Critical Policy Studies

ISSN: 1946-0171 (Print) 1946-018X (Online) Journal homepage: https://www.tandfonline.com/loi/rcps20

Governance-driven democratization

Mark E. Warren

To cite this article: Mark E. Warren (2009) Governance-driven democratization, Critical Policy
Studies, 3:1, 3-13, DOI: 10.1080/19460170903158040
To link to this article: https://doi.org/10.1080/19460170903158040

Published online: 30 Nov 2009.

Submit your article to this journal

Article views: 3774

View related articles

Citing articles: 63 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rcps20
Critical Policy Studies
Vol. 3, No. 1, April 2009, 3–13

ARTICLES
1946-018X
1946-0171
RCPS
Critical Policy Studies
Studies, Vol. 3, No. 1, Aug 2009: pp. 0–0

Governance-driven democratization
Mark E. Warren*
Critical
M.E. Warren
Policy Studies

University of British Columbia, Department of Political Science, C425-1866 Main Mall, Vancouver,
BC, V6T 1Z1, Canada

While democratic reforms of electoral institutions continue, their capacities for deepening
democracy are limited by the large, pluralized, complex, and territorial political units
through which they work. Much of the institutional innovation in democracy has
shifted elsewhere – into ‘governance’, the development and administration of public
policy. The democratic potentials of governance reside in the potentially responsive
linkages between what governments do and what citizens receive. From the perspective
of democratic theory, however, this ‘governance driven-democratization’ is a frontier.
Concepts inherited from participatory democratic theory such as citizen engagement
and direct democracy obscure more than they reveal. Most of the new experiments
engage a relatively few citizens. Many involve only self-selected stakeholders and
activists, and so by-pass broader public interests, or generate new forms of exclusion.
In this paper, I specify the concept and domains of governance-driven democratization,
and suggest a method for critical assessment. We can ask, on a case by case basis,
about opportunities and dangers as measured by the democratic values of inclusion of
the affected, empowerment, representation, and deliberation.
Keywords: democracy; democratization; participation; governance

We are witnessing a new and important development in the history of democracy: policy
and administration are moving into the front lines of the project of democratization. On
the face of it, this development comes as something of a surprise to those who have
viewed democratization as the mission of class agents and social movements, or as a mat-
ter of establishing and reforming electoral processes and the rule of law. Who would have
thought that policy and policy-making – the domain of technocrats and administrators –
would move into the vanguard of democratization? And yet it is in this domain – not in
electoral democracy – that we are seeing a rebirth of strongly democratic ideals, including
empowered participation, focused deliberation, and attentiveness to those affected by
decisions. I shall refer to these developments as governance-driven democratization.
I hasten add that I am not suggesting that governance-driven democratization is replac-
ing other forms and spaces of democracy – such as electoral democracy, public sphere dis-
course, social movement and advocacy democracy, and so on. To the contrary, it is likely
that governance-driven democratization is a response to democratization in these other
areas. Innovations in electoral democracy continue to develop – though slowly, no doubt
because most potentials of this form have already been achieved in the developed democra-
cies, and it is now overtaxed by complex societies, globalizing, marketizing societies, as

*Email: warren@politics.ubc.ca

ISSN 1946-0171 print/ISSN 1946-018X online

© 2009 Institute of Local Government Studies, University of Birmingham
DOI: 10.1080/19460170903158040
http://www.informaworld.com
4 M.E. Warren

well as new social capacities for advocacy and influence over the past several decades (Cain
et al. 2003). From the perspective of governments, these new capacities can either produce
political gridlock, or they can be harnessed through new kinds of institutions and mecha-
nisms which have been identified with the umbrella term ‘governance’. I should also say
that although these developments are growing most rapidly within electoral democracies,
they are probably driven by broad social and economic developments that are quite inde-
pendent of electoral democracy as we know it. In China, for example, experiments in
governance – many of them genuinely democratic, most not – are developing hand over
fist, despite the authoritarian rule of the Communist Party (He 2006). The likely reason is
simply functional: in China as elsewhere, market-driven development and complexity gen-
erate social capacities for opposition and cooperation in ways that are similar to those in the
developed democracies, unleashing potentials for democratization that are quite autono-
mous from the regime-level democratization with which we are more familiar.
Let me also say by way of introducing the concept of governance-driven democratization
that I refer to this phenomenon as ‘democratization’ not because everything that happens
within this domain is ‘democratic’ – but rather that it is a domain that holds out the promise
of an important and deep trajectory of democratic development. The term is teleological,
hopeful, contingent, and critical rather than empirically descriptive or logically necessary.
The term names a domain of political experiments that may have democratic potentials, by
which I shall mean, simply, that they increase the chances that those potentially affected
by collective decisions can influence those decisions.
My first task, then, is simply to locate governance-driven democratization and to specify
its characteristics. Then I shall offer a couple of thoughts as to why this domain has
become so dynamic in the last couple of decades, followed by a list of democratic oppor-
tunities as well as some of the challenges and dangers. I will conclude with a comment
about critical research agendas.
Let me begin, then, simply by locating governance-driven democratization in what
might be called the space of democratic possibilities within today’s complex, differenti-
ated societies (Warren 2003). Governance-driven democratization names only one kind of
political space among many, and its opportunities and challenges have much to do with
the space it occupies (see Table 1). In one dimension, we can think of politics (and thus
possible sites of democracy) as being located primarily within the state, within society, or
within the economy. The significance of these distinctions is that they capture the fact that
developed societies make collective decisions and organize collective actions in three
basic ways: (1) through administrative power organized through the state; (2) through
norms expressed and organized through the associations of civil society; and (3) through
money organized by markets (Habermas 1987). As societies differentiate along these
lines, so we should expect politics to differentiate as well. This same dimension should
reflect the fact that liberal-democratic constitutional states separate electoral institutions
such as legislatures from non-electoral institutions such as executive agencies and courts,
leading to distinctive patterns of politics within each. Likewise, with the increase in refer-
endums and other forms of direct decision-making, we should distinguish between repre-
sentative and direct democracy.
In a second dimension, we can distinguish between institutionalized politics, such as
occurs through elections and political parties, and non-institutionalized politics of the sort
exemplified by new social movements and civil society organizations. The significance of
this distinction is that the success of democracy depends both on the existence of protected
space for non-institutionalized forms of politics, such as protest and public debate, as well
as political institutions that are sensitive enough to the ‘output’ of non-institutionalized
 Critical Policy Studies 5

Table 1. Domains of democracy.

Institutionalized
Domain focus Centred Devolved, reflexive Non-institutionalized
State: Parties, interest Political protest,
Representative groups, corporatist demonstration, public
democracy representation discourse and
influence, networks
Direct democracy Referenda, town Political protest,
meetings demonstration, public
discourse and
influence, networks
Non-elected Public hearings, Devolved and Political protest,
institutions deliberative stake- deconcentrated demonstration, public
(agencies, courts) holder rule-making, administration, citizen discourse and
citizen juries, use of courts enabled influence, networks
town-hall meetings by legal standing

Society Devolution of public Political protest,

functions to demonstration, public
associations discourse and
influence, networks
Democracy within
associations
Economy Collective bargaining, Direct action
stakeholder
representation,
workplace democracy
Source: Adapted from Warren 2004.

politics to formulate these outputs into collective agendas and decisions (Habermas 1996,
chaps 7–8). In order to account for the de-centring impact of liberal-democratic governance
on decision-making, we should also distinguish between ‘centred’ and ‘devolved’ and
‘reflexive’ forms of institutions (Teubner 1983). Centred institutions receive input, pro-
cess it, issue authoritative decisions, and then organize collective actions. Representative
institutions combined with executive agencies operate in this manner. Devolved or reflex-
ive institutions develop the conditions under which parties to a conflict can work out their
differences and monitor solutions more directly. An example would be laws establishing
collective bargaining, since they serve to equalize the powers of labour relative to man-
agement and define fair processes without deciding or imposing outcomes.
Combining these dimensions in Table 1 gives us a rough map of the potential spaces
of democracy, from direct action in the economy, discourse and demonstration in public
spheres, to the more traditional politics of political parties. The circled cells identify the
domain within which governance-driven democratization is a possibility – that is, the
domain of non-electoral institutions of government, of both the older, centralized variety
as well as the newer decentred and reflexive kind. It is within this domain that we are see-
ing a rapid development of what are often called ‘citizen engagement’ and ‘public engage-
ment’ processes – that is, everything from the from the public hearings and mandatory
public comment periods that emerged after World War II, to the stakeholder meetings that
began to spread in the 1980s, and to newer consensus conferences, town hall meetings,
citizen juries, citizen assemblies, deliberative polling, online dialogues, deliberative
6 M.E. Warren

planning, participatory budgeting, study circles, planning cells, collaborative learning, and
even participatory theatre (Fung 2003, Hajer and Wagenaar 2003, Smith 2009, Gastil
2008, Parkinson 2006). There are, most probably, nearly one hundred named processes.
Typically, these processes use the languages of participation and deliberation; they are
designed for particular policy problems; they bypass the formal institutions of democracy,
and they do not involve protest, lobbying, or obstruction.
In drawing this map, I want to make what may be an obvious point. The domain of
governance-driven democratization has certain characteristics which will limit the kinds
of democratization that can occur here, but also bring possibilities that cannot be gained in
other domains. We could think about this map using an ecological metaphor: we need to
identify the characteristics of the species so we can identify its niche, comparative advan-
tages, and vulnerabilities.
Four of these characteristics are as follows: first, governance-driven democratization is
a response to democracy deficits. Second, it tends to be elite-driven. Third, its evolution is
often de-linked from electoral democracy. Fourth, it has an important capacity to bring into
existence dynamic, serial, and overlapping peoples and constituencies based on the ‘all
affected’ principle in contrast to pre-defined and relatively static territorial constituencies.
With respect to the first characteristic, I do not think it is a stretch to interpret governance-
driven democratization as a response to deficits in standard electoral representative demo-
cracy. The general idea is that there is a lot of political work in today’s societies that elect-
oral democracy is unable to do. Part of this story is a functional one: the capacities of
political institutions are being outstripped by the development of the societies in which
they are embedded. The story is well-known, but it is worth repeating: the incapacities of
electoral democracy are inherent and growing with respect to policy development and
implementation. Under the standard model of democratic self-rule, citizens elect repre-
sentatives who then make policy on their behalf. They legislate goals and means, and then
pass these broad directions over to administrators, who then further develop and execute
policy. But these linkages that should establish democracy are broken at almost every
step: territorial electoral constituencies do not match issue constituencies, which are
increasingly non-territorial. Electoral systems only roughly reflect citizen preferences, and
they do so in highly aggregated, information-poor ways. Legislatures have very low
policy-making capacities, and so pass most decisions to executives and executive
agencies, which are then often left to guess what constituents want. These problems are
multiplied, of course, by influence markets, corporatism, and bureaucratic power and inertia
(Bobbio 1987). The bureaucratic form of policy development and implementation has
inherent limitations as well: rule-based hierarchy tends to be rigid and information poor.
Then add to this story several broad facts of social and economic development:
today’s societies are enormously complex, not just technically, but also politically.
Although it was technical complexity that bothered earlier generations of democratic theo-
rists from Schumpeter through Sartori, it is political complexity that is especially chal-
lenging for the standard model (Beck 1997). Today’s societies are increasingly populated
by groups with resources – not just wealth and market power, but also social movements,
cultural and ethnic groups, and groups that arise spontaneously in response to issues and
policy initiatives. The world as we know it is becoming more rather than less pluralized.
And this pluralization is now deeply rooted in dense civil societies, which are growing
algebraically not only in the numbers of associations, but in their relative size (Warren
2001). Pluralized civil societies provide the social infrastructure of social capacities for
organization and advocacy, which are essential to articulating preferences and supporting
a public sphere, but also tend to overwhelm the capacities of electoral institutions.
 Critical Policy Studies 7

Finally, add two familiar trends to this picture. One is the intensification and globaliza-
tion of markets, migration, security regimes, environmental issues and the like. Globaliza-
tion threatens the relevance of territorially-based electoral democracy, simply because the
scope and locations of issues fails to match the organization of constituencies (Bohman
2007). The other is the general post-modernization of culture, particularly in the developed
world (Inglehart and Welzel 2005). Generation by generation, people are becoming less def-
erential to authority and have more small-d democratic instincts. They are now more likely
to volunteer for causes; they are more interested in self-government; and they are keenly
attentive to their powers over their own biographies. Moreover, organized groups have
increasingly gained expert capacities themselves, and are able to challenge government (and
corporate) expertise, often with the aid of relatively new sunshine and freedom of informa-
tion laws. They have been increasingly successful at prying open the older corporatist net-
works of technocrats and experts introducing contestation into the domain of knowledge.
The gap between the functional incapacities of electoral democratic institutions and
the demands and aspirations of citizens has been showing up, as is well known, in high
levels of distrust with, and disaffection from the political system. Among the developed
democracies these effects vary by country, with Denmark on one extreme and the US on
the other. But the trends are detectable and well-documented in almost every developed
democracy (Newton 2008).
From a political perspective, I do not think these trends add up to anything like the
legitimation crisis predicted in the 1970s (Habermas 1975). The reason is that there are
many points of adjustment and deflection in the developed democracies, so the broad
legitimation pressures show up not as a general system crisis, but rather issue by issue and
policy by policy, in protests over airport expansion, medical coverage, poverty issues,
changes in regulation of genetically-modified organisms, forest management, struggles over
neighbourhood development, energy pricing, and so on. So rather than general system cri-
ses, we have something like pluralized ungovernability, driven by functional incapacities
of government agencies, as well as overlapping or competing jurisdictions, or failures to
anticipate spill-over consequences and linkages among issues (Beck 1997). This is why, I
think, much of the interesting action in democracy and democratization has moved into
the domains of administration and policy.
These points bring us to a second characteristic of governance-driven democratization.
If a previous recent wave of democratization could be said to reside in the development of
civil society and the public sphere driven by citizen activists, this next arena – governance-
driven democratization – is very much about elite responses. What is most remarkable
about the last decade or so is the interest administrators have taken in new democratic
processes, with the active encouragement of a now large industry of consultants and
democratic process entrepreneurs, such as America Speaks, Deliberative Polling, and Cit-
izen Juries. The functional reason is simply that policy-makers are on the front lines of this
new pluralized ungovernability, which shows up in several ways. First, policy-makers find
that the legitimacy generated by electoral democracy does not carry over to issue-segmented
constituencies. Second, they find that any given policy generates opposition which derails
their capacities to plan. Third, they often find they are short on information and enforce-
ment capabilities, which often only can be expanded by engaging with advocates – most
notably in the environmental area, but also in labour policy, public health, and elsewhere.
And, finally, if policy makers fail to anticipate problems, formal democracy remains
powerful enough to impose solutions, but often in ways that are clumsy and costly – as
with the cases of medical research tissue-banking in the UK, and stem-cell research in the
US. Thus, policy-makers increasing speak the language of trust, collaboration, citizen
8 M.E. Warren

engagement, and dialog, in part to avoid the alternative politics of opposition, protest, grid-
lock, and imposed solutions. Athough there is certainly no overall plan, the strategy amounts
to a functional compensation for low global legitimacy of electoral democracy by generat-
ing legitimacy ‘locally’ – issue by issue, policy by policy, and constituency by constitu-
ency.
The third feature of governance-driven democratization follows: it is often de-linked
from electoral democracy. The logics that drive governance-driven democratization are
related to policy development, planning, and administration rather than to the global legit-
imacy of elected governments. That said, elected governments have become increasingly
aware that electoral legitimacy does not translated into policy-specific legitimacy. Thus,
legislation has for some time directed agencies to establish processes for ‘public input’ or
required ‘community representation’ during policy development. From this perspective,
governance-driven democratization can be seen as supplementary to electoral democracy,
shoring up its functional weaknesses.
But the de-linking from electoral democracy is also evident in the emergence of gov-
ernance-driven democratization in places where competitive, multi-party democracy does
not exist at all. In China, the Communist Party has been actively encouraging new ‘con-
sultative’ and ‘deliberative’ processes. In a now-famous experiment in Wenling City, for
example, officials used deliberative polling to set annual budget priorities (Mo and Chen
2005). The process has been so popular that city officials now find they cannot go back to
the old ways, for fear of generating protests. The Chinese case suggests that the conditions
for governance-driven democratization are broader than deficits in electoral democracy. In
China, attention to governance is driven by a development agenda, combined with the ten-
dencies of development to produce multiple actors with credible capacities for opposition.
The Chinese Communist Party is, no doubt, hoping that these democratic experiments will
head off regime democratization in the Western sense, while – like their Western counter-
parts – they hope to generate legitimacy, policy by policy, in ways that will replace the
now-bankrupt socialist ideology. Whether the Party will be successful remains to be seen.
The point here, however, is that the forces that drive this trajectory are deeply rooted in
pluralization, contestation, and social capacities unleashed by social and economic devel-
opment rather than by pressures for electoral democratization or reform. From this per-
spective, the Chinese case is more comparable to the developed democracies than to the
less-developed electoral democracies.
Fourth, it follows from the policy-focused nature of governance-driven democratiza-
tion that constituencies may come into existence in ways that are quite distinct from electoral
democracy. In electoral democracy, ‘the people’ are those who live within the boundaries
of a state, and they are represented (typically) through territorial constituencies. Governance-
driven democratization is less bound by this logic. ‘Peoples’ are, in effect, brought into
existence in response to issues, and often dissolve when issues are resolved. Constituen-
cies are dynamic, overlapping, issue-focused, and serial. Individuals may belong to many
distinct ‘peoples’ at any given point in time, in a way that is, potentially, more adaptable to
the ‘all affected’ principle of democracy than are territorial constituencies.
This sketch of the governance-driven democratization landscape suggests two broad
judgments. The first is critical: if the democratic potentials of governance-driven democra-
tization are to be captured, we need to be clear that these forms of democracy have inherent
limitations. In this domain, for example, agendas are typically elite-defined, while policy-
driven issues often make it difficult to encompass full political programs in the way that
coherent political parties can within legislatures. Protest and opposition are ill-suited to
the form, though without these possible obstructions there would be little reason for its
 Critical Policy Studies 9

existence. And although the language of ‘citizen participation’ and ‘public engagement’ is
ubiquitous, the limits of size and scale operate here as in other areas – and so participatory
claims are inevitably exaggerated.
The second broad judgment, however, is that if we view governance-driven democra-
tization as a response to the rise of advocacy, social movements, and civil society more
generally, we may be witnessing a transformation of democracy as dramatic and important
as the rise of mass, electoral democracy in the nineteenth century. Viewed as a whole, these
developments promise to alter definitions of peoples and boundaries, open new venues of
direct participation, alter forms of representation, inject new forms of deliberation into
government, and harness the enormous collective capacities of civil society.
It seems to me, then, that we need to be as imaginative about these possibilities and
opportunities as we are critical about their dangers and limitations. The first opportunity I
have already suggested. Because governance-driven democratization is not tied to stati-
cally-defined electoral constituencies, it has the potential to include all those affected by
an issue. Because constituencies are defined by issue rather than territory, it is possible to
draw democratic processes closer to the ‘all-affected’ principle – the basic norm of demo-
cracy that holds that those potentially affected by a collective decision should have some
influence over the decision (Goodin 2007). This quality is an artefact, as it were, of the
drivers of governance-driven democratization – that is, specific issues and policies which
can evoke and provoke those affected. This quality of governance-driven democratization
is the democratic side of the elite-led nature of the domain. Though elite constitution of
peoples is not, on the face of it, a particularly democratic way of constituting peoples,
within the broader environment of electoral institutions, advocacy, and social activism, we
could look at elites as intermediaries in an on-going process of constituency formation that
is potentially more responsive to those affected by issues than territorially-defined elect-
oral constituencies can be.
Of course, challenges and dangers are part of the same package. Without electoral
mechanisms of accountability and operating under mandates that are often vague, elites
have discretion in how they define relevant constituents when they devise ‘public engage-
ment’ processes. Elite discretion can produce at least three kinds of anti-democratic dan-
gers. First, elites can frame agendas in ways that simply fail to capture the issues, and in so
doing fail to include the relevant constituencies. Second, because elites are often respond-
ing to opposition, they are most responsive to well-organized stakeholders rather than to
all affected, thus biasing constituency definition toward those with wealth, education, and
power (Fung 2003). Third, if elites open processes to all those who are interested, proc-
esses of self-selection tend to be biased toward the intensively interested and well-
organized, and biased against the unorganized, as well as against latent public goods
(Dalton et al. 2003). So the challenges here will be to conceive and devise recursive
agenda-setting and constituency definition processes that approximate the all-affected
principle, including broad unorganized publics.
A second opportunity is in the area of empowerment. Whereas voting for representa-
tives is a limited and non-specific form of empowerment, many of the new governance-driven
processes promise to empower participants directly and specifically – an ironic potential,
perhaps, since empowerment often builds on elite capacities to define agendas precisely
and narrowly enough so they can lead to decisions. But this democratic opportunity typi-
cally comes with an imbalance between elites and participants: the elites who generate
processes are often operating under powers formally granted by states, and they have the
benefit of this kind of authority. The authority of participants in ad hoc processes is less
clear, particularly if their decisions involve deploying the powers of states. So, in fact,
10 M.E. Warren

most new processes are advisory rather than empowered. The challenges here will have to
do with conceiving forms of empowerment that are consistent with electoral democracy.
The challenges are not insurmountable: the Canadian provinces of British Columbia and
Ontario, for example, recently constituted citizens’ assemblies on electoral reform, with
the power to set the agenda for a referendum, thus leaving the final judgment to citizens
(Warren and Pearse 2008, cf. Cutler and Fournier 2007).
A third opportunity is in the domain of representation. These opportunities are, of
course, linked to the question of constituency formation. But they are worth highlighting
in their own right. Democracy entrepreneurs often use the language of ‘participation’ and
‘engagement’ to describe these processes. In my view, democratic theorists should be
focusing more closely on representative relationships, since the fact remains that most
people do not participate in these processes (Warren 2008). What makes them distinct is
that a few citizens represent other citizens. They are usually selected or self-selected rather
than elected. What matters most from the perspective of democratic representation is that
the processes include an inclusive sample of interests, values, views, and opinions of those
potentially affected, as well as have the capacity to render considered advice or decisions
that responds to those affected. And, again ironically, because governance-driven democ-
ratization is elite-led, elites have the opportunity to design representative processes that
are neither electoral in nature, nor self-selecting – these being key features of electoral
democracy and civil society-based advocacy democracy. Representation can be designed
to include marginalized people and unorganized interests, as well as latent public interests.
That said, for strategic reasons, elites typically prefer stakeholder representation, since
stakeholders hold the keys to governance. So even if potentials for new forms of represen-
tation exist, on average the rise of governance-driven democratization threatens more
rather than less inequality (Dalton et al. 2003). The more points of access there are to gov-
ernment, the more advantaged are those with organization and resources. Thus, a key chal-
lenge for democratic theorists will be to identify alternative forms of representation that
would balance these tendencies, and then figure out why elites should want to adopt them.
A fourth opportunity is that of deliberative leadership by ordinary citizens. I will say
less about deliberation because there’s no shortage of theory. I will note, however, that gov-
ernance-driven democratization provides some specific kinds of deliberative opportunities
that exist neither in the electoral arena nor in the public sphere. In electoral democracy,
deliberation is often undercut by the strategic features of campaigns, elections, and party
politics. Public sphere deliberation runs the gamut of possibilities, and is essential to public
opinion formation. But public discussion is often poorly linked to the learning, problem-
solving, and sustained attention necessary for policy decisions. Both limitations can be mit-
igated by designed ‘minipublics’ which are de-linked from both the strategic elements of
electoral politics, and the unfocused elements of broad publics (Goodin and Dryzek 2007).
Well-designed and well-facilitated, minipublics can capture the innovative and progressive
capacities of deliberation, and do so ways that are broadly representative of public interests.
The challenges here have less to do with designing high deliberative quality into minipub-
lics – we mostly know how to do that. The challenges have more to do with linking mini-
publics back to broader publics, since learning and deliberation transforms participants
from citizen representatives into public opinion vanguards which may become as distant
from public opinion as any other kind of representative (Chambers 2009).
The final class of opportunities has to do with the key role that governance-driven
democratization innovations can play in connecting state power to civil society – a point
that returns to the basic thesis, namely, that the rise of governance-driven democratization has
much to do with the functional incapacities of electoral democracy in complex societies.
 Critical Policy Studies 11

Again, the way I am constructing the thesis, governance-driven democratization is a

response to the increasing strength and pluralism of society, reflected in its increasingly
self-conscious and organized characteristics. That said, connections between society and
state operating through electoral democracy, as I have suggested, are often global, low on
information, and non-specific. In contrast, connections working through advocacy and social
movements are important for pressure, but often lead to political gridlock. Governance-
driven democratization, which would neither be necessary nor possible without these
developments, can be structured in ways that governments can capture the potentials
within civil society for organization, information, energy, and creativity (Hajer and
Wagenaar 2003). Examples are now legion, in social and medical services, urban plan-
ning, environmental issues, community development, policing, public schools, and other pol-
icy areas. I see the challenges from a democratic perspective in two areas. First,
governments can co-opt civil society organizations in such a way that they lose their
capacities to represent their constituencies. Selective inclusion can develop into new kinds
of exclusive and informal corporatist structures. Second, civil society capacities are
unevenly distributed, in such a way that any general policy to, say, deliver services in
cooperation with civil society organizations is likely to favour those areas and segments of
society where civil society is robust.
In conclusion, I am not making an argument so much as painting a picture, and then
calling the picture governance-driven democratization. If there is an argument, it is both
optimistic and cautious. On the optimistic side, I think there are radically democratic
potentials in this emerging phenomenon. ‘Governance’ can inject into modern societies
and government democratic ideals that were thought to be impossible: more citizen parti-
cipation, more government responsiveness, better representation, more deliberation. On
the cautious side, these are potentials, not necessities. Hence the need for critical theory in
a generic sense: we need to look at these emerging institutions and practices, and ask
about their immanent potentials. To identify these potentials, we will probably need to use
a two-step strategy. On the one hand, we need to become more general in the way we
think about democratic norms – for example, thinking about ‘democracy’ as any set of
arrangements that enables people to develop into self-developing, self-governing beings.
We should not identify democratic norms with any particular institutional mechanism,
such as elections, majority rule, deliberation, or even participation. These are ways and
means, not ends. On the other hand, we need to be quite specific in looking for ways in
which democratic norms find life in new institutions and practices, and to keep an open
mind about how these might evolve into practices that contribute to self-development and
self-government – including those new institutions devised by elites in response to
political and policy gridlock (Fung 2007, Smith 2009).
There is also a more specific kind of theoretical task. Sorting out the good from the
bad in these policy-defined arenas of governance means that we will probably need to
develop a somewhat complex middle-level body of theory to help guide more ‘local’
analyses and critical assessments – particularly those that might distinguish more from
less democratic innovations. So we need to develop approaches that will allow us to assess
the fit between processes and issues (Warren 2001, Fung 2007). What kinds of processes
are likely to generate better rather than worse outcomes – more legitimacy, justice, or
effectiveness, say – given the characteristics of the issues and the constraints of time and
money? We are not starting from scratch. We have much general knowledge about demo-
cratic mechanisms such as voting schemes, kinds of representation, and deliberative proce-
dures. We also now have lots of specific knowledge about the effects of various
combinations. We know, for example, that combining experts with lay citizens over time
12 M.E. Warren

within a deliberative context can overcome many of the constraints of technical complex-
ity. We know that processes which allow citizens to self-select will bias the process
toward organized, high-resource interests, and that random selection can produce a closer
approximation of informed public opinion. We have detailed knowledge of particular
designs, such as the British Columbia Citizens’ Assembly or 21st Century Town Hall
Meetings. We know that ‘enclave deliberation’ may reinforce undesirable kinds of preju-
dice (Sunstein 2002, chap. 1). But we cannot yet begin with an issue and a set of goals, and
then conceive processes that are appropriate for the particular demands of the issues in
relation to the goals.
Relating issue characteristic to answers to each design dimension should maximize the
chances that the process will achieve the normative goals for any given issue. We will then
be able to deliver specific assessments about what counts as a better or worse process from
the perspectives of democracy – or, if we wish, from any other normative perspective,
though this is the one I choose. In short, if I am right that governance-driven democratiza-
tion is now one of the most dynamic vectors of progress in democratization, critical policy
analysis will have moved to the front lines of democratic reform and innovation.

Acknowledgements
This paper is a revised version of a talk presented to the Interpretation in Policy Conference, Univer-
sity of Essex, UK, 19 June 2008. My thanks to Frank Fischer, He Baogang and Albert Weale for
comments on the original draft.

Notes on contributor
Mark E. Warren holds the Merilees Chair for the Study of Democracy in the Department of Political
Science, University of British Columbia. He is interested in new forms of citizen participation and
democratic representation, the relationship between civil society and democracy, and the corruption
of democratic relationships. Warren is author of Democracy and Association (Princeton University
Press, 2001), editor of Democracy and Trust (Cambridge University Press, 1999), and co-editor of
Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge University
Press, 2008).

References
Beck, U., 1997. The reinvention of politics: rethinking modernity in the global social order. Ritter,
M., trans. Cambridge: Polity Press.
Bobbio, N., 1987. The future of democracy: a defence of the rules of the game. Minneapolis: University
of Minnesota Press.
Bohman, J., 2007. Democracy across borders: from Dêmos to Dêmoi. Cambridge, MA: MIT Press.
Cain, B., Dalton, R., and Scarrow, S., eds, 2003. New forms of democracy? The reform and transfor-
mation of democratic institutions. Oxford: Oxford University Press.
Chambers, S., 2009. Rhetoric in the public sphere: has deliberative democracy abandoned mass
democracy? Political theory, 37 (3), 323–350.
Cutler, F., and Fournier, P., 2007. Why Ontarians said no to MMP. Globe and Mail, 25 October,
Opinions.
Dalton, R.J., Cain, B.E., and Scarrow, S.E., 2003. Democratic publics and democratic institutions.
In: B. Cain, R. Dalton, and S. Scarrow, eds. New forms of democracy? The reform and transfor-
mation of democratic institutions. Oxford: Oxford University Press, 250–275.
Fung, A., 2003. Survey article: recipes for public spheres: eight institutional design choices and their
consequences. Journal of political philosophy, 11 (3), 338–367.
Fung, A., 2007. Democratic theory and political science: a pragmatic method of constructive
engagement. American political science review, 101 (3), 443–458.
Gastil, J., 2008. Political communication and deliberation. Thousand Oaks, CA: Sage
 Critical Policy Studies 13

Goodin, R.E., 2007. Enfranchising all affected interests, and its alternatives. Philosophy and public
affairs, 35 (1), 40–68.
Goodin, R.E., and Dryzek, J., 2006. Deliberative impacts: the macro-political uptake of mini-publics.
Politics and society, 34 (2), 219–244.
Habermas, J., 1975. Legitimation crisis. T. McCarthy, trans. Boston: Beacon Press.
Habermas, J., 1987. The theory of communicative action. Vol. 2. T. McCarthy, trans. Boston: Beacon
Press.
Habermas, J., 1996. Between facts and norms: contributions to a discourse theory of law and democracy.
W. Rehg, trans. Cambridge, MA: MIT Press.
Hajer, M.A., and Wagenaar, H., eds, 2003. Deliberative policy analysis. Cambridge: Cambridge
University Press.
He, B., 2006. Participatory and deliberative institutions in China. In: E. Leib and B., He eds. The
search for deliberative democracy in China. New York: Palgrave, 2006, 176–196.
Inglehart, R., and Welzel, C., 2005. Modernization, cultural change, and democracy: the human
development sequence. Cambridge: Cambridge University Press.
Mo Yifei and Chen Yiming. 2005. Democratic deliberation: the innovation from Wenling. Beijing:
Central Compliance and Translation Press.
Newton, K., 2008. Trust and politics. In: D. Castiglione, J.W. van Deth, and G. Wolleb, eds. The
handbook of social capital. Oxford: Oxford University Press, 241–272.
Parkinson, J., 2006. Deliberating in the real world. Oxford: Oxford University Press.
Smith, G., 2009. Democratic innovations: designing institutions for citizen participation. Cambridge:
Cambridge University Press.
Sunstein, C., 2002. Designing democracy: what constitutions do. Oxford: Oxford University Press.
Teubner, G., 1983. Substantive and reflexive elements in modern law. Law and society review
17 (2), 239–285.
Warren, M.E., 2001. Democracy and association. Princeton: Princeton University Press.
Warren, M.E., 2003. A second transformation of democracy? In: B. Cain, R. Dalton, and S. Scarrow,
eds. Democracy transformed: expanding political opportunities in advanced industrial democ-
racies. Oxford: Oxford University Press, 223–249.
Warren, M.E., 2004. A second transformation of democracy? In: B. Cain, R. Dalton, and S. Scarrow,
eds. New forms of democracy? The reform and transformation of democratic institutions.
Oxford: Oxford University Press, 223–249.
Warren, M.E., 2008. Citizen representatives. In: M.E. Warren, and H. Pearse, eds. Designing delib-
erative democracy: the British Columbia Citizens’ Assembly. Cambridge: Cambridge University
Press, 50–69.
Warren, M.E., and Pearse, H., eds, 2008. Designing deliberative democracy: the British Columbia
Citizens’ Assembly. Cambridge: Cambridge University Press.