domain. The
second feature is yet to be proven relevant, relying
upon a heavily contested empirical and theoretical
notion of how agents are meaningfully connected to
one another across the globe. Indeed, it appears to
rely upon a liberal individualist conception of agency
that does not actually derive (and cannot be derived)
from O'Neill's constructivism. This can be shown, I
believe, by applying J. L. Mackie's questions con-
cerning the status of ethics and universality to
17
O'Neill's argument.
Taking issue with the philosophical requirement
that practical reason be universalisable, it is clear
that O'Neill is suggesting we must focus upon what is
morally relevant and discount those aspects of our
lives that are `arbitrary' in some way. For her, local
and bounded traditions and practices are examples of
such arbitrariness and fall far short of the require-
ment of practical reason because they do not and
cannot take adequate account of all relevant and
connected others (the links between agents across
such particularised contexts). This follows, she main-
tains, from the logical structure of moral reasoning
 CONSTRUCTING INTERNATIONAL COMMUNITY
itself and not from some substantive, idealising lib-
eral-cosmopolitan standpoint. I do not believe that
this is the case. Following Mackie's seminal line of
reasoning here is particularly enlightening. He be-
gins by posing the same question as O'Neill (albeit in
a different form), but when he explores the logical
structure of moral reasoning he comes to a very
different conclusion.
Do the desires and especially the sufferings of
other people, if known to me, constitute a reason
for me to do something, if I can, or try to do
something to satisfy those desires or to relieve
those sufferings? It would be natural to say that
they constitute some reason; how strong a reason,
how easily over-ruled by other considerations,
18
may be a matter of dispute.
For Mackie, moral judgements are universalisable:
any truly moral statement concerning an action is
nonsensical if it cannot hold for any relevantly similar
action. As he notes, the key to understanding the logic
19
of morality is the idea of `relevantly similar'; he goes
on to argue convincingly that there are three stages of
universalisation that attempt to account for it. It is
worth quoting him at some length here:
There are . . . different kinds or stages of uni-
versalisation. In each of them a moral judgement
is taken to carry with it a similar view about any
relevantly similar case. But the first stage rules out
as irrelevant only the numerical difference be-
tween one individual and another; the second
stage rules out generic differences which one is
tempted to regard as morally relevant only be-
cause of one's particular mental or physical qua-
lities or condition, one's social status or resources;
the third stage rules out differences which answer
to particular tastes, preferences, values and ideals.
It is at most the first stage . . . that is built into
the meaning of moral language: the correspond-
ing logical thesis about the second stage is more
controversial, while that about the third stage
20
would be plainly false.
If this argument holds, and I believe it does, then
O'Neill is not merely outlining the necessary logical
257
structure of ethics. Rather, she is insisting that
accounts of ethics that do not give ethical priority
to individuals are subordinate to her cosmopolitan
conception of ethics which clearly does. This is not a
justifiable position for a constructivist to hold.
O'Neill's claim is that particularist practices, value
systems and politics are morally irrelevant, that the
treatment of (for example) non-Muslims in a funda-
mentalist Muslim state and in a liberal state are
relevantly similar. This argument employs all three
stages of universalisation and, while it may be in-
tuitively powerful it is, as Mackie points out, a
21
peculiarly liberal stance.
The crucial point to note here is that what is
hidden by the phrase `relevantly similar' is the con-
struction or portrayal of the international political
arena. O'Neill does not make enough of her recogni-
tion that `connection will peter out at differing
boundaries for different agents and in activities of
22
differing sorts in different circumstances.' Rather,
in order to underpin her cosmopolitanism O'Neill
claims that neither particularism nor Rawlsian lib-
eralism can provide principles relevant to the mod-
ern circumstances of justice. Her principal claim is
that the modes of existence they privilege are not
accessible, and cannot adapt to become accessible, to
all relevant agents in world politics. For O'Neill,
anything less than an individual-centred, cosmopo-
litan conception of agency would be so restrictive
that it simply could not generate principles of ethics
that relate to the world as we know it. This critique
of non-cosmopolitan reasoning relies upon her argu-
ment that we can abstract from all the contingencies
and detail of political pluralism to a core of shared
23
reasoning. Both this claim and the claim that we
are in fact connected to all other human individuals
as individuals (and not as, say, citizens of a state) are,
though, highly contestable.
For O'Neill, the constructivist projects (that she
convincingly shows to be necessary) share a common
aim. It is `in the end no more, but also no less, than
an attempt to work towards a result that is neither
unattainable for those whose lives are both led using
the categories and norms of restricted milieu and
connected to those of others with differing categories
and norms, nor unsustainable for what they could
24
become.' However, I believe that the more limited
 PETER SUTCH
projects of developmental communitarianism take
more than satisfactory account of this aim. In doing
so, they can avoid O'Neill's criticism of particularism
and build a bridge between the Westphalian, state-
centric system of international politics and liberal
international relations theory, thus making the latter
consistent with the realities of international politics
and its development.
Introducing Developmental
Communitarianism: Constructing
an International Community
I now wish to present three examples of the devel-
opmental communitarian method in international
relations theory. One of the principal advantages of
developmental communitarianisms is that they begin
their constructivist projects by attempting to con-
struct a viable image of international community ± of
the `relevantly similar'. Although I cannot provide a
full account of the conclusions of Frost, Walzer and
Rawls, I do want to demonstrate some of the reason-
ing behind their positions. Not only do they offer a
construction of ethical principles, they also provide
an account of how these principles become valued by
the actors in world politics. This second task is a vital
aspect of their constructivist project because it re-
fuses to assume that the first principles of practical
reason have moral and political validity regardless of
all other considerations yet it still maintains (con-
trary to O'Neill's opinion) the theoretical where-
withal to condemn injustice in world politics.
I have chosen to label Frost, Walzer and Rawls as
`developmental communitarians' for two primary
reasons. Firstly, I wish to stress how far removed
are the justifications of their liberal ideas from those
of liberal cosmopolitanism. Cosmopolitan liberals
believe, as noted in the previous section, that the
first principles of liberalism should apply to all in-
dividuals irrespective of their historical and political
situation. For the developmental communitarians,
however, the justifications of the first principles of
liberalism are themselves contingent upon history,
upon consensus or the shared development of a
background justification for the principles of inter-
national politics. This I believe to be a vital point,
and one which feeds into the second reason for my
258
having chosen the term. The `communitarian/norm-
oriented method' of these thinkers is not an argu-
ment about the essential primacy of community over
the individual but of the justice or injustice of the
dominance of certain principles of international
politics. Their disagreement with liberal-cosmopoli-
tanism does not focus upon the notion of universa-
lisability but on the idea of the `relevantly similar'. In
this regard, denying the reality of claims to sover-
eignty, self-determination or human rights does not
make sense of international politics. To argue, for
example, that sovereign borders and the rights of
citizenship hinder the cause of international justice
assumes that we are justified in including a strong
notion of transnational community in our theory.
The strong abstraction necessary to make this argu-
ment is not thought to be warranted. Rather, argue
the developmental communitarians, we may abstract
to the existent plurality of political communities and
to the consensus upon principles of international
justice implicit in the actions and norms they share.
We can go on to show how important the norms of
human rights and other liberal principles are to the
existence of international politics only if this fact is
clearly an uncontested assumption of world politics.
This distinction makes a significant difference to
claims to have legitimately reached broadly liberal
principles of international justice. Indeed, as earlier
demonstrated, it is hard to see how O'Neill's theory
can reach the cosmopolitan conclusions she does
without the aid of some strong liberal assumptions.
The limited conclusions of the developmental com-
munitarians seem more suited to the realities of the
pluralistic-yet-connected world. The liberalism of
Walzer, Rawls and, to a more limited extent, Frost
is a conclusion rather than an assumption and it is
this fact, one that results from their general method,
that marks them off from traditional liberal and
cosmopolitan theory.
Frost: Social Norms and Human Rights
The basic tension in contemporary liberal interna-
tional relations theory is portrayed with appealing
clarity in Frost's Ethics in International Relations,
which charts the development of the norms of
sovereignty, human rights and democratisation.
 CONSTRUCTING INTERNATIONAL COMMUNITY
For Frost, the starting point for an analysis of inter-
national politics is the `modern state domain of
discourse', an image of contemporary practices that
makes a realistic attempt to construct international
community by recognising the tensions between the
remnants of the Westphalian order and a post-
25
Westphalian system. He explains these tensions
by referring to the settled norms that are shared within
this practice. He lists those norms that he considers
within the modern state domain of discourse under
the subheadings `sovereignty' norms (S), `interna-
tional law' norms (L), `modernisation' norms (M)
and `domestic' norms (D). Summarising them, Frost
claims that that the following goods have the status
of `settled':
S1. The preservation of the society of states.
S2. State sovereignty.
S3. Anti-imperialism.
S4. The balance of power.
S5. Patriotism.
S6. Protecting the interests of a state's citizens.
S7. Non-intervention.
S8. Self-determination.
L1. International law.
L2. Ius ad bellum.
L3. Ius in bello.
L4. Collective security.
L5. Economic sanctions (under specified circum-
stances).
L6. The diplomatic system.
M1. Modernisation.
M2. Economic cooperation.
D1. Democratic institutions within states.
26
D2. Human rights.
The first step in Frost's procedure for looking into the
background for the modern state domain of discourse
is to start with what everybody knows. Essentially,
this process leads to the creation of the above list of
settled norms. It is important to remember that this
seemingly comprehensive list is (or could be) made
up of some strikingly different conceptions of what
each of the norms calls for. It is also quite clear that
there is a certain `tension' between those goods that
concern sovereignty and those that concern the
27
rights of individual persons. Frost argues that lib-
259
eral theorists are right to suggest that `it may be
argued that the primary question for a normative
theory of international relations is whether the system
of sovereign states is a justifiable way of organising
the government of the world.' Yet he recognises that:
On reflection, what is striking about the foregoing
list is the primacy of the first two items on the list.
Most of the other settled goods are in some way
derivatives of them . . . Thus any satisfactory
background theory will have to justify the settled
belief that the preservation of a system of sover-
28
eign states is a good.
Frost's general approach to the philosophy of
international ethics, one that coheres with the cen-
tral tensions of contemporary international relations
theory outlined above, really gets to the heart of the
matter. He recognises that in order to follow the
cosmopolitan line of reasoning one must question
the settled status of S1 ± and for Frost so much in
international politics only makes sense in relation to
S1 that to reject its importance would be to radically
misunderstand the nature of international rela-
29
tions.
The central point of Frost's position could easily
be misread as pleading a conservative case with the
correlative argument that it is never going to be
possible to challenge the settled status of S1. How-
ever, a more generous reading ± and one that Frost's
more recent articulations of his theory appears to
30
endorse ± would argue that it is more a case of
making any theoretical transition of norms justified
in reference to S1 to a system of norms justified (for
example) in reference to D2 compatible with a
practical (and just) transition from a world organised
around S1 to a world organised around D2. Captur-
ing this transition, rather than defining and defend-
ing the old international order or the potential
transnational order, would seem to be a vital task,
vital because it would be inappropriate to build an
ethical theory suited only to an imagined or inde-
fensible theory of international politics.
The resources within the developmental commu-
nitarian tradition of liberal theory are well suited to
this task. The work of thinkers such as Walzer and
Rawls has advantages over much cosmopolitan and
 PETER SUTCH
communitarian writing. Their work is constructivist
in that it attempts to avoid the difficulties of philo-
sophical foundationalism while actively combating
the ethical scepticism that is often seen as a con-
sequence of foundationalism's crisis in theoretical
justification. It therefore attempts to theorise inter-
national politics on premises that are both philoso-
phically secure and able to be the object of some kind
of consensus among the plurality of international
actors. Such an approach, combined with a focus on
the competing claims of sovereignty and universal
rights, would suggest that their ideas really should
demand the close attention of international relations
theory.
Rather than pursue Frost's reading of contempor-
31
ary international relations theory here, I wish to
turn to a brief examination of the work of Rawls and
Walzer which, I maintain, follows a similar theore-
tical pattern to Frost's. For Rawls, this theoretical
and political process would, over time, lead the
actors in the international sphere to realise indepen-
dently that toleration of reasonable pluralism is the
chief virtue in international ethics. This realisation
(in the most optimistic interpretation of Rawls's `Law
of Peoples') would lead to the move from the modus
vivendi of international law to a fully overlapping
consensus on the value of toleration and thus to a
new understanding of human rights suited to liberal-
ism at the turn of the millennium. In Walzer's less
confident brand of developmental communitarian-
ism this process would lead (at the very best) to a
shared understanding (and, more optimistically, a
shared articulation) of a `moral minimum' which
could again equip liberalism with a normative stand-
point capable of tackling contemporary global chal-
lenges.
Rawls and Walzer:
Consensus and Reiteration
Rawls's Amnesty lecture `The Law of Peoples' (here-
after: LOP) is often taken as an incomplete expres-
sion of ideas that have their roots firmly in liberal
political culture. It is therefore viewed as tangential
to the needs of international relations. Nothing
could be further from the truth. Originally entitled
`Constructivism and Universality' the essay is con-
260
cerned to explore precisely those problems faced by
international relations theory today. The liberalism
that Rawls transplants from his Political Liberalism is a
political mechanism for dealing with the question of
pluralism in international politics and examines the
guiding concepts of international relations theory,
testing their legitimacy in the face of prevailing
conditions of world politics. Similarly, the perceived
difficulties in Walzer's work have led to an insuffi-
cient examination of some very interesting ideas.
Walzer's work is either viewed as statist, commu-
nitarian and therefore relativist, or as failing even
these standards and relying on covert and illegiti-
mate liberal principles. His use of `universal rights to
life and liberty' has attracted more than their fair
32
share of criticism. However, his reworking of his
`minimalist morality' and its subsequent portrayal as
`reiterative universalism' as opposed to `covering law
universalism' gets right to the heart of contemporary
33
international relations theory. The principal rea-
son why Rawls and Walzer are treated variously as
failed universalists or failed particularists is because
the aim of reconciling political life within an inter-
national system with an emerging global politics is
inadequately conceptualised.
Interestingly, O'Neill herself criticises non-cosmo-
politan liberal theory ± and particularly that of Rawls
and Walzer ± on many points. The most important
here are her claim that Rawls can only make his law
of peoples attack crucial problems in world politics
by relying (covertly) upon idealised liberal assump-
tions about human agency and the suggestion (aimed
at Walzer and MacIntyre) that any contemporary
international political theory that takes account of
pluralism when constructing an account of ethical
universality is doomed to inaccessibility and thus
political irrelevance. The idea, common to liberal
literature, is that only a cosmopolitan account of
universality provides principles of international jus-
tice. However, the developmental communitarian
account of universality offered by Walzer and Rawls
is not rendered ineffectual by its reliance upon norm-
governed rather than a priori Kantian principles of
practical reason and it does not rely upon concealed
liberal assumptions to construct a useful theory of
international justice. In particular Rawls, in moving
towards the communitarian camp in an attempt to
 CONSTRUCTING INTERNATIONAL COMMUNITY
construct a conception of the relevantly similar and
in modifying his Kantian predicates to make the
liberal criterion of universality a realistic require-
ment in political theory, avoids both these criticisms
and develops a redefined liberal international rela-
tions theory.
O'Neill's case against Rawls is that his famous use
of the modifier `reasonable' in relation to the fact of
pluralism illegitimately assists the work of his theory
of justice. This, however, is not the case. At the most
basic level in Rawlsian constructivism there is a
division between the Kantian and communitarian
elements of his thought. For example, the regulative
role of what Rawls calls Kant's theory of `constitutive
autonomy' is abandoned in favour of the political
34
value of a shared public life. In separating his
constructivism from Kant's `deeper', comprehensive
approach, Rawls notes that `the full significance of a
constructivist political conception lies in its connec-
35
tion with the fact of reasonable pluralism.'
recognition of the importance of this `fact' means
that the in-built cosmopolitan universalism of con-
stitutive autonomy is replaced by the notion of what
Rawls calls `doctrinal autonomy, a conception that
gives full weight to the autonomy of all `appropriate
36
conceptions of society and the person.' Here,
Rawls rejects the notion that constructivism should
provide universality (for which read cosmopolitan-
37
ism) from the start. The analytic idea of the
reasonable links a very thin conception of practical
reason with the idea of the relevantly similar, and
tells us how and why peoples may enter into the
political process in international politics. For Rawls
an appropriate account of practical reason allied to a
norm-governed framework (the construction of the
relevantly similar) provides `both the form that
practical reason requires for its application . . .
[and] the context in which practical questions and
38
problems arise.' The idea of the reasonable ensures
that the conception of agency we use in our con-
struction is accessible to all in the relevant domain.
It is precisely because Rawls must take account of all
appropriate accounts of agency that he believes the
cosmopolitan project to be insecure and he thus
focuses upon peoples as politically organised rather
than as individual agents. This move towards a
particularist stance is indicative of Rawls's general
shift towards a communitarian position. For Rawls,
even though (or perhaps because) international
politics is characterised by a plurality of reasonable
yet incompatible social systems (what he terms the
`fact' of reasonable pluralism), human rights, or a
version of that doctrine, can ± and should ± be
supported by any legitimate state. We must remem-
ber that Rawls believes that the background condi-
tions are central to the construction of any principles
of justice. The spur, if any, to greater cooperation or
to justice in world politics must be seen to stem from
world politics itself and not from Rawls's liberal
position. For this reason, he reconstructs the idea
of human rights and its role in liberal international
relations theory.
For Rawls,
The fact of reasonable pluralism is not an un-
fortunate condition of human life, as we might say
A of pluralism as such, allowing for doctrines that
are not only irrational but mad and aggressive. In
framing a political conception of justice so it can
gain an overlapping consensus, we are not bend-
ing it to existing unreason, but to the fact of
reasonable pluralism, itself the outcome of the
free exercise of human reason under conditions of
39
liberty.
The idea implicit in the emerging political culture of
international society is that societies capable of
international cooperation are already contributing,
and will continue to develop a workable system of
human rights. The recognition that a plurality of
states, some comprehensively liberal, some politi-
cally so, and some non-liberal or hierarchical can
(and do) share political principles informs the inter-
40
national ethics of Rawls and Walzer. Neither of
them believes that there exists an international
consensus sufficiently robust to ground a cosmopo-
litan or egalitarian human rights project. Neverthe-
less both thinkers agree that a `minimal' conception
of universal rights plays an important role in world
politics. In trying to reconcile human rights and
pluralism, Rawls argues in favour of the develop-
mental potential of a constitutional (political) con-
sensus and Walzer for a minimalist morality formed
out of repeated encounters with international pro-
261

blems. For Walzer `the value of minimalism lies in
the encounters it facilitates, of which it is also a
product. But these encounters are not ± not now at
least ± sufficiently sustained to produce a thick
41
morality.' Similarly for Rawls the value of inter-
national justice stems from a consensus on a need to
fix the political situation and put issues of justice
beyond particularist interest, not upon any consensus
strong enough to underwrite cosmopolitanism.
For Rawls the basic principles of the law of peoples
that have evolved historically as a modus vivendi are
unsurprisingly traditional. But the key to Rawls's
project in LOP lies in his reworking of his seventh
principle, that `peoples are to honour
42
rights.' The empirical claim that since The Second
World War human rights have been on the agenda
of international relations theory is supplemented by a
justification of their prominence, as for Rawls the
existence of liberal/cosmopolitan human rights is
unstable in the sense that it lacks the normative
justification appropriate to the international
43
sphere. The idea of justification that Rawls applies
to international relations is modelled on that found
in Lecture IV of his Political Liberalism and he
completes his sketch of a law of peoples by arguing
for two further conditions that lend normative and
practical stability to his project. The first of these two
conditions is that the international order should be
stable `in the right way'; that is, they pursue stability
for reasons of justice rather than of mere accommo-
44
dation (modus vivendi). The second condition is
that the law of peoples should ideally be the object of
an overlapping consensus: people would
come to affirm the moral value of the law of peo-
45
ples. The conditions of stability and consensus
represent the philosophical justification of the se-
ven-point law of peoples that Rawls argues is con-
sistent with both liberal and non-liberal perspectives.
Elements of international law that are controversial
(peculiarly liberal) in character (human rights for
example) are likely to receive whatever justification
they get from a variety of incompatible philosophical
or doctrinal sources. It is here that the relationship
between stability and consensus shows us the source
and potential of Rawls's theory. Stability is the goal
of society, a condition in which we know that those
with whom we share political space will comply with
PETER SUTCH
the institutions of justice. Beyond modus vivendi exist
conditions of society which, if the forms of consensus
(constitutional or ideally overlapping) occur, offer
the possibility of a greater transnational interaction.
It is interesting to see how Rawls cashes out the
relationship between the idea of stability and the
differing forms of consensus in his international
relations theory. Human rights, in Rawls's ideal
portrayal of a liberal international society, represent
a `thick' conception of rights founded upon speci-
fically liberal ideas. They signify that each person is
`assigned a certain status, the status of members of
46
some society who possess rights as human persons.'
human Yet human rights in the fully extended law of
peoples represent a `minimum standard of well-
ordered political institutions for all peoples who
belong, as members in good standing, to a just
47
political society of peoples.' Rawls is acutely aware
that much of international society does not hold a
western individualist conception of the person and
therefore assigns rights differently, and this fact
alters his discussion of stability and consensus.
The form of consensus reached between liberal
and non-liberal states is likely to be analogous to
the idea of constitutional consensus in Political
Liberalism where
a constitution satisfying certain basic principles
establishes . . . procedures for moderating poli-
tical rivalry within society . . . Where there is
agreement on certain basic political rights and
liberties . . . there is disagreement . . . as to the
ideally more exact content and boundaries of these
48
rights.
Constitutional consensus arises, argues Rawls, as
the only workable solution to endless destruction
and strife, and can be thought of as the product of
what Rawls calls the `dramatic shift' in international
49
law since the Second World War. The importance
of this stage in Rawls's theory is both its historicity
and its dynamism, its potential. For him:
At the first stage of constitutional consensus the
liberal principles of justice, initially accepted
reluctantly as modus vivendi . . . tend to shift
citizens' comprehensive doctrines so that they
262
 CONSTRUCTING INTERNATIONAL COMMUNITY
at least accept the principles of a liberal constitu-
tion. To this extent the [state's principles] are
reasonable if they were not before: simple plur-
alism moves towards reasonable pluralism and
50
constitutional consensus is reached.
Let us just look at the three requirements Rawls
places upon international society, his construction of
international community, and see whether it seems
reasonable to use these ideas to support a construc-
tivist theory of international justice. The image of
international politics that Rawls presents is one
where:
1. the fact of reasonable pluralism must lead to
consensus upon the need to fix the political
situation and put such issues beyond particu-
larist interest;
2. the application of principles of justice pays
heed to the guidelines of public reason. These
guidelines and rules must be specified in re-
ference to forms of reasoning available to all;
3. (dependent on 1 and 2) those involved believe
the principles are just and thus the consensus
51
works effectively.
Rawls's use of the `reasonable', it would seem, is itself
a move to ensure the accessibility of principles of
international justice to those whose autonomy does
not rely upon a politically western and ethically
Kantian heritage.
The aims of Walzer's moral theory also explicitly
engage with the problems of how we may link
theories of international ethics with an appropriate
conception of international society. His approach
sums up much of what Frost and Rawls can be seen to
be engaged in:
Advocates of liberal enlightenment confront ad-
vocates of communal tradition; those who aspire
to global reach confront those who yearn for local
intensity. We all know one another's lines . . . I
want to argue from within what I, and many
others, have taken to be the opposing camp; I
want to take my stand among the universalists
and suggest that there is another universalism, a
non-standard variety, which encompasses and
263
perhaps even helps to explain the appeal of moral
52
particularism.
In making his case Walzer contrasts `covering-law
universalism' with `reiterative universalism'. The
central difference between them is neatly sum-
marised by Miller, who writes that
`covering-law universalism' affirms the rightness
of a particular way of life, and sees one nation as
the bearer of that way of life . . . `Reiterative
universalism' . . . recognises that, subject to cer-
tain minimal constraints imposed on all, there are
a number of radically different and valuable forms
53
of life which flourish in different places.
Our main focus here is on the content of those
`minimal constraints' and upon their construction.
For Walzer we cannot use an inaccessible account of
agency to ground a `thick' covering-law universal-
54
ism. Rather, reiterative universalism relies on
something much weaker, on `overlapping expecta-
55
tions' about the behaviour of others. Walzer's
`rights to life and liberty' stem from reiterated claims
made by significantly different agents over time and
in response to specific problems in international
politics.
One reason why Rawls and Walzer are often
sidelined in liberal international relations theory is
that what they believe they are obliged to allow to
stand as legitimate domestic arrangements can take
many forms, some of which may not appeal to our
liberal conception of legitimacy. As Walzer observes,
`by democratic standards, most states throughout
human history have been oppressive and illegiti-
mate, but those are not necessarily or even usually
the standards by which they are judged by their own
56
people.' Given this, plus the contestability of the
foundational assumptions of liberal human rights
theory, Walzer (and Rawls) want to promote inter-
national interaction while guarding against the uni-
versal imposition of what Walzer calls an `alien
57
maximalist morality'. To this end, his primary
principle that expresses moral minimalism in inter-
national affairs is self-determination, a concept ab-
stracted from its many reiterations or occurrences to
signify the protection of the political liberty of
 PETER SUTCH
58
members of a plurality of communities. From here
the rights to life and liberty develop to be repeated
and reiterated at times of international and domestic
tyranny. They do not (yet?) form a transnational
59
moral Esperanto, but the political consequences of
recognising such values are genuinely progressive.
For Rawls, as for Walzer, the central idea is that the
benefits of true stability only derive from a fair
construction of principles of justice among consent-
ing and dependable parties. The rights this process
makes explicit, argues Rawls, are politically neutral
in the sense that they are not specific to western
60
liberal culture.
Conclusion
There is a lot that liberals do not like about the
international relations theory of Frost, Walzer and
Rawls. Prime among these dislikes are the limitations
that these thinkers have to place upon the role of
international justice. For example, it may appear that
the human rights that non-cosmopolitan liberals
develop have no teeth: they cannot do the work
1. See, in particular, Rawls (1996) and (1999), chapters
24 and 25.
2. See, for example, Beitz (1979); P. Jones, `International
Human Rights: Philosophical or Political?', in Caney,
George and Jones (eds) (1996); Moellendorf (1996);
Paden (1997); and Pogge (1994).
3. Cosmopolitanism is a well-defined and well-under-
stood tradition. For the purposes of this chapter,
Brown's analysis stands as a perfect example of what
I have in mind: Brown (1992), pp. 23±51.
4. Frost (1996).
5. See, particularly, Walzer (1977) and (1994).
6. Brown (1992), p. 120.
7. O'Neill (1996).
8. Ibid., especially chapters 2 and 3.
9. Ibid., especially chapter 4.
10. Ibid., chapters 1 and 2.
11. T. Pogge, `Cosmopolitanism and Sovereignty',
Brown (ed.) (1994), pp. 89±122, at p. 48.
12. These points must be considered subject to all the
controversy surrounding the empirical and theoretical
debates over such issues as interdependence and glo-
balisation, and to the philosophical and political
debates concerning the constitution of a moral com-
munity.
we would wish a theory of human rights to do. This
charge is drawn from the long history of human
rights theory that insists that it is possible to develop
a conception of international justice from outside
the practice of international politics. But to do so is
to ignore the constitution of the international com-
munity and thus of the philosophical notion of the
`relevantly similar', thereby risking the alienation of
liberalism from the real world of international rela-
tions. The power of human rights discourse is sig-
nificantly weakened if its justification belongs to one
particular culture, even more so when it appears to
foster the desires of the established hegemonic block.
The liberalism of the developmental communitar-
ians offers a way beyond this problem. If the outline
of this developing liberal discourse does fulfil the
requirements of constructivism in ethics, then it is
certainly time to admit the non-Kantian liberalism of
Frost, Walzer and Rawls into mainstream interna-
tional relations theory. As for the practice: limited
and progressive liberalisation of international poli-
tics may not be all we can hope for, but it may be all
that we are justified in demanding.
Notes
13. O. O'Neill (1994), `Justice and Boundaries', in Brown
(ed.) (1994), pp. 69±88, at p. 85.
14. O'Neill (1996), p. 63.
15. Ibid., p. 51.
16. Ibid., p. 74.
17. As discussed in Mackie (1977).
18. Ibid., p. 78.
19. Ibid., p. 83.
20. Ibid., pp. 97±8.
21. Ibid., p. 98.
22. O'Neill (1996), p. 117.
23. See, for example, O'Neill (1988).
24. O'Neill (1996), p. 64.
25. Frost (1996), p. 79.
26. Ibid., pp. 111±12.
27. Ibid., p. 137.
28. Ibid., p. 112.
in 29. Ibid., p. 112 n. 7, pp. 229±30.
30. Frost (1997a), (1997b).
31. This task is completed in Sutch (2000), forthcoming.
32. See, for example, Doppelt (1980), Luban (1980).
33. Walzer (1994); see also Walzer (1990), p. 509.
34. Rawls (1996), pp. 98±100.
35. Ibid., p. 90.
36. Ibid., p. 98.
264
 CONSTRUCTING INTERNATIONAL COMMUNITY

37. Cf. Rawls (1999), p. 532. level of consensus would play a role in international
38. Rawls (1996), p. 110. relations, but as a vital element of Rawls's conceptual
39. Ibid., p. 144; see also p. xxiv. mechanism (and if appropriately modified) I think it is
40. The language is, of course, Rawls's but the ideas are fascinating at least to attempt the transition.
implicit in Walzer's communitarianism and in early 49. Ibid.; see also Rawls (1999), p. 535.
stages of Frost's work. 50. Rawls (1996), pp. 163±4.
41. Walzer (1994), pp. 18±19. 51. Ibid., pp. 161±2.
42. Rawls (1999), p. 540. 52. Walzer (1990), p. 509.
43. Ibid., p. 542. 53. Miller (1995), p. 10.
44. Ibid. 54. Walzer (1990), pp. 522±3.
45. Ibid., p. 543. For a fuller account of stability and 55. Ibid.
consensus, see Rawls (1996), Lecture IV, section 2. 56. Walzer (1980), p. 128.
46. Ibid., p. 566. 57. Ibid., pp. 519±20.
47. Ibid., p. 552. 58. Walzer (1994), pp. 67±8.
48. Rawls (1996), pp. 158±9. I am aware that there is no 59. Ibid., p. 7.
textual evidence to support the claim that this middle 60. Rawls (1999), p. 552.

265
This page intentionally left blank
 Part Eight
NEW DIRECTIONS
FOR LIBERAL THINKING
This page intentionally left blank

LIBERALISM AND POST-
COMMUNISM
Introduction
The last hundred years have been a troubled time
for liberalism. An idea whose prospects seemed so
bright in the last fin de sie
Á cle was drowned in the
blood of the First World War and the left and right
authoritarianisms that emerged out of that conflict.
This has now all changed. `Suddenly', Ernest Gell-
ner notes, `the century of troubles and tribulations
of liberalism is all over. A century that
looked as if it were to witness liberalism's demise,
in fact documents its almost embarassed triumph. It
does not quite know what to do with it.'
victory was given intellectual form by
Fukuyama in his thesis on `the end of history',
arguing that with the fall of communism there was
no longer any sustained programmatic universal
2
alternative to liberalism. There has always been
a type of liberal historicism that favoured the view
of an evolution through stages in human affairs,
rejecting cyclical theories (Spengler, Toynbee), or
those that thought in terms of stasis and self-
perpetuating order of a Chinese sort. Yet many
impeccable liberals (like Dahrendorf) found Fu-
kuyama's radical restatement of a liberal historicism
profoundly offensive and ± unliberal. Some sort of
victory for liberalism was, however, recorded with
the fall of communism. The question was not only
what to do with this victory, but for the post-
communist countries the problem remained: how to
do it? Which liberal tradition is the most appro-
20
Richard Sakwa
priate for each specific country? And how can
liberalism be established from scratch?
The post-communist transitions have raised the
most profound questions, not only about the current
status of liberalism, democracy and economic mod-
ernisation, but also by forcing old issues once again to
the fore. In a way post-communist transformation is a
replay of the struggle for liberalism in Europe from
the late seventeenth to the nineteenth centuries.
often Similar fears about the dangers of excessive state
power have been voiced, while the establishment of
secure property rights was at the centre of the
1
This European and post-communist social revolutions.
Francis Indeed, post-communism forces us to rethink the
relationship between the two faces of liberalism:
political democracy and the market economy. Emer-
ging out of the command economy it is understand-
able that the predominant belief at the fall of
communism was that without economic freedom
all other freedoms, civil and political, are threatened.
It was soon realised, however, that freedom (espe-
cially if confined to the economic sphere) is not the
sum of liberalism.
 Merquior, we can consider liberal-
Following Jose
ism as ideally comprising the following four elements:
`(1) the (negative) freedom of not being subjected to
arbitrary interferences; (2) the (positive) freedom of
participating in public affairs; (3) the (interior) free-
dom of conscience and beliefs; and (4) the (personal)
3
freedom of self-development for each individual.'
This is a full-bodied liberalism drawing on all five
269
 RICHARD SAKWA
of the liberal traditions identified by Jaguaribe (clas-
sical, conservative, social, sociological and neo),
whereas the liberalism prevalent under post-commun-
ism is predominantly of the neo-liberal type. The
ideologists of neo-liberalism are Von Mises, Hayek,
Nozick and Milton Friedman, all condemning state
regulation in favour of the unrestrained operation of
market forces and a radical individualism: prioritising
(1) and underplaying (2)±(4).
While it is easy to condemn the excesses of neo-
liberalism, especially in a post-Thatcher/Reagan era,
it is more difficult to sustain a defence of the
rationality of the market and the enterprise as the
most efficient allocater of resources while simulta-
neously condemning the neo-liberal demonisation of
the state and the attempt to impose `the law of the
jungle on societies whose stability was obtained
thanks to the healthy effects of the welfare state.'
These predicaments will be examined on the basis of
developments in the post-communist world, primar-
ily in Russia and Poland.
From the Wreckage of Communism
The emergence of political liberalism in Russia under
Gorbachev in 1987±88 during `perestroika' (`restruc-
turing') was as a distinctive addition to socialism; only
later did it emerge as an autonomous force in its own
6
right to triumph between 1991 and 1992.
triumph, however, was ambivalent. Nodari Simonia
notes that `the defeat of the attempted coup in
August 1991 was no victory for democracy. Political
democracy was the dominant tendency, but there
was neither the social nor the economic base neces-
sary for the establishment of democracy. A civil
7
society had not yet been created.' He goes on to
argue that `while the remnants of the ideological
nomenklatura waged rearguard battles against Russian
democracy, an economic nomenklatura moved into
8
position to attain political power.' It is for this
reason, according to Jerry Hough, that the end of
communism in the USSR, although no less of a
revolution than the events of 1917 that had brought
the Bolsheviks to power, was relatively bloodless.
Within a year organised liberalism appeared to
have exhausted itself, and since then has been at a
low ebb. According to Lev Timofeev, one of Russia's
leading liberals, the mistake had been to ally with the
4
democrats from the old Communist Party of the
10
Soviet Union (CPSU) in the late 1980s. By
1995 he had become convinced that there was no
basis for a liberal democratic society in Russia. The
problem was no longer the power of selfish ruling
elites, but the conservative people, above all the
11
patriarchal-minded mass of peasants. It is this sort
of contemptuous attitude towards their own people
that has characterised Russian liberalism. Sogrin
takes a more balanced view, arguing that while there
were the preconditions for liberal democracy in
Russia, the objective reasons for the failures and
shortcomings of liberalism can be found in the
specific liberal democratic model that was adopted
12
and imposed by Russian liberals. No distinction
was drawn between liberalism and democracy, and
5
only from around 1990 did liberalism begin to
13
emerge as an autonomous political tendency. Even
then, the liberal trend failed to develop into a
genuinely popular movement and source of mass
14
political action.
Jerzy Szacki, a professor of sociology at Warsaw
University, argues that while the philosophy of
economic liberalism as such appealed only to a
minority of dissidents, the political value of what
he calls `protoliberalism' as an alternative `neglected
15
path of anti-communism' actually increased. He
This argues that liberalism was a phenomenon of the
western world with few roots in Eastern Europe,
and that communism, far from suppressing indigen-
ous liberal development, artificially fostered it as its
alternative political vision. As David Ost puts it,
`when the enemy became the state, the opposition
16
became liberals.' The fall of communism would
allow the anti-liberalism of Catholicism, in Szacki's
view, to reassert itself (or, perhaps, any one of the
numerous anti-liberal traditions of the area). He fails
to realise that in modern pluralist societies even
intrinsically authoritarian faiths can, if freely adopted
by its members and when part of a broader patch-
work of equally autonomous interests, strengthen the
liberal heterogeneity of society.
9
The argument, nevertheless, is an important one.
Liberalism appeared less a systemic necessity and
outgrowth of the society than an instrument with
which to assault the party-state. Many have argued
270
 LIBERALISM AND POST-COMMUNISM
that the social basis for liberal democracy was, quite
17
simply, lacking in Poland, and if that was the case,
then in Russia it stood no chance whatsoever. In
Russia it was Andrei Sakharov who became the
symbol of the liberal opposition to the regime, yet
paradoxically it was Solzhenitsyn, usually charac-
terised as an authoritarian patriot, who advocated
a gradualist strategy of self-liberation from commun-
18
ism while the `liberal' Sakharov espoused a more
revolutionary break with the past. It is out of para-
doxes such as these that post-communist liberalism
was born. In Hungary by 1998 the Prime Minister,
Victor Orban, even went so far as to argue that
certain elements of West European liberalism could
not be `simply transferred' to Central Europe, and the
attempt to do so might often do `more harm than
19
good'. Faced by rising crime rates the response is
understandable, but his argument that the ruling
Federation of Young Democrats-Hungarian Civic
Party was now `much more than a liberal party'
suggested a deeper unease with liberalism itself.
We should thus examine liberalism as a transforma-
tive challenge, a set of ideas and a complex of
institutions.
Liberalism and Capitalist Transformation
The central contradiction of post-communist liberal-
ism is that as an economic philosophy it was pre-
dominant, and its political ideals of civil rights,
political pluralism, the separation of powers, toler-
ance and advocacy of non-militaristic solutions to
domestic and foreign problems became the main
currency of debate, but as an organised force it
has tended to be weak.
Politically Organised Liberalism
The relationship between liberalism and mass de-
mocracy is at the best of times problematic, but in
post-communist countries it has fallen prey to the
technocratic temptation. In 1991 the liberal concept
of a parliamentary constitutional state prevailed over
residual notions of a socialist alternative. But, as in
Germany in 1919, `politically organised Liberalism,
the Liberal parties, played no part at all in this
victory; and the victory drew Liberal democracy
271
itself into a political danger zone, where the risks
20
were far greater than ever before.' The liberal
`party' in the anti-communist revolutions of 1989±
91 took the form of an inchoate social movement
whose leadership was amorphous and its policies
limited to the negative programme of opposition
to the decaying communist regime. Its positive pro-
gramme was a transformative one that assaulted the
society from which it sought to draw its democratic
legitimacy: transformative legitimacy ran counter to
democratic legitimacy.
This was particularly the case in Russia. The main
liberal `party' in the December 1993 State Duma
elections, Yegor Gaidar's Democratic Choice of
Russia (DVR), won 15.51 per cent of the party list
vote, coming second to the 22.92 per cent for
Vladimir Zhirinovskii's Liberal Democratic Party
of Russia (LDPR) (which of course is not a liberal
party). Rather than organising as an autonomous
force, liberals relied on the benevolence of President
Boris Yeltsin and sought to use the state to promote
their policies. They thus became the first `party of
power', until losing out in political intrigues and
defeat at the polls. Indeed, it appeared that there
were more liberals in the entourage of the relatively
authoritarian Yeltsin presidency than in the parlia-
ment itself.
For liberals like Gaidar, it was fear of losing power
in 1994 rather than attempts to institutionalise their
dominance that led them to activise their attempts
to create a party to support their programme. The so-
called democratic movement and the liberals asso-
ciated with the Yabloko party fared dismally in the
regional elections between September 1996 and into
21
early 1997, not winning a single governorship.
Federal party politics counted for very little in these
elections in the regions, and even less in the repub-
lics, especially where titular ethnic elites dominated
political life. Regime-type systems have proliferated
in the regions and republics where more often than
not both liberalism and democracy have been com-
promised.
In the December 1995 elections the combined
liberal vote was only 16.2 per cent. Yet, as Steven
Fish has argued, the reason for the liberals' crushing
defeat lies not in any profound anti-liberalism of the
Russian people as such, nor in the institutional
 RICHARD SAKWA
framework, the electoral system, lack of resources or
disunity among the liberal parties. Rather, he argues,
the answer lies in the deficiencies of the liberals
themselves, and in particular their poor organisation
(consistently underestimating the importance of
party organisation), unprofessional and miscon-
ceived campaigns, and the legacy of the behaviour
22
of liberals in power. Organised liberalism in the
December 1999 parliamentary elections was even
more fragmented and disorganised, grouped around a
number of leading personalities whose alliances were
as fragile as the differences between them were
minimal. The Right Cause (Pravoe Delo) grouping
brought together some of Russia's leading liberals,
including Boris Nemtsov, Anatolii Chubais and
Boris Fedorov, together with `the right-wing liberal'
Yegor Gaidar, `the odious architect of the attempts to
23
implement ``shock therapy''.' According to Buzga-
lin, the author of the just-quoted sentiments, Right
Cause was dangerous for Russia because of the
`communist fanaticism' with which they sought to
implement their plans. Few in Russia would forget
the role that some of the members of Right Cause
played in creating the system of oligarchical capit-
alism, accompanied by the immiseration of large
sectors of society. All of this is true, yet the trajectory
of post-communist liberalism has a more profound
meaning.
Marketisation and Liberalism
Like the communist regimes that they replaced, the
liberalism that emerged in the post-communist states
was forced to create the grounds for its social ex-
istence by a massive act of economic transformation.
Broader humanistic concerns were thus subverted by
the economism implicit in this transformative mis-
sion. Bellamy has noted the shift from ethical to
24
economic theories of liberal democracy, but under
post-communism this has taken extreme forms. Pol-
itics has been subordinated to the alleged economic
imperative of social transformation. Old tensions
within the liberal tradition, above all between social
liberalism, with its more positive view of liberty and
state activism, and neo-classical liberalism, stressing
the redemptive powers of the market, have burst out
once again. Torn between social democracy and new
272
right thinking, it is unclear what remains of a dis-
tinctively liberal agenda.
Economic liberalism became the guiding principle
of post-communist transformations. The reasoning
behind `shock therapy' was outlined by Leszek Bal-
cerowicz, arguing against any evolutionary exit out of
the old state-planned system in favour of a complete
25
and radical institutional break with the past. The
issue of sequencing was crucial for him, insisting that
systemic market-oriented reforms had to be achieved
during the period of `extraordinary politics' of high
political mobilisation early on in the process. An
even more robust defence of economic liberalism
came from Vaclav Klaus, Prime Minister of the
Czech Republic from 1991 to 1997, where he praised
the virtues of the untrammelled market and con-
26
servative liberal principles, although in practice he
presided over a semi-reformed system that allowed its
own type of crony capitalism to emerge. Szacki's
warnings in this context about the shallowness of
true political liberalism in the region need to be
taken seriously. The balance drawn between eco-
nomic liberalism, liberal political ideas, the collec-
tive rights of the nation and the persistent support for
many features of the old state socialist welfare system
differed between states, but throughout the region
none could avoid the dilemmas posed by these issues.
Transformative Liberalism
These problems were particularly acute in Russia
where, according to Sogrin, a `radical liberal' ten-
dency predominated, characterised by a utopianism
that ignored the character of Russian society and
27
civilisation. From November 1991 liberals like
Gaidar, Chubais and Fedorov dominated the Russian
government. From 2 January 1992 they launched a
`shock therapy' economic reform with the liberal-
isation of some prices. The economic legacy of this
period is still debated, but the immediate result was
near-hyper inflation, a slump in production and a
record deterioration in standards of living. More
positively, queues disappeared and market principles
began to be introduced in the economy. By the end
of the year Yeltsin had sacrificed the liberal refor-
mers, and from December 1992 Victor Chernomyr-
din took over as Prime Minister. Andrei Kozyrev, the
 LIBERALISM AND POST-COMMUNISM
liberal foreign minister, stayed on until the Decem-
ber 1995 parliamentary elections saw the commu-
nists become the largest grouping in the Duma, but
in the economic sphere only Chubais survived to
mastermind the privatisation of the economy. This is
one of the most spectacular shifts in ownership
relations in history; today, less than a quarter of
the national economy is in state hands. Mass redis-
tribution of ownership was accompanied by equally
spectacular corruption.
Chubais was appointed head of the presidential
administration following Yeltsin's re-election
June±July 1996, and was made a First Deputy Prime
Minister in March 1997. This time, together with
the other First Deputy Prime Minister, Nemtsov, he
sought to end the domination of the `old economic
nomenklatura', to break up the huge natural mono-
polies and the utilities, and to pursue housing and
28
pensions reform. Although yoked together, Chu-
bais and Nemtsov represent two very different trends
in Russian liberalism. Chubais is a radical liberal,
whereas Nemtsov represents a tendency dubbed by
Sogrin as `national liberalism' ± although it is un-
likely that a direct reference is intended to Germa-
ny's National Liberal party established in 1867, and
whose left wing seceded in 1880. While respecting
traditional liberal values of private property, the
market, pluralism and political democracy, Russian
national liberals sought to replace abstract or Wes-
tern liberal models with one more suited to Russian
conditions. Disenchanted with Western democracy,
national liberals began to favour authoritarian gov-
ernment to democracy. They regarded the social
liberal idea of a universal welfare state as inappropri-
ate for Russia, favouring instead a carefully targeted
social policy. A national liberal economic model
equally rejects the monetarism of the radical liberals
and favours a more active role for the state in
national economic development and a degree of
29
protection for national production. A representa-
tive of this tendency was Alexander Lebed, who
sought to free the state from bureaucratic, nomenk-
latura and criminal dominance, forces that had
usurped the fruits of reform by distorting the mar-
30
ket. National liberalism represented the adaptation
of liberalism with a social conscience to Russian
conditions.
While there is no doubt that Nemtsov is an
economic liberal, in politics he is more of a national
liberal with a statist bent. An admirer of Peter the
Great, he noted in an interview in February 1997:
I'm a liberal economist and support the idea of a
strong state in politics. Yegor Gaidar, for instance,
is a liberal in economics and a liberal in politics,
in the bad sense of this word. In Russia, such
31
liberalism is ruinous.
in He advocated using Russia's financial resources to
buy up the city and naval base of Sevastopol. `His-
torical justice should be restored by capitalist meth-
ods,' he suggested. As Peter Rutland notes, `Nemtsov
may be a liberal, but he is a liberal with a Russian
32
face.' The same could be said of Sergei Kirienko,
Prime Minister from April to August 1998, when he
struggled to transform oligarchical capitalism into a
more open version. His failure to deal with the legacy
of nearly a decade of mismanagement led to the
partial default of 17 August 1998, and his own
dismissal six days later.
The `Sin of Constructivism'
To achieve the desired socio-economic results and to
generate market-oriented behaviour by economic
actors, the post-communist state intervened on a
massive scale in the early years of the transition, thus
compromising, both politically and economically,
many of the principles of the new `liberal' order.
33
This `sin of constructivism', in the words of Szacki,
is reminiscent in an inverted form, not surprisingly,
of earlier `grand designs' of the communist sort. The
Polish sociologist Radzislawa Gortat also referred to
34
`liberalism as constructivism', describing how lib-
eralism in Poland was born in a way that contra-
dicted its nature. Thus in its struggle to create
capitalist institutions the Tadeusz Mazowiecki gov-
ernment relied on the authority of the state to
achieve the desired results. The liberal reforms,
lacking a social interest to act on its behalf, relied
on an external agency to articulate an inevitably
synthetic and derivative version of the social ideal.
In the absence of social interests to articulate and
promote the programme, the Balcerowicz plan of late
273
 RICHARD SAKWA
1989, and the Gaidar model of social transformation
in Russia from late 1991, could only be the creation
35
of the state.
It is well-known that Hayek condemned Keynes's
advocacy of government intervention in the econ-
omy to correct the market as presumptious `con-
structivism'. As Bellamy puts it (following John
Gray), Hayek distinguishes `between the notion of
society as a spontaneous order embodying evolution-
ary reason, on the one hand, and constructivist
36
rationalist theories of society, on the other.'
his Road to Serfdom of 1944, Hayek had condemned
western governments for subverting the law and
replacing it with legislation. But what spontaneous
order could there be in post-communist conditions?
Liberals were forced to adopt the guise of revolu-
tionaries and act not as the defenders of a sponta-
neously generated social order, but as a small sect of
priests who claimed a privileged understanding of the
laws of history and social development. Solutions
were imposed upon society and not derived from
society.
This should be placed in context, and the notion
of post-communist neo-liberal reformers as `Bolshe-
viks of the market' should not be taken too far. In
form, content and methods the differences are so
large as to render analogies almost meaningless. As
Andrzej Walicki notes:
One of the most conspicuous characteristics of
post-communist Poland is the omnipresent con-
tradiction between the radical, uncompromisingly
anti-communist ideology and the principle of
compromise that has enabled a smooth, peaceful
37
transfer of power.
The new authorities were revolutionary in their
actions but liberal in their methods. Yet the charge
of constructivism is accurate, and one that tainted
post-communist liberalism (above all in Russia) with
the stamp of illiberality, if not authoritarianism and
hypocrisy.
After the radical phase a reaction set in marked by
a return to the politics of gradualism and a reconsi-
deration of the social values of reform. In Poland this
took the form of the Finance and Deputy Prime
Minister (from May 1994 to February 1997) Grzegorz
Kolodko's `Strategy for Poland' document of 1994.
The Balcerowicz group, it argued, had an ideological
commitment to liberal doctrine, pursuing its pro-
gramme in a centralised, technocratic manner that
treated the public as passive objects of transition.
This was, according to Jerzy Hausner, an `imperative'
method of post-communist change, that gave way
later to the `interactive' approach favoured by Ko-
lodko. The `interactive' approach sought to ground
ongoing processes of social change in a continuing
In discursive relationship between the authorities and
the public. Rejecting the view that there was `no
alternative' to the transition to an Anglo-Saxon style
economy, the `interactive' approach recognised a
variety of capitalisms and that the role of the state
would be subject to negotiation rather than a priori
38
rejected.
It was this more interactive approach that was
favoured by the liberal opposition in Russia, and in
particular Grigorii Yavlinskii's `social liberal' Yablo-
ko party. However, there appeared to be no devel-
oped alternative to radical liberal economic policies.
Attempts to devise alternatives, even by the `demo-
39
cratic opposition' in Yabloko, in practice failed.
These social liberals emphasised the social basis of
any viable political or economic order, while at the
same time insisting that it had to be rooted in the
national interest. They took the universals of liberal-
ism and sought to present them in a patriotic idiom.
The behaviour of the radical liberals in power in
Russia was characterised not only by constructivism
but also by corruption, the denigration of patriotism,
the rupture of relations with the former Soviet states,
and by a narrow dogmatism that appeared to regard
the dismantling of state capacity as the pinnacle of
the liberal project. The social liberals promised to
continue the reforms but in a more socially oriented
way. They would remain tempted, however, by the
sin of constructivism.
Constrained Liberalism
Kryzstof Bielecki, the liberal Polish Prime Minister in
1991, has talked in terms of `constrained liberal-
40
ism'. Andrzej Rychard refers to what he calls
`blockages' in legitimation `which stem from the
elite's usage of outdated symbols and discourses in
274
 LIBERALISM AND POST-COMMUNISM
the construction of liberal democracy', failing above
all to negotiate with, and often even to explain to,
the populace over the strategies for reform. It instead
struck hard during the breathing space offered by the
period of `extraordinary politics' to impose ready-
made solutions, often imported from abroad.
notes that `the communist order existed but was
[indiscriminately] rejected while the post-communist
order has not been shaped but is [naively] ac-
42
cepted'. The Mazowiecki government's liberal-
minded Industry Minister's statement during his
confirmation hearings before the Sejm in September
1989 is instructive: `I represent subjects that do not
43
yet exist.' The absence of any organised social force
to carry out the liberal reforms translated into what
Jadwiga Staniszkis called a `revolution from above' in
which the state promoted the (temporary) demobi-
lisation of society in order to achieve the economic
conditions that would allow greater public inclu-
44
sion.
Constrained liberalism has yet another and per-
haps more profound meaning. Janine Holc has ar-
gued that liberal discourse is itself part
construction of a very specific type of democratic
45
subject in post-communism. As she notes, `all
states ± indeed, all communities ± are based on
assumptions about who properly belongs in them',
and in Poland between 1990 and 1993 a very specific
`liberal' narrative became dominant that explicitly
linked `private property, individual rights, and the
``shock therapy'' of the free market, and it implicitly
constructed an ideal and quite fixed ``citizen'' of the
46
new Polish democracy.' Bielecki himself, as Prime
Minister in 1991, was the most radical of these liberal
`identity-builders'. Not only was the state being
rebuilt, but the notion of what made up the citizen
was being reconstructed. Holc seeks to break away
from traditional institution-centred discussions of
transition to examine the change in subjectivities
involved. She sums up her argument as follows:
The rise and fall of the Liberal movement in
Poland illustrates the manner in which a vision
or philosophy of government that assigns indivi-
dual civil and economic rights to its citizens is not
inherently liberating or democratic for those sub-
jects. Rather, any discourse of authority that seeks
to `fasten' identities to fixed, naturalised subject
positions inevitably disciplines those subjects and
limits possibilities for political action. Further-
more, discourses that seek to attain totality, even
a democratic totality, do so by covering over
41
He discursive instabilities, by mistaking a unified
47
subject for a fluid ensemble of subject positions.
Thus the liberal citizen was to be grounded in
economic liberty, specifically private property as
the basis for civil society and the emancipatory
potential of economic empowerment. Post-commu-
nist liberals thus appeared to be more Lockean than
Locke himself, `in implying not only that liberty will
be achieved by acting on the right to property, but
48
through property itself'. A radical disjuncture was
achieved in what had previously united Solidarity,
the combination of social power and individual
rights. In Poland and elsewhere social power was
consigned to some non-liberal limbo, together with
state activism in its entirety, while the liberal subject
was constrained by an eviscerated model of what the
of the citizen of this liberal utopia was to look like.
Liberalism and Post-communist Democracy
Philippe Schmitter has noted that:
Liberalism, either as a conception of political
liberty, or as a doctrine about economic policy,
may have coincided with the rise of democracy.
But it has never been immutably or unambigu-
49
ously linked to its practice.
This question is particularly acute in post-communist
countries, where the very `givenness' of a structured
society is in question. Implicit in many discussions is
the `Kirkpatrick principle', the view that in the
transition from authoritarianism to democracy, so-
ciety can spring back into shape, whereas in the
transition from totalitarianism, society is destroyed to
such a degree that it has to become an object of the
transition process itself.
The fate of liberalism in the post-communist
world is dependent on the ability of democracy itself
to survive. This is nowhere more evident than in the
sphere of party formation. Throughout the post-
275
 RICHARD SAKWA
communist world the emergence of this quintessen-
tial aspect of a modern liberal democracy was a more
protracted, fragmented and tenuous process than in
the other `third wave' democratising states in Latin
America and Southern Europe. It might be noted
that in America and elsewhere parties by no means
developed at the same time as parliamentarianism
and democracy. According to Kagarlitskii, the `plur-
alism of elites' allowed capitalism in its `primary
accumulation of capital' phase to preserve liberal
political institutions while keeping the people away
from decision-making:
The interaction of a harsh and authoritarian
structure of executive power with a parliamentary
elite was an ideal recipe for oligarchical rule,
characteristic for the liberal-market capitalism
50
of the nineteenth century.
It was this sort of system, marked by a large number of
`independent' politicians elected largely due to their
financial resources, that was created by what Kagar-
litskii calls the `liberal-authoritarian' regime in Rus-
sia, a system that was ideal for the needs of the
corrupted Russian bureaucracy, striving to gain for-
mer state property, and for western elites who sought
a power strong enough to suppress any popular
resistance to the `reforms' while avoiding the open
authoritarianism typical of Latin America. The sys-
tem also corresponded to national traditions, where
nascent liberal institutions were always grafted on to
the authoritarian Russian state, whether this was in
the late Tsarist period or in the late Soviet epoch,
and acted as no more than a supplement to funda-
51
mentally non-democratic systems.
In Russia this tension has taken extreme forms.
`Perestroika' was a battleground between the liberal
part of the nomenklatura and the nascent civil so-
ciety. If `perestroika' was accompanied by the emer-
gence of `civil society' movements dealing with local,
environmental and civil rights issues, the triumph of
the liberal nomenklatura and intelligentsia in August
1991 was followed very quickly by the stifling of this
social insurgency. In addition, throughout the 1990s
the status and material position of the liberal profes-
sions and intellectual life in general declined sharply,
damaging the core supporters of liberalism, while the
276
financial crash of August 1998 effectively wiped out
the nascent middle class. As Kagarlitskii puts it:
the retreat from communist ideology was accom-
panied by the centralisation of power in the hands
of the local bosses. The organs of local self-
government, already very weak, were quite simply
abolished until revived as part of the centre's
attempts to outflank the powerful regional bosses.
In conditions of economic crisis enterprise admin-
istrations strengthened their control over work-
52
ers.
How can we explain this and in what terms can we
describe it? There are three substantive approaches
to the issue.
1. Illiberal democracy
In a recent article Fareed Zakaria has implicitly taken
up Raymond Aron's view that the concept of `con-
stitutional pluralism' is in many ways more accurate
than the formula of `liberal democracy'. Zakaria
distinguishes between liberal democracy, defined as
`a political system marked not only by free and fair
elections, but also by the rule of law, a separation of
powers and the protection of basic liberties of speech,
assembly, religion and property', which he calls
constitutional liberalism, and illiberal democracy.
In the latter `democratically elected regimes, often
ones that have been re-elected or reaffirmed through
referenda, are routinely ignoring constitutional limits
on their power and depriving their citizens of basic
53
rights and freedoms'. For Zakaria, the regular sta-
ging of relatively fair, competitive, multi-party elec-
tions might make a country democratic, but it does
not ensure good governance. For him, constitutional
liberalism
is not about the procedures for selecting govern-
ment, but rather government's goals. It refers to
the tradition, deep in Western history, that seeks
to protect an individual's autonomy and dignity
against coercion, whatever ± state, church, or
society . . . It is liberal because it draws on the
philosophical strain, beginning with the Greeks,
that emphasises individual liberty. It is constitu-
 LIBERALISM AND POST-COMMUNISM
tional because it rests on the tradition, beginning
54
with the Romans, of the rule of law.
In practice, even relatively free elections in coun-
tries like Kazakhstan `have resulted in strong execu-
tives, weak legislatures and judiciaries, and few civil
55
and economic liberties.' Zakaria argues that while
constitutional liberalism can lead to democracy,
democracy does not necessarily lead to constitutional
liberalism. Thus the government of Hong Kong
under British rule was epitomised by constitutional
liberalism, protecting the rights of citizens and the
rule of law, but it was far from democratic. The
Central European post-communist states are nego-
tiating the passage to democracy more successfully
than the former Soviet states, it is argued, because
they went through a long phase of liberalisation
without democracy in the nineteenth century that
grounded the rule of law and property rights into
social practices. The problem with this approach is
that it assumes that history is destiny and provides no
mechanism for a society to break out of `path de-
pendency'. It appears that if a country misses the bus
of constitutional liberalism it has little chance of
getting on the bus called democracy.
2. Delegative democracy
In a similar vein, Giullermo O'Donnell has argued
that in weakly established democracies a leader can
become so strong that he or she can ignore those who
they are meant to represent. O'Donnell characterises
these countries as having `delegative' rather than
representative democracy, with the electorate alleg-
edly having delegated to the executive the right to
do what it sees fit, `constrained only by the hard facts
of existing power relations and by a constitutionally
56
limited term of office'. Thus a government emerges
that is `inherently hostile to the patterns of repre-
sentation normal in established democracies' by
`depoliticising the population except for brief mo-
ments in which it demands its plebiscitary support.'
This sort of democracy is, according to O'Donnell,
underinstitutionalised, `characterised by the
stricted scope, the weakness, and the low density
of whatever political institutions exist. The place of
well-functioning institutions is taken by other non-
formalised but strongly operative practices ± clien-
58
telism, patrimonialism, and corruption.' The
notion of delegative democracy has clear applica-
tions to Russia, and has been used fruitfully in
59
analysing regional politics. However, the concept
has clear limitations when applied to Russia, ignor-
ing precisely the political consequences of imposing
a capitalist economic system. It is the combination of
political reform and economic transformation that
allowed the whole process to be hijacked by sectoral
and other special interests, not the flaws in the
democratic method as such.
3. Regime democracy
As Merquior noted of Brazil, the concentration of
60
economic power `is a freedom-killing exercise', an
observation that is equally true of Russia and other
post-communist countries. These countries do not
lack freedom; they lack liberalism. The new freedom
has allowed an oligarchical, corporate and corrupt
capitalism to emerge.
Regime-type politics are characteristic of all the
post-communist states of the former USSR (but not
Central Europe). This covers not only features like
enhanced powers for the executive, the prevalence
of personalised leadership where the personality
defines the contours of the political system, the
manipulation of parliaments, weak judicial over-
sight, seamless links between the political and new-
ly-enriched socio-economic elites, the lack of
distinction between public and private posts, and
above all the lack of accountability of elected lea-
ders. The very nature of post-communism in this
region has taken on the colouring of what postmo-
dernists call a `decentred' politics. Although the
powers of the executive everywhere have been en-
hanced, these are not classical presidentialist regimes
of the Latin American type. Politics is too unstruc-
tured, institutions too fluid and the personages too
constrained by the emerging class system and ethnic
57
contradictions to allow the emergence of a defined
balance of powers. Instead regime politics is defined
re- by flux within the parameters of the exigencies of the
transitional process itself: they are not arbitrary but
neither are they ordered as part of a structured
political order (Ordnungspolitik). Its codes are not
277
 RICHARD SAKWA
formalised but they are recognised. The legitimacy
inherent in the electoral process comes into contra-
diction with the functionality of the transformative
impulse. In short, the political process in Russia and
elsewhere lacks a subject, focused neither on person-
alities nor institutions but on fluid but ritualised
relations between universally objectified processes
of transition.
Universalism and Post-communism
Although the domestic sources of liberalism may be
relatively weak in most post-communist countries,
the international context is increasingly dominated
by the political counterpart of globalisation, what we
might call `universalisation'. It is under this nominal
rubric that the most important questions of our day
are discussed. This is the question of the extent to
which the universal norms, as formulated above all
in the 1948 United Nations Universal Declaration of
Human Rights, are applicable to all countries, irre-
spective of their specific traditions. From this per-
spective globalisation can be challenged not so much
from a `left' perspective as from a universal one, in
which a rich array of international and national
organisations established in the postwar years can
be mobilised to establish a set of international norms.
In the European context the main instruments of this
universalism are the Council of Europe (CE) and the
Organisation for Security and Cooperation in Europe
(OSCE). Russia joined the CE in February 1996, and
on 20 February 1998 the Duma voted to ratify the
European Convention on Human Rights, allowing
Russian citizens to file appeals to the European
Court. By that time there were already 800 com-
plaints from Russian citizens pending with the Eur-
opean Commission on Human Rights, the body that
61
refers cases to the European Court.
Universalism has reforged a link between moder-
nisation and development. While some Asian coun-
tries, as Huntington has noted, have been able to
separate modernisation from westernisation, by
adopting western techniques but not capitulating
62
to western culture in its entirety, for post-commu-
nist countries this question is particularly acute.
Although for most Eastern European countries the
`return to Europe' is not perceived as particularly
278
problematic, for Russia (and some other post-Soviet
states) post-communism has once again raised the
question that has racked the country for at least the
last three hundred years: is Russia part of Europe or a
separate and distinct civilisation sufficient unto itself?
All of these countries, moreover, have been faced by
the `Nuremberg question', the degree to which the
perpetrators of the crimes of the old regime can be
63
brought to book. Whether there is a distinctively
`liberal' approach to this aspect of universalism is
unclear, other than a commitment to the rule of law.
Too often decommunisation became an aspect of the
partisan persecution of political opponents.
More broadly, the question is often posed whether
any universal agenda of human and liberal rights is
no more than a covert form of imposing Western
values on other societies. Some post-communist
leaders have explicitly argued that liberalism is not
suitable for their societies at this stage of develop-
ment. Familiar arguments about distinctive paths
and `third ways' are being mobilised by post-com-
munist leaders (for example, President Islam Kari-
mov in Uzbekistan and the Mayor of Moscow, Yurii
Luzhkov) and leaders of the developing countries
(for example, Prime Minister Mahathir Mohammed
of Malaysia). Both the `East' and the `South' are now
joined by a series of common dilemmas that trans-
cend issues of globalisation and focus on the cultural
and political identity of the nations themselves. The
former Second and Third Worlds are still far from
one, but when it comes to civilisational issues and
developmental trajectories they do face some com-
mon problems.
Post-communist Liberalism
Liberalism is an extended meditation on the legit-
imate sphere of state activism and the proper scope
for public authority. As with any living idea, answers
are always in the process of being rethought. Liberal-
ism under post-communism becomes post-commu-
nist liberalism.
Liberalism and the Post-communist State
The opening of John Stuart Mill's On Liberty has in
effect become the slogan of the post-communist
 LIBERALISM AND POST-COMMUNISM
revolutions: as countries become democratic, he
noted, people believed that `too much importance
has been attached to the limitation of power itself.
That (it might seem) was a resource against rulers
whose interests were habitually opposed to those of
the people.' With the people in charge, there was
allegedly no need for such caution: `The nation did
not need to be protected against its own will. There
64
was no fear of tyrannising over itself.' Mill of course
goes on to argue that
the limitation . . . of the power of government
over individuals loses none of its importance
when the holders of power are regularly accoun-
table to the community, that is, to the strongest
65
party therein.
Zakaria quotes this argument approvingly.
peoples emerging from communism had learnt the
bitter lesson that it is precisely the attempt to fulfil
the entire potential of democracy that leads to
absolute servitude. Since the French revolution de-
mocracy has been associated with undivided popular
sovereignty, forgetting Tocqueville's warning of the
`tyranny of the majority'. These lessons are now
taken to heart, but there are other processes that
undermine the emergence of clear conclusions.
As we know, the advance of democracy led to the
emergence of a state that is far from minimal and a
liberalism that embraced interventionism. While
Hayek's The Road to Serfdom, with its anti-construc-
tivist thesis that any state activism in the social
sphere provides scope for the development of tota-
litarianism, became popular reading in post-commu-
nist countries, the experience of postwar democracy
in Japan and the West demonstrated that `planned
capitalism and the welfare state helped to avoid
totalitarianism because they contributed decisively
to the neutralisation of political movements that
carry a model of totalitarian socialism.'
post-communism `economic liberalism
on an extreme, virulent form at times, in which
antistatism ± one of the most lucid positions ± turns
into a generalised statephobia, often accompanied by
68
antidemocratic sentiments.' The post-communist
transitions coincided with the predominance of the
globalisation paradigm, itself a recipe
minimalism, reducing the legitimate scope of state
activity. In the post-communist world the anti-statist
strain in liberalism became an over-determined as-
sault on the very existence of purposive state action,
something that China managed to avoid because of
its distinctive traditions and refusal to accept the
agenda of `transition' and liberal democracy.
Gottfried Dietze has distinguished between `liber-
alism proper', defined as `liberty without restriction',
and `proper liberalism', a balance between personal
liberties and social restrictions that allows the en-
joyment of the one with the public order of the
69
other. Clearly, it is a vulgar `liberalism proper' that
has exerted a profound influence in post-communist
Russia accompanied by the weakening of law en-
forcement, the privatisation of security (for those
who can afford it) and lawlessness. There is a pre-
66
The valent sense of lack of order and bespredel (absence of
limits) that gives rise, as Fish notes, to the absence of
rights (bespravie) ± `albeit a very different form of
70
bespravie than existed under the Soviet regime,' but
an absence that in many ways is more substantive
than earlier since it is experienced at the level of
daily life in a personal way. The denigration of the
state that had accompanied the assault against the
communist order was perpetuated in a simplified
`liberalism proper' that disregarded problems of pub-
lic order and personal security. Post-communist lib-
erals discovered that the threat to individual liberty
could come as much from an anarchic society as from
an overweaning state or a sovereign people. The
emergence of national liberalism was a response to
this, but more common was support for neo-com-
munist, nationalist and other quasi-authoritaran so-
lutions.
Liberalism and Post-communist Social Democracy
Under post-communism the focus has been on re-
67
Under storing the elements of classical constitutional lib-
still takes eralism, the critique of political power and the
attempt to impose restraints on despotism. With
equal force post-communist liberalism has had to
face the question of the social sources of despotism,
the `tyranny of the majority' and of the masses on the
individual. In this context the social liberalism
for state espoused by T. H. Green and his successors ± an
279
 RICHARD SAKWA
evolution that in the United States has redefined the
very meaning of liberalism to denote interventionist
welfare state policies ± has been largely absent.
Sogrin notes that at the basis of the reformist policies
of Russian radicals `lay a simplified Western liberal-
ism, stripped off practically all the humanistic for-
71
mulae of the twentieth century.' The insouciance
of the majority of post-communist liberals in the face
of the social traumas of an almost unprecedented
scale in peacetime did much to discredit the liberal
idea in its entirety.
Irving Kristol may well be right to argue that `a
72
conservative is but a liberal attacked by reality'
we can with equal force argue that a liberal is but a
conservative attacked by conscience. The problem
under post-communism, however, is that this con-
science is not well-developed. What liberalism there
is tends to emerge out of an abstract intellectualism
rather than the commitment to the broad range of
liberal values. The reconciliation between the so-
cialist striving for social justice and the liberal con-
cern with individual liberty has not yet been
achieved, weakening both in the process. The very
notion of equality, which in a sense could be re-
garded as the ground for both socialism and sub-
stantive (social) liberalism, has been delegitimated
by the experience of Soviet socialism. A `radical
liberal' approach predominated that considered
any attempt to achieve substantive equality of out-
comes, rather than simply `equal starting possibili-
ties', as an outbreak of neo-Soviet `equality-
73
mongering' (uravnilovka).
As far as Dzarasov was concerned, a group of
`anarcho-liberals' had come to power in Russia.
While extreme liberals in economic affairs, they were
at the same time destructive of the integrity of the
state. He argued that capitalism was artificially
foisted on Russia, and that the population does
not recognise the legitimacy of property acquired
in the course of `mafia-nomenklatura privatisation'.
He insisted that Russia lacked the preconditions for a
purely capitalist type of development. The Russian
bourgeoisie in his view was never concerned with the
good of the country, and that the backbone of the
country was represented by hired labour (workers
and employees in the trade unions), `honest' entre-
preneurs and scientific-technical intelligentsia. It is
for this reason that he insisted that the ideology of
Russia's modernisation could not be liberalism but
76
social democracy. Dzarasov emerged as the most
cogent exponent of the view that social democracy
was the only way forward for Russia, remaining loyal
to a modified version of convergence theory, a
compromise (synthesis) of capitalism and socialism,
Marxism and liberalism. Russian liberalism, in his
view, was sustained by its ability to convince people
that all evil had arisen from communism, and that all
things good came from the example and the recom-
77
mendation of the West. A turn away from the
but West, however, thus ran the risk of a rejection of
liberalism in its entirety.
It was only after successive electoral defeats and
political marginalisation, according to Sogrin, that
78
social liberalism from 1994 became prominent.
The social liberals, in sharp contrast to the radical
liberals, began to stress issues of social order, became
supporters of `civilised nationalism' and sought to
bridge the gulf that had opened between liberalism
79
and democracy. The problem, however, is deeper
than this. While classical liberalism may well have
advanced an individualistic view of society, today
the main actors in advanced democracies are col-
lective: parties, trade unions, churches, lobby groups
and the government itself. In Russia these collectives
have not yet become `interest groups' on the western
model; their identity is still Soviet, focused on the
internal management of their affairs rather than on
the external advancement of their interests. The
constituency for social liberalism remains passive
and ideologised, rather than advocatory and struc-
74
tured.
Benedetto Croce is the arch-exponent of a liberal
historicism, reabsorbing the thinking of Giambattista
Vico into mainstream liberalism, but he is also
notable for his attempt to `re-moralise' liberalism,
endowing it with an ethical charge that was lacking
75
in economic utilitarian approaches. Croce is `Fu-
kuyama with a conscience'. This trend is important
in Russia, although lacking sustained formal expres-
sion. The conscience of liberalism remains abstract,
while the transformative challenge is immediate and
urgent; the liberal historicist tendency has triumphed
at the expense of its ethical content. Neo-liberalism
has predominated over ethical liberalism.
280
 LIBERALISM AND POST-COMMUNISM
Liberalism, the Nation-State and Nationalism
For Roland Axtmann, the only serious basis for
liberal democracy is the nation-state, but at the same
time, it was enlightenment and liberal values that
80
sustained the nation-state. Liberalism is the most
universal of ideologies yet, paradoxically, it is the
most national in form ± a feature recognised by John
Stuart Mill, Jules Michelet and Giuseppe Mazzini as
they became the intellectual proponents of liberal
81
nationalism. An appropriate combination of lib-
eralism and nationalism, however, is far from easy, as
the Frankfurt Assembly in March 1849 discovered to
its cost. After the collapse of the Assembly the link
between liberalism and nationalism in Germany was
severed, opening the way to the horrors that fol-
82
lowed. Liberalism's relationship with authority in
pre-revolutionary Russia was equally ambivalent,
espousing a progressivistic view of history while
suspicious of the radical populism and socialism that
sought to make historical progress their own. No-
where were these tensions more apparent than in the
person of Paul Milyukov, the effective leader of
organised Russian liberalism in the form of the
Constitutional Democratic Party (Kadets). Liberal-
ism in Russia, it appeared, was subverted even before
it had begun, unable to sustain itself as an indepen-
dent political force or to reconcile the democratic
and liberal impulses. Ironically, it was Milyukov's
pragmatic application of liberalism to Russian cir-
cumstances that brought him down when in 1917 he
embraced `state consciousness (gosudarstvennost) and
sent off his ill-fated Note reasserting imperial war
83
aims. More broadly, the democratic revolution in
Russia was shorn entirely of liberal coloration and
took an extreme Jacobin form. The tension between
liberalism and democracy was to persist into the post-
communist era.
Liberalism requires home grown liberals to pro-
mote it, specific socio-economic conditions to
thrive, national cultures to be sustained, and char-
acteristic institutions and individuals to practise it.
There is clearly a need for a new synthesis of liberal-
ism and nationalism under post-communism; if lib-
eralism remains too internationalist, it will fail to
find a domestic constituency whose bonds of alle-
giance would forge a new synthesis of a free society
281
with a viable state. This is a harder task than might
appear at first sight; how is it possible to reconcile
liberal individualism with the collectivism that is an
intrinsic part of nationalist discourse? In Russia the
issue takes on particular force since the collectivist
ethos is the core of tradition, of history and of
politics.
For Will Kymlicka there is not such a stark con-
trast between the two principles as some of liberal-
84
ism's critics have suggested, but finding a balance
between collective traditions and individual rights
remains a problem in many communities. In Dage-
stan, for example, the enormous burden (both fi-
nancial and emotional) of traditional rituals marking
the passage of each stage of life has led many to
escape into the relative simplicity of the Wahhabi
form of Islam. Throughout Russia the struggle be-
tween the abstract liberal principles embodied in the
constitution of December 1993 and myriad forms of
inherited collectivist communities (both Soviet and
pre-Soviet) represents a profound reorientation of
the culture. The tension between the two principles
cannot be wished away, but at the same time it
should not be exaggerated. As the burgeoning lit-
erature on identity formation suggests, cultures (like
identities) are not stable unchanging entities but are
in a constant process of adaptation and renegotiation
85
with the challenges of modernity. Stephen Welch
has made the same point with regard to the mut-
86
ability of political culture. The very resurgence of
communitarian thinking suggests that liberalism
cannot be reduced to the sovereign individual,
and that communities can make enhanced claims
on an individual without infringing the principles of
liberal individualism. The problem, of course, is: at
what point does communitarianism become collec-
tivism (in the negative sense of the term, that is
when the community makes demands that under-
mine individual autonomy in threatening ways)? At
what point do the rights of the collective and the
principle of collective rights lead to an anti-liberal
collectivism?
The rich vein of post-communist paradoxes does
not end here. While few would now defend the
concept of empire (and for early post-communist
liberals it was clear that democracy and empire in the
old USSR were incompatible), the disintegration of
 RICHARD SAKWA
the empire was not a sufficient, through probably a
necessary, condition for the emergence of democ-
racy. Only the republics appeared to be an appro-
priate vessel in which democratic institutions could
be constructed. As Dankwart Rustow argued, stable
87
borders are a prerequisite for democracy. Thus out
of the detritus of empire nation-states were to be
born ± this was one side of the liberal project ± but no
concessions were to be made to nation-state nation-
alism ± the other side of liberalism. Even though
Carlton Hayes had long ago argued that liberalism
88
and nationalism were far from antithetical, man-
oeuvring between these two positions provoked
many painful contortions. The commitment to de-
mocracy was an intrinsic part of the formal move-
ment for national rebirth in Russia, and this perhaps
made the democratic project rather more resistant to
both nation-state nationalism and ethnic national-
ism than in some other post-communist countries.
Yet the susceptibility was there, as evidenced for
example by the nationalist turn taken by the Russian
Christian Democratic Movement despite its commit-
ment to `enlightened patriotism'. Another example
is that of the Jews in Russia, traditionally associated
with the country's liberal and (especially in the pre-
revolutionary era) radical movements, who gradually
associated more with Russian patriotic ideas. Ac-
cording to Dmitry Shlapentokh, `the emergence of
this strange and unknown political species, red-to-
brown Jews, was indicative not so much of a trend
among Jews as of a deep crisis in Russian pro-Wes-
89
tern liberalism.'
Some key questions remain unanswered. Above
all, can an effective Russian state of whatever type be
constructed in the absence of agreement on Russia's
historical border, and can a liberal Russia be estab-
lished if the country retains imperial pretensions
outside Russia itself? These are the questions faced
by the national liberals in the most acute form
possible, perhaps trying to reconcile the irreconcil-
able. The powers given up by the communist state, to
paraphrase Hoffmann's argument about the Eur-
opean Union, `have gone not to the new central
90
institutions, but to the market.' The liberalism that
was successful in reducing the state's power, `has
91
created a formidable anonymous new power', char-
acterised by the eruption of crime and corruption.
282
The market, moreover, even in the best of times is
not a source of loyalty and allegiance (except perhaps
for the most rabid neo-liberals) but the source of
enrichment (for some); people do not die for the
market but they do for the state.
The attempt to combine patriotism and liberal
values gave rise to the national liberal tendency, yet
no adequate synthesis occurred, leaving the field
open to radical rejectionists of the whole new order.
Alexander Prokhanov is one of the most eloquent of
those who condemn Russia's renewed westernisa-
tion. He insists that `today's Westernisers in Russia
are the liberals and radicals. They are criminals; they
92
are destroying Russia.' A similar viewpoint is ad-
vanced by Alexander Tarasov when he details the
deaths resulting from the events of 3±4 October
1993. `This', he argues, `is what liberalism looks like
93
in action.' The sentiment of pre-fascist Italian
nationalist intellectuals who `viewed liberalism as
too individualistic and horribly bourgeois, while
94
democracy encouraged mediocrity and corruption'
is shared by many citizens of the post-communist
world. Many, while not rejecting liberalism as such,
have adopted anti-western, and in particular anti-
American, views; modernisation based on Western
95
models has lost its attractiveness. In the early post-
communist years the neo-liberal economic reformers
systematically dismantled Russia's ties with its former
partner republics, and the language of national in-
terests was alien to them as they pursued a globalising
agenda. Nevertheless, liberals gradually came to
appreciate the importance of national identity in
the state-building endeavour. A de-ethnicised de-
mocracy did not necessarily mean a denationalised
one. It remains to be seen what liberalism with a
Russian face will really look like.
Individualism and Civil Society
The shift in post-communism is from the aspiration
to achieve an organic collectivist society to an
individualist `civil society'. Civil society was the
slogan of the anti-communist revolution, of course,
and its central features at that time were the ad-
vancement of negative liberty for the individual and
a public sphere where these individuals organised in
civil associations could gather to influence policy.
 LIBERALISM AND POST-COMMUNISM
Neither has yet been entirely achieved, and indeed
the terrain of post-communist civil society
tended to take on some of the features of Hobbes's
`war of all against all', a state of nature of pre-liberal
and pre-contract societies. There is no guarantee
that the free market on its own leads to a free politics.
From the liberal perspective more broadly the
relationship between the individual and the collec-
tive, the personality (lichnost) and the community,
is not resolved in general, let alone in Russia. As
David Ost puts it in the Polish context, `liberalism is
no friend of community', and he goes on to argue
that liberals `are in crisis everywhere in Eastern
Europe. It is not so much economic as political
96
liberalism that is in danger.' `Negative liberty
demands the creation of a social space in which
the choices, preferences and will of each individual
97
are sovereign.' The `political right to do wrong' is
not yet entirely conceded, either from the national
or the ideological perspective. The agreement to
disagree was the basis of the modern liberalism
emerging out of the Reformation and Counter-
Reformation, and the necessity of the principle
has been amply confirmed by the ideological wars
of the twentieth century. The balance between
individual civil and human rights, state power
and social responsibility is far from having been
achieved in post-communist societies.
There is another aspect to post-communist indi-
viduation, discussed above in the context of con-
strained liberalism. The process has sometimes been
described as the transformation of the individual
from `comrade' to `citizen'. This is indeed part of
the truth, but an equally profound parallel transfor-
mation is taking place: from `comrade' to `customer',
in certain cases missing out the citizenship stage
altogether, and certainly everywhere undermining
the full political entitlements denoted by the con-
cept of citizenship. The will of the customer takes
priority over the will of the planner, production takes
the place of distribution, and interdependence is
placed above independence ± but where is the
emancipation of which Merquior spoke? While re-
ligions hold that there is a natural subject situated
within the religious community itself and indepen-
dent of the particularities of the state-society rela-
tionship (i.e. that this subject is pre-political), the
consumerist version of citizenship has analogously
has withdrawn the subject from the full-blooded political
community, but this time it is post-political. The
social sphere vacated by the communist state has
been filled not by political emancipation but by the
tyranny of a corrupted neoliberal market.
Conclusion:
Systemic or Asystemic Liberalism
Does liberalism lack profound roots in the societies
themselves? Is liberalism a systemic prerequisite of
post-communist societies, or was it imposed simply as
a byproduct of the anti-communist struggle and as a
function of aspirations to join the international
system? If the former, then despite many tribulations
it is here to stay, but if the latter, it is fragile and
liable to be swept away by more militant rejectionist
nationalist agendas. Was Szacki indeed right to argue
that liberalism was a specific western phenomenon
sustained only by the role it played in the anti-
communist struggle? He probably exaggerates the
argument, especially since he fails to define what
he means by liberalism, preferring instead to stress its
situational character. Even in Russia liberalism has a
long history, going back at least to Herzen and
sustaining a vigorous literature over legal and local
government reform, and under communism it was far
more than an instrumental ideology to be discarded
at the first opportunity. He is right to argue that
political liberalism can accommodate various types
of property rights, but underestimated the problems
facing economic reformers in marketisation ±
although this is not to say that neoliberal economic
reforms were the only way forwards. Attempts to
devise `third path' strategies, however, like those of
98
Gay and Alekseeva have not been convincing. It is
not at all clear whether more gradualist and organic ±
indeed, more liberal ± methods would in the long run
have secured the future for liberalism as well as the
`illiberal' methods actually implemented.
Post-communist liberalism has, paradoxically,
used the power of the state to implement its eco-
nomic programme ± indeed, as Szacki would argue,
hardly a demonstration of the triumph of liberalism.
Political liberalism, however, has not been ne-
glected: the `universalist' principles of the post-com-
283
 RICHARD SAKWA
munist international community have been rein-
forced by the attempts by all Eastern European states
(including Russia) to return to Europe and the world.
However, their implementation has often fallen far
short of declarations, giving rise to an illiberal and
constrained democracy of the regime system. The
democratisation of liberalism has yet to be secured.
At the same time, the basic principles of liberalism ±
individualism, tolerance, private property, pluralism,
restraints on state power ± have gained broader
support than people like Szacki and others who insist
on a primordial collectivism and authoritarianism in
Poland or Russia would concede. The communist
regimes did not simply perpetuate traditional pat-
terns of authoritarianism; communist authoritative-
ness was based on thoroughly modern principles.
Indeed, in yet another of those paradoxes that mark
the post-communist world, in the social sphere the
communist period helped clear the way for liberalism
by destroying the old aristocratic ethos of the old
Russian and Polish ruling classes.
However weak post-communist liberalism might
be, in Poland it was able to force the adaptation of
the post-communist communists to its programme,
while in Russia it has defeated (so far) the great
Notes
1. Gellner, `Introduction', in Gellner and Cansino (eds)
(1996), pp. 1±4, at p. 3.
2. Fukuyama (1989), (1992).
3. Helio Jaguaribe, `Merquior and Liberalism', in Gellner
and Cansino (eds) (1996), pp. 21±36, at p. 24; in the
same volume see also Roberto Campos, `Merquior the
Liberalist', pp. 65±76, at p. 72.
4. Ibid., pp. 21±36.
5. Ibid., p. 32.
6. V. V. Sogrin, `Liberalizm v Rossii: peripetii i perspek-
tivy', Obshchestvennyie nauki i sovremmenosti, 1, 1997,
p. 13.
7. Nodari A. Simonia, `Priorities of Russian Foreign
Policy and the Way It Works', in Dawisha and
Dawisha (eds) (1995), p. 17.
8. Ibid., p. 18.
9. Hough (1997), pp. 1±2 and passim.
10. Lev Timofeev, `Bol'shaya rastrata, in Izvestiya, 26
February 1993.
11. Lev Timofeev, `Vse russko pechal'no', in Izvestiya, 31
October 1995.
12. Sogrin, `Liberalizm v Rossii', p. 14 (see note 6 above).
284
power communism of Gennadii Zyuganov and the
Communist Party of the Russian Federation (CPRF).
As Sogrin puts it:
Radical liberalism was able to destroy the supports
of communism and helped Russia achieve a
breakthrough into market relations. But it became
increasingly clear that the radical liberal approach
to reconstructing Russia on the classical Western
model is nothing less than a utopia, and that in
Russian conditions radical liberalism leads to the
socio-economic polarisation of society and the
poverty of the majority and enshrines the power
99
of bureaucratic nomenklatura capitalism.
In its place a social liberalism emerged, but what was
really required according to him was a national
liberalism. The challenge perhaps is even greater
than this: to combine the political constitution of
liberty, the social basis for equal citizenship, the
patriotic consensus for national integration, and
the developmental basis to allow the conversion
of constrained transformative liberalism into a gen-
uine emancipatory liberalism.
13. Ibid., p. 15.
14. This is examined by Pantin (1997).
15. Szacki (1995), chapter 5.
16. Ost (1991), p. 91.
17. See ibid.
18. Solzhenitsyn (1974).
19. Radio Free Europe/Radio Liberty, `Newsline', 10 Au-
gust 1998.
20. Schieder (1962), p. 39.
21. See Russia Briefing, 12 February 1997, pp. 6±7.
22. Fish (1997).
23. Aleksandr Buzgalin, `Right Cause: You Are Not Right,
Gentlemen', Jamestown Foundation Prism, vol. V
(14), part 6, 16 July 1999.
24. Bellamy (1992).
25. Balcerowicz (1995).
26. Klaus (1997).
27. Sogrin, `Liberalizm v Rossii' p. 17 (see note 6 above).
28. Rustam Narzikulov, Nezavisimaya gazeta, 11 March
1997.
29. Sogrin, `Liberalizm v Rossii' p. 21 (see note 6 above).
30. Lebed's ideas are outlined in Izvestiya, 24, 31 May
 LIBERALISM AND POST-COMMUNISM
1996 and Za derzhavu obidnu (Moscow: Moskovskaya
pravda, 1985). An analysis of his views is Vladimir
Polushin, General Lebed ± zagadka Rossii (Moscow,
1996).
31. Cited in Institute for East±West Studies, Russian
Regional Report, vol. 2 (11), 20 March 1997.
32. Ibid.
33. Szacki (1995), p. 150.
34. Radzislawa Gortat, `Liberalizm jako Konstruktywizm',
in Przegland Spoleczny 6, 1992, cited in Szacki (1995),
p. 154.
35. This is above all the argument of David Ost, `Shap-
ing a New Politics in Poland: Interests and Politics
in Post-Communist Europe', Programme on Central
and East Europe Working Paper 8 (Centre
European Studies, Harvard University, 1991), pp.
4±5.
36. Bellamy (1994), p. 420. For John Gray's view, see
Gray (1986), chapter 1. And for Hayek's own sum-
mary of his position, see Hayek (1993), volume 1, p. 5.
37. Walicki (1991), p. 106.
38. Jerzy Hausner and Jacek Kuron, Strategia Negocjacy-
jina w Procesie Transformacji Gospodarki (Warsaw:
Stefan Batory Foundation, 1994), p. 26. I am grate-
ful to Nick Spiro for drawing my attention to this
debate.
39. Andrei Kolesnikov, `Liberalizm umer, da zdravstvuet
liberaliz!', Novoe vremya, 12 (1997), p. 15, cites the
case of St Petersburg when Vladimir Yakovlev re-
placed the `liberal' Anatolii Sobchak as mayor but
was soon forced to pursue similar policies.
40. Jan Krzysztof Bielecki interviewed by Nick Spiro,
London, 2 August 1996.
41. This theme is explored in the doctoral dissertation by
Nick Spiro, The Politics of Economic Transition: Shock
Therapy in Poland, 1990±1991, University of Kent at
Canterbury, June 1998.
42. Rychard (1992), p. 144.
43. Cited by Jadwiga Staniszkis in Staniskis (1991), p.
184.
44. Ibid.
45. Holc (1997).
46. Ibid., p. 402.
47. Ibid., p. 427.
48. Ibid., p. 416.
49. Zakaria (1997), p. 23.
50. Boris Kagarlitskii, `Levye v Rossii: nadezhdy, neuda-
chi, bor'ba', in Svobodnaya mys', 11, 1994, p. 33.
51. Ibid.
52. Boris Kagarlitskii, `Vremya neopravdavshikhsy na-
dezhd', Svobodnaya mysl', 3, 1997, p. 10.
53. Zakaria (1997), p. 22.
54. Ibid., pp. 25±26.
55. Ibid., p. 28.
56. O'Donnell (1994), p. 59.
57. O'Donnell (1993), p. 1367.
58. O'Donnell (1994), p. 59.
59. For example, Vladimir Gel'man, `Regional'nye rezhi-
my: zavershenie transformatsii?', Svobodnaya mysl', 9,
1996, pp. 13±22; `Stanovlenie regional'nykh rezhi-
mov v Rossii', NG-stsenarii, 19 September 1996, p.
5.
60. As put by Roberto Campos in Gellner and Cansino
(eds) (1996), p. 73.
61. Radio Free Europe/Radio Liberty, `Newsline', 20 Feb-
ruary 1998.
62. Huntington (1993) and his later book (1996).
63. See, above all, Borneman (1997); Kritz (ed.) (1995);
Pogany (1998).
64. Mill (1972), p. 71.
for 65. Ibid., pp. 72±3.
66. Cited in Zakaria (1997), p. 30.
 Merquior, `A Panoramic View on the Renais-
67. Jose
sance of Liberalisms', in Gellner and Cansino (eds)
(1996), pp. 7±20, at pp. 11±12.
68. Ibid., p. 9.
69. Dietze (1985), p. 1; Fish, (1997), pp. 208±9, draws on
Dietze to make the same point.
70. Fish (1997), p. 208.
71. Vladimir Sogrin, `Zapadnyi liberalizm i Rossiiskie
reformy', Svobodnaya mysl', 1, 1996, p. 43.
72. Quoted by Merquior, `A Panoramic View on the
Renaissance of Liberalisms', in Gellner and Cansino
(eds) (1996), p. 7.
73. Sogrin, `Liberalizm v Rossii', p. 16 (see note 6 above).
74. Soltan Dzarasov, `Liberalizm ili sotsial-demokratizm?',
in Svobodnaya mysl', 4, 1996, p. 28.
75. Ibid., p. 26.
76. Ibid., p. 31.
77. Ibid., p. 31.
78. Sogrin, `Liberalizm v Rossii', p. 18 (see note 6 above).
79. Ibid., p. 19.
80. Axtmann (1996).
81. See Hoffmann (1995), p. 162.
82. The point is made by Robert Taylor, `Springtime for
Hitler?', in Prospect, March 1998, p. 13.
83. Stockdale (1996).
84. Kymlicka (1989).
85. For a particularly rich debate of the issue, see Rajch-
man (1995).
86. Welch (1993).
87. Rustow (1970).
88. Hayes (1949).
89. Dmitry Shlapentokh, ` ``Red-to-Brown'' Jews and Rus-
sian Liberal Reform', in The Washington Quarterly,
21(4), Autumn 1998, p. 112.
90. See Hoffmann (1995), p. 165.
91. Ibid., p. 175.
92. Alexander Prokhanov interviewed, `Western Values
Âs', in Transitions, August 1997, p. 77.
are Cliche
93. Aleksandr Tarasov, `Striptiz liberalov', in Svobodnaya
mysl', 10, 1997, pp. 106±114, at p. 110.
285
 RICHARD SAKWA

94. Richard Bellamy, `How Not to Make Italians', in
Times Literary Supplement, 22 December 1995, p. 12.
95. Sogrin, `Zapadnyi liberalizm i Rossiiskie reformy, pp.
32±3 (see note 71 above).
96. Ost (1991), p. 85.
97. Gregory R. Johnson, `Modernity and Postmodernity in
 Merquior', in Gellner and Can-
the Thought of Jose
sino (eds) (1996), pp. 37±63, at p. 55.
98. Gay and Alekseeva (1996).
99. Sogrin, `Liberalizm v Rossii', p. 20 (see note 6 above).

286
 21
LIBERALISM,
ECOCENTRISM AND PERSONS
Brian Baxter
I
Throughout its long history liberalism has been
concerned almost exclusively with matters of human
well-being in so far as these are affected by social,
economic and political arrangements. Until very
recently virtually nothing has been said by liberals
about the issue to which Green political theorising so
vehemently directs our attention, namely that of
human relations with the natural context within
1
which all human action takes place. The reason
for this neglect does not seem hard to find. It appears
to stem from the fact that the fundamental value of
liberalism is that of the pursuit by individuals of
personal projects which they have freely chosen for
themselves.
Although on this planet these individuals com-
prise solely members of the human species, in fact for
liberalism it is specifically personhood, in whichever
species that is to be found, which is the crucial
element. Persons are, by definition, entities able to
exercise certain kinds of choice over the pursuit of
their life-plans, and it is persons with which liberal-
ism has exclusively concerned itself. Non-human
creatures and entities are noticed by liberalism pri-
marily as resources for human persons. At most the
capacity for suffering which some of them possess has
been regarded as a source of certain kinds of limita-
tions upon how precisely persons may put them to
use ± limitations we call, revealingly, `humaneness'.
From this person-centred standpoint has emerged
287
the call of liberals for the neutrality of the state with
respect to substantive views of the good life. Indi-
vidual persons are to be allowed, as far as is compa-
tible with their all being allowed to do so, to pursue
their own distinctive view of the good life. Each
individual counts equally with all others, and no
one's view of the good life is to be accorded greater
weight than those of other individuals. For the
liberal, the heart of morality is this impartiality
between persons.
Different individuals will have different value
commitments. Certainly, some may seek to attribute
non-instrumental value to the non-human, natural
world, and to seek to live lives in which the im-
plications of such valuing are worked out. But some
may deny any such value, and seek to treat the non-
human world purely on the basis of what conduces to
human well-being. For liberalism, such arguments
about which values should command our allegiance
are, of course, to be allowed and encouraged, and the
possibility of persuading others to our view of things
must always be left open (with some tricky issues
about toleration's limits remaining an ongoing mat-
ter of debate). De facto universal agreement on
substantive values is, thus, compatible with liberal-
ism provided the possibility of disagreement breaking
out is always left open by, for example, the main-
tenance of free speech, freedom of conscience and
the rest.
The view of values which is operating here is one
which sees them as anthropogenic, originating in
 BRIAN BAXTER
human (person) acts of valuing, and as anthropo-
centric, with human well-being as the fundamental
touchstone of what is good. Thus, even individuals
who attribute intrinsic value to nature are inter-
preted as really finding a kind of `higher' (aesthetic,
spiritual, non-practical) instrumental value for hu-
man beings in the natural world.
On this view, then, liberalism treats environmen-
tal concerns in the way it treats any other substan-
tive value positions: they are located within the
structures of individual preference and left to the
processes of liberal democratic politics for resolu-
tion. Whether they should lead to wholesale
changes in human practices is a matter upon which
liberalism, as such, has no view, save in one regard:
it is committed to opposing any changes which
threaten its fundamental value of individual choice
of life-project. Only when such changes are neces-
sary to avert clear danger to the very survival of
human beings (persons) would liberalism permit
them, temporarily and reluctantly, until the danger
is past.
At least some Green political theorists will be
unhappy with this standpoint, for it appears to
accord no fundamental value to the non-human
world of nature. In particular, the position known
as ecocentrism, to which many Green thinkers have
given their allegiance, directly contradicts the liberal
fundamental value postulate, for it enunciates the
following rival basic proposition:
[E] The non-human natural world has intrinsic
value, independent of its instrumental impor-
tance for human well-being.
In other words, ecocentrism rejects the liberal inter-
pretation of the attribution of intrinsic value to
nature as really being the attribution to it of various
`higher' forms of instrumental value for human
beings.
However, ecocentrism is not the only possible
basis for the development of a Green political theory.
This is just as well, for ecocentrism faces rather severe
difficulties which I will try to explore. These centre
on the problem of establishing an intrinsic connec-
tion between principle E above and a conception of
human well-being or flourishing. This connection is
288
clearly needed in order to establish for individual
human beings some compelling sense that acting in
accordance with E will achieve for them, and the
individuals about whom they care most, a form of
human life which will be satisfying and fulfilling.
Without some such claim ecocentrists will lack a
connection between their fundamental value-postu-
late E and what Thomas Nagel has identified as the
2
`personal' standpoint, which always characterises
the consciousness of self-aware beings such as our-
selves.
Liberalism, by contrast, makes this connection
quite straightforwardly, for its basic value postulate
amounts to an endorsement of the importance of
precisely that standpoint. It is true that, as liberalism
also contends, we can, and must, also adopt the
`impersonal' or universal standpoint, which is what
we do when we enunciate for general acceptance
liberalism's own fundamental value postulate. But for
much of the time such impersonal conceptions are
weak sources of human motivation and always run
the risk of being trumped by the requirements of the
3
`personal' standpoint.
The difficulty for the ecocentrist, of course, is that
on the face of it principle E sets out precisely to
eschew the `personal' standpoint entirely and to bid
people to adopt a point of view with respect to non-
human nature which completely leaves behind hu-
man interests and concerns, including their personal
ones.
Ecocentrism, then, needs a way of fusing the
personal and impersonal standpoints. The view to
be defended below is that the key to this lies in
developing the concept, whose importance to liber-
alism has already been noted, of personhood and
relating it to non-human nature. In the course of
doing this, however, we discover that the apparently
clear-cut distinction between anthropocentric (or
personcentric) and ecocentric positions disappears,
and that reasons can be found within the liberal
perspective for endorsing at a fundamental level the
kinds of value which Green theorists have so ar-
dently brought to political theorists' attentions. The
aims of liberalism and Green political theory turn out
not to be so completely at odds as appeared to be the
case at first sight, provided that Greens accept the
abandonment of pure ecocentrism, and liberals ac-
 LIBERALISM, ECOCENTRISM AND PERSONS
cept Green arguments for a revised formulation of
personhood.
II
Before developing this position, however, we need
first to consider some arguments explored by Robyn
4
Eckersley, which seek to elucidate and defend the
ecocentric position in a rather different way, one
which might be seen as reinterpreting such key
liberal values as autonomy by finding ecocentric
versions of them. It is a feature of the suggestions
she explores that they operate in such a way that the
intrinsic value of non-human nature is directly con-
nected to human flourishing. Hence, if such argu-
ments work, then the above-noted difficulties faced
by ecocentrism ought to disappear. As will now be
contended, these arguments do not succeed and thus
a different mode of reconciling the personal and
impersonal standpoints will be required.
Eckersley's discussion begins with the observation
that ecocentrism takes its cue from the ecologically
informed philosophy of the internal relatedness of
life forms within ecosystems, according to which
`there are no absolutely discrete entities and no
absolute dividing lines between the living and the
5
non-living.' Acceptance of this viewpoint in turn
leads to an ethic of general emancipation, according
to which all beings should be permitted to fulfil
themselves in accordance with their natures ± a
principle which will apply both to human beings
and to non-human nature. We should note that such
a principle clearly chimes with the liberal concern
for human self-development embodied in its funda-
mental postulate even though, as we have seen,
liberalism restricts its concerns in this regard solely
to human beings.
However, the ecological principle of intercon-
nectedness appealed to by Eckersley does not seem
to be straightforwardly related to the principle of
general emancipation which is supposed to unite
human and non-human flourishing. For all that
`interconnectedness' shows, it seems possible for
the flourishing of some beings to require that others
do not flourish (parasites, for example); or for the
flourishing of some beings to require only that others
exist, but not that they should flourish, either in-
289
dividually or as a species (domestic animals in factory
farms, for example).
Also, the kind of interconnectedness between
ourselves and non-human life-forms is important
when we are considering our attitudes to non-human
nature. The science of ecology suggests the connec-
tions are physical and biological, via `energy path-
ways' for example. However, it is arguable that
ecocentrists really need to establish interconnected-
ness between human beings and non-human nature
at the level of value or meaning. The focus should
thus be on the kinds of meaning ± moral, aesthetic,
religious, and so on ± which non-human nature can
have for people. Only thus is it going to be possible to
establish the right kind of connection between prin-
ciple E and the concept of human flourishing, which is
value-laden and permeated with a concern with
meaning.
One suggestion along these lines which Eckersley
explores is the ecofeminist claim that at least be-
tween one group of human beings ± women ± and
non-human nature there is a connection of mean-
6
ing. Women, with their nurturing, caring and life-
giving natures, are more `in tune' than men are with
natural processes. Thus they will find the caring and
nurturing attitude to non-human nature required by
the full recognition of E to be congenial and con-
ducive to their own fulfilment as women.
There may be truth in the claim that human
beings whose self-understanding involves caring
and nurturing will find the forms of life enjoined
by ecocentrists to be congenial in this way. But it is
clear that this is a contingent fact about them, if it is
a fact at all. There seems to be no incoherence in the
idea of a being whose caring and nurturing instincts
are directed only to its own kind, as is true of nearly
all non-human forms of life, for example. And in the
human case it is at least arguable that caring and
nurturing comes most naturally, for both males and
females, though on occasion for neither, with respect
7
to their own offspring and relations.
A more general kind of approach to the problem is
explored by Eckersley via direct consideration of the
concept, hotly debated between liberals and socia-
lists, of autonomy. According to her, the key move is
to expand the concept of the `self' whose autonomy
is characterised in the classic Kantian terms as `living
 BRIAN BAXTER
8 12
in accordance with self-imposed principle.' What
this expansion seems to mean is, in effect, accepting
that non-human nature has its own capacity for
autonomy. Non-human life-forms have their own
ways of living and modes of being. Respect for
autonomy in general thus requires respect for the
modes of living of such beings and underpins the
project of `general emancipation' already alluded to
above.
The focus on autonomy, and the expansion of its
application mooted by Eckersley, allow her to estab-
lish a link between principle E and human flourish-
ing. She argues that since ecocentrism is committed,
by its acceptance of E and the expanded concept of
autonomy, to the project of general emancipation for
all creatures, it necessarily also supports the condi-
tions conducive to the exercise by humans of their
autonomy. Giving this concept in turn what is in
effect a socialist gloss, she then suggests that this
commits ecocentrists to the pursuit of social justice in
9
human societies.
Now these ideas are not pursued very far by
Eckersley and it is, of course, a serious problem with
the suggestion she makes that it is not entirely clear
what is meant by extending the Kantian sense of
autonomy (the `following of self-imposed principle')
to creatures which do not seem able to impose
principles upon themselves in any sense.
Another attempt at elucidating how one might
reasonably arrive at an extension of the concept of
autonomy to non-humans, thus motivating accep-
tance of the general emancipation principle, is via
what Eckersley refers to as `Autopoietic Intrinsic
10
Value' (AIV) theory. In her account of it, the
key claim of AIV is that non-human nature is
capable of being self-renewing and that self-renewing
entities are `ends-in-themselves'. This is another of
Kant's classical characterisations of autonomous
beings ± beings which possess intrinsic, or non-
11
instrumental, value. Hence we should extend
the concept of autonomy to non-human nature,
even to entities such as ecosystems, for they too
possess self-producing or renewing capacities.
However, Kant's argument to persuade autono-
mous agents to value the autonomy of others begins
with the demonstration of the importance of one's
own autonomy, and then turns on the view of
290
autonomy as an exercise of rationality. This may
not be the only possible concept of rationality, but it
is one which excludes the attribution of autonomy to
creatures believed to be devoid of the requisite kinds
of rationality. Kant's argument is that since we must
value our own rationality, as a condition of the
meaningfulness of our own existence, and since
rationality is a matter of consistency ± not making
a distinction without a difference ± we cannot value
our own rationality without thereby committing
ourselves to valuing the rationality of others, for
to fail to do that would be making a distinction
without a difference, and thus failing to value ra-
13
tionality.
There is a problem about trying to rework this
argument if we do not understand autonomy as an
exercise of rationality, but instead regard it as synon-
ymous with the concept of self-renewal/production.
Certainly one can intelligibly claim that we must
each value our own self-renewing capacities, for we
cannot have meaningful lives if we go out of ex-
istence. However, this in no way commits us to the
valuation of the self-renewing capacities of other
entities. We may, of course, value their self-renewing
capacities for instrumental reasons of our own, but that
does not amount to viewing them as possessing
intrinsic values, and in any case is a value-attribution
exercise which requires us to consider each case on its
own (instrumental) merits. We are not committed to
a blanket valuing of all self-renewing entities by our
instrumental valuation of the self-renewing capacities
of some of them. Thus Kant's argument works, if it
does, because of the peculiar logic of the concept of
rationality. It is plain that this concept is not in-
volved in the idea of self-renewing/producing beings
and so it is hard to see how we can be argued into
accepting the intrinsic value of entities just in so far as
14
they instantiate this concept.
A further view, labelled `transpersonal ecology' by
15
Eckersley, does not aim at argument to convince us
to accept E, she tells us, but rather attempts to
produce psychological conversion to the ecocentrist
viewpoint. It does this by developing an extended
`sense of self'. That is, in the manner adumbrated by
Arne Naess, transpersonal ecologists seek to encou-
rage us to identify with non-human nature so that its
16
flourishing and ours become one and the same.
 LIBERALISM, ECOCENTRISM AND PERSONS
There are some obvious problems with this idea,
especially given that it is supposed to lead us to
identify not just with aspects of non-human nature
which possess a mentality and resemble us in various
respects, such as primates and higher mammals, but
also with entities which we can only grasp via the
possession of quite sophisticated theory and which
(on most views of the matter) are devoid of men-
tality, such as species and ecosystems. One can
certainly care about such things: seek to preserve
them, respecting and marvelling at them. But how
feasible is it to identify with them, to regard them as
in some sense part of ourselves?
There is one way in which this might be con-
ceived to happen which I will now outline, although
it is a way with which ecocentrists may not be
entirely happy. It is possible to undergo an extension
of the sense of self with respect to at least some
entities devoid of mentality. People, for example,
often seem to experience this with respect to their
automobiles. However, it appears to be a condition
of this happening that (1) there should be physical
control of the entity concerned via the human body
(the automobile is experienced as an extension of
the human body); (2) the external appearance of the
entity needs to be regarded as a direct expression of
the human self in question (it `makes a statement'
about the owner).
If one were to generalise this idea to the relations
between the human self and non-human nature then
it implies (1) that we identify with non-human
nature in so far as it is viewed as an extension of
our physical selves via our manipulation of it, which
suggests in turn the necessity for artefactuality; (2)
that we need to convert it into a symbolic expression
of our personal psychology, again requiring that it be
`made over' by us. These ideas are, of course, central
to the Hegelian/Marxist theme of the overcoming of
alienation by praxis.
On the face of it, these suggestions are unwelcome
to the ecocentrist position because they look to be a
version of the Enlightenment theme of rendering
non-human nature entirely subservient to human
purposes and needs when what is needed, according
to ecocentrists, is respect for non-human nature
because of its intrinsic value. But it may be feasible
to argue that the Enlightenment interpretation of
291
the key ideas is not the only one available. This may
not seem possible at first because we appear to be
faced with two contradictory requirements:
1. that we `make over' the world so that it
expresses ourselves to ourselves and to others;
2. that we respect non-human nature by main-
taining its processes intact as far as possible.
Here, (a) seems to imply that we countenance
wholesale interference with non-human nature,
whereas (b) seems to imply that this be kept to a
minimum. However, it is arguable that in the con-
text of our present world, (b) is to be regarded as
being as much a manifestation of human making and
self-expression as would be the effort to convert the
whole of non-human nature into a global garden
city.
This is because, on any account, wilderness will
not continue to exist, species will not survive, the
atmosphere and oceans will not remain unpolluted
unless human beings make strenuous and protracted
efforts to secure these aims. If we do successfully
achieve them, then they can unproblematically
count as manifestations of praxis. And, as is inherent
in the very idea of praxis, we will have remade
ourselves in the course of this `making over' of our
world. It is perhaps not too strained to suggest that it
will be precisely a sense of our identity with non-
human nature, whose (continued) existence will
then be in a clear sense our achievement and an
expression of our values, which will then be attained.
In a nutshell, a world which shows as few obvious
effects of human artefactuality as possible will, in
modern conditions, be the largest artefact we can
achieve. Such an artefact will be conceivable by us as
an extension of our (collective) personality and as
something which reflects that personality back to us.
Non-human nature and the human world will then
have been totally fused.
There are, however, various problems with this
hypothesis. From the ecocentric point of view it
perhaps still looks too anthropocentric, as if it in-
volved the claim that the only way in which we can
be brought to recognise the value of non-human
nature is by making it an instrument of our human
personality. Also, the hypothesis posits an enlarged
 BRIAN BAXTER
sense of self as the result of the implementation of the
Green project. But it does nothing to show people
why they should now embark upon such a protracted
and difficult project in a situation where large num-
bers of them do not have such a wider sense of self.
One point must be made at this juncture, now that
we have surveyed various suggestions for establishing
the connection between principle E and human
flourishing. We have looked at ecofeminist argu-
ments, autopoietic intrinsic value theory and trans-
personal ecology. I have indicated why they should
be regarded as not adequate to do the job required.
However, it should be noted that they all seek to
establish a connection between principle E and hu-
man flourishing by seeking to effect an identity,
partial or whole, between human beings and non-
human nature. This suggests that the ecocentric/
anthropocentric distinction has ceased to have any
real meaning. The defence of principle E turns out
always to involve appeal to something of prime
importance for human beings too, so that the idea
of non-human nature's having intrinsic value, irre-
spective of human concerns, has become very un-
clear. When we put non-human nature at the centre
we find that we put ourselves there too. The real
issue, therefore, is which of the ideological rivals has
the most adequate conception of humanity, not
whether ecocentrism is superior to anthropocentrism
or personcentrism. As will shortly emerge, this also
suggests a way in which the fundamental liberal
value postulate may be interpreted so as to recognise
and absorb the point of the ecocentric endeavour.
However, before we reach that point we still need
an argument to establish the connection between E
and the concept of human flourishing which can
give people the motivation to undertake the Green
form of praxis. In the next section I will try to find
such an argument.
III
The point made at the end of the last section is
recognised in effect by Eckersley when she suggests
that what is required is a new, ecological concept of
the self, in which persons are related to, not set apart
17
from, non-human nature.
Two points immediately arise. One is the cau-
292
tionary thought that the `ecological' concept of the
person ought to be developed out of existing con-
cepts of the person with which people are reasonably
18
familiar. A concept which makes no contact with
existing concepts is unlikely to have the appeal
necessary to motivate action. The second is the
question of how persons, as opposed to the animal
species Homo sapiens, are to be shown to be part of
nature. As members of that animal species we are
interconnected with non-human nature. But what is
the significance of that fact for the persons which we
also are?
To answer this question we need to say rather
more about what a person is. The following char-
acterisation, drawn from the European tradition of
philosophical and social-scientific thought to be
found in thinkers such as Hobbes, Kant, Rousseau,
Hegel and Marx, picks out the features essential to
personhood. Some of them are of direct relevance to
our problem:
A person is a self-conscious being, with the
following abilities all being implicit in the posses-
sion of self-consciousness:
1. to formulate intentions and determine how to
carry them out efficiently;
2. to develop a concept of self-interest and to
pursue self-interest so conceived;
3. to commit itself to values and to criticise itself
in the light of failure to realise those values to
which it has committed itself;
4. to differentiate between itself and others of its
kind;
5. to participate with others of its kind in creating
a culture of shared values and the associated
customs, traditions and institutions which em-
19
body those values.
It is a corollary of these features of personhood that
the prime focus of interest for persons is other persons
of the same kind. For it is from them that the values,
principles and conceptions come in the light of
which the individual person formulates its projects,
20
criticises and assesses itself.
Thus it is natural for persons to regard the non-
person context within which they are embedded as
 LIBERALISM, ECOCENTRISM AND PERSONS
primarily, or essentially, vehicles for interpersonal
interaction. Hence one might well argue that it is not
anthropocentrism, conceived of as a viewpoint with
its roots in Judeo-Christian and Enlightenment-phi-
losophical modes of thought, which is the main cause
of human persons' attitude of domination towards
nature. Rather, if the above characterisation of
persons is correct, then persons of whatever species,
wherever they live in space and time, will tend to
have that attitude towards the non-person world.
Thus, if any attitude is the problem here it is the
attitude referred to at the start of this discussion as
`personcentrism'.
If this charge is correct, then the question ob-
viously arises as to whether persons can ever avoid
personcentrism. It might be the case, for example,
that members of the human species could only avoid
personcentrism by seeking to avoid being persons
entirely, perhaps by trying to become purely `natural'
beings, like non-human animals, locked into some
suitable ecological niche, living only for the moment
and at the level of instinct. The cost of doing so
would appear to be the destruction of the kind of
mental life essential to personhood. However, as
long as the above-noted abilities remain intact the
suppression of their exercise will be a very tall order.
The question which needs to be addressed, then, is
whether it is possible for persons intelligibly to
conceive of the non-person world as something more
than a mere vehicle for interpersonal interaction. It
would clearly be very helpful for the ecocentric case
if it were possible to establish a link between non-
human nature and personhood which was intrinsic
to and ± at least in part ± determining of the very
nature of personhood. Further, this should ideally be
an intrinsic connection whose recognition by per-
sons would lead naturally to humility before non-
human nature and a compassionate concern for its
existence and well-being.
The beginnings of an answer may well be found in
ability (5) in the characterisation of persons noted
above. Persons are essentially culture-creators. Is it
possible to argue that non-human nature is essen-
tially that out of which culture is created? If it were
so, then if non-human nature were to be removed
there would be no person-created culture, and with
no culture there would be no persons at all.
293
On the face of it this looks to be a highly im-
plausible claim. Imagine that our world was made
completely over into an entirely urbanised form,
with monocultural agriculture to sustain it, with
wilderness and wildlife existing only in zoos, genetic
banks, on film and in virtual-reality theatres. This
surely would be a description of a world in which
human culture had become all-pervasive, rather than
one in which it had become impossible.
At this point, however, the possibility arises of
formulating what one might call the transcendental
21
deduction of the category of culture. The key claim
here is that the resources for human culture (the
culture of a particular group of persons) must be non-
human (non-personal). This is because it is impos-
sible for there to be a source of culture which was
entirely person-based. A world consisting only of
persons would have no wherewithal to formulate
values, principles or purposes. This in turn is because,
as the analysis at the start of this section indicates,
`personhood' can only be formulated in purely formal
terms. If, per impossibile, a group of persons were to
exist who were only persons, and not embodied in
some animal species located in some ecosystem on
some planet, they could not generate culture because
they could not generate anything. Persons require to
be embodied to become determinate enough to
develop the self-consciousness, intentionality, value
22
and cultures which give them substantial existence.
The content and meaning of their purposes and
values then comes from their embodiment in a
context of extra-personal nature. The categories
which human beings have used to establish their
particular aims and values are thus necessarily drawn
from the animalian predicament of embodied per-
sons inhabiting a non-human natural context ±
`mighty hunter, feeder of my people'; `great builder';
scientist (`tamer of nature'); artist and priest (inter-
preters of nature's meanings). Non-human nature,
then, has to be viewed as a concrete matrix of
opportunity and opposition to give specific content
to human values and purposes, and to furnish the
essential context within which human persons can
`crystallise out' a sense of what they are, derived from
their joint construction of a culture of artefacts and
symbols, necessarily using their experiences and
interpretations of non-human nature to do so.
 BRIAN BAXTER
This is the theme of `praxis overcoming aliena-
tion' again, but with the additional claim now
emerging that it is essential to culture and to person-
hood that what is alien should never be `made over'
in its entirety, for then the material out of which
culture, and hence personhood, is made disappears
and when that happens culture and personhood
disappear too.
However, isn't this claim overblown? Even if the
above argument is successful, does it not show only
that persons need non-human nature to get culture
under way? Surely, however, once cultural creation is
initiated then there is no essential need for non-
human nature to remain as a sustainer of culture?
Why should not the cultural life of persons achieved
in the maximally person-transformed world outlined
above simply continue for as long as there are human
beings in existence?
We need to try to show, then, that human cultures
cannot be sustained purely from cultural materials by
means of one, or both, of the following two processes:
1. reflection, exploration and experimentation
internal to a culture;
2. encounters with other cultures: dialectical in-
terplay; absorption; collage.
A culture is a more-or-less dynamic complex of
interrelated elements encompassing the following
categories: language, art, religion and philosophy,
science and technology, economic and political
systems, customs and traditions. (More rudimentary
cultures will not have developed under science, but
all cultures develop the other categories to a degree.)
The first point to make is that many cultural
activities under these headings do require a context
of non-human nature directly in order to exist at all
or to be fully intelligible. Within our own culture, for
example, such widely popular activities as rambling,
hill-walking, mountaineering, bird-watching and
hunting need non-human nature (albeit a nature
which is often extensively made over by human
hands) to be feasible. Some scientific disciplines ±
biology, ecology, ornithology, zoology and botany,
for example ± take non-human nature as their direct
object of study. It is clear from the very idea of `sites
of special scientific interest' that such disciplines
294
have an interest in as wide a range of species and
habitats continuing to exist as possible to provide
objects of study, sources of data and arenas for the
testing of theories. Arguably it is from many of the
disciplines and subdisciplines in those sciences
which directly study non-human nature that we
are obtaining insights into our own nature as embo-
died persons. Many forms of art, too, need non-
human nature as a source of subject-matter and of
design ideas, even in the practical design of artefacts.
Obviously, this applies to landscape painting, but it is
essential to the development of any art with a strong
visual component, including architecture and dance.
The endless variety of nature is essential here, as a
continuing source of forms, designs and motifs.
The example of the art of painting may help to
elucidate and support this claim. Can paintings be
made entirely out of the experience of other paint-
ings? It is certainly arguable that painters initially
acquire the idea of what painting is from their
encounters with existing paintings, encounters
which they can then use as launching-pads for their
23
own works. But it is clear that the painter needs to
return unfailingly to non-human nature in order to
replenish the wellsprings of painting. Even the de-
velopment of new schemata and new artistic voca-
bularies requires the painter to have encounters with
non-human nature in order to provide the elements
experienced under the schemata, and this is no less
true of non-representational painting.
In reply to these considerations the point might be
made that such aspects of our own culture as appear
to need direct confrontation with non-human nature
conceived as such are by no means universal: not
everyone is interested in them. Also, there are plenty
of ingredients in our culture ± many games, for
example ± which do not appear to need the con-
tinued flourishing of non-human nature to provide
them with meaning and purpose. So are not these
points too weak to establish the necessity for human
culture of the existence of non-human nature? Surely
activities such as those mentioned may fade away
without there being any significant diminution in
the cultural possibilities left to us?
However, the existence of such cultural activities
alerts us to the possibility of a Rawlsian-style defence
24
of non-human nature, and thus takes us in the
 LIBERALISM, ECOCENTRISM AND PERSONS
direction of the absorption by liberalism of the main
thrust of the ecocentrist argument. It is possible to
argue that activities such as those mentioned are
clearly part of the self-understanding and flourishing
of those who take part in them. None of us can know
for certain that they might not become part of our
own conditions for flourishing, even if they are not
so at the moment. Also, such people as find them
essential to their sense of self are not members of a
separate group of human beings. They are inter-
spersed among the general population and are the
objects of caring feeling as children, spouses, parents
and so on. Thus, even if these activities in our
culture carry no meaning for a given individual, that
individual may find himself caring for others for
whom they are extremely important.
So we all have an interest in maintaining the non-
human natural context within which such cultural
activities are to be carried on. Rawlsian persons in
the `original position' would undoubtedly want to
preserve as varied a natural context as possible to
permit the possibility of engaging in such meaning-
conferring activities, for themselves or for those for
whom they might care. More generally, many com-
plete forms of life, such as those centred on hunting
and fishing, need the specific contexts and oppor-
tunities provided by non-human nature in order to
exist. The customs, traditions, art-forms, social ar-
rangements and religious beliefs associated with such
lifestyles depend, therefore, derivatively on such a
context. Across the world, the spread of western
cultural forms, the degradation of the natural envir-
onment and the sheer increase in human numbers
are attacking the underpinning in non-human nat-
ure of such diverse `experiments in living'.
However, it might be claimed that no human
cultural form has to exist to secure human well-being
and that diminution in the range of cultural forms is
not necessarily a problem. For, after all, people are
adaptable and within a generation or so the grand-
children of fisherfolk can happily exist as urban
25
sophisticates.
Why, then, should there be as wide a variety of
human cultures as possible, including those directly
dependent on specific contexts of non-human nat-
ure? Two reasons may be given. The first is that the
variety is needed in order to allow the possibility of
295
exploring all the ways in which one may lead a truly
human life. Even if some such experiments in living
have their pernicious aspects, they add to the sum
total of our self-knowledge and may contain redeem-
ing aspects for human culture which would be
otherwise unobtainable (for example, consider the
way in which blues and jazz emerged out of the slave
experience in the USA).
The obverse of this point is that no single cultural
formation can be entirely adequate to all the sides of
our human personality. This is, of course, a central
liberal theme but one which can be turned to good
account by Green theorists concerned to defend the
integrity and variety of non-human nature.
The second reason is that any single culture
which does not periodically, or continuously, re-
plenish itself by the second process noted above ±
namely, encounter with other cultures ± runs the
serious risk of becoming `played out'. The possibi-
lities of purely internal development are probably
limited. A culture which does not receive jolts from
either direct contact with non-human nature or
from contact with other cultures faces the strong
possibility of decay. A single, homogeneous world
culture which we may be in the process of creating
would, on this view, be a disaster for human cultural
development as a whole. Hence, we need to sustain
as many diverse experiments in living as this tiny
planet will allow, including those which require
healthy non-human natural environments within
which alone they can develop.
In a nutshell, then, the personcentric argument for
the maximum diversity of non-human nature is that
it is needed for the maximum diversity of human
culture. The latter in turn is essential to the con-
tinued health of human cultural development. Re-
cognition of these points by liberalism requires no
more than an interpretation of traditional liberal
defences of the diversity of human cultures and
subcultures via grasp of the implications of the
necessarily embodied nature of persons. It is the
latter which the defenders of Green political theory,
such as ecocentrists, have securely grasped. Accep-
tance of it by liberalism enables the importance of
the non-human world to be grasped from within the
liberal tradition in a way which enables an argument
to be mounted for the defence of the multiplicity of
 BRIAN BAXTER
that world. Pure ecocentrism is rejected in this
version of liberalism, but a fusion of the personal
and impersonal standpoints is achieved in a way
which maintains the strong motivational impetus
which is liberalism's strong point. Further, non-hu-
man nature ± on this interpretation of liberalism ± is
at least as central a concern as the human (person)
well-being with which it is inextricably entwined.
IV
It has, of course, long been a liberal theme that only
within an economically dynamic capitalist system is
human diversity properly catered for. The trouble
with this view of that diversity is that it is concep-
tualised too individualistically. Individual flourishing
only occurs within cultural formations. Diversity of
modes of flourishing, therefore, requires diversity of
cultures and, as we have seen, that in turn needs
diversity of non-human nature within which experi-
ments in living can take place. On this view a single,
homogeneous technosphere, of the kind which
Green theorists fear that liberal-supported capitalism
is setting in train, is an impoverished environment
for persons, however materially well-off they may be.
This line of argument suggests that the personcentric
version of liberalism developed above may be used to
formulate liberal reasons for moving away from un-
trammelled capitalism. In this respect, however, this
version of liberalism has honourable antecedents
within the liberal tradition for many liberal theorists,
from Mill to Rawls, have cast doubt upon the
necessary connection between the basic value pos-
tulate of liberalism and the capitalist system.
There is, however, a point critical of Green po-
litical theory in general, and the `greened' person-
centric version of liberalism in particular, which
emerges from these arguments for the diversity of
environments within non-human nature and the
diversity of cultures which they underpin. Such
political theories, it may be claimed, seem them-
selves to be aiming for a single, stable, ecologically
defensible cultural formation. Do not such theories,
like all totalising ideologies, foreclose on certain
possibilities of human cultural development? Cannot
296
the first round of arguments in the previous section
be turned against such theories so as to argue for the
possibility of, for example, patterns of urban organi-
sation within which very unecological modes of life
may be sustained (for at least some of our loved ones
may find these to be terribly important for their self-
understanding)?
Clearly, what is needed at this stage by the
proponent of the personcentric argument for respect-
ing the variety of non-human nature's world is a
principle to deal with this objection. This principle
will require that the maximum variety of human
cultures is to be allowed compatible with the pre-
servation of that diversity of non-human nature
which is needed to underpin cultures. That is, any
culture which requires the eradication of the variety
of nature is to be ruled out.
Obviously, some cultures permitted under this
principle will have features which are morally objec-
tionable on other grounds. However, the above
principle at least allows the possibility of experiments
in living which require a specific context of non-
human nature within which to operate, even if not
all such experiments may be ones with which it
would be in the moral interest of human beings to
persist. To put the point in another way, the person-
centric argument for the variety of human cultures
does not preclude moral criticism, and attempted
termination, of some of the experiments in living
which natural variety makes possible.
Thus from ecocentrism liberalism is able to ex-
tract an argument for claiming that human persons
would do well to maintain non-human nature in as
rich and varied a form as possible, so as to provide
the largest set of possibilities for contexts within
which human culture can find its significance and
resources. Since human culture as it has recently
been developing has been extensively destroying
non-person nature, human persons need to learn
to restrain their cultural acts so as not to cut off the
branch on which they are sitting. Human culture
must limit itself so as to preserve itself, for only then
can human persons guarantee their continued flour-
ishing as persons, which is the fundamental liberal
26
aim.
 LIBERALISM, ECOCENTRISM AND PERSONS
Notes
1. Brian Barry (1995) and Marcel Wissenburg (1998) are
the two contemporary liberal theorists who have most
clearly shown an awareness of, and willingness to
engage with, the arguments of Green political theor-
ists.
2. See, for example, Nagel (1991), pp. 4±5: `The im-
personal standpoint in each of us produces . . . a
powerful demand for universal impartiality and equal-
ity, while the personal standpoint gives rise to indi-
vidualistic motives and requirements which present
obstacles to the pursuit and realisation of ideals . . .
My claim is that the problem of designing institutions
that do justice to the equal importance of all persons,
without making unacceptable demands on indivi-
duals, has not been solved ± and that this is so is
partly because for our world the problem of the right
relation between the personal and impersonal stand-
points has not been solved.' The problem identified by
Nagel is here supposed by him to arise with respect to
traditional political theories, but it plainly arises also
for ecocentrism.
3. John O'Neill (1993, p. 159) offers an Aristotelian/
Marxist argument to connect individual human well-
being and our capacity to appreciate disinterestedly
the intrinsic value of the natural world (as in scientific
study of it, for example). The latter is claimed, cor-
rectly I think, to be our characteristically human
capacity and thus in exercising it we are at our most
fully and satisfyingly human. The argument, however,
is open to the objection that, as Nagel has claimed, our
ability to ascend to the level at which we can view the
world from the impersonal standpoint of morality and
science is, for most of us, a fitful and fragile business.
The arguments developed later in this chapter might
be seen as an attempt to connect the appreciation of
the intrinsic value of non-human nature with indivi-
dual human well-being in a more robust way, via the
conception of oneself as embedded in a culture. They
are, therefore, intended as complementary to the
Aristotelian/Marxist positions.
4. Eckersley (1992), chapter 3.
5. Ibid., p. 49.
6. Ibid., pp. 63±71.
7. As one of the commentators on an earlier version of
this chapter noted, Eckersley is herself critical of the
proposition that women are more `in tune' with
natural processes than men.
8. Eckersley (1992), p. 54.
9. Ibid., p. 56.
10. Ibid., pp. 60±1.
11. This point is attributed to Fox (1990), p. 172.
12. Kant (1948), p. 91.
297
13. This, at least, is my understanding of what Kant is
saying in a notoriously difficult, though penetrating,
discussion.
14. It has been argued by, for example, Regan (1988) that
at least some non-human animals may be devoid of
rationality and thus autonomy, in the Kantian sense,
and yet still be worthy of moral consideration. He
argues that those animals which are `subjects of a life'
possess rights of the same sort and to the same degree
as do human beings. The concept of `preference
autonomy' applies to such creatures. Whatever the
merits of this case it is clear, as Regan himself recog-
nises, that it extends the scope of human moral
concern to only a relatively few members of the
non-human realm. It also rests upon a concept of a
`subject of a life' whose meaning and moral weight are
likely to be highly controversial. Many may find it
persuasive but it would be preferable if a concept of
autonomy could be found which is applicable to non-
humans and which did not lend itself to being morally
trumped by the Kantian concept of autonomy, invol-
ving rationality, which applies only to human beings
(as far as we know).
15. Eckersley (1992), pp. 61±3.
16. See Naess (1989).
17. Ibid., p. 54.
18. I am here endorsing the general immanentist position
in moral and political debate espoused, for example, by
Walzer (1983), p. xiv.
19. Different philosophers have stressed some of these
elements more than others. Self-interest and means-
end rationality (points (1) and (2)) were of particular
importance to Hobbes and he gave a very specific
account of how (5) is possible and why it is necessary.
Self-consciousness and the differentiation of oneself
from others were of great concern to Kant. Point (5)
was given a non-individualistic basis in Rousseau, a
theme developed by Hegel and Marx and taken
further in more recent socialist and conservative
thinking.
20. This is a theme at least as old as Rousseau's Discourse
on the Origins of Inequality and has been the focus of
much of the discussion among recent communitarian
theorists.
21. The allusion here is, of course, to Kant's attempt to
establish what is necessary for there to be the possi-
bility of experience. Here I try to establish what is
necessary for there to be the possibility of culture.
22. These claims may gain plausibility from the arguments
put forward by P. F. Strawson to establish the logically
primitive nature of the concept of a person, in which a
person is essentially conceived of as an entity to which
 BRIAN BAXTER
both mental and physical attributes are to be ascribed.
See Strawson (1959).
23. This is the theme of Sir Ernst Gombrich's account of
the history of western visual art in terms of `schema-
and-correction', presented in Gombrich (1977).
24. Rawls (1972), chapter 3.
25. Ted Benton has considered the claim that habitat
protection, while it may be necessary to preserve the
well-being of non-human animals, is not important in
the human case because the direct relations of causal
dependency which link habitat conditions, patterns of
social life and individual well-being in the case of non-
human animals simply do not apply to humans (Benton,
1993, p. 174). He replies that human adaptability is
neither limitless nor devoid of definite causal condi-
298
tions, even though we cannot say in advance of an
analysis of a human community's mode of life (especially
the material culture) what those conditions will be (p.
174). Further, he makes a claim for the idea of an
environmental right to preserve the `ecological integrity
of a sufficient geographical terrain for the living of that
social life' (p. 175). These arguments are complementary
to my own, which might be regarded as furnishing
reasons why even people who are not themselves mem-
bers of a community whose culture is under threat may
view such a threat as being directed towards them too.
26. I would like to acknowledge the very helpful sugges-
tions made by Mark Evans concerning the revision of
this chapter for the present volume. Any faults which
remain are, of course, my own.
 22
PROLEGOMENON TO A LIBERAL
THEORY OF THE GOOD LIFE
Mark Evans
I
The possibility and desirability of a perfectionist
liberalism, which encourages a liberal state to favour
1
one or more conceptions of the `good' or `good life'
over others, has been much debated in contemporary
political thought ever since John Rawls popularised
the opposing idea of liberalism as an essentially anti-
perfectionist doctrine. Sometimes, the perfectionist
case rests upon the claim that liberalism's funda-
mental political principles cannot be plausibly con-
ceptualised in abstraction from specific conceptions
2
about how it is good for people to live. Alone, the
`conceptual inevitability' argument may not be en-
ough to win the day. Anti-perfectionists can admit
that liberal political principles unavoidably lean
towards certain ideas of the good and away from
others. But they view this bias with regret, as some-
thing to be avoided as far as possible when promoting
3
liberalism's fundamental values. Hence anti-perfec-
tionism can be recast as a politics governed not by an
implausible commitment to `total neutrality' but a
tendency towards neutrality, going no further in fa-
vouring particular good-life conceptions than is ne-
cessary to secure the basic liberal political values.
Perfectionist liberalism need not be so readily
rolled over, though. Joseph Raz, one of its most
prominent followers, rejects `abstinence-where-pos-
sible' from good-life bias. He contends that a liberal
state can and sometimes should actively promote
certain good-life conceptions beyond those implicit
299
in the basic principles to which anti-perfectionists
would restrict their concerns. He argues that being
autonomous is not sufficient to make a life good, for
autonomy must be exercised in pursuit of valuable
projects for it to have value. Believing that the
options from which people autonomously choose
their projects are, in general, socially defined and
determined, Raz urges that to promote autonomy as a
good, the liberal state may support those which are
4
indeed valuable. By offering liberal politics this
responsibility, his theory has significantly challenged
anti-perfectionist liberalism's efforts to equate itself
with liberalism in toto.
What is striking, though, about much of the
interest Raz has renewed in the idea of perfectionist
liberalism is that the acceptance of engagement in
practice with certain conceptions of the good has not
been matched with a parallel examination of them in
theory. He himself says little about which ideals a
perfectionist liberal state might be inclined to fa-
5
vour. Another liberal critique of anti-perfectionism,
William Galston's Liberal Purposes, also strives to
restrict itself to a `minimal perfectionism', hardly
venturing beyond the modal assumptions about
the good that anti-perfectionists might concede they
6
need to make. George Sher's Beyond Neutrality
develops an account of the good centred upon the
pursuit and achievement of `near-universal and near-
unavoidable goals' but it is noticeably reluctant to
discuss in detail the possible content of these goals
7
and their political ramifications. Peter Digeser's Our
 MARK EVANS
Politics, Our Selves? puts forward a Raz-type case for
the liberal state actively supporting certain concep-
tions of the ideal self to aid individuals' `self-crafting'
but once more provides little clue as to which of such
conceptions might be worthy of the liberal state's
8
support.
It is almost as if an extreme caution in deliberating
upon the nature of the good life as a political
concern, which is turned into a regulatory political
principle by anti-perfectionists, also overcomes per-
fectionist liberal theorising. The oddity of this char-
acteristic can be more fully appreciated when one
looks back at earlier liberals such as John Stuart Mill,
T. H. Green, Leonard Hobhouse or John Dewey.
These thinkers offered substantive views on how it
was good to lead one's life which they freely inter-
wove with their political reflections. The sharp
disjunction between political thinking and reflec-
tions upon ideals of personal conduct, which char-
acterises so much contemporary liberal theory, is
noticeably absent in many of its forebears.
Although it will be suggested that they should
decline to follow such precedents to the letter, this
chapter urges that today's perfectionist liberals can
and should seek to retrieve something of this tradi-
tion's spirit (thus it grants the perfectionist case as its
starting point). It hopes to encourage a conversation
about the good as something that liberals shouldn't
be afraid to join for fear of being thought `illiberal',
`arbitrary', `authoritarian' or worse. It proposes some
elements of a perfectionist ideal that they could
support, defending their liberal credentials and in-
dicating the role that liberal politics might play in
promoting them. As a prolegomenon, the chapter
can do no more than indicate some of the possibi-
lities here (hence no irony should be read into the
skimpiness of what is proposed!). But if its sugges-
tions seem fruitful, it can perhaps energise a more
complete and rigorous discussion of the type that has
unfortunately and unnecessarily become alien to
contemporary liberalism.
II
Before explaining how to rehabilitate good-life the-
orising in liberalism, we should reflect upon the
possible reasons why perfectionist liberals shy away
300
from it. It is hardly obvious, after all, that the mere
fact that liberals used to indulge thus is a compelling
reason for their descendants to do so today.
Contemporary perfectionist liberals are likely to
share with anti-perfectionists a much deeper sensi-
tivity to human diversity than we find among their
predecessors, who typically operated with detailed
metaphysical beliefs about human nature which
enabled such confident talk about the good life's
composition. Given this sensitivity, such determi-
nate and prescriptive beliefs are now much more
difficult to entertain. Perfectionists don't go as far as
anti-perfectionists in using `diversity' as a reason for
avoiding `good-life favouritism' in politics. But they
might nevertheless be reluctant to theorise concep-
tions of the good because of a doubt that any theory
could meaningfully identify specific ideals in abstrac-
tion from the myriad options that are actually avail-
able in society. Perhaps the good life is something
that is pluralistic in content, or heavily determined
by particular social context and hence capable of
formulation only by those to whom it applies as they
reflect upon their own circumstances. On this view,
there would be little for philosophers to say about the
good life which would carry any weight over and
above the citizenry's deliberations. This might lead
perfectionists towards what can be called a formal or
`open-textured' perfectionism, which is deliberately
not given much content a priori. Instead, it specifies
the form of the good in highly abstract, generalised
terms alone, rendering it compatible with highly
diverse substantive embodiments which individuals
or groups must formulate for themselves. Those who
adopt this approach may therefore insist that their
distinction from anti-perfectionism largely comes in
their practice, not the content of their theories. For
the diversity-sensitive open-textured perfectionist,
the most naturally liberal option could then be a
`democratic perfectionism' which would leave the
deliberation over, and choice of, the perfectionist
ideal's content to the political arena itself.
9
As Raz notes, it is not difficult to appreciate how
liberals might be swayed by these intuitions. But they
are not strong enough to render it inappropriate for
liberal political theory to formulate and examine
candidates for perfectionist promotion. Even if we
accept (as this chapter does) that the choice of
 PROLEGOMENON TO A LIBERAL THEORY OF THE GOOD LIFE
whether to pursue perfectionist policies, and which
ones, if so, should be left to a democratic process, it is
only when political theory is thought to be solely in
the business of identifying what principles must be
adopted by the state that the case for abstinence from
perfectionist enquiries seems compelling. Yet not
even theorists of justice, the staple diet of all liberal
political thought, necessarily view the fruits of their
deliberations as claiming to be compulsory. Some of
them, at least, also restrict themselves to theorising
what principles of justice might be adopted by a
liberal-democratic citizenry. They recognise that
there could be reasonable alternatives which might
10
be embraced with equal validity. They do not
think that their function is to prejudge, and circum-
vent the need for, citizens' own choices.
So in the way that theorists of justice use political
theory to articulate and analyse the candidate con-
ceptions of justice to be offered up for approval or
rejection by citizens themselves, I contend that
perfectionist liberal political theory can also propose
and scrutinise the possible candidate conceptions of
the good which might be presented as political
options. To sketch one such candidate we need to
clarify the main concepts it will employ. We must
then attempt to characterise its possible content in a
fashion which has substantive implications (i.e.
leading to both a theory and a practice which is
bolder on good-life questions than anti-perfection-
ism and the cautious perfectionisms described ear-
lier) but yet performs the delicate balancing act of
remaining faithful to the liberal ethos.
III
One way of thinking about the structure of a perfec-
tionist liberalism (hereafter: `PL') is to reflect upon
how it might diverge from anti-perfectionist liberal-
ism (`APL'). APL is typically constructed by sharply
distinguishing `conceptions of the good' and what we
can call the `principles of liberal justice', those
fundamental values that a liberal state is centrally
concerned to enforce. This distinction relies upon
some version of the belief that the moral basis of the
latter can be intelligibly described as being drawn not
from any particular conception of the good alone but
from something impartial with respect to such con-
301
ceptions that all in the relevant domain could
accept. Now, as indicated above, PL often contends
that principles of liberal justice are not obtainable
from anything other than a substantively, distinc-
tively liberal conception of the good. However, it is
still possible and certainly desirable for it to preserve
some morally and politically relevant distinction
between `justice' and `the good' in the following
sense. Principles of liberal justice have the specific
function of organising basic social and political
institutions and if they unavoidably express certain
assumptions about how it is good to live, these
nevertheless pertain to what are deemed to be
people's `essential' interests (and it is the business
of a theory of justice to propose what these may be).
Because they are `essential' the principles that these
interests prompt have first claim on the state's
attentions. Familiarly, liberals generally assume that
people's primary interest lies at least partly in being
equally free and hence they prioritise a scheme of
basic liberties. But they regard these principles as
providing only a framework of rules rather than what
Rawls calls a `comprehensive doctrine' which orien-
tates activity in many more, if not all, aspects of
11
individual and social life.
The `gaps' in the framework can be filled by
conceptions of the good, which may be thought of
as those ideals of how an individual might conduct
her life, which projects to adopt and/or character-
istics/virtues to acquire within the boundaries of
justice. Qua liberal, PL will not prioritise its favour-
itism in good-life ideals over the need to secure
justice. But it rejects the idea that the politics of
liberal justice automatically rules out all perfectionist
concerns.
We now need to consider what kind of perfection-
ism a liberal might consistently adopt. Here, it helps
to think about how a liberal might understand
`perfectionism'. The term has been used in confus-
ingly vague and varied ways. It can be `tightly'
defined as a doctrine which holds the pursuit of
human perfection or excellence (however that may
be understood) to be the object of morality. A
slightly slacker definition drops the exacting notion
of perfection and describes the goal to be the devel-
opment of human beings' potentials and capacities to
a full and flourishing degree (again variously under-
 MARK EVANS
12
stood). Contemporary liberal debate has tended to
use the term more loosely still to denote a doctrine
which simply advocates the promotion of a specific
ideal, or set of ideals, of how it is good to live (with
no necessary reference to either `perfection' or `flour-
ishing').
Although we are proceeding with the assumption
that PL will have more to say about the good than
the potentially vacuous notion of an `open-textured'
perfectionism, to retain its liberal identity it still has
to be in some way organised around a respect for
human diversity. We shouldn't expect it to be overly
detailed in its prescriptions and thus the loose defi-
nition is probably the most appropriate (although, as
I will later suggest, ideas of `flourishing' are not
necessarily absent). PL's ideals are also likely only
to be `partially comprehensive' (again following
13
Rawls ), declining to provide a complete, rigorously
specific blueprint for all areas of life but focusing
instead only upon certain aspects raised in the
question of how to live well.
Two final points before sketching a candidate PL.
A potential misapprehension concerning the perfec-
tionist project is that it necessarily posits a universal
ideal of the good, which is said to hold true invar-
iantly and without regard to time or place. It will be
clear, though, that many of my examples of PL's
concerns look very culturally specific. I suggest, then,
that even if we wish to retain universalist application
for principles of liberal justice (which not all liberals
wish to do anyway) it would seem appropriate to
make any perfectionist elements of a liberal doctrine
much more context-sensitive and localised.
Secondly, recognition of cultural variability plus
an acknowledgement of the role that socialisation
plays in the adoption of lifestyles strongly indicates
that PL rests upon a theory of the self that stresses its
malleability, or plasticity, over any intrinsic or es-
sential features. Instead, the self can be seen as much
less determinately constituted in its intrinsic nature,
awaiting instead the imprint of a culture for identity-
formation. PL therefore assumes that politics has
some legitimate interest in influencing the forces
of socialisation that shape the self. Political inter-
vention to encourage alternative influences upon the
construction of one's character need not be seen as
qualitatively distinct in any significant way from the
302
other social influences already and inevitably exer-
cised upon it. Consequently, a view of power which
admits that its effects can be `constitutive' and not
just `repressive' would seem to be an invaluable
14
conceptual tool for this type of theory. It helps
to remove the automatic link between the exercise of
power in the application of perfectionist policy and
the idea that it is necessarily an act of repression, a
link which might otherwise make liberals altogether
more suspicious of the perfectionist case.
IV
My candidate PL initially divides its concerns into
two. I call them `self-pertaining' and `other-pertain-
ing' issues, distinguishing them from the terminology
of `self-regarding' and `other-regarding' acts which is
popularly used to explicate Mill's famous harm prin-
15
ciple. The latter isolates what is and what is not of
potential concern to the state, whereas the politics of
PL has an interest in both self-pertaining and other-
pertaining acts. By `pertaining', then, I do not mean
to identify aspects of life which in some way are
either `of (political) concern' only to `self' or of
`others' too. Rather, a question about a `self-pertain-
ing' aspect of life relates to how one is living one's life
regardless of its effect on others. That's not to say a
self-pertaining aspect of life doesn't affect others and
nor does it mean that others may not raise legitimate
concerns about its effect upon them. If I ask the
question `is X a good thing to do?' part of the answer
may well rest upon the effect that X has on others.
But part of it can also refer simply to what good it
does to the self alone ± and this is the self-pertaining
dimension.
This aspect can be illustrated by thinking about
the following: whether to work hard to gain quali-
fications and get a good job or reject hard work for a
life of indolent pleasure; whether to produce and/or
consume great poetry or pornography in the privacy
of your own home; whether physically to live health-
ily rather than unhealthily; whether to be a well-
informed citizen willing and able to exercise one's
political rights competently, or know nothing about
politics and care even less; whether to live in squalor
or cleanliness; whether to lead a life that is harmo-
niously and happily balanced between work and play
 PROLEGOMENON TO A LIBERAL THEORY OF THE GOOD LIFE
or in which work is so onerous that life is constantly
stressful; whether to regard sex as an activity to be
engaged in freely and lovingly or as a commodity to
be bought and sold; whether to live a life of drug-
induced stupor or alert and lively mental sobriety.
Although all of these pairings may display other-
pertaining aspects, particularly when questions of
justice are involved (for example if the pornography
in question is paedophiliac), PL can also reflect upon
them as questions about the intrinsic quality of how
one's life is conducted, regardless of the effect upon
others, when justice is no longer at issue (unlike
APL, whose political concern ceases at this point).
PL is concerned first to specify those respects in
which politics may legitimately influence the life-
styles that people live from the self-pertaining per-
spective, and then to propose some principles by
which the value of various ways of life might be
determined.
`Other-pertaining' aspects of life are identified
solely with the effect that they have on others. PL
identifies which ones may rightfully become matters
of political concern, again proposing standards by
which to evaluate the effects to determine whether
they should be permitted, regulated or prevented
where possible. As we shall see, the kinds of beha-
viour which might pertain to others are not simply
what might `harm' them, which is the criterion for
Mill's `other-regarding' acts. Let us assume that these
are largely covered by the principles of justice. PL
proposes that there are other respects in which
people might justifiably expect or require certain
forms of behaviour from each other and in which
PL may grant politics a stake.
To illustrate this second aspect, think of the
following: drunken and foul-mouthed behaviour in
the streets; public billboards displaying hard-core
pornography; playing music very loudly at the ex-
pense of neighbourhood peace and quiet; advertising
work in the sex industry as a career opportunity in
job centres and school career offices; running a
workshop in your own garage which generates un-
pleasant though not noxious or toxic smells that
nevertheless disgust your neighbours; engaging in
sexual intercourse on the upper deck of a crowded
bus.
Let us grant that there are instances of such
303
activities which do not fundamentally harm those
who `witness' them in a way that would make them
matters for justice. In these instances people suffer no
physical injury and their own ability to choose their
own conception of the good need not be fundamen-
tally impugned, for example. Instead, they suffer
`offence', `embarrassment', `revulsion' or otherwise
16
experience such activities as unwelcome nuisances.
PL proposes that what effect they do have may
nevertheless warrant state intervention.
Perhaps under the influence of the general ten-
dency to neutrality, the `other-pertaining' category
has been drastically under-theorised in much recent
liberalism and it might therefore immediately strike
some liberals as too counter-intuitive to be some-
thing with which they should engage. But quite apart
from the fact that this category is present in some
extremely influential members of the wider liberal
17
tradition, I think that its neglect is extraordinarily
blinkered in another way. For it is a category that is
familiarly operationalised as law in actual liberal-
democratic states. Whatever is or is not tolerated in
purely private conduct, all such states will regulate
non-harmful conduct in public spheres and spaces to
a greater degree. To take one simple example: what-
ever sexual conduct is permissible in private, nearly
all of it is prohibited as a public performance. And I
think most liberals (including, if they are honest,
anti-perfectionists) would regard the existence of
some such restrictions as a good thing.
I am not invoking the crass argument that what-
ever is actually done by liberal-democratic states
must therefore be permitted by liberals' political
theory (plenty of such states' activities are unsanc-
tioned and indeed condemned by the theory). But I
am arguing that the fact that we have this category in
political practice is a very strong prima facie reason for
thinking that liberals can and should embrace it
theoretically. I suggest most liberals believe that
there are certain things which can be said and done
in private which, even though they do not `harm'
anyone, are rightfully prohibited in public. They
don't want the state to forbid your viewing of ob-
scene material in your front room, your nudist pro-
clivities in the kitchen and your boorish, foul-
mouthed drunkenness when you're at a friend's
house. But they do want it to forbid you from putting
 MARK EVANS
your obscene photos on a public billboard, from
walking naked up and down the high street or from
being drunkenly foul-mouthed at the bus
Transferring these activities into the public realm
is what makes these acts of rightful concern to others.
PL can therefore articulate a theory of `acceptable
public conduct' (very generally understood) to help
identify what should fall into this category.
Overall, then, PL seeks to support ways of life
that are `dignified' or `decent', `worthwhile' or `ful-
filling'. The ideal to inform this commitment could
take one or both of the following forms: (a) a
conception of what is good for people, specifically
to encourage its adoption; (b) a conception of what
is bad in people's conduct of their lives (`undigni-
fied', `shameful', `worthless', `unfulfilling') to discou-
rage its pursuit. This obviously implies
counterposition some notion of what is good for
people, but it could be compatible with various
conceptions of the good life and is thus perhaps
not as tightly prescriptive as (a).
V
It is now time to begin delivering on the promise to
sketch some definite content for PL. Here are three
possible components.
1. The conception of physical well-being
Liberal justice may invoke an obligation to provide
adequate health care for all regardless of circum-
stances, but it does not demand that the state make
everyone healthy. Nevertheless, the liberal state can
(and often does) campaign to promote physical well-
being, even though many people's conceptions of the
good life (smoking, drinking, prolonged sunbathing,
unprotected promiscuous sex, sloth) may be (some-
times knowingly) to their physical detriment. PL can
affirm that people ought to be free to determine how
to manage their own health, but it proposes that the
state has a legitimate interest in informing people of
18
the desirability of physically healthy lifestyles.
state may judge that ignorance over health matters
leads too many people to misjudge what is good for
them. PL allows it to propagandise for healthy life-
styles even with respect to people who are adequately
informed yet have not been willing to change their
unhealthy ways.
stop.
2. A conception of psychological well-being
Whereas it may be relatively easy to identify how it is
good to live in terms of physical health given the
availability of objective information on the subject,
it may be more difficult to do so with regard to some
questions of psychological health. Although we
might agree that there are certain psychological
disorders which can be diagnosed, the broader no-
tion of the `ideal state of mind' is likely to be
altogether more nebulous and variable than `ideal
physical condition' is for any one individual. I do not
therefore envisage PL being too prescriptive in this
by regard (avoiding such banalities as `everyone should
be happy and carefree!' It would be absurd to have a
state that campaigned against sullenness in the same
way it campaigned against poor diet). However,
there is some opportunity for it to posit certain
psychological conditions which it is not good for
people to experience and which its policies may
sensibly address.
Here is just one example to illustrate this possi-
bility. A wearily familiar refrain of many free-market
libertarians in particular over recent years has been
that the culture of competition in contemporary
capitalist society, which has left people increasingly
insecure in their jobs and often having to work ever
harder to try to keep them, is nevertheless a good
thing. They argue that such pressure increases pro-
ductive efficiency and hence brings material benefits
to the economy. On this view it is good for employ-
ees in the sense that they are compelled to become
more `effective' in their work.
Leaving aside the arguments that such employ-
ment trends do not unambiguously increase produc-
tive efficiency, PL could maintain that serious
questions about the quality of life are raised in these
practices regardless of their material effectiveness. It
could conceptualise some elements of ideal psycho-
The logical well-being to identify as undesirable certain
levels of stress which, for example, exacerbate feel-
ings of alienation and distress, and destroy the
capacity for a well-balanced personality to flourish
(not least if longer working hours in such a stressful
304
 PROLEGOMENON TO A LIBERAL THEORY OF THE GOOD LIFE
environment disable a person from properly experi-
encing life's other facets). Thus, (2) could be in-
voked as at least part of the justification for policies
which attempt to regulate working practices against
the interests of `economic efficiency.'
3. The ethical character-ideal
This is the third and most broad category, which
specifies ethical standards by which it is thought
good for people to live. This can be subdivided into
at least five (interlinkable) categories:
1. the virtues and skills which facilitate the re-
quirements of good citizenship;
2. the virtues and skills which facilitate autono-
my;
3. the qualities which comprise an appropriately
fulfilling life for humans to lead;
4. the virtues which constitute admirable character
and the qualities which constitute a dignified
life (`a life fit for a person to lead');
5. The requirements that constitute socially ac-
ceptable conduct.
I treat (2) ± (4) as largely self-pertaining matters; (1)
and (5) identify other-pertaining issues, though both
draw upon, and have consequences for, the self-
pertaining too.
The first two categories might be thought of as
those which primarily constitute the kind of char-
acter-ideal which is most directly favoured by liberal
principles of justice (the conception of the good from
which perfectionists might argue such principles are
necessarily drawn). Category (1) specifies the virtues
± the ideal dispositions and characteristics ± that are
required by citizens to discharge the social role of
citizenship well: civility, toleration, fair-mindedness,
cooperativeness, among others. PL may hold that, in
its ideal of politics, public affairs should always be
conducted with such virtues to the fore (and its ethic
can be the basis for attacking political conduct which
fails to exhibit them). But the liberal state would
have the largest scope for directly promoting virtue-
based forms of conduct in the kind of education it
provides for its children and it is on this matter that I
shall restrict my comments regarding this category.
305
Pupils are not only educated in certain academic
disciplines; they are also schooled (if often only
indirectly) in specific norms of conduct. Well-man-
nered, cooperative, trustworthy, respectful of
authority . . . a whole set of virtues and vices are
assumed in the ways they are expected to behave.
When we focus on the increasingly popular notion of
19
civic education in particular, we can see this
project coalescing around a character-ideal of `good
citizen' as being the intended end-product.
Now APL can reconcile itself to the idea of en-
couraging such characteristics in so far as they are
valuable for strictly political ends alone. It might also
reluctantly admit that fitting people for liberal politics
could have knock-on or overspill effects upon other
20
spheres of life. Thus teaching tolerance and equality
of respect for the political arena might introduce the
exercise of such virtues in people's private relation-
ships. On my conception, PL might not do much more
than APL to promote good citizenship but it may not
have the same degree of reluctance about its overspill.
It could regard it as good and justifiable for a politically
mandated programme of civic education to encourage
the development not just of good (tolerant, fair-
minded and reasonable) citizens but good people in
general, who display the virtues of good citizenship in
all their dealings with others.
Of course, it would be desperately naive to expect
education to yield exactly the same type of good
citizen/person in all children. Nor, in all probability,
should PL even want it to or regret that it does not.
We might be pleased at the fact that no such method
of socialisation is going to avoid the production of
highly diverse characters, some of whom will be rude,
dishonest, ruthless, uncooperative and so on. It is a
good thing that we don't have a society in which all
such vices have been knocked out in one's youth
(quite apart from the value of variety, maybe many
necessary and great things can only be done by
people who display what we would ordinarily think
of as unwelcome characteristics: ruthlessness, say).
But still, this does not mean there is no moral worth
or practical point in schools seeking to set such
examples. There is no paradox in attempting to
impart some character-ideals generally even when
it is not necessarily to be regretted that some will not
live by them.
 MARK EVANS
Promoting autonomy (Category (2)) requires the
encouragement of both the taste and the aptitude for
autonomous choosing, and the provision of a social
environment which facilitates it. PL could accept
that the perfectionist liberal state should tolerate
some ways of life which are not grounded in auton-
omy and, in the name of providing a variety of
options, even allow it actively to sponsor some of
them (for we mustn't assume that the state cannot
equally support mutually incompatible forms of life
21
for individuals ). But its preference for autonomy
implies the possibility that it views these as less good
ways of living (note this is crucially distinct from
viewing them as bad, even though it does not adopt a
relativist or sceptical approach to how lifestyles
might be evaluated). Non-autonomous lifestyles
may have a harder battle to survive than their
autonomous counterparts in a perfectionist liberal
society where people are encouraged to choose for
themselves. But this is not something that PL ne-
22
cessarily regrets.
Realistically, PL must recognise that autonomy is
always `bounded' in numerous ways: choices are
inescapably limited by one's context. For example,
in societies where one generally has to get a job to
live a minimally decent life, employment opportu-
nities are often severely restricted in particular loca-
tions and contexts (if they are available at all).
Practical limitations may mean that an autonomous
good life is always going to be a matter of degree and/
or episodic in character. That does not mean,
though, that one must always accept whatever lim-
itations happen to pertain. Some of the restrictions
upon people's choices may be intolerable and PL's
theory could therefore provide a guide to identify
situations in which autonomy has become unaccep-
tably restricted. I discuss an example of this in
section VII.
An important corollary of this function is that PL
can strive to widen and secure a range of valuable
options from which people may choose their auton-
omous projects. This is the policy made familiar to us
by Raz, whose perfectionism largely concentrates on
the possibility that the state can subsidise various
kinds of activity and lifestyle deemed to be worthy of
people's support (particularly when they may struggle
23
to survive without such help). No one is forced by
306
PL to take up such options. If people want to live
lazy, slovenly lives passively consuming vast quan-
tities of lager whilst watching pornographic videos,
the state cannot forbid them to do so (it need not
subsidise them, though; if people have no source of
income but welfare payments, the liberal state may
require them actively to look for work, or undertake
training programmes, to qualify for such benefits).
One consideration in identifying what might
count as a valuable option is provided by Category
(3). This retrieves from many prominent currents in
the liberal tradition the idea that the good life for
humans is one spent developing their potentials and
capacities. PL could favour activities that are devel-
opmental (creative, intellectual, physically demand-
ing) rather than enervating. Once again, it is
primarily in education that the perfectionist liberal
state is most likely to be able to act in support of this
ideal. It could, for example, attempt to gear a curri-
culum towards the development of a wide range of
intellectual, creative, aesthetic, emotional and phy-
sical capacities well beyond the functional require-
ments of a `flexible labour market' (`personal
transferable skills'), the kind of economic imperative
that may come to inform education if the state does
not take care otherwise. Category (3) may also
facilitate a critique of the wider social and cultural
environment to ascertain whether it is encouraging
or frustrating the development of individuals in
desirable directions. It may thus provide the grounds
for a radical condemnation of a culture that, say,
fosters banality and stifles creativity and fulfilment.
VI
Category (4) has close links to Category (1), in
which it was argued that PL would welcome the
overspill of the good-citizenship virtues into people's
characters more generally. The notion of what
counts as an `admirable character' could, then, sim-
ply build upon what is implicit in the citizenship
character-ideal. It is potentially very wide, ranging
across large sets of ideal dispositions and character-
istics. To specify its content PL could, for example,
follow Joel Feinberg in drawing upon Hume's four-
fold categorisation of features which are (a) pleasing
to us; (b) pleasing to others; (c) useful to us; (d)
 PROLEGOMENON TO A LIBERAL THEORY OF THE GOOD LIFE
useful to others (inverting each to obtain the set of
vices which could be discouraged). PL would then
think it, on balance, good for people to be, say, clean
rather than dirty ((a)); tactful and tolerant rather
than rude and insensitive ((b)); prudent and moti-
vated to better oneself rather than reckless or in-
dolent ((c)); trustworthy and cooperative rather
than deceitful and unreliable ((d)) ± the list could
24
go on.
Now as ideals for conduct in spheres of life
stretching well beyond the political, these features
may appear to be of no possible concern to any
liberal politics ± beyond, again, the kinds of char-
acteristics that state-sponsored education may try to
inculcate in children. Certainly, liberals would find
ridiculous the idea of a state enforcing rules requiring
any such behaviour. But it is far from clear that an
ethic governing such matters is and/or ought to be
wholly absent from liberalism's political morality. To
explain just one possible reason why, we can add a
sub-category to (4): (4a) The ethical standards of public
officials.
This issue has received practically no attention in
contemporary liberalism, but it points to what I take
to be a familiar way in which liberal politics may
serve to promote substantive ethical character-ideals.
It entails the identification of standards of conduct
which, though their opposites are not to be legally
proscribed, it is right for citizens to expect their
public officials in particular to exhibit. These func-
tionaries have a responsibility, if not a duty, to
support these ideals. If they fail to live up to them,
they set inappropriate examples of self-pertaining
behaviour to those whom they serve. The facts that
they can be condemned for doing so and that citizens
have some legitimate expectation that they should
not behave thus give credence to the idea of a self-
pertaining morality being a proper object of political
25
concern.
Popular reactions to the revelations concerning
President Clinton's relationship with Monica Le-
winsky illustrate this claim well. It is doubtless true
that the charge of perjury, which led to Clinton's
impeachment and which is primarily a matter of
criminal justice, is what concerned people upper-
26
most. Many people thought that his personal life
did not affect his ability to govern. But still, the way
307
he conducted himself in these self-pertaining matters
was widely thought to have been a poor example to
set to the nation and that the public had a right to
expect his behaviour, qua the leading public official,
to be better. His adultery, lying to his family and the
American people, the inappropriateness of the re-
lationship with a very junior employee barely older
than his own daughter: these are not illegal activ-
ities, but a certain widely shared conception of the
good nevertheless condemned them, and deemed
them particularly improper for the leading figure
in public life.
We might have been similarly condemnatory of
such behaviour if Clinton had been the chief ex-
ecutive of a burger bar company. But it is not clear
that citizens have the same legitimate expectation of
ethical conduct from someone who does not hold
public office. Of course it is true that people are
nowadays very cynical about politicians' motiva-
tions, suspecting (sometimes rightly) some hypocri-
tical Machiavellianism behind their protestations of
virtue. Also, too much genuine moral sincerity from
their mouths can eventually sound just tiresomely
sanctimonious and PL can recognise that the realities
of political life may well make it no place for moral
saints. But the very fact that there is a widespread
cynicism about the moral quality of political life
seems to imply some wish that it should really be
more virtuous, and PL could give voice to that wish.
If it is therefore thought right that political repre-
sentatives should display certain forms of conduct
and avoid others to set suitable ethical examples in a
way that would not be so publicly expected of the
boss of McClinton's, then political theory may have
a legitimately vested interest in scrutinising this
specifically political concern. PL can posit possible
ideals to inform these expectations of conduct.
To conclude this section, I offer just a few remarks
on the concept of a `dignified' life which was in-
cluded in Category (4). I treat this as distinguishable
from the notion of having an admirable character
and, again, it would be surprising if liberals of any
hue ± if they reflected hard on their ethical commit-
ments ± would want to resist adopting some such
notion to put to political work. PL merely embraces
this commitment openly and seeks to articulate it
fully.
 MARK EVANS
It is best illustrated by an example of Sher's: that of
drug addicts, living from fix to fix in excrement-
stained squalor, hopelessly dependent upon their
dealers who ruthlessly take sexual advantage of
27
them. Of course there may be questions of severe
injustice in this situation which take priority in its
condemnation. PL could also attack it for the loss of
autonomy and opportunities for leading a genuinely
fulfilled life. But on top of all this (if we are talking
about rich addicts who continually affirm an auton-
omous desire both to take drugs and experience all
that their effects entail), it could also judge such a life
to be simply too degrading, undignified and unfit for
humans to live. And, in self-pertaining judgements,
it could employ this distinctive ethic of dignity in
campaigns against its adoption.
VII
Finally, Category (5). Here we should recall that
other-pertaining judgements refer to phenomena
which become political concerns because of how
they affect other people in terms other than the harm
they may cause them. Often, their effect is sufficient
to justify legal restraint of the activity in question.
We have already rehearsed the argument that it is
consistent and desirable for PL to address other-
pertaining questions, but is there a distinctively
liberal way of isolating which issues they may ad-
dress? In other words, can PL identify them in a way
that would distinguish the liberal from non-liberal
approaches?
Here is an inadequately brief reply which hope-
fully is enough to suggest a distinctively liberal slant
to this issue. Although it has been the subject of
much conceptual attack, the classical distinction
between public and private realms of life can remain
operational for liberals in a very obvious way. Activ-
ities done without the presence of non-participating
others (in one's own home, paradigmatically) are
private in the way that the same activities done in
the high street ± publicly ± are not. On this con-
ception, PL holds that other-pertaining activities
have to be public: they really can or do physically
confront other people who otherwise have no part in
them. This is distinguished from those conservative
views which argue that there are private activities ±
308
typically sexual ± that are other-pertaining even
when no one else directly witnesses it (they think
it permissible legally to prohibit certain kinds of
consensual sexual activity even when done in the
privacy of one's own home because society's moral
28
fabric is supposedly damaged by them ).
In its self-pertaining deliberations, we have seen
that PL offers some recommendations as to how
politics might encourage or discourage people to live
privately in certain ways. But questions of legal
prohibition have not here been broached. Only
when certain activities move into the public realm
are they raised ± and this is a crucial part of what
makes this a liberal conception. In so far as other-
pertaining judgements inform permissible restric-
tions and prohibitions, it is largely informed by a
conception of how it is bad to live and act in the
public realm. Sometimes this draws upon self-per-
taining judgements about how it is good and bad to
live for oneself. At other times, it is only the location
of the activity in the public realm that generates any
negative evaluation of it. Given that sexual conduct
is a significant topic in PL's morality, we can use
some examples of such to illustrate this point.
With one important potential exception to be
explained shortly I take it that PL would, in general,
have nothing to say about any form of private
consensual adult sexual activity that did not pose
physical dangers to anyone else. But questions of
public offence are clearly raised as far as sex in public
is concerned. Thus, part of PL's concern in Category
(5) is to identify those types of activity which,
although all well and good in private, are wholly
inappropriate in public such that their public per-
formance may rightfully be punished.
The important exception provides an example
where a self-pertaining judgement is made about
an act which directly contributes towards the for-
mulation of an other-pertaining judgement about
that act when publicly performed. This is when
sex is treated as a commodity, engaged in not for
love or pleasure but for money: prostitution and
pornographic modelling. Let us leave aside the very
real problems of injustice in the plight of the (often
very poor) women in the sex industry, which PL
could happily admit removes any grounds for con-
demning the women. But this is not to say it cannot
 PROLEGOMENON TO A LIBERAL THEORY OF THE GOOD LIFE
condemn the kind of work they do and this is where
a self-pertaining judgement drawn from an ethic of
dignity, as suggested in (4), might be made.
Consider a woman who is forced by economic
circumstance into prostitution. Aside from the jus-
tice questions, the `dignity' issue can feature in the
following way. Most of us basically have to sell our
labour-power to live and many may have little
meaningful choice as to what kind of job can be
taken. This is an important way in which autonomy
is always constrained, as observed in section V. Yet if
the only option for an unemployed woman in a job
market is to become a secretary or a clerk, we might
regret the lack of choice but not consider that any
truly damaging harm has been done to her dignity as
a human being. However, if the only option left was
to fall into prostitution, then a liberal is much more
likely to judge that her life choices had become
unacceptably limited and that there was something
wrong with the socio-economic system that had left
her with such dismal prospects. This is where the
ethic of dignity informs an ethically and politically
crucial evaluation.
Now for the way in which this self-pertaining
judgement feeds into other-concerned legal restric-
tions on the sex industry. Imagine that PL follows
many liberals in thinking it desirable more fully to
decriminalise prostitution and pornography for prag-
matic reasons. In such instances, it would still hardly
require liberals to think it now acceptable to have
explicitly pornographic advertising billboards, broth-
els in any part of a town (next to an infant school or
old people's home?) and advertisements for the sex
industry in school careers centre advice packs and
Job Centre vacancy boards. For PL, there may re-
main something in the ethical quality of prostitution
and pornography, marking them out as less than fully
dignified, which justifies their exclusion from the
immediate, unavoidable public realm in which we all
necessarily move. That lack of ethical quality gen-
erates sufficient offence for many to desire not to
have to encounter it against their wishes in public. It
is this perfectionist liberal ethic that says it is right for
the law to try to spare us all (not just children and old
ladies) the sight of streetwalkers, kerb crawlers and
images of explicit sex on billboards or prime-time
television, but not other types of person and activity
309
against whom some might take offence. I may be as
disgusted at the sight of people wearing orange flared
trousers and open-toed sandals as others are at the
sight of explicit sexual imagery, but PL's ethic is what
identifies my sartorial tastes as inappropriate for the
legally controllable category of public offence.
VIII
Of course, all of this ± and particularly, I suspect,
some of the suggestions in the previous section ± is
eminently contestable. And it is right that it should
be so. It should be recalled, after all, that this theory
was not designed to settle questions that are in fact
only properly decidable in the political realm itself.
But this chapter has been designed to show that it is
appropriate for perfectionist liberals not only to raise
these questions but to venture some answers, elabor-
ating the assumptions upon which they are offered
and attempting to structure them into a coherent
ethical outlook. In so doing, it has hopefully forced
liberals to think harder about some of the ways in
which ethical judgements of this kind are made in
practice and to evaluate for themselves whether ±
despite what the tendency to neutrality might
prompt them to think ± it is in fact desirable for
liberals to preserve them in their theorising.
Still, it would be naive not to admit that serious
reservations may remain about the whole project.
There is no space to address them all, but I conclude
with a few remarks in PL's defence. I have not
claimed that my version of PL is the only way of
characterising a liberal theory of the good life, so the
conversation I advocate might well lead away from
some of the elements I have proposed. For example, I
can well envisage an argument holding that the gist
of the comments concerning the sex industry reflect
an unnecessarily ± illiberal? ± `prudish' attitude,
unfairly denigrating its work. I am happy to concede
the conversation on this matter is far from over.
But one might also object that part of liberalism's
very point has been to try to reduce the state's
interest in such apparently paternalist matters and
that PL threatens a dangerously reactionary reversal
of this trend. Of course it is true that liberalism has
mounted frontal assaults on many of the taboos and
conformist norms that permeated western cultures
 MARK EVANS
until fairly recently. Yet it is wrong to conclude from
this that liberalism, as many of its enemies often
charge, is necessarily committed to abolishing all
such standards and expectations of behaviour. Lib-
eralism is certainly a politics of toleration ± but one is
usually said to tolerate when one puts up with some-
thing that one doesn't much like. This therefore
assumes a standard of evaluation in itself, which PL
can articulate. And I have argued that liberalism
need not be (and in practice is not) open-endedly
tolerant; there are limits. Such perfectionist judge-
ments are familiarly operationalised in political prac-
tice and, again, I insist that there is simply no
compelling reason to think that, by reflecting upon
this fact, liberals would wish to abandon such prac-
tice altogether (they may want to alter some of its
content, but that is a different matter).
It is also incorrect to think that, by permitting in
principle the possibility of perfectionist politics, one
would inadvertently open the door again to the kind
of bigotry that liberalism has worked so hard to
undermine. This claim usually proceeds with a `slip-
pery-slope' argument: `legislating against public of-
fence X is wrong because it would encourage
legislation against Y and it would be wrong to
legislate against Y.' This argument implies that an
anti-X policy in itself has no serious drawback (if it
did have such, then we wouldn't need to condemn it
for the consequences it may have; its intrinsic flaws
would suffice). In principle, though, this argument
fails if one can show that we need not necessarily
slide down the slope at all. And if anti-X has much to
commend it, then the argument in its favour could
easily outweigh the slippery-slope argument.
Given that we are to leave the final decisions
about perfectionist ideals to democratic deliberation,
there is always a danger that PL's politics would
simply legitimise the enforcement of existing illiberal
majoritarian prejudices. There is a lurking worry
about the tyranny of the majority here and perhaps
29
a prudent (or, in Judith Shklar's sense, a `fearful'
liberalism, designed to reduce the opportunities for
such unwelcome uses of power, would argue that,
attractive though PL sounds in theory, we cannot
trust those who exercise power to do so wisely
enough to encourage it in practice.
It is unclear, though, that this fear is enough to
justify APL. Of course PL could be abused but then
so can, for example, the democratic ideal itself. Many
dictatorships have co-opted `democracy' to `justify'
the most terrible tyranny, but we don't regard this as
sufficient reason to abandon the ideal, only to be
more vigilant about its operation. And the dangers of
PL's abuse can be lessened precisely through the
careful theorisation of conceptions of the good,
rather than abandoning the matter in its entirety
to whatever instant, perhaps vulgarly prejudicial,
opinions might be thrown into democratic debate.
A properly thought-through PL, examined thor-
oughly and its moral positions exactingly contested
both by professional political theorists and by a
democratic citizenry in the political arena, would
seem to be the best guarantee we could hope for to
avoid such tyranny.
It may ultimately be the case that a liberal polity
would decide against an active perfectionist agenda,
in which case the articulation of such an ethic would
appear not to be necessary. But because liberal-
democratic states in practice do wander into perfec-
tionist territory, it is surely more incumbent upon
APL to say why it is not good for states to do so
(without caricaturising PL by raising the spectre of
some patently hideous Orwellian totalitarianism as
their reason). To add a final, brief remark to hint at
PL's possibilities (and to suggest as-yet uncharted
future paths for research into it): it may be that the
better-off denizens of the rich, capitalist west may be
ultimately unable to sustain their kind of materia-
listic lifestyles for two reasons. One is that the
demands of redistributive global justice will become
so overwhelming that cosseted indifference will no
longer be possible. The second is that the need to
avoid environmental catastrophe will require noth-
ing short of a profound change in the resource-
ravaging lifestyles of the rich nations. Both points
suggest the need to rethink fundamentally the ma-
terially extravagant conceptions of human flourish-
) ing encouraged in such societies. And to the extent
that political means may be required to facilitate
these ethical changes, the attempted indifference of
APL will clearly have to be abandoned.
So it is obvious that PL has been pushed by the
tendency to neutrality onto the defensive far too
much. It would infinitely profit liberalism for it to
310
 PROLEGOMENON TO A LIBERAL THEORY OF THE GOOD LIFE
become more emboldened, and to think harder and
more openly about the way in which politics might
Notes
1. I use the terms `good' and `good life' loosely and
somewhat interchangeably here. Sher (1997), pp.
37±44 discusses some of the vagaries of `conception
of the good.'
2. For example Galston (1991), part II.
3. See, for example, Rawls (1996), pp. 152, 199±200.
4. Raz (1986), chapter 14 develops most of the points
summarised here.
5. Raz talks of values being grounded in `concrete social
forms', suggesting a rather conventionalist account of
perfectionism's content. Even if this conservative bent
satisfies liberals ± and I will suggest it may not ± this is
still too insubstantial a formulation for us to have
much idea of where Raz's perfectionism could lead us.
6. Galston (1991), p. 177.
7. Sher (1997).
8. Digeser (1995).
9. Raz (1986), p. 162.
10. See, for example, Rawls (1999), pp. 581±3. Rawls
defends `justice as fairness' as a morality for political
fundamentals, but here recognises that it is only one of
many `reasonable' conceptions that might be adopted.
11. Rawls (1996), pp. 13, 175.
12. See Hurka (1993), pp. 3±4.
13. Rawls (1996), pp. 13, 175.
14. This quasi-Foucauldian approach to power as consti-
tutively disciplinary of the self is discussed in Digeser
(1995), chapters 1 and 4. Space permits no more than
a mention of this conceptual possibility; its articula-
tion and analysis is, though, a highly promising can-
didate for perfectionist liberalism's research agenda.
15. Although this specific terminology is F. H. Bradley's
rather than his own, Mill's distinction is implicit in his
On Liberty, Mill (1972), p. 78.
16. It might be argued that, for consistency's sake, liberal-
ism needs to reconceptualise `harm' in order to justify
to itself the idea that these are proscribable offences.
The problem with this is that it threatens to stretch
the concept of `harm' to breaking point. Quite apart
from the fact that it is hardly clear that the coercive
repression of something becomes permissible simply by
dint of redefining a concept, expanding `harm' to
311
contribute to what is already that most challenging
of projects: leading a good life in the modern world.
include `offence' does not remove any problems over
arguing about the proscribability of offence; it just
shifts them.
17. Mill (1972), p. 167; Hart (1963), p. 47; Feinberg
(1985).
18. In one sense, this might be an other-concerned activ-
ity where the state (that is the taxpayer) shoulders the
cost of treating illness. But it is primarily self-con-
cerned in that it is left up to individuals to decide
whether they wish to lead healthy lifestyles.
19. A familiar feature of American school life, civic
education is now being incorporated into Britain's
national curriculum as well.
20. Rawls (1996), pp. 199±200.
21. Raz (1986), pp. 161±2, 425.
22. For an argument urging liberals to reconcile them-
selves to the fact that their doctrine will unavoidably
discriminate heavily against certain kinds of cultures
in modern society, see Fitzmaurice (1993).
23. Raz (1986), pp. 417±18.
24. Feinberg (1988), pp. 277±81.
25. If it is posited that immoral conduct by its public
officials diminishes the polity as a whole, rather than
just simply sets bad examples to individuals, then such
behaviour arguably becomes other-pertaining too. On
this account, we would all be directly tarnished ethi-
cally by the antics of our elected representatives, as a
citizenry.
26. Note, though, that certain forms of lie ± such as
perjury ± are criminalised even though it is not clear
that people are always and necessarily `harmed' (in a
classic liberal sense) by them. The wrongfulness of
lying might therefore be derived at least in part from a
substantive ethical character-ideal which is separate
from the `harm' considerations that are primary in
grounding principles of justice. If it is right that types
of lying remain criminalised (as with perjury), then
perhaps we should admit that this is another example
of perfectionist considerations in politics.
27. Sher (1997), p.179.
28. Famously argued by Devlin (1965), chapter 1.
29. Shklar (1989).
 BIBLIOGRAPHY
Ackerman, B. (1991), We The People: Foundations, Cam-
bridge, MA: Cambridge University Press.
Ackerman, B. (1994), `Rooted Cosmopolitanism', Ethics,
104, pp. 516±35.
Adorno, T. and Horkheimer, M. (1986), Dialectic of En-
lightenment, 2nd edn, trans. J. Cumming, London: Verso.
A'Haleem, A. M. (1992), `Claiming Our Bodies and Our
Rights: Exploring Female Circumcision as an Act of
Violence in Africa', in M. Schuler (ed.), Freedom from
Violence: Women's Strategies from Around the World , New
York: Women, Law and Development, OEF Interna-
tional.
Albertini, M. (1999), Nazionalismo e federalismo, Bologna:
Biblioteca Federalista/Fondazione Europea Luciano Bo-
lis.
Alexander, J. C. and Lara, M. P. (1996), `Honneth's New
Critical Theory of Recognition', New Left Review 220,
pp. 126±36.
Anderson, P. (1992), A Zone of Engagement, London:
Verso.
Arneson, R. (1989), `Equality and Equal Opportunity for
Welfare', in Philosophical Studies, 56, pp. 77±93.
Avineri, S. (1972), Hegel's Theory of the Modern State,
Cambridge: Cambridge University Press.
Axtmann, R. (1996), Liberal Democracy into the Twenty-
First Century: Globalization, Integration and the Nation-
State, Manchester: Manchester University Press.
Babbit, S. (1993), `Feminism and Objective Interests: The
Role of Transformation Experiences in Rational Delib-
eration, in L. Alcoff, and E. Potter (eds), Feminist
Epistemologies, New York: Routledge.
Bader, V. (1997), `Transnational Citizenship', in Political
Theory, 25(6), pp. 771±813.
Balcerowicz, L. (1995), Socialism, Capitalism, Transforma-
tion, Budapest: Central European University Press.
Barry, B. (1995), Justice as Impartiality, Oxford: Oxford
University Press.
Beetham, D. (1991), The Legitimation of Power, London:
Macmillan.
312
Beitz, C. (1979), Political Theory and International Relations,
Princeton, NJ: Princeton University Press.
Beitz, C. (1989), Political Equality: An Essay in Democratic
Theory, Princeton, NJ: Princeton University Press.
Bellamy, R. (1992), Liberalism and Modern Society, Cam-
bridge: Polity.
Bellamy, R. (1994), ` ``Dethroning Politics'': Liberalism,
Constitutionalism and Democracy in the Thought of F.
A. Hayek', British Journal of Political Science, 24(4), pp.
419±41.
Bellamy, R. (1999), Liberalism and Pluralism: Towards a
Politics of Compromise, London: Routledge.
Bellamy, R. (2000), `Dealing with Difference: Four Models
of Pluralist Politics', in Parliamentary Affairs, 53(1), pp.
198±217.
Bellamy, R. and Castiglione, D. (eds) (1996), Constitu-
tionalism in Transformation, special edition of Political
Studies (44(3)).
Bellamy, R. and Castiglione, D. (1998), `Between Cosmo-
polis and Community: Three Models of Rights and
Democracy within the European Union', in D. Archi-
bugi, D. Held and M. Koehler (eds), Re-Imagining
Political Community, Cambridge: Polity.
Bellamy, R. and Hollis, M. (1995), `Liberal Justice: Poli-
tical and Metaphysical', Philosophical Quarterly, 45, pp.
1±19.
Bellamy, R. and Hollis, M. (eds) (1999), Pluralism and
Liberal Neutrality, London: Frank Cass.
Benhabib, S. (1992), Situating the Self, Cambridge: Polity.
Benhabib, S. (ed.) (1996), Democracy and Difference,
Princeton, NJ: Princeton University Press.
Benhabib, S., Cornell, D. (eds.) (1987), Feminism as
Critique, Cambridge: Polity.
Benson, P. (1991), `Autonomy and Oppressive Socialisa-
tion', in Social Theory and Practice, 17, pp. 385±408.
Benton, T. (1993), Natural Relations: Ecology, Animal
Rights and Social Justice, London: Verso.
Berlin, I. (1969), Four Essays on Liberty, Oxford: Oxford
University Press.
 BIBLIOGRAPHY
Berlin, I. (1978), Concepts and Categories: Philosophical
Essays, ed. H. Hardy, London: Hogarth Press.
Berlin, I. (1991), The Crooked Timber of Humanity, ed. H.
Hardy, London: Fontana.
Berlin, I. and Williams, B. (1994), `Pluralism and Liberal-
ism: A Reply', in Political Studies, 42(2), pp. 306±9.
Bertram, C. (1997), `Political Justification, Theoretical
Complexity, and Democratic Community',
107, pp. 563±83.
Bideleux, R. (1996), `In Lieu of a Conclusion: East Meets
West', in R. Bideleux and R. Taylor (eds), European
Integration and Disintegration: East and West, London:
Routledge, pp. 281±95.
 anglaise devant l'inte-
Bideleux, R. (1999a), `L'ambiguite
gration', in A. Landuyt (ed.), Europe: Fe
 de
 ration ou
nations, Paris: Sedes, pp. 97±120.
Bideleux, R. (1999b), `Europeanization Versus Democra-
tization in East-Central Europe, Ten Years On', in
Society and Economy in Central and Eastern
(Budapest), 21(3), pp. 10±62.
Bideleux, R. and Jeffries, I. (1998), A History of Eastern
Europe, London: Routledge.
Blackburn, S. (1984), Spreading the Word, Oxford: Oxford
University Press.
Blackburn, S. (1993), Essays in Quasi-Realism, Oxford:
Oxford University Press.
Blackburn, S. (1996), `Securing the Nots: Moral Episte-
mology for the Quasi-Realist', in W. Sinnott-Armstrong
and M. Timmons (eds), Moral Knowledge?, New York:
Oxford University Press, pp. 82±100.
Blease, W. L. (1913), A Short History of English Liberalism,
London: T. Fisher Unwin.
Bobbio, N. (1996), The Age of Rights, Cambridge: Polity.
Boddy, J. (1989), Wombs and Alien Spirits: Women, Men
and the Zar Cult in Northern Sudan, Madison, WI:
University of Wisconsin Press.
Boesche, R. (1987), The Strange Liberalism of Alexis de
Tocqueville, Ithaca, NY: Cornell University Press.
Borneman, J. (1997), Settling Accounts: Violence, Justice and
Accountability in Postsocialist Europe, Princeton,
Princeton University Press.
Boucher, D. and Kelly, P. (eds) (1994), The Social Contract
from Hobbes to Rawls, London: Routledge.
Bradley, I. (1980), The Optimists: Themes and Personalities
in Victorian Liberalism, London: Faber & Faber.
Brown, C. (1992), International Relations Theory: New
Normative Approaches, London: Harvester Wheatsheaf.
Brown, C. (ed.) (1994), Political Restructuring in Europe:
Ethical Perspectives, London: Routledge.
Brownlie, I. (ed.) (1981), Basic Documents on Human
Rights, 2nd edn. Oxford: Clarendon.
Brubaker, R. (1994), Citizenship and Nationhood in France
and Germany, Cambridge, MA: Harvard University
Press.
Bullock, A. and Shock, M. (eds) (1956), The Liberal
Tradition from Fox to Keynes, London: A. & C. Black.
Bulmer, S. (1993), `The Governance of the European
Union: A New Institutionalist Approach', in Journal
of Public Policy, 13(4), pp. 351±80.
Burke, E. (1961), Reflections on the Revolution in France,
New York: Doubleday.
Cairns, A. and Williams, C. (eds) (1985), Constitutional-
ism, Citizenship and Society in Canada, Toronto: Uni-
Ethics, versity of Toronto Press.
Caney, S., George, D., Jones, P. (eds) (1996), National
Rights, International Obligations, Oxford: Westview Press.
Canovan, M. (1996), Nationhood and Political Theory,
Cheltenham: Edward Elgar.
Canovan, M. (1999), `Trust the People! Populism and the
Two Faces of Democracy', in Political Studies, 47(1), pp.
2±16.
Cassirer, E. (1951), The Philosophy of the Enlightenment,
Boston, MA: Beacon.
Claude, I. (1955), National Minorities: An International
Europe Problem, Cambridge, MA: Harvard University Press.
Cohen, G. A. (1989), `On The Currency of Egalitarian
Justice', Ethics, 99, pp. 906±44.
Cohen, G. A. (1994), `Amartya Sen's Unequal World',
New Left Review, 203, pp. 117±29.
Cohen, G. A. (1986), `Self-Ownership, World Ownership,
and Equality: Part II', in E. F. Paul, F. D. Miller Jr., J.
Paul, J. Ahrens (eds), Marxism and Liberalism, Oxford:
Blackwell, pp. 77±96.
Cohen, G. A. (1995), Self-Ownership, Freedom and Equal-
ity, Cambridge: Cambridge University Press.
Cohen, G. A. (1996), `Incentives, Inequality and Com-
munity', in P. King (ed.), Socialism and the Common
Good, London: Frank Cass.
Cohen, J. (ed.) (1996), For Love of Country, Boston, MA:
Beacon.
Cohen, J. and Rogers, J. (1983), On Democracy, New York:
Penguin.
Cohen, J. and Rogers, J. (1995), Associations and Democ-
racy, London: Verso.
Cohen. R. (1991), Negotiating Across Cultures, Washing-
NJ: ton, DC: United States Institute of Peace Press.
Collingwood, R. G. (1940), An Essay on Metaphysics,
London: Oxford University Press.
Connor, W. (1984), The National Question in Marxist-
Leninist Theory and Strategy, Princeton, NJ: Princeton,
University Press.
Constant, B. (1988), Political Writings, ed. B. Fontana,
Cambridge: Cambridge University Press.
Cooke, M. (1997), `Authenticity and Autonomy: Taylor,
Habermas and the Politics of Recognition', Political
Theory, 25(2), pp. 258±88.
Croce, B. (1946), Politics and Morals, London: Allen &
Unwin.
Croce, B. (1973), Etica e politica, Bari: Laterza.
Crocker, D. (1992), `Functioning and Capability: The
Foundations of Sen's and Nussbaum's Development
Ethic', Political Theory, 20(4), pp. 584±612.
313
 BIBLIOGRAPHY
Crowder, G. (1994), `Pluralism and Liberalism', Political
Studies, 42(2), pp. 293±305.
Cunningham, F. (1987), Democratic Theory and Socialism,
New York: Cambridge University Press.
Dagger, R. (1997), Civic Virtues: Rights, Citizenship and
Republican Liberalism, Oxford: Oxford University
Press.
D'Agostino, F. (1993), `Transcendence and Conversation:
Two Conceptions of Objectivity', American Philosophical
Quarterly, 30(2), pp. 87±108.
Dahl, R. (1985), A Preface to Economic Democracy, Cam-
bridge: Polity.
Dahl, R. (1994), `A Democratic Dilemma: System Effec-
tiveness Versus Citizen Participation', Political Science
Quarterly, 109(1), pp. 23±34.
Dahrendorf, R. (1974), `Citizenship and Beyond: The
Social Dynamics of an Idea', Social Research, 41, pp.
673±701.
Daniels, N. (1996), Justice and Justification, Cambridge:
Cambridge University Press.
Dawisha, A. and Dawisha, K. (eds), (1995), The Making of
Foreign Policy in Russia and the New States of Eurasia, New
York: M. E. Sharpe.
Derrida, J. (1989), Spectres of Marx, trans. by P. Kamuf,
London: Routledge.
Devigne, R. (1994), Recasting Conservatism: Oakeshott,
Strauss, and the Response to Postmodernism, New Haven,
CT: Yale University Press.
Devlin, P. (1965), The Enforcement of Morals, London:
Oxford University Press.
Diamond, M. and Kipnis, K. (1998), `Pediatric Ethics and
the Surgical Assignment of Sex', Journal of Clinical
Ethics, 9(4), pp. 398±410.
Dietze, G. (1985), Liberalism Proper and Proper Liberalism ,
Baltimore, MD: Johns Hopkins University Press.
Digeser, P. (1995), Our Politics, Our Selves?, Princeton, NJ:
Princeton University Press.
Dinnerstein, D. (1976), The Mermaid and the Minotaur:
Sexual Arrangements and Human Malaise, New York:
Harper Colophon.
Donnelly, J. (1989), Universal Human Rights in Theory and
Practice, Ithaca, NY: Cornell University Press.
Doppelt, G. (1980), `Statism Without Foundations', Phi-
losophy and Public Affairs, 9 (4), pp. 398±403.
Doyle, M. (1983a), `Kant, Liberal Legacies, and Foreign
Affairs: Part 1', Philosophy and Public Affairs, 12(3), pp.
205±35.
Doyle, M. (1983b), `Kant, Liberal Legacies, and Foreign
Affairs: Part 2', Philosophy and Public Affairs, 12(4), pp.
323±53.
Doyle, M. (1986), `Liberalism and World Politics', Amer-
ican Political Science Review, 80, pp. 1151±63.
Dreze, J. and Sen, A. (1989), Hunger and Public Action,
Oxford: Oxford University Press.
Dryzek, J. (1990), Discursive Democracy, Cambridge: Cam-
bridge University Press.
314
Durkheim, E. (1982), The Rules of Sociological Method,
trans. W. D. Hall, ed. by S. Lukes, London: Macmillan.
Dworkin, R. (1981a), `What is Equality? Part 1: Equality of
Welfare', Philosophy and Public Affairs, 10(3), pp. 185±246.
Dworkin, R. (1981b), `What is Equality? Part 2: Equality of
Resources', Philosophy and Public Affairs, 10(4), pp. 283±
345.
Dworkin, R. (1985), A Matter of Principle, Cambridge,
MA: Harvard University Press.
Dworkin, R. (1995), `Constitutionalism and Democracy',
European Journal of Philosophy, 3, pp. 2±11.
Dyzenhaus, D. (1997), Legality and Legitimacy, Oxford:
Oxford University Press.
Eagleton, T. (1990), The Ideology of the Aesthetic, Oxford:
Blackwell.
Eckersley, R. (1992), Environmentalism and Political Theory:
Towards an Ecocentric Approach, London: UCL Press.
Eden, R. (1982), `Doing Without Liberalism: Weber's
Regime Politics', in Political Theory, 10(3), pp. 379±407.
Elshtain, J. B. (1981), Public Man, Private Woman, Prin-
ceton, NJ: Princeton University Press.
Engell, J. (1982), The Creative Imagination: Enlightenment to
Romanticism, Cambridge, MA: Harvard University
Press.
Evans, M. (1985), Simone de Beauvoir: A Feminist Mandar-
in, London: Tavistock.
Evans, M. (1999), `Is Public Justification Central to Lib-
eralism?', Journal of Political Ideologies, 4(1), pp. 117±36.
Fausto-Sterling, A. (2000), Sexing the Body, New York:
Basic Books.
Feinberg, J. (1985), Offense to Others, New York: Oxford
University Press.
Feinberg, J. (1988), Harmless Wrongdoing, New York:
Oxford University Press.
Ferguson, M. (1985), First Feminists: British Women Writers
1578±1799, Bloomington, IN: Indiana University Press.
Ferrara, A. (1998), Reflective Authenticity: Rethinking the
Project of Modernity, London: Routledge.
Fish, M. S. (1997), `The Predicament of Russian Liberal-
ism: Evidence from the December 1995 Parliamentary
Elections', Europe-Asia Studies, 49(2), pp. 191±220.
Fitzmaurice, D. (1993), `Autonomy as a Good: Liberalism,
Autonomy and Toleration', Journal of Political Philoso-
phy, 1(1), pp. 1±16.
Flax, J. (1995), `Race, Gender and the Ethics of Differ-
ence', Political Theory, 23(3), pp. 500±10.
Forsyth, M. (1987), Reason and Revolution: The Political
Á s, Leicester: Leicester University
 Sieye
Thought of Abbe
Press.
Fox, W. (1990), Towards a Transpersonal Ecology, Boston,
MA: Shambhala.
Franco, P. (1990), `Michael Oakeshott as a Liberal The-
orist', Political Theory, 18(3), pp. 411±36.
Fraser, N, (1995), `From Redistribution to Recognition?
Dilemmas of Justice in a Post-Socialist Age', New Left
Review, 212, pp. 68±93.
 BIBLIOGRAPHY
Fraser, N. and Nicholson, L. (1988), `Social Criticism
Without Philosophy: An Encounter Between Feminism
and Postmodernism', Theory, Culture and Society, 5(2/3),
pp. 373±94.
Frazer, E. (1996), `Feminism and Liberalism', in J. Mea-
dowcroft (ed.), The Liberal Political Tradition, Chelten-
ham: Edward Elgar.
Freeden, M. (1978), The New Liberalism: An Ideology of
Social Reform, Oxford: Clarendon.
Freeden, M. (1986), Liberalism Divided: A Study in British
Political Thought 1914±1939, Oxford: Clarendon.
Freeden, M. (1996), Ideologies and Political Theory: A
Conceptual Approach, Oxford: Clarendon.
Freeden, M. (1998), `Is Nationalism A Distinct Ideology?',
Political Studies, 46 (4), pp. 748±65.
Frost, M. (1996), Ethics in International Relations: A Con-
stitutive Theory, Cambridge: Cambridge University Press.
Frost, M. (1997a), Layers of Justice: The Emerging Structure
of Global Ethics, paper presented to the University of
Wales conference on Political and International The-
ory, Gregynog.
Frost, M. (1997b), The Role of Human Rights in Under-
standing World Politics, paper presented to the University
of Manchester MANCEPT conference on Political
Thought and International Affairs.
Fukuyama, F. (1989), `The End of History?', The National
Interest, 16 (summer), pp. 3±18.
Fukuyama, F. (1992), The End of History and the Last Man,
London: Penguin
Fuss, D. (1989), Essentially Speaking: Feminism, Nature and
Difference, New York: Routledge.
Galston, W. (1991), Liberal Purposes, Cambridge: Cam-
bridge University Press.
Garton Ash, T. (1998), `Europe's Endangered Liberal
Order', in Foreign Affairs, 77(2), pp. 51±65.
Gauthier, D. (1986), Morals By Agreement, Oxford: Clar-
endon.
Gay, W. and Alekseeva, T. A. (1996), Capitalism With a
Human Face: The Quest For a Middle Road in Russian
Politics, London: Rowman & Littlefield.
Gellner, E. (1983), Nations and Nationalism, Oxford:
Blackwell.
Gellner, E. and Cansino, C. (eds) (1996), Liberalism in
 G. Merquior,
Modern Times: Essays in Honour of Jose
Budapest: Central European University Press.
Gerth, H. H., Wright Mills, C. (eds) (1991), From Max
Weber, new edn, London: Routledge.
Gifford, E. A. (1994), ` ``The Courage to Blaspheme'':
Confronting Barriers to Resisting Female Genital Mu-
tilation', UCLA Women's Law Journal, 4, pp. 329±64.
Gilligan, C. (1982), In a Different Voice, Cambridge, MA:
Harvard University Press.
Goldman, A. and Kim, J. (eds) (1978), Values and Morals,
Dordrecht: Reidel.
Gombrich, E. (1977), Art and Illusion, 5th edn, London:
Phaidon.
315
Goodin, R. (1988), `What Is So Special About Our Fellow
Countrymen?', Ethics, 98, pp. 663±86.
Goodin, R. (1995), `Political Ideals and Political Practice',
British Journal of Political Science, 25(1), pp. 37±56.
Gray, J. (1986), Hayek on Liberty, 2nd edn, Oxford: Black-
well.
Gray, J. (1989), Liberalisms, London: Routledge.
Gray, J. (1993), Post-Liberalism: Studies in Political Thought,
London: Routledge.
Gray, J. (1995a), Enlightenment's Wake, London: Routle-
dge.
Gray, J. (1995b), Berlin, London: Fontana.
Gray, J. (1997), Endgames: Questions in Late Modern
Political Thought, Cambridge: Polity.
Gray, J. (1998), `Global Utopias and Clashing Civilisa-
tions: Misunderstanding the Present', International Af-
fairs, 74(1), pp. 149±63.
Green, P. (1985), Retrieving Democracy, London:
Methuen.
Greenfeld, L. (1992), Nationalism: Five Roads to Modernity,
Cambridge, MA: Harvard University Press.
Griffin, S. (1994), `Political Philosophy versus Political
Theory: The Case of John Rawls', Chicago-Kent Law
Review, 69, pp. 691±707.
Gros, D. and Steinherr, A. (1994), `In Favour of EMU:
A Manifesto of European Economists', in A. Steinherr
(ed.), 30 Years of European Monetary Integration: From
the Werner Plan to EMU, London: Longman, pp. 74±
79.
Gunning, I. (1997), `Uneasy Alliances and Solid Sister-
hood: A Response to Obiora's ``Bridges and Barri-
cades'' ', Case Western Reserve Law Review, 47 (2), pp.
445±60.
Gurr, T. (1993), Minorities at Risk: A Global View of
Ethnopolitical Conflict, Washington, DC: Institute of
Peace Press.
Gutmann, A. (ed.) (1994), Multiculturalism and the Politics
of Recognition, 2nd edn, Princeton, NJ: Princeton Uni-
versity Press.
Gutmann, A. and Thompson, D. (1996), Democracy and
Disagreement, Cambridge, MA: Harvard University
Press.
Habermas, J. (1979), Communication and the Evolution of
Society, trans. T. McCarthy, London: Heinemann.
Habermas, J. (1992), The Structural Transformation of the
Public Sphere, trans. T. Burger and F. Lawrence, Cam-
bridge: Polity.
Habermas, J. (1993), `Struggles for Recognition in Con-
stitutional States', European Journal of Philosophy, 1(2),
pp. 128±55.
Habermas, J. (1995), `Rawls's Political Liberalism', Journal
of Philosophy, 92(3), pp. 109±31.
Habermas, J. (1996), Between Facts and Norms, trans. W.
Rehg, Cambridge, MA: MIT Press.
Habermas, J. (1998a), A Berlin Republic: Writings on Ger-
many, Cambridge: Polity.
 BIBLIOGRAPHY
Habermas, J. (1998b), The Inclusion of the Other: Studies in
Political Theory, Cambridge, MA: MIT Press.
Haddock, B. (1993), `Georg Wilhelm Friedrich Hegel:
Philosophy of Right', in M. Forsyth, M. Keens-Soper,
J. Hoffman (eds.), The Political Classics: Hamilton to Mill,
Oxford: Oxford University Press, pp. 104±36.
Haddock, B. (1996), `Michael Oakeshott: Rationalism in
Politics', in M. Forsyth and M. Keens-Soper (eds), The
Political Classics: Green to Dworkin, Oxford: Oxford
University Press, pp. 100±20.
Hallowell, J. (1946), The Decline of Liberalism as an Ideol-
ogy, London: Kegan Paul, Trench, Trubner and Co. Ltd.
Hampshire, S. (1983), Morality and Conflict,
Blackwell.
Hampshire, S. (1989), Innocence and Experience, London:
Penguin.
Hampshire, S. (1991), `Justice is Strife', Proceedings and
Addresses of the American Philosophical Association, 65(3)
(Presidential Address).
Harris, A. (1990), `Race and Essentialism in Feminist Legal
Theory,' in Stanford Law Review, 42, pp. 581±616.
Harriss, J. (ed.) (1995), The Politics of Humanitarian Inter-
vention, London: Pinter.
Harsanyi, J. (1975), `Can the Maximin Principle Serve
as a Basis for Morality? A Critique of John Rawls's
Theory', American Political Science Review,
594±606.
Hart, H. L. A. (1961), The Concept of Law, Oxford:
Clarendon.
Hart, H. L. A. (1963), Law, Liberty and Morality, London:
Oxford University Press.
Hartz, L. (1955), The Liberal Tradition in America, New
York: Harcourt, Brace & World Inc.
Haslett, D. W. (1985), `Does the Difference Principle
Really Favour the Worst Off?', Mind, 94, pp. 111±15.
Hayek, F.A. (1939), `Economic Conditions of Inter-State
Federation', New Commonwealth Quarterly (September),
pp. 131±49.
Hayek, F. A. (1960), The Constitution of Liberty, London:
Routledge & Kegan Paul.
Hayek, F. A. (1967), Studies in Philosophy, Politics and
Economics, London: Routledge & Kegan Paul.
Hayek, F. A. (1993), Law, Legislation and Liberty, 3 vols in
1, London: Routledge.
Hayek, F. A. (1976), The Road to Serfdom, London:
Routledge & Kegan Paul.
Hayes, C. J. (1949), The Historical Evolution of Modern
Nationalism, London: Macmillan.
Heater, D. (1990), Citizenship: The Civic Ideal in World
History, Politics and Education, London: Longman.
Hegel, G. W. F. (1969), Jenaer Realphilosophie, Hamburg:
Meiner.
Hegel, G. W. F. (1991), Elements of the Philosophy of Right,
ed. A. W. Wood, Cambridge: Cambridge University
Press.
Held, D. (1995), Democracy and the Global Order: From the
Modern State to Cosmopolitan Governance , Cambridge:
Polity.
Held, D. (1996), Models of Democracy, 2nd edn, Cam-
bridge: Polity.
Helgerson, R. (1992), Forms of Nationhood: The Elizabethan
Writings of England, Chicago: University of Chicago
Press.
Hennis, W. (1988), Max Weber: Essays in Reconstruction,
London: Allen & Unwin.
Herman, B. (1985), `The Practice of Moral Judgment',
Journal of Philosophy, 82 (8), pp. 414±36.
Hindess, B. (1996), Discourses of Power, Oxford: Blackwell.
Oxford: Hirschman, A. (1994), `Social Conflicts as Pillars of
Democratic Market Society', Political Theory, 22, pp.
203±18.
Hirst, P. (1994), Associative Democracy: New Forms of
Economic and Social Governance, Cambridge: Polity.
Hobhouse, L. T. (1924), Social Development: Its Nature and
Conditions, London: Allen & Unwin.
Hobhouse, L. T. (1964), Liberalism, Oxford: Oxford
University Press.
Hobson, J. A. (1909), The Crisis of Liberalism, London: P.
S. King & Son.
Hobson, J. A. (1928), `Review of G. de Ruggiero, The
History of European Liberalism', Journal of Philosophical
69, pp. Studies, 3, pp. 378±80.
Hoffman, S. (1995), `The Crisis of Liberal International-
ism', Foreign Policy, 98, pp. 159±77.
Holc, J. (1997), `Liberalism and the Construction of the
Democratic Subject in Postcommunism: The Case of
Poland', Slavic Review, 56(3), pp. 401±27.
Holmstrom, N. (1994), `Humankind(s)', in M. Matthen and
R. X. Ware (eds), Biology and Society: Reflections on Meth-
odology, Calgary: University of Calgary Press, pp. 69±105.
Honneth, A. (1995) The Struggle for Recognition: The Moral
Grammar of Social Conflicts, Cambridge: Polity.
Hont, I. (1994), `The Permanent Crisis of a Divided
Mankind: ``Contemporary Crisis of the Nation State''
in Historical Perspective', in Political Studies, 42, Special
Issue on `Contemporary Crisis of the Nation State?', pp.
166±231.
hooks, b. (1981), Ain't I A Woman?, London: Pluto.
Hough, J. (1997), Democratization and Revolution in the
USSR, 1985±91, Washington, DC: Brookings Institu-
tion Press.
Hulme, P. and Jordanova, L. (eds) (1990), The Enlight-
enment and its Shadows, London: Routledge.
Hume, D. (1966), Enquiries Concerning the Human Under-
standing and Concerning the Principles of Morals, ed. L.A.
Selby-Bigge, Oxford: Oxford University Press.
Hume, D. (1969), A Treatise of Human Nature, ed. E. C.
Mossner, Harmondsworth: Pelican.
Huntington, S. (1993), `The Clash of Civilizations?',
Foreign Affairs, 72(3), pp. 23±49.
Huntingdon, S. (1994), The Clash of Civilizations and the
Remaking of World Order, New York: Simon & Schuster.
316
 BIBLIOGRAPHY
Hurka, T. (1993), Perfectionism, Oxford: Oxford Univer-
sity Press.
Hutchinson, J. and Smith, A. D. (eds) (1994), Nationalism,
Oxford: Oxford University Press.
Ignatieff, M. (1994), Blood and Belonging, London: Vin-
tage.
Isaac, J. (1988), `Republicanism vs. Liberalism?: A Recon-
sideration', in History of Political Thought, 9, pp. 349±77.
Jacob, M. (1994), `The Mental Landscape of the Public
Sphere: A European Perspective', Eighteenth-Century
Studies, 28 (4), pp. 95±113.
Jahanbegloo, R. (1993), Conversations with Isaiah Berlin,
London: Peter Halban.
James, S. M. (1998), `Shades of Othering: Reflections on
Female Circumcision/Genital Mutilation', Signs, 23(4),
pp. 1031±48.
Jardine, A. (1985), Gynesis: Configurations of Women and
Modernity, Ithaca, NY: Cornell University Press.
Johnson, P. (1993), `Feminism and the Enlightenment',
Radical Philosophy, 63, pp. 3±12.
Kaldor, M. (1999), New and Old Wars, Cambridge: Polity.
Kamenka, E. (ed.) (1983), The Portable Karl Marx, Har-
mondsworth: Pelican.
Kant, I. (1948), Groundwork of the Metaphysic of Morals,
translated by H.J. Paton as `The Moral Law', London:
Hutchinson.
Kant, I. (1963a), Kant on History, ed. L. White Beck, New
York: Bobbs-Merrill.
Kant, I. (1963b), Lectures on Ethics, ed. L. Infield, New
York: Harper & Row.
Kant, I. (1991), Political Writings, ed. H. Reiss, Cambridge:
Cambridge University Press.
Kauffman, L. (ed.) (1989), Gender and Theory: Dialogues on
Feminist Criticism, Oxford: Blackwell.
Kedourie, E. (1993), Nationalism, 4th edn, Oxford: Black-
well.
Kekes, J. (1993), The Morality of Pluralism, Princeton, NJ:
Princeton University Press.
Kekes, J. (1996), `Cruelty and Liberalism', Ethics, 106, pp.
834±44.
Kelley, G. A. (1995), `Who Needs a Theory of Citizen-
ship?', in R. Beiner (ed.), Theorising Citizenship, New
York: State University of New York Press.
Kersting, W. (1993), Wohlgeordnete Freiheit: Immanuel
Kant Rechts- und Staatsphilosophie, Frankfurt am Main:
Suhrkamp Verlag.
Klaus, V. (1997), Renaissance: The Rebirth of Liberty in the
Heart of Europe, Washington, DC: Cato Institute.
Kouba, L. and Muasher, J. (1985), `Female Circumcision in
Africa: An Overview', African Studies Review, 28(1), pp.
95±110.
Kritz, N. J. (ed.) (1995), Transitional Justice: How Emerging
Democracies Reckon with Former Regimes , 3 vols, Wash-
ington, DC: United States Institute of Peace Press.
Kronowitz, R. (1987), `Towards Consent and Co-opera-
tion: Reconsidering the Political Status of Indian Na-
317
tions', Harvard Civil Rights ± Civil Liberties Law Review ,
22(2), pp. 507±87.
Krouse, R. and McPherson, M. (1988), `Capitalism, ``Prop-
erty-Owning Democracy'' and the Welfare State', in A.
Gutmann (ed.), Democracy and the Welfare State, Prin-
ceton NJ: Princeton, University Press, pp. 79±105.
Kymlicka, W. (1989), Liberalism, Community and Culture,
Oxford: Clarendon.
Kymlicka, W. (1990), Contemporary Political Philosophy,
Oxford: Oxford University Press.
Kymlicka, W. (1995), Multicultural Citizenship, Oxford:
Oxford University Press.
Kymlicka, W. (1996), `Concepts of Community and Social
Justice', in F. Olser Hampson and J. Reppy (eds), Earthly
Goods: Environmental Change and Social Justice, Ithaca,
NY: Cornell University Press, pp. 30±51.
Kymlicka, W. (1998), `Is Federalism an Alternative to
Secession?', in P. Lehning (ed.), Theories of Secession,
London: Routledge, pp. 111±50.
Kymlicka, W. and Shapiro, I. (eds) (1997), Ethnicity and
Group Rights, New York: New York University Press.
Ladrech, R. (1994), `Europeanisation of Domestic Politics
and Institutions: The Case of France', Journal of Com-
mon Market Studies, 32(1), pp. 69±88.
Laffan, B. (1999), `Democracy and the European Union',
in L. Cram, D. Dinan and N. Nugent (eds), Develop-
ments in the European Union, London: Macmillan, pp.
329±49.
Lamont, J. (1994), `Pareto Efficiency, Egalitarianism and
Difference Principles', Social Theory and Practice, 20, pp.
311±25.
Landes, J. (1988), Women and the Public Sphere in the Age of
the French Revolution, Ithaca, NY: Cornell University
Press.
Laponce, J. (1987), Languages and Their Territories, Tor-
onto: University of Toronto Press.
Laponce, J. (1993), `The Case for Ethnic Federalism in
Multilingual Societies: Canada's Regional Imperative',
Regional Politics and Policy, 3(1), pp. 23±43.
Laquer, T. (1990), Making Sex: Body and Gender from the
Greeks to Freud, Cambridge, MA: Harvard University
Press.
Larmore, C. (1996), The Morals of Modernity, Cambridge:
Cambridge University Press.
Laski, H. J. (1936), The Rise of European Liberalism, Lon-
don: Allen & Unwin.
Laslett, P. (ed.) (1956), Philosophy, Politics and Society,
Oxford: Blackwell.
Layne, C. (1994), `Kant or Cant: The Myth of Democratic
Peace', International Security, 19(2), pp. 5±49.
Lijphard, A. (1968), The Politics of Accommodation: Plur-
alism and Democracy in the Netherlands, Berkeley, CA:
University of California Press.
Lind, E. A. and Tyler, T. (1988), The Social Psychology of
Procedural Justice, New York: Plenum.
Linklater, A. (1992), `What is a Good International
 BIBLIOGRAPHY
Citizen?', in P. Keal (ed.), Ethics and Foreign Policy, St
Leonards, NSW: Allen & Unwin, chapter 2.
Linklater, A. (1998), The Transformation of Political Com-
munity, Cambridge: Polity.
Lloyd, G. (1984), Man of Reason, London: Methuen.
Locke, J. (1991), Two Treatises of Government, ed. P.
Laslett, Cambridge: Cambridge University Press.
Lodge, J. (1994), `Transparency and Democratic Legiti-
macy', Journal of Common Market Studies, 32(3), pp.
343±68.
Loveland, I. (1996), `Parliamentary Sovereignty and the
European Community: The Unfinished Revolution?',
Parliamentary Affairs, 49(4), pp. 517±35.
Lovibond, S. (1989), `Feminism and Postmodernism', New
Left Review, 178, pp. 5±28.
Lovibond, S. (1994), `Feminism and the `Crisis of Ration-
ality', New Left Review, 207, pp. 72±86.
Luban, D. (1980), `Just War and Human Rights', in
Philosophy and Public Affairs, 9(2), pp. 160±81.
Luhmann, N. (1981), The Differentiation of Society, New
York: Columbia University Press.
Lyotard, J. (1993), `The Other's Rights', in S. Shute and S.
Hurley (eds), On Human Rights, New York: Basic Books,
pp. 135±47.
Macedo, S. (1989), `Capitalism, Citizenship and Commu-
nity', in E. F. Paul, F. D. Miller Jr, J. Paul and J. Ahrens
(eds), Capitalism, Oxford: Blackwell.
MacIntyre, A. (1985), After Virtue, 2nd edn, London:
Duckworth.
Mackie, J. L. (1977), Ethics: Inventing Right and Wrong,
Harmondsworth: Penguin.
Majone, G. (1992), `Regulatory Federalism in the Eur-
opean Community', Government and Policy, 10, pp. 77±
101.
Malachowski, A. (ed.) (1990), Reading Rorty, Oxford:
Blackwell.
Manin, B. (1987), `On Legitimacy and Deliberation',
Political Theory, 15(3), pp. 338±68.
Mann, M. (1993), `The Nation-State in Europe and Other
Continents: Diversifying, Developing, Not Dying', Dae-
dalus, 122(3), pp. 115±40.
Mansfield, E. D. and Snyder, J. (1995), `Democratization
and the Danger of War', International Security, 20(1), pp.
5±38.
Marcuse, H. (1964), One-Dimensional Man, Boston, MA:
Beacon.
Martin, R. (1985), Rawls and Rights, Lawrence, KS: Uni-
versity Press of Kansas.
Martin, R. (1993), A System of Rights, Oxford: Clarendon.
Martin, R. and Shenoy, P. (1983), `Two Interpretations of
the Difference Principle in Rawls's Theory of Justice',
Theoria, 49, pp. 113±41.
Mason, A. (1993), Explaining Political Disagreement, Cam-
bridge: Cambridge University Press.
McDowell, J. (1998), Mind, Value and Reality, Cambridge,
MA: Harvard University Press.
318
McGarry, J. (1998), `Demographic Engineering: The
State-Directed Movements of Ethnic Groups as a Tech-
nique of Conflict Resolution', Ethnic and Racial Studies,
21(4), pp. 613±38.
McGrew, A. (1997), The Transformation of Democracy?,
Cambridge: Polity.
Mendus, S. (1990), `Time and Chance: Kantian Ethics and
Feminist Philosophy', Morrell Discussion Paper, Depart-
ment of Politics, University of York.'
Mendus, S. (1992), `Losing the Faith: Feminism and
Democracy', in J. Dunn (ed.), Democracy: The Unfin-
ished Journey, Oxford: Oxford University Press, pp. 207±
19.
Merkel, W. (1994), `Deepening and Widening: The Limits
of European Integration', in M. Szabo (ed.), The Chal-
lenge of Europeanisation in the Region: East-Central Eur-
ope, Budapest: Kirali & Co., pp. 32±60.
Meyers, D. T. (1989), Self, Society and Personal Choice,
New York: Columbia University Press.
Meyers, D. T. (1993), `Cultural Diversity: Rights, Goals
and Competing Values', in D. T. Goldberg and M.
Krausz (eds), Jewish Identity, Philadelphia: Temple Uni-
versity Press.
Meyers, D. T. (1994), Subjection and Subjectivity: Psycho-
analytic Feminism and Moral Philosophy, New York: Rout-
ledge.
Meyers, D. T. (ed.) (1997), Feminists Rethink the Self,
Boulder, CO: Westview Press.
Meyers, D. T. (2000), `Intersectional Identity and the
Authentic Self? Opposites Attract!', in C. Mackenzie
and N. Stoljar (eds), Relational Autonomy: Feminist
Perspectives on Autonomy, Agency and the Social Self,
New York: Oxford University Press.
Michelman, F. (1995), `On Regulating Practices with
Theories Drawn from Them: A Case of Justice as Fair-
ness', in I. Shapiro and J. W. DeCrew (eds), Nomos, vol.
37: Theory and Practice, pp. 309±41.
Mill, J. S. (1972), `Utilitarianism', `On Liberty' and `Con-
siderations on Representative Government', ed. H. B. Ac-
ton, London: J. M. Dent.
Miller, D. (1976), Social Justice, Oxford: Clarendon.
Miller, D. (1995), On Nationality, Oxford: Oxford Uni-
versity Press.
Minow, M. (1990), Making All The Difference: Inclusion,
Exclusion and American Law, Ithaca, NY: Cornell Uni-
versity Press.
Moellendorf, D. (1996), `Constructing the Law of Peoples',
Pacific Philosophical Quarterly, 77, pp. 132±54.
Moi, T. (1992), Simone de Beauvoir: The Making of an
Intellectual, Oxford: Blackwell.
Moody-Adams, M. (1994), `Culture, Responsibility, and
Affected Ignorance', Ethics, 104, pp. 291±309.
Mouffe, C. (1992), `Citizenship and Political Identity',
October, 61, pp. 28±32.
Mouffe, C. (1993), The Return of the Political, London:
Verso.
 BIBLIOGRAPHY
Mugo, M. G. (1997), `Elitist Anti-Circumcision Discourse
as Mutilating and Anti-Feminist', Case Western Reserve
Law Review, 47(2), pp. 461±80.
Mulhall, S. and Swift, A. (1996), Liberals and Commu-
nitarians, 2nd edn, Oxford: Blackwell.
Mu
È llerson, R. (1997), Human Rights Diplomacy, London:
Routledge.
Naess, A. (1989), Ecology, Community and Lifestyle, trans. D.
Rothenberg, Cambridge: Cambridge University Press.
Nagel, T. (1991), Equality and Partiality, Oxford: Oxford
University Press.
Narayan, U. (1997), Dislocating Cultures, New York: Routledge.
Nardin, T. (1983), Law, Morality and the Relation of States,
Princeton, NJ: Princeton University Press.
Neill, T. P. (1953), The Rise and Decline of Liberalism,
Milwaukee, WI: Bruce Publishing.
Newman, G. (1987), The Rise of English Nationalism: A
Cultural History 1740±1830, London: Weidenfeld &
Nicholson.
Nozick, R. (1974), Anarchy, State and Utopia, New York:
Oxford University Press.
Nussbaum, M. (1992), `Human Functioning and Social
Justice', Political Theory, 20(2), pp. 202±46.
Nussbaum M. (1995), Poetic Justice, New York: Beacon.
Nussbaum, M. and Glover, J. (eds) (1995), Women,
Culture and Development, Oxford: Clarendon.
Nussbaum, M. and Sen, A. (eds) (1993), The Quality of
Life, Oxford: Clarendon.
Oakeshott, M. (1962), Rationalism in Politics, London:
Methuen.
Oakeshott, M. (1975a), On Human Conduct, Oxford:
Clarendon.
Oakeshott, M. (1975b), Hobbes on Civil Association, Ox-
ford: Blackwell.
Oakeshott, M. (1996), The Politics of Faith and the Politics of
Scepticism, ed. T. Fuller, New Haven, CT and London:
Yale University Press.
Obiora, L. A. (1997), `Bridges and Barricades: Rethinking
Polemics and Intransigence in the Campaign Against
Female Circumcision', Case Western Reserve Law Review,
47(2), pp. 275±378.
O'Brien, C. C. (1988), Godland: Reflections on Religion and
Nationalism, Cambridge, MA: Harvard University Press.
O'Donnell, G. (1993), `On the State, Democratization and
Some Conceptual Problems (A Latin American View
with Glances at Some Postcommunist Countries)',
World Development, 21(8), pp. 1355±69.
O'Donnell, G. (1994), `Delegative Democracy', Journal of
Democracy, 5(1), pp. 55±69.
Ogbu, M. A. (1997), `Comment on Obiora's ``Bridges and
Barricades'' ', Case Western Reserve Law Review, 47(2),
pp. 411±22.
Okin, S. M. (1994), `Gender Inequality and Cultural
Difference', Political Theory, 22(1), pp. 5±24.
Okin, S. M. (1998), `Feminism, Women's Human Rights
and Cultural Difference', Hypatia, 13(2), pp. 32±52.
319
Oneal, J. and Ray, J. L. (1997), `New Tests of the
Democratic Peace: Controlling for Economic Interde-
pendence', Political Research Quarterly, 50, pp. 1950±
85.
O'Neill, J. (1993), Ecology, Policy and Politics: Human Well-
Being and the Natural World, London: Routledge.
O'Neill, O. (1979), `The Most Extensive Liberty', Proceed-
ings of the Aristotelian Society, 80, pp. 45±59.
O'Neill, O. (1986), Faces of Hunger, London: Allen &
Unwin.
O'Neill, O. (1988), `Abstraction, Idealization and Ideology
in Ethics', in J. D. G. Evans (ed.), Moral Philosophy and
Contemporary Problems, Cambridge: Cambridge Univer-
sity Press, pp. 55±69.
O'Neill, O. (1996), Towards Justice and Virtue, Cambridge:
Cambridge University Press.
Orton, W. J. (1945), The Liberal Tradition, New Haven,
CT: Yale University Press.
Ost, D. (1991), `The Crisis of Liberalism in Poland', Telos,
89, pp. 85±94.
O'Sullivan, N. (1989), `The New Right: The Quest for a
Civil Philosophy in Europe and America', in R. Eatwell,
N. O'Sullivan (eds), The Nature of the Right, London:
Pinter, pp. 167±91.
O'Sullivan, N. (1997), `Difference and the Concept of the
Political in Contemporary Political Philosophy', Political
Studies, 45(4), pp. 739±54.
Outram, D. (1995), The Enlightenment, Cambridge: Cam-
bridge University Press.
Paden, R. (1997), `Reconstructing Rawls's Law of Peoples',
Ethics and International Affairs, 11, pp. 217±32.
Pantin, I. (1997), `Dramatic Confrontation of Democracy
and Liberalism in Old and New Russia', Current Politics
and Economics of Russia, 7(1), pp. 23±38.
Parekh, B. (1995), `Politics of Nationhood', in K. von
Benda-Beckmann and M. Verkuyten (eds), Nationalism,
Ethnicity and Cultural Identity in Europe, Utrecht: Eur-
opean Research Centre on Migration and Ethnic Rela-
tions.
Parekh, B. (1997), `Rethinking Humanitarian Interven-
tion', International Political Science Review, 18, pp. 49±69.
Parker, P. L. (ed.) (1912), Character and Society, London:
Williams & Norgate.
Pateman, C. (1970), Participation and Democratic Theory,
New York: Cambridge University Press.
Pateman, C. (1988), `The Theoretical Subversiveness of
Feminism', in C. Pateman and E. Grosz (eds), Feminist
Challenges, Sydney: Allen & Unwin, pp. 1±10.
Peffer, R. G. (1990), Marxism, Morality and Social Justice,
Princeton, NJ: Princeton University Press.
Peffer, R. G. (1994), `Towards a More Adequate Rawlsian
Theory of Social Justice', Pacific Philosophical Quarterly,
75(3/4), pp. 251±71.
Penz, P. (1992), `Development Refugees and Distributive
Justice: Indigeneous Peoples, Land and the Developmen-
talist State', Public Affairs Quarterly, 6 (1), pp. 105±31.
 BIBLIOGRAPHY
Penz, P. (1993), `Colonization of Tribal Lands in Bangla-
desh and Indonesia: State Rationales, Rights to Land,
and Environmental Justice', in M. Howard (ed.), Asia's
Environmental Crisis, Boulder, CO: Westview Press.
Peterson, M. (ed.) (1977), The Portable Thomas Jefferson,
Harmondsworth: Pelican.
Pettit, P. (1997), Republicanism: A Theory of Freedom and
Government, Oxford: Oxford University Press.
Phillips, A. (1993), Democracy and Difference, Cambridge:
Polity.
Pocock, J. G. A. (1957), The Ancient Constitution and the
Feudal Law, Cambridge: Cambridge University Press.
Pogany, I. (1998), Righting Wrongs in Eastern Europe,
Manchester: Manchester University Press.
Pogge, T. (1994), `An Egalitarian Law of Peoples', Philo-
sophy and Public Affairs, 23(2), pp. 195±224.
Pogge, T. (1995), `How Should Human Rights Be Con-
ceived?', Jahrbuch fu
È r Recht und Ethik, vol. 3, pp. 103±20.
Primus, R. A. (1999), The American Language of Rights,
Cambridge: Cambridge University Press.
Rajchman, J. (ed.) (1995), The Identity in Question, Lon-
don: Routledge.
Ramsbotham, O. and Woodhouse, T. (eds) (1996), Hu-
manitarian Intervention in Contemporary Conflict , Cam-
bridge: Polity.
Rawls, J. (1972), A Theory of Justice, Oxford: Oxford
University Press.
Rawls, J. (1990), `Justice as Fairness: A Restatement'
(unpublished mimeo).
Rawls, J. (1996), Political Liberalism, 2nd edn, New York:
Columbia University Press.
Rawls, J. (1999), Collected Papers, ed. S. Freeman, Cam-
bridge, MA: Harvard University Press.
Raz, J. (1986), The Morality of Freedom, Oxford: Claren-
don.
 aume, D. (1995), `Justice Between Cultures: Autonomy
Re
and the Protection of Cultural Affiliation', University of
British Columbia Law Review, 29(1), p. 121.
Rees, J. C. (1971), Equality, London: Macmillan.
Rees, J. C. (1985), John Stuart Mill's `On Liberty', Oxford:
Clarendon.
Regan, T. (1988), The Case for Animal Rights, London:
Routledge.
Renteln, A. D. (1990), International Human Rights: Uni-
versalism versus Relativism, London: Sage.
Rhees, R. (1970), Discussions of Wittgenstein, London:
Routledge & Kegan Paul.
Richardson, H. (1994), Practical Reasoning About Final
Ends, Cambridge: Cambridge University Press.
Ritchie, D. G. (ed.) (1954), Natural Rights, London: Allen
& Unwin.
Robertson, C. (1996), `Grassroots in Kenya: Women,
Genital Mutilation, and Collective Action 1920±
1990', Signs, 21(3), pp. 615±42.
Roemer, J. (1985), `Equality of Talent', Economics and
Philosophy, 1, pp. 151±87.
320
Roemer, J. (1986), `Equality of Resources Implies Equality
of Welfare', Quarterly Journal of Economics, 101, pp.
751±84.
Roemer, J. (1994), A Future for Socialism, Cambridge, MA:
Harvard University Press.
Rorty, A. O. (1994), `The Hidden Politics of Cultural
Identification', Political Theory, 22(1), pp. 152±66.
Rorty, R. (1985), `Solidarity or Objectivity?', in J. Rajch-
man and C. West (eds), Post-Analytic Philosophy, New
York: Columbia University Press.
Rorty, R. (1989), Contingency, Irony and Solidarity, Cam-
bridge: Cambridge University Press.
Rorty, R. (1998), Truth and Progress: Philosophical Papers
Volume 3, Cambridge: Cambridge University Press.
Ruddick, S. (1989), Maternal Thinking: Towards a Politics of
Peace, London: Women's Press.
Ruggiero, G. de (1927), The History of European Liberalism,
London: Oxford University Press.
Russett, B. (1995), `Correspondence: ``And Yet It Moves'',
International Security, 19(4), pp. 164±75.
Rustow, D. A. (1970), `Transitions to Democracy: Toward
a Dynamic Model', Comparative Politics, 2(3), pp. 337±
63.
Ryan, A. (ed.) (1979), The Idea of Freedom, Oxford:
Oxford University Press.
Rychard, A. (1992), `Politics and Society After the Break-
through: The Sources and Threats to Political Legiti-
macy in Post-Communist Poland', in G. Sanford (ed.),
Democratisation in Poland 1988±90, London: Macmillan,
pp. 133±54.
Saage, R. (1973), Eigentum, Staat und Gesellschaft bei
Immanuel Kant, Stuttgart/Berlin: Kohlhammer.
Sachs, D. (1981), `How to Distinguish Self-Respect from
Self-Esteem', in Philosophy and Public Affairs, 10(4), pp.
346±60.
Sandel, M. (1982), Liberalism and the Limits of Justice,
Oxford: Blackwell.
Sandel, M. (1984), `The Procedural Republic and the
Unencumbered Self', Political Theory, 12(1), pp. 81±96.
Sandel, M. (1996), Democracy's Discontent: America in
Search of a Public Philosophy, Cambridge, MA: Harvard
University Press.
Sapiro, V. (1992), A Vindication of Political Virtue: The
Political Theory of Mary Wollstonecraft, Chicago: Uni-
versity of Chicago Press.
Saward, M. (ed.) (forthcoming), Innovations in Democratic
Theory, London: Routledge.
Scanlon, T. (1982), `Contractualism and Utilitarianism',
in A. K. Sen and B. Williams (eds), Utilitarianism and
Beyond, Cambridge: Cambridge University Press, pp.
103±28.
Schaar, J. H. (1981), Legitimacy in the Modern State, New
Brunswick, NJ: Transaction Books.
Scheman, N. (1993), Engenderings: Constructions of Knowl-
edge, Authority and Privilege, New York: Routledge.
Schieder, T. (1962), The State and Society in our Times:

th
Studies in the History of the 19 and 20
London: Thomas Nelson & Sons.
Schmitt, C. (1985), The Crisis of Parliamentary Democracy,
trans. E. Kennedy, Cambridge, MA: MIT Press.
Schnapper, D. (1994), La communaute
 des citoyens: sur
 e moderne de la nation, Paris: Gallimard.
l'ide
Schweickart, D. (1993), Against Capitalism, Cambridge:
Cambridge University Press.
Scott, J. (1996), Only Paradoxes to Offer, Cambridge, MA:
Harvard University Press.
Scruton, R. (1990), The Philosopher on Dover Beach: Essays,
Manchester: Carcanet.
Seglow, J. (1998), `Universals and Particulars: The Case of
Liberal Cultural Nationalism', Political Studies, 46(5),
pp. 963±77.
Seligman, A. B. (1992), The Idea of Civil Society, New
York: Free Press.
Sen, A. K. (1981), Poverty and Famines, New York: Oxford
University Press.
Sen, A. K. (1982), Choice, Welfare and Measurement,
Oxford: Blackwell.
Sen, A. K. (1984), Resources, Values and Development,
Cambridge, MA: Harvard University Press.
Sen, A. K. (1985), `Well-Being, Agency and Freedom',
Journal of Philosophy, 82(4), pp. 169±221.
Sen, A. K. (1992), Inequality Reexamined, Cambridge, MA:
Harvard University Press.
Shaheed, F. (1994), `Controlled or Autonomous: Identity
and the Experience of Network, Women Living Under
Muslim Laws', Signs, 19(4), pp. 997±1019.
Shaheed, F. (1995), `Networking for Change: The Role of
Women's Groups in Initiating Dialogue on Women's
Issues', in M. Afkhami (ed.), Faith and Freedom: Wo-
men's Human Rights in the Muslim World, Syracuse, NY:
Syracuse University Press, pp. 78±103.
Shaw, P. (1992), `Rawls, the Lexical Difference Principle
and Equality', in Philosophical Quarterly, 42, pp. 71±7.
Sher, G. (1975), `Justifying Reverse Discrimination in
Employment', in Philosophy and Public Affairs, 4(2),
pp. 159±70.
Sher, G. (1997), Beyond Neutrality, Cambridge: Cambridge
University Press.
Shklar, J. (1989), `The Liberalism of Fear', in N. Rosen-
blum (ed.), Liberalism and the Moral Life, Cambridge,
MA: Harvard University Press, pp. 21±38.
Shue, H. (1980), Basic Rights: Subsistence, Affluence and US
Foreign Policy, Princeton, NJ: Princeton University Press.
Singer, P. (1974), Democracy and Disobedience, Oxford:
Oxford University Press.
Singer, P. (1993), How Are We To Live? Ethics in an Age of
Self-Interest, London: Mandarin.
Skinner, Q. (1998), Liberty Before Liberalism, Cambridge:
Cambridge University Press.
Smith, A. (1976), An Inquiry into the Nature and Causes of
the Wealth of Nations, vol. 1, Oxford: Clarendon.
Smith, M. (1996), `The EU as an International Actor', in
BIBLIOGRAPHY
th
Centuries, J. Richardson (ed.), European Union: Power and Policy-
Making, London: Routledge, pp. 247±62.
Solzhenitsyn, A. (1974), Letter to the Soviet Leaders, Lon-
don: Index on Censorship.
Soper, K. (1990), Troubled Pleasures, London: Verso.
Soper, K. (1997), `Naked Human Nature and the Drap-
eries of Custom', in E. Yeo (ed.), Mary Wollstonecraft and
200 Years of Feminism, London: Rivers Oram Press, pp.
207±21.
de Sousa Santos, B. (1996), Towards a New Common Sense,
London: Routledge.
Spelman, E. V. (1988), Inessential Woman, Boston, MA:
Beacon Books.
Spinelli, A. (1957), `The Growth of the European Move-
ment Since World War II', in C. G. Haines (ed.),
European Integration, Baltimore, MD: Johns Hopkins
University Press, pp. 40±63.
Spivak, G. (1991), `Remembering the Limits', in P. Os-
borne (ed.), Socialism and the Limits of Liberalism, Lon-
don: Verso, pp. 227±39.
Spivak, G. (1993), Outside in the Teaching Machine, Lon-
don: Routledge.
Squires, J. (ed.) (1993), Principled Positions, London: Law-
rence & Wishart.
Staniskis, J. (1991), The Dynamics of the Breakthrough in
Eastern Europe, Berkeley, CA: University of California
Press.
Stanley, F.G. (1961), The Birth of Western Canada: A
History of the Riel Rebellions, Toronto: University of
Toronto Press.
Stockdale, M. K. (1996), Paul Miliukov and the Quest for a
Liberal Russia 1880±1918, Ithaca, NY: Cornell Univer-
sity Press.
Strauss, L. (1968), Liberalism Ancient and Modern, New
York: Basic Books.
Strawson, P. (1959), Individuals: An Essay in Descriptive
Metaphysics, London: Methuen.
Sunstein, C. (1991), `Preferences and Politics', Philosophy
and Public Affairs, 20, pp. 3±34.
Sunstein, C. (1993), The Partial Constitution, Cambridge,
MA: Harvard University Press.
Sussman, E. (1998), `Contending with Culture: An Ana-
lysis of the Female Genital Mutilation Act of 1996', in
Cornell International Law Journal, 31(1), p. 193.
Sutch, P. (2000), `Human Rights as Settled Norms: Mer-
vyn Frost and the Limits of Hegelian Human Rights',
Review of International Studies, forthcoming.
Szacki, J. (1995), Liberalism after Communism, Budapest:
Central European University Press.
Tamir, Y. (1993), Liberal Nationalism, Princeton, NJ:
Princeton University Press.
Taylor, C. (1989), Sources of the Self, Cambridge: Cam-
bridge University Press.
Taylor, C. (1995), Philosophical Arguments, Cambridge,
MA: Harvard University Press.
Temple-Lang, J. (1996), `Community Constitutional Law',
321
 BIBLIOGRAPHY
in B. Laffan (ed.), Constitution-Building in the European
Union, Dublin: Institute of European Affairs, pp. 124±
40.
Tickell, T. (1822), Poems of Thomas Tickell, The British
Poets, vol. XXVII, College House: C. Whittingham.
Tomaselli, S. (1985), `The Enlightenment Debate on
Women', History Workshop Journal, 20, pp. 101±44.
Toubia, N. (1995), Female Genital Mutilation: A Call for
Global Action, New York: Rainbo.
Touraine, A. (1997), What is Democracy?, Boulder, CO:
Westview Press.
Tsagarousianou, R., Tambini, D. and Bryan, C. (eds)
(1998), Cyberdemocracy: Technology, Cities and Civic
Networks, London: Routledge.
Tully, J. (1995), Strange Multiplicity: Constitutionalism in an
Age of Diversity, Cambridge: Cambridge University
Press.
Turpel, M. E. (1990), `Aboriginal Peoples and the Cana-
dian Charter: Interpretive Monopolies, Cultural Differ-
ences', Canadian Human Rights Yearbook, 6, pp. 3±45.
Tuveson, E. L. (1960), The Imagination as a Means of Grace,
Berkeley, CA: University of California Press.
Van Parijs, P. (1991), `Why Surfers Should Be Fed: The
Liberal Case for an Unconditional Basic Income', in
Philosophy and Public Affairs, 20(2), pp. 101±31.
Van Parijs, P. (1992), `Basic Income Capitalism', Ethics,
102, pp. 465±84.
de Varennes, F. (1996), Language, Minorities and Human
Rights, Dordrecht: Kluwer.
Vincent, A. (1990), `The New Liberalism in Britain 1880±
1914', Australian Journal of Politics and History, 36(3), pp.
388±405.
Waldron, J. (ed.) (1984), Theories of Rights, Oxford: Ox-
ford University Press.
Waldron, J. (1993), Liberal Rights, Cambridge: Cambridge
University Press.
Waldron, J. (1999), The Dignity of Legislation, Cambridge:
Cambridge University Press.
Walicki, A. (1991), `From Stalinism to Post-Communist
Pluralism: The Case of Poland', New Left Review, 185,
pp. 92±121.
Walker, M. U. (1998), Moral Understandings: A Feminist
Study in Ethics, New York: Routledge.
Walley, C. (1997), `Searching for ``Voices'': Feminism,
Anthropology, and the Global Debate over Female
Genital Operations', Cultural Anthropology, 12(3), pp.
405±38.
Walzer, M. (1977), Just and Unjust Wars, New York: Basic
Books.
Walzer, M. (1983), Spheres of Justice, Oxford: Blackwell.
Walzer, M. (1990), `Nation and Universe', in G. Peterson
(ed.), The Tanner Lectures on Human Values, vol. XI, ed.
G. B. Petersen, Salt Lake City, UT: University of Utah
Press, pp. 509±56.
Walzer, M. (1994), Thick and Thin: Moral Arguments at
322
Home and Abroad, Notre Dame, IN: University of Notre
Dame Press.
Ward, I. (1997), `Law and the Other Europeans', Journal of
Common Market Studies, 35(1), pp. 79±96.
Warnke, G. (1995), `Discourse and Feminist Dilemmas of
Difference', in J. Meehan (ed.), Feminists Read Haber-
mas: Gendering the Subject of Discourse , London: Rout-
ledge, pp. 247±61.
Weber, M. (1994), Political Writings, eds P. Lassman and R.
Speirs, Cambridge: Cambridge University Press.
Welch, S. (1993), The Concept of Political Culture, London:
Macmillan.
Wenar, L. (1995), `Political Liberalism: An Internal Cri-
tique', Ethics 106, pp. 32±62.
Wiggins, D. (1988), `Truth, Invention and the Meaning of
Life', in G. Sayre-McCord (ed.), Essays on Moral Rea-
lism, Ithaca, NY: Cornell University Press, pp. 127±65.
Williams, A. (1995), `The Revisionist Difference Princi-
ple', Canadian Journal of Philosophy, 25, pp. 257±82.
Williams, H. (1983), Kant's Political Philosophy, Oxford:
Blackwell.
Winch, P. (1958), The Idea of a Social Science, London:
Routledge & Kegan Paul.
Winch, P. (1972), `Nature and Convention', in his Ethics
and Action, London: Routledge & Kegan Paul, pp. 50±
72.
Wissenberg, M. (1998), Green Liberalism: The Free and
Green Society, London: UCL Press.
Wittgenstein, L. (1968), Philosophical Investigations, trans.
G. E. M. Anscombe, Oxford: Blackwell.
Wittgenstein, L. (1969), On Certainty, eds G. E. M.
Anscombe and G. H. von Wright, Oxford: Blackwell.
Wolff, J. (1996), An Introduction to Political Philosophy,
Oxford: Oxford University Press.
Wollstonecraft, M. (1989), The Works of Mary Wollstone-
craft, vol. 5, 2nd edn, eds J. Todd and M. Butler, New
York: New York University Press.
Yolton, J., Rogers, P., Porter, R. and Stafford, B. (eds)
(1991), The Blackwell Companion to the Enlightenment,
Oxford: Blackwell.
Young, I. M. (1989), `Polity and Group Difference: A
Critique of the Ideal of Universal Citizenship', Ethics,
99, pp. 251±74.
Young, I. M. (1990), Justice and the Politics of Difference,
Princeton, NJ: Princeton University Press.
Young, I. M. (1995), `Gender as Sociality: Thinking About
Women as a Serial Collective', in L. Nicholson and S.
Seidman (eds), Social Postmodernism: Beyond Identity
Politics, Cambridge: Cambridge University Press, pp.
187±215.
Zakaria, F. (1997), `The Rise of Illiberal Democracy',
Foreign Affairs, 76 (6), pp. 22±43.
Zolo, D. (1992), Democracy and Complexity, Cambridge:
Polity.
 INDEX

abstraction, 76±7, 80, 199, 201, 256 Baumeister, A., 11±12

Ackerman, B., 54, 61, 128 Baxter, B., 13, 16
Addison, J., 204 Beitz, C., 53
A' Haleem, A. M., 216 Bellamy, R., 12, 272, 274
Albertini, M., 226, 230 Benson, P., 213, 219, 221n34
Alekseeva, T. A., 283 Bentham, J., 167
Alexander, W., 203 Benton, T., 298n25
alienation, 212, 291 Berlin, I., 57, 69, 137±42, 145, 146n3, 147n19,
Anderson, P., 140, 142 161n43, 163, 170n11
anti-essentialism, 75, 78±82, 200 Bideleux, R., 13
anti-foundationalism, 147±59, 162±3 Bielecki, K., 274±5
anti-perfectionism, 6, 8, 299±301, 303, 305, 310 Blackburn, S., 152±3
see also impartiality; neutrality Brown, C., 254
Appiah, K. A., 97 Brubaker, R., 59±60
Aristotelianism, 56, 188, 192, 193n1, 297n3 Buchanan, A., 129
Aristotle, 56, 141, 182 bureaucracy, 177, 182±3, 284
Arneson, R., 124, 129 Burke, E., 201±2, 204
Aron, R., 276 Buzgalin, A., 272
Augustine, Saint, 52
Austin, J., 167 Callinicos, A., 129
authenticity, 95±8, 213±14, 220 Canovan, M., 11
authority, 154±6, 160n8, 167±8, 185 capitalism, 8, 17±18, 114±15, 117±18, 120±1,
autonomy, 13, 24, 28±9, 31, 32n42, 32n51, 54, 84, 125±9, 131n40, 134n43, 229±31, 236, 272±5,
90±1, 116, 162, 180, 185, 189±90, 192, 198± 280, 283±4, 296
9, 200, 204, 208±20, 221n50, 222n61, 233, Carens, J., 44
261, 276±7, 289±90, 297n14, 299, 305±6, 308 Carter, E., 204
Axtmann, R., 281 Cassirer, E., 199
categorical imperative, 83
Babbit, S., 213, 219±20 Chernomyrdin, V., 272±3
Bader, V., 61 Chubais, A., 272±3
Balcerowicz, L., 272±4 Cicero, 67
Barber, B., 71 citizenship, 10±11, 37±8, 41, 51±61, 61n3, 75±84,
Barry, B., 128, 297n1 197±8, 202, 227, 231, 283±4, 305, 306

323
 INDEX

civil association, 13, 63±4, 167, 225±39 186n16, 189±90, 197±8, 200±1, 205, 225±39,
civil society, 8, 25, 37, 162±3, 177, 192, 236, 247, 243±5, 263, 269±70, 272±3, 275±7, 280±1,
282±3 300, 303
Clinton, B., 307 deontology, 188
Cohen, G. A., 8, 120±2, 124, 129 Derrida, J., 200
Cohen, J., 129 Dewey, J., 54, 300
Cohen-Tanugi, L., 231 dialogue, 98±9, 155±6
communism, 49n21, 66±7, 113, 125±8, 238, Diderot, D., 202±3
252n31, 269±84 Dietze, G., 279
communitarianism, 9±10, 16, 22, 30, 94, 100n5, difference, ethics/politics of, 11, 46, 75, 77, 79,
116±17, 182, 253±64, 281, 297n20 80±1, 115±16, 205, 206n26
Condorcet, M. J. A. N., 202 difference principle, 12, 101±10, 110n3, 115±16,
consensus, 10, 28, 67±8, 71±2, 95, 98, 156, 162± 118±23, 130n14
3, 169, 176±7, 179, 184, 189±90, 214, 231, Digeser, P., 299±300
258, 261±2, 265n45, 265n48 Dinnerstein, D., 78
conservatism, 32n38, 66, 68, 73n22, 94, 297n19, Doyle, M., 241±51
308 Dryzek, J., 183
Constant, B., 56±7, 65, 188±9 Dworkin, R., 5, 25, 31, 54, 90, 128
constitutionalism, 49n31, 57, 179, 182±5, 188±92, Dzarasov, S., 280
225, 237, 262, 276±7
see also constitution(s) Eckersley, R., 289±90, 292, 297n7
constitution(s), 44±5, 168, 175, 177, 180±1, 185, ecocentrism, 13, 288±93, 295±6
247, 262 ecologism (`Green' theory), 13, 16, 182±3, 287±96
see also constitutionalism education, 38, 40±1, 53, 76, 89, 96, 180, 215,
constructivism 217±19, 221n36, 305±7, 311n19
as meta-ethical theory, 103±4, 107, 141, 188± egalitarianism, 101±9, 112n30, 113±16, 119±20,
92, 253±7, 260±1, 264 128±9
as form of political doctrine, 273±4 see also equality

contractarianism, 9, 29, 104 Eisenberg, A., 40

cosmopolitanism, 11, 51±3, 55, 60±1, 61±2n3, 71, Elshtain, J. B., 78
226, 229, 232, 239, 244, 248, 253±7, 260±2, Elster, J., 129
264n3 Engels, F., 15
counter-Enlightenment, 12±13, 140, 200 Enlightenment, 6, 10±12, 76, 197±205, 205n1,
Croce, B., 25±6, 29, 280 205±6n1, 206n14, 263, 291, 293
Cudd, A., 110 equality, 6, 30, 35, 68, 71, 75±7, 80±1, 83±4, 94,
culture, 8, 18±19, 35±47, 47n5, 49n24, 93±9, 138, 101±10, 111n9, 115, 117±18, 121, 123, 183±
162, 165±6, 183, 185, 199, 203, 208±20, 4, 198, 248±9, 280
293±6, 306 see also egalitarianism

essentialism, 78±9, 200, 205

Dahl, R., 230 ethnocentrism, 78, 152, 157±8, 205
Dahrendorf, R., 57, 269 eurocentrism, 35, 47, 49n33
Daniels, N., 128 European Union, 13, 225±39, 240n15
de Beauvoir, S., 200 Evans, M., 12, 129, 146n4
deconstruction, 200 exploitation, 123±5
de la Barre, P., 202
democracy, 7, 15, 16±17, 39±40, 42, 67, 82, 99, Fedorov, B., 272
115±18, 126±7, 136±7, 143±4, 175±85, Feinberg, J., 306

324
 INDEX

feminism, 11±12, 75±84, 197±205, 289 Hallowell, J., 27±9

Ferrara, A., 100n15 Hampshire, S., 137±8, 141
Fish, S., 271±2, 279 Harris, A., 75, 78±9, 111n8
Flax, J., 75, 78 Hart, H. L. A., 162±3, 166±8
foundationalism, 27, 29, 148±51, 153, 158±9, Hartz, L., 26, 28
160n10, 162±3, 167, 170, 198, 255, 260 Haslett, D. W., 112
Fraser, N., 99, 201 Hausner, J., 274
fraternity, 120, 130n25 Hayek, F., 9, 65, 72n4, 73n22, 73n27, 137, 183,
Freeden, M., 6, 11 225, 231±2, 234±5, 237±8, 239n1, 270, 274,
freedom, 6, 7, 23, 27, 35, 72, 76, 89, 93, 97, 145, 279, 285n36
184, 198, 247±9, 269±70, 287 Hayes, C., 282
see also liberty/liberties Hegel, G. W. F., 26, 125, 165, 242±3, 248, 292,
French Revolution, 70, 201, 279 293n19
Freud, S., 210 Held, D., 17
Friedman, M., 270 Herder, A., 95
Frost, M., 254, 258±9, 263±4 Herman, B., 83
Fukuyama, F., 6, 14, 18, 242±4, 246, 248±51, 269, Herzen, A., 283
280 Hiley, C., 149
fundamentalism (religious), 15, 161n36, 200, 257 history, theory(ies) of, 14±15, 22±3, 25±7, 242±3,
249±50, 269
Gaidar, Y., 271±3 Hobbes, T., 164, 166±7, 169, 283, 292, 297n19
Galston, W., 299 Hobhouse, L. T., 22±7, 29±30, 178, 300
Garton Ash, T., 239 Hobson, J. A., 23, 28, 30
Gay, W., 283 Hoffmann, S., 282
Gellner, E., 269 Holc, J., 275
genital cutting, 208±20, 221n50 Holmstrom, N., 129
Geras, N., 129 Honneth, A., 91±3, 97
Gilligan, C., 85n15 Hont, I., 70
globalisation, 16±17, 186n16, 229, 231, 238, 279 hooks, b., 75, 78
Goodin, R., 73n46, 179 Hough, J., 270
Gorbachev, M., 270 human rights, 7, 35±47, 49n24, 49n33, 51±3, 57±
Gortat, R., 273 9, 63, 65, 69±70, 148±59, 160n23, 231, 253,
Gottlieb, R., 129 258±62, 264
Gray, J., 3, 4, 8, 13, 61, 66±7, 69, 73n39, 162±3, see also group rights; minority rights; rights

170n11, 227, 232, 274, 285n36 humanitarian intervention, 70±1, 149, 156±7,
Green, T. H., 57, 280, 300 161n39
group representation, 10, 75, 81±2, 181, 185 Hume, D., 164, 166, 204, 243, 306
group rights, 11, 34±47, 49n24, 49n33, 55, 238 Humeanism, 156
see also human rights; minority rights; rights Huntington, S., 278
Guignon, D., 149 Hutcheson, F., 204
Gutmann, A., 128
identity, 8, 66, 75, 81±4, 89±99, 200, 205, 231,
Habermas, J., 53, 76, 82±4, 85n50, 135±6, 185, 291, 302
188±92, 201, 206n22 Ignatieff, M., 69
Haddock, B., 12, 73n35, 240n47 impartiality, 10, 24, 26, 68, 76±7, 80, 82±3, 90,
Hague, W., 228 184, 288, 297n2
Hailsham, Lord, 237 see also anti-perfectionism; neutrality

325
 INDEX

imperfect duty/obligation, 76, 82±3 law/legal system(s), 43, 156, 164, 167±8, 176±81,
individualism, 7, 10, 35, 38, 52, 184, 255, 281, 185, 186n41, 190±2, 193n12, 225±6, 231±9,
282±3 245, 248, 260±2, 303
individuality, 8, 16, 18±19 Layne, C., 241
Islam, 210, 215, 281 Lebed, A., 273
legitimation crisis, 17, 274±5
Jacob, M., 201 Lenin, V. I., 69
Joad, C. E. M., 27 Levine, A., 129
Johnson, P., 200 libertarianism, 7, 9
judicial review, 177, 181 liberty/liberties, 23, 26, 57, 67, 71, 101, 115±17,
see also Supreme Court 119, 175±8, 186n37, 188±91, 225±6, 228,
justice, 7±9, 16, 22, 24, 29, 30±1, 35±47, 47n5, 236±7, 263±4, 280, 282±3
76, 80, 82, 90±9, 101±9, 110n1, 110n2, 110± see also freedom

11n6, 113±31, 135±8, 141, 143, 179±82, 185, Linklater, A., 53

188±9, 233, 253±4, 257, 262, 264, 280, 301, Lloyd, S., 129
311n10 Locke, J., 29±30, 32n54, 191, 243
justification, 11±12, 29, 94±5, 143, 154, 156, 158, Loveland, I., 231
164, 189, 262 Luzhkov, Y., 278
Lyotard, J.-F., 165
Kagarlitskii, B., 276
Kaldor, M., 236, 240n54 McBride, W., 129
Kant, I., 51, 54, 61n2, 75, 83±4, 135, 146, 198, Macaulay, C., 76, 203
200, 203±4, 207n34, 241±51, 252n16, 261, MacDonald, J. A., 38
290, 292, 297n13 MacDonald, M., 151
see also Kantianism McDowell, J., 160n13
Kantianism, 4, 6, 28, 53, 75, 82, 83±4, 139, 156, Machiavelli, N., 138, 182, 242
158, 163, 170, 203, 252n31, 253±4, 260±1, MacIntyre, A., 260
263, 289±90, 297n14, 297n19, 297n21 Mackie, J. L., 160n16, 256±7
see also Kant, I. Mann, M., 228
Karimov, I., 278 market(s), 57, 120, 128±9, 175±7, 182±3, 229,
Kekes, J., 163 236±7, 243, 250, 269, 272±5, 282±2
Keynes, J. M., 274 Marshall, T. H., 58
Kirienko, S., 273 Martin, R., 12, 110
Klaus, V., 272 Marx, K., 15, 292, 297n19
Kolodko, G., 274 Marxism, 12, 26, 69, 113±32, 280, 297n3
Kozyrev, A., 273 maximin reasoning, 103, 106±9, 111n23, 119
Kristol, I., 280 Mazowiecki, T., 273, 275
Kymlicka, W., 10±11, 55, 95, 99n2, 106, 129, 281 Mazzini, G., 281
Mendelssohn, M., 200
Lamont, J., 110, 111n11 Mendus, S., 77
Lamont, N., 228 Merquior, J., 269, 277, 283
Landes, J., 199 Meyers, D. T., 13
language, 38, 40±1, 48n20, 93, 96, 165 Michelet, J., 281
Laquer, T., 202 Mill, J. S., 30, 35, 55, 69, 93, 163, 197, 204, 245,
Larmore, C., 137, 141±2, 144, 191 278±9, 281, 296, 300, 302, 311n15
Laski, H., 24, 32n60 Millar, J., 203
Lassman, P., 12 Miller, D., 100n30, 129, 263

326
 INDEX

Milyukov, P., 281 Pettit, P., 178

minimum morality, 140±1, 163, 167, 260, 262±3 Phillips, A., 80
minority rights, 35±47, 48n6, 48n20, 99n2 Plato, 18
see also group rights; human rights; rights pluralism, 8, 12, 26, 82, 93, 99, 135, 136±44, 162,
Mises, L. von, 270 170n11, 175±85, 189±90, 200, 257
Mohammad, M., 278 `reasonable', 138±9, 141, 144, 162, 261, 263
Montesquieu, C., 65 Pogge, T., 52±3, 61n3, 129, 255
Moody-Adams, M., 217, 219 political liberalism, 10, 29, 30, 31, 32n51, 135±8,
Mouffe, C., 156 141, 143±4, 162±3, 188±92
multiculturalism, 10±11, 26, 35±47, 53, 177, political philosophy/theory, 3±6, 13, 19, 19n1, 22,
186n16, 231 116, 135±6, 139, 149±50, 159, 162, 169, 241,
255±6, 259
Naess, A., 290 politics, conception(s) of, 30, 57, 63±4, 69, 75,
Nagel, T., 5, 128, 288, 297n2, 297n3 139, 142, 156, 164
Nardin, T., 233±4, 239n1 Popper, K., 29
nationalism, 11, 15, 24, 28, 51±61, 63±72, 142, post-communism, 13, 269±84
280±4 postmodernism, 130n44, 198±201, 205
nationality, 11, 35±47, 51±61, 62n20, 63±72, 176 practical reason, 12, 170, 255±6, 260±1
Nemtsov, B., 272±3 see also public reason; rationality; reason

neutrality, 6, 31, 287, 299 pragmatism, 12, 27, 148±59

see also anti-perfectionism; impartiality praxis, 291±2, 294
Nielsen, K., 129 progress, 13±14, 23±7, 29±30, 76, 162, 175, 249±
Nietzsche, F., 156 50
non-cognitivism, 150±3, 156, 159, 160n16, 160± progressivism, 6, 18, 32n38
1n29, 161n34 projectivism, 152±3, 158
Nozick, R., 9, 130n24, 137, 270 Prokhanov, A., 282
Nussbaum, M., 53, 75, 80, 85n14, 128, 161n32 public/private distinction, 28, 75, 78, 97, 181,
188±92, 198, 225, 302±3
Oakeshott, M., 63±7, 72, 73n22, 167, 169, 225, public reason, 6, 28, 30, 263
232±4, 238, 239n1 see also practical reason; rationality; reason

objectivity, moral, 150±1, 158

O'Donnell, G., 277 Rasmussen, D., 12
Okin, S. M., 75, 80, 129 rational choice theory, 111n18, 212±13, 221n20
O'Neill, O., 83, 254±8, 260±1 rationality, 11, 28, 146, 162, 204, 290
Orban, V., 271 see also practical reason; public reason; reason

original position, 104±5, 109, 189, 295 Rawls, J., 3, 5±6, 8±10, 12±13, 20n18, 23, 25, 28±
Ost, D., 270, 283, 285n35 31, 36, 54, 90, 101±10, 110n1, 110n2,
Outram, D., 200, 206n26 110n4, 110±11n6, 111n8, 111n9, 111n18,
111n20, 111n23, 112n24, 112n26, 110n29,
Parekh, B., 68, 73n49 112n30, 113±29, 130n14, 135±9, 141±6,
Pareto optimality, 107, 119 147n51, 156, 160n8, 161n36, 162±3, 170,
Pateman, C., 75±7 188±92, 253±4, 258±64, 296, 299, 311n10
Peffer, R. G., 12 Raz, J., 6, 184, 299±300, 306, 311n4, 311n5
perfect duty/obligation, 76, 82±3 realism, moral, 150, 152±3, 160n13
perfectionism, 13, 16, 64, 299±311, 311n5 reason, 12, 24, 26±7, 76, 199±200, 203±4, 206n14
personcentrism, 13, 287±8, 292±3, 295±6 see also practical reason; public reason;

personhood, 287±8, 292±4, 297n19, 297±9n22 rationality

327
 INDEX

 aume, D., 47
Re Shklar, J., 12, 65, 72n16, 135, 144±6, 310
recognition, 12, 89±99, 100n2, 100n3, 180, 205, Shlapentokh, D., 282
242, 248 Shue, H., 128
Redwood, J., 228 Ás, Abbe
Sieye Â, 56
Rees, J. C., 35 Simonia, N., 270
Regan, T., 297n14 Simpson, O. J., 68
Reiman, J., 129 Skinner, Q., 178
relativism, 9, 116±17, 152, 171, 202, 205 Smith, A., 185n5, 204
republicanism, 12, 18, 175±85, 186n7, 188±92, Smith, M., 228
244±6, 249 socialisation, 91, 151, 213±14, 221n34, 302, 305
rights, 7, 92, 94, 115±17, 123, 126, 165, 175±7, socialism, 7, 16, 18, 113±15, 117±18, 122, 124±9,
181, 185, 189±91, 197, 216, 236, 275, 278, 131n46, 198, 279±80, 297n19
283, 297n25 Sogrin, V., 272±3, 280, 284
see also group rights; human rights; minority Solzhenitsyn, A., 271
rights Soper, K., 11±12
Rights of Man, 1789 Declaration, 59, 64±5, 70 Spelman, E. V., 75, 78±9
Roemer, J., 124±5, 129 Spinelli, A., 226, 230
romanticism, 18±19, 24, 140, 204 Spivak, G., 200
Roosevelt, F. D., 191 Staniskis, J., 275
Rorty, R., 7, 12, 142, 148±59, 160n4, 160n10, state, 68±70, 227, 231, 247±8
160n13, 160n23, 162±3 Stoicism, 52±3
Rousseau, J.-J., 67, 203±4, 292, 297n19, 297n20 Strauss, L., 30, 142
Ruddick, S., 78 Strawson, P., 297±8n22
Ruggiero, G. de, 23±9, 54 Supreme Court, 28, 43±4, 45
Rustow, D., 282 see also judicial review

Rutland, P., 273 Sutch, P., 13

Rychard, A., 274±5 Szacki, J., 270, 272±3, 283±4

Sakharov, A., 271 Tarasov, A., 282

Sakwa, R., 13 Taylor, C., 20n27, 23, 54±5, 91, 95±8, 100n3,
Sandel, M., 9 100n15, 100n19, 100n21, 100n23, 100n30
Sapiro, V., 198 Taylor, H., 197
Scanlon, T., 5, 128 teleology, 188±90, 205
Schaar, J., 70±1 Temple-Lang, J., 231
Scheffler, S., 128 Thatcher, M., 228
Schmitt, C., 137, 140, 143, 161n42 Tickell, T., 70
Schmitter, P., 275 Timofeev, L., 270
Schumpeter, J., 57, 242 Tocqueville, A. de, 18, 65, 245, 279
Schweickart, D., 121, 129 toleration, 145, 154±5
Scott, J., 202 Tomaselli, S., 203
Seglow, J., 11±12 Tully, J., 176
self-esteem, 91±4, 100n9 Tuveson, E. L., 204
self-respect, 92, 100n2, 100n9, 115, 119
Sen, A. K., 9, 124, 128 United Nations, 36±8, 53±4, 65, 68, 80, 156
Shaftesbury, Third Earl of, 204 Universal Declaration of Human Rights, 35, 37,
Shenoy, P., 110 65, 149, 278
Sher, G., 299, 308 universalisability, 254±7

328
 INDEX

universalism, 52, 59, 61, 71, 75±84, 140, 199±200, Wenar, L., 162±3
239, 253±7, 260±1, 263, 278 Wiggins, D., 153
utilitarianism, 55, 106, 112n26, 163, 193n1 Williams, A., 106, 110
Williams, B., 137, 141
Van Parijs, P., 128±9 Williams, H., 13
Vico, G., 280 Winch, P., 164±5, 167
Vincent, A., 11 Wissenberg, M., 297n1
Voltaire, 204 Wittgenstein, L., 164±5, 167
Wolff, J., 106
Waldron, J., 128 Wolff, R. P., 129
Walicki, A., 274 Wollstonecraft, M., 76, 197±9, 201±2, 204±5,
Walker, A., 213 206n6, 206n8
Walker, G., 49n31
Walley, K., 221n36 Yavlinskii, G., 274
Walzer, M., 9, 54, 116, 129, 136, 254, 258±64, Yeltsin, B., 271±3
297n18 Young, I. M., 10, 55, 75±7, 80±3, 85n23, 85n42,
Ward, I., 236 100n11
Warnke, G., 85n50
Weber, M., 17, 135±6, 138, 141±5, 146n4, Zakaria, F., 276±7, 279
147n19, 147n27, 147n38, 182 Zhirinovskii, V., 271
Weimar `Problem', 135, 137±8, 143±4, 146 Zyuganov, G., 284
Welch, S., 281

329