Admir Skodo, Ed., Other Logics - Alternatives To Formal Logic in The History of Thought and Contemporary Philosophy-2014

Other Logics
Philosophy of
History and Culture
Edited by
Michael Krausz, Bryn Mawr College
Advisory Board
Annette Baier, University of Pittsburgh

Purushottama Bilimoria, Deakin University, Australia
Cora Diamond, University of Virginia
William Dray, University of Ottawa
Nancy Fraser, New School for Social Research
Clifford Geertz†, Institute for Advanced Study, Princeton
Peter Hacker, St. John’s College, Oxford
Rom Harré, Linacre College, Oxford
Bernard Harrison, University of Sussex
Martha Nussbaum, University of Chicago
Leon Pompa, University of Birmingham
Joseph Raz, Balliol College, Oxford
Amélie Rorty, Harvard University
VOLUME 33
The titles published in this series are listed at brill.com/phc

Other Logics
Alternatives to Formal Logic in the History of
Thought and Contemporary Philosophy
Edited by
Admir Skodo
LEIDEN | BOSTON
Library of Congress Cataloging-in-Publication Data
Other logics : alternatives to formal logic in the history of thought and contemporary philosophy / edited by
Admir Skodo.
pages cm. — (Philosophy of history and culture, ISSN 0922-6001 ; VOLUME 33)
Includes bibliographical references and index.
ISBN 978-90-04-27003-9 (hardback : alk. paper) — ISBN 978-90-04-27018-3 (e-book) 1. Logic—History.
I. Skodo, Admir, editor of compilation.
BC15.O84 2014
160—dc23
2013050314
This publication has been typeset in the multilingual ‘Brill’ typeface. With over 5,100 characters covering
Latin, ipa, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities.
For more information, please see brill.com/brill-typeface.
issn 0922-6001
isbn 978 90 04 27003 9 (hardback)
isbn 978 90 04 27018 3 (e-book)
Copyright 2014 by Koninklijke Brill nv, Leiden, The Netherlands.

Koninklijke Brill nv incorporates the imprints Brill, Brill Nijhoff, Global Oriental and Hotei Publishing.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided
that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive,
Suite 910, Danvers, ma 01923, usa. Fees are subject to change.
This book is printed on acid-free paper.

Contents
Acknowledgments vii
Notes on Contributors viii
Introduction 1
Admir Skodo
PART 1
Perspectives from the History of Thought 15
1 Proving the Principle of Logic: Quentin Meillassoux, Jean-Luc Nancy,

and the Anhypothetical 17
Christopher Watkin
2 The Self, Ideology, and Logic: F.C.S. Schiller’s Pragmatist Critique of
and Alternative to Formal Logic 32
Admir Skodo
3 Language, Truth, and Logic: Heidegger on the Practical and Historical
Grounds of Abstract Thought 51
Aaron James Wendland
4 The Obstacle: Jacques Lacan’s Critique of the Formal Logical
Representation of the Real 66
Ervik Cejvan
5 Collingwood’s Logic of Question and Answer against the
Relativization of Reason 81
Christopher Fear
PART 2
Perspectives from Contemporary Philosophy 101
6 Representationalist Logic 103
Frank Ankersmit
7 On Logical Aliens 123
Alessandra Tanesini
8 The Heart of Metaphysical Pluralism and the Consistency Dilemma:
A Critical Analysis of the Possibility of Incompatible Truths 148
Thord Svensson
vi contents
9 The Logic of “Oughts” and the Bindingness of Past Practice:

A Critique of Normative Judgment Internalism through a
Reading of King Lear’s Act I 169
Karim Dharamsi
10 First-Order Logic, Incongruism, and Anti-Formalism 188
Anders Kraal
11 Zombies, Selves, and the Possibility of Afterlife 207
Johan Modée
12 The Logocentric Predicament and the Logic of Question
and Answer 221
Giuseppina D’Oro
Indices
Name 235
Subject 237
Acknowledgments
This book is the final outcome of an international project funded by Stiftelsen

Riksbankens Jubileumsfond (Bank of Sweden Tercentenary Foundation), initially enti-
tled “Selves and Logics: Historical and Philosophical Perspectives on Logic from
Dialectical to Formal Logic.” We in the editorial committee wish to express our grati-
tude to the Foundation for their generous support. The editorial committee has been
composed of Professor Catharina Stenqvist (project leader), Dr. Admir Skodo, Ervik
Cejvan (doctoral candidate), and Professor David Dunér. The work of the entire com-
mittee has been invaluable in organizing the project from its initial to its final stage.
The committee also wishes to thank all the participants not only for their fine papers,
but also for their participation at the conference held in Höör, Sweden, in February
2012. Finally, the committee wishes to thank its editors at Brill, Jennifer Pavelko and
Julia Berick, for all their help and assistance.
Notes on Contributors
Frank Ankersmit
is Professor Emeritus of Intellectual History and Historical Theory at the
University of Groningen. A renowned philosopher of history and intellectual
historian, Ankersmit’s publications include Narrative Logic (1983), History and
Tropology (1994); Aesthetic Politics (1996) Historical Representation (2001), and
most recently Meaning, Truth, and Reference in Historical Representation
(2013).
Ervik Cejvan
is a doctoral candidate in the philosophy of religion at Lund University. He is
currently finishing his doctoral thesis on the concept of madness in the
thought of Plato, Lacan, and Bataille. He is the co-editor of and chapter
contributor to the book Tillvarons utmaningar (2012).
Giuseppina D’Oro
is Reader in Philosophy at Keele University. She has published widely on the
philosophy of action and the philosophy of science. Key publications include
Collingwood and the Metaphysics of Experience (2002), and a number of
articles in refereed philosophical journals.
Karim Dharamsi
is Associate Professor of Philosophy at Mount Royal University. He has
previously published articles and book chapters on moral philosophy.
Christopher Fear
is a doctoral candidate in Philosophy and Politics at the University of Exeter.
He is finishing his doctoral thesis on Collingwood’s philosophy of history. An
article on Collingwood’s philosophy of history has recently been published by
the journal History of the Human Sciences (2013).
Anders Kraal
is a Postdoctoral Fellow at the University of Calgary. Kraal works on the
philosophy of logic and theology. He has published articles on these topics in
the journals Philosophy Compass and History and Philosophy of Logic.
notes on contributors ix
Johan Modée
is a Senior Lecturer at Malmö University College. He works on the philosophy
of religion and the philosophy of science. Modée has published numerous
articles and book chapters, and is the co-editor of Mänskliga rättigheter och
religion (2011), and Frihet och gränser (2006).
Admir Skodo
is a Postdoctoral Scholar at the University of California at Berkeley, and
President of the UC Berkeley Humanities and Social Sciences Association.
Skodo works on modern European cultural and intellectual history. He has
published articles in numerous refereed journals. Skodo is currently finishing
a book on new idealist philosophy and post-World War II English
historiography.
Thord Svensson
is a doctoral candidate in the philosophy of religion at Lund University. He is
currently finishing his doctoral thesis on religion and metaphysical pluralism.
Svensson is the main editor of and chapter contributor to the book Tillvarons
Utmaningar (2012).
Alessandra Tanesini
is Professor of Philosophy at Cardiff University. She has published on the
philosophy of feminism, Wittgenstein, and epistemology. Key publications
include An Introduction to Feminist Epistemologies (1999), and Wittgenstein:
A Feminist Introduction (2004).
Christopher Watkin
is Senior Lecturer and Convenor of French Studies at Monash University.
Watkin works on contemporary French philosophy. Key publications include
Difficult Atheism: Post-theological Thinking in Alain Badiou, Jean-Luc Nancy
and Quentin Meillassoux (2011).

is a doctoral candidate in Philosophy at Oxford University. He will soon take
up a Postdoctoral Fellowship in philosophy at the University of Johannesburg.
He works on the philosophy of Heidegger, and is the co-editor of and chapter
contributor to the recent book Wittgenstein and Heidegger (2013).
Introduction
Admir Skodo
There is a widespread assumption in today’s western philosophical culture that

logic is an inherently formal, monolithic, universal, and ahistorical science that
best describes (and often prescribes) the nature of thought and reasoning. This
collection of essays challenges this assumption by showing, in various ways,
that logic is inextricably situated in multi-dimensional cognitive, conceptual,
and extra-logical contexts.1 In short, we aim to show that, why, and how we
should move beyond the formalism of logic on the one hand, and embrace the
idea of “other” logics as a philosophically legitimate one, on the other.
Since there is a prevailing and still largely unquestioned view of formal logic
as the best systematic account of rational thought irrespective of historical
and social context, our aim is to make a valuable contribution to the human
sciences in critiquing such a position, and offering alternatives to it. This intro-
duction will, first, briefly discuss this view we are critiquing by means of a
critical survey of some of the representative literature on the topic. Second, it
will highlight the ways in which the studies in this volume are related to that
literature.
Three distinct and often, though not necessarily, related views of logic that
tend to set the conceptual boundaries for what is to count as logic, both past
and present, we call purist, perfectionist, and universalist (or absolutist). These
names are here used as a means to render comprehensible some of the key
themes underpinning much of the literature on the philosophy of logic that all
the authors in this book view as problematic. Since all chapters in this volume
seek to question these (entangled) views, we need to show what they mean.
The purist view assumes that logic is a wholly self-contained system that
consists of, on the one hand, atomistic thought elements expressed in lan-
guage by means of propositions; and on the other hand, rules of transforma-
tion of these elements. Logic in this view is not constituted by concrete acts of
thinking, and the socio-cultural context of those acts; just the opposite, logic is
said to constitute such acts. The perfectionist agrees with the purist, but argues
that there are other ways of thinking than those artificially constructed by pur-
ist logicians such as Giuseppe Peano, George Boole, Frederic Fitch, Gottlob
Frege, and Bertrand Russell; the perfectionist therefore insists that among the
1 Cf. Anssi Korhonen, Logic as a Universal Science: Russell’s Early Logicism and its Philosophical
Context (Basingstoke and New York: Palgrave Macmillan, 2013).
© koninklijke brill nv, leiden, ��4 | doi ��.��63/9789004270183_��2

2 skodo
different conceptions of thinking, formal logic is the best, for it alone is con-
ducive to perfect reasoning. The universalist or absolutist, finally, agrees with
both the purist and the perfectionist, but adds that since there are imperfect
ways of reasoning, it is the duty of logic to combat and replace them with one
true logic. Such views, in various degrees and combinations, can be found in
works on the philosophy of logic, in common textbooks of logic used in teach-
ing at the undergraduate and graduate levels,2 as well as in scholarly journals
devoted to the study of logic, such as Journal of Symbolic Logic and History and
Philosophy of Logic.
For instance, we find such views propagated in recent handbooks of the his-
tory of logic, as in Handbook of the History of Logic from 2004.3 This work covers
the period 1685 to 1900, and sweepingly argues for a “mathematical turn” in
logic, even discussing the “imperial designs of mathematicians,” which, once
these designs began acting as maps for logicians for understanding the ter-
rain that is human thought, became the imperial designs of logicians. Leibniz,
Frege, and Russell are seen as the heroes of this plot, which, moreover, confi-
dently asserts that regardless of their different social, cultural, political, and
personal contexts, logicians during this period thought about the same con-
cepts and principles, saw the same problems, and even offered the same solu-
tions to those problems. In other words, logic in this book is assumed to be a
monolithic entity, impervious and independent from concrete human life, and
yet the only true standard of all human rational thought.
Such a purist and universalist view of logic has a strange effect in the hand-
book: Kant and Hegel are included, but not as having worked on logic in the
“real” sense of the term. Thus, on this view “Kant is not a major contributor to
the development of formal logic. He fails, too, in his most conspicuous efforts
to build his transcendental logic on clues provided by formal logic.”4 Mary
Tiles, the author of the chapter on Kant, argues that Kant was much broader in
his view of logic than the definition of the handbook as a whole would have it,
and she argues that he was rightly so. Therefore, she continues, Kant should be
seen as a revisionist of logic, who posited a “transcendental logic,” as opposed
2 Jon Barwise and John Etchemendy, Language, Proof, and Logic (Chicago: The University of
Chicago Press, 2002); John P. Burgess, Philosophical Logic (Princeton: Princeton University
Press, 2009).
3 Handbook of the History of Logic: Volume 3: The Rise of Modern Logic from Leibniz to Frege,
ed. Dov M. Gabbay & John Woods (Amsterdam: Elsevier, 2004).
4 Mary Tiles, “Kant: From General to Transcendental Logic,” in Handbook of the History of
Logic: Volume 3: The Rise of Modern Logic from Leibniz to Frege, ed. Dov M. Gabbay & John
Woods (Amsterdam: Elsevier, 2004), 85–130, 85.
Introduction 3
to a “general logic,” the former of which fruitfully seeks to overcome flaws of

rationalism and empiricism, and arrive at a logic in the intersection between
reason and sensation, form and matter, which is “necessary for thought of con-
crete individuals.”5 But why, then, is Kant given a place in this volume, as he is
understood as the maker of a (transcendental) logic that is at odds with formal
logic, and seeks to overcome its flaws, which stem from a debilitating rational-
ism and empiricism?
This question reveals the dominance of a purist concept of logic, and so any
other logic that deviates from it (such as Kantian logic) immediately triggers
the need for some sort of justification, which often amounts to stating that it
is not really logic (only formal logic is “real” logic), but nonetheless useful for
other contemporary philosophical debates. This is indeed what we find in the
chapter on Kant, but it is indicative of the way contemporary philosophers of
logic treat logic historically. Thus, the value of Kant’s transcendental logic is
interpreted (ahistorically) as Wittgensteinian, and so, for instance, “inference”
in Kant is “not based on comparison of ideas, but [as in Wittgenstein] on the
application of a rule.”6 Not only does such an interpretation debar logics that
deviate from formal logics from even being considered valid or useful analyses
of thought and reasoning, but it also carries a thoroughly unhistorical attitude
which sees in the past only the reflection of the present, and so amounts to a
tautological assertion of present philosophical beliefs.
The chapter on Hegel in Handbook of the History of Logic is very much in
the same spirit as the chapter on Kant, thereby contradicting the overall ambi-
tions of the handbook, and at the same time showing how marginalized ideal-
ist logics have become, and how much intellectual authority formal logic in
the analytical tradition commands.7 The chapter on Hegel argues that accord-
ing to Hegel, logic is finite and imperfect, and so for Hegelian logic: “Human
reasoning is nothing else but this dynamic of lived experience as it has been
distilled into the essence of thought.”8 However, the chapter adds that there
is a big debate between those Hegelians who argue that “logic is essentially a
5 Tiles, “Kant,” 102.

6 Tiles, “Kant,” 102.
7 John W. Burbidge, “Hegel’s Logic,” in Handbook of the History of Logic: Volume 3: The Rise of
Modern Logic from Leibniz to Frege, ed. Dov M. Gabbay & John Woods (Amsterdam: Elsevier,
2004), 131–177. Contrast with Errol E. Harris, An Interpretation of the Logic of Hegel (Lanham:
University Press of America, 1983); G.R.G. Mure, A Study of Hegel’s Logic (Oxford: Clarendon,
1950); W.J. Mander, British Idealism: A History (Oxford: Oxford University Press, 2011),
551–552.
8 Burbidge, “Hegel’s Logic,” 170.
4 skodo
metaphysics,” and those who propound a non-metaphysical reading.9 Again,

what we are presented with is an interpretation of idealist logic as not really
being logic, that is, an accurate description of human reasoning and thinking.
The broad picture we are presented with in the Handbook boils down to invit-
ing us to understand real logic as stemming from analytical philosophy, which
takes on the role of expressing timeless accounts, valid regardless of historical
context or context of thought. If this is not enough to convince the reader of
the dominance of the three assumptions about logic adumbrated above, in
the more recent collection of essays entitled The Development of Modern Logic
we witness purism and universalism coming to the fore in the contest over
the meaning of logic.10 This book is intended as a rival to the widespread 1962
The Development of Modern Logic, which shares with The Development the
same purist and universalist attitude,11 which is why we will focus only on
the more recent one. It is strikingly evinced in this collection that the purist-
universalist conception of logic corresponds to some extent to, and so is based
on, the way people actually reason:
When we state in everyday language that a person’s logic fails, we nor-

mally mean that the rules of valid reasoning, which ought to guide our
thinking, are not in action for some reason. The word ‘logic’ of our every-
day language can usually be analyzed as ‘the collection of rules that guide
correct thinking or reasoning.’ That collection is assumed to be known
naturally; a rational human being follows these rules in normal circum-
stances, even if he or she could not formulate them, that is, express them
in language.12
Moreover, it is claimed that this concept of logic corresponds to the way peo-
ple have reasoned in the past (philosophers in the main). Thus, in a passage
worth quoting in full:
9 Burbidge, “Hegel’s Logic,” 170.

10 Leila Haaparanta, “Introduction,” in The Development of Modern Logic, ed. Leila
Haaaparanta (Oxford: Oxford University Press, 2009), 3–11.
11 William Kneale and Martha Kneale, The Development of Logic (Oxford: Clarendon Press,
1962). See also Ernst Mayr, “When is Historiography Whiggish?,” Journal of the History of
Ideas 51 (1990), 301–309, 304. Mayr emphasizes that Kneale and Kneale’s strategy in writ-
ing the book has been to trace the origins of only the philosophy of logic that resonates
with the philosophy of logic of their day.
12 Haaparanta, “Introduction,” 3.
Introduction 5
In each period in the history of logic, researchers called logicians have

been interested in concepts or terms that are not empirical, that is, whose
meanings are not, at least not incontestably, based on sensuous experi-
ence, and that can be called logical concepts or terms. What concepts or
terms have been regarded as logical has varied in the history, but interest
in them unites Aristotle, William of Ockham, Immanuel Kant, and Frege as
well as logicians in the twentieth and twenty-first centuries. Other points of
interest have been the so-called laws of thought, for example, the law of
non-contradiction and the law of excluded middle. A third theme that
unites logicians of different times is the question of the validity of rea-
soning [emphasis added].13
To put it succinctly: “modern logicians believed that there is one and only one
true logic.”14 The Development, however, argues that this particular purism-
universalism has been rejected after Leibniz. But what has been rejected is
not the purist-universalist disposition, but rather the belief that natural lan-
guage harbors logic in its purity and universality. After Leibniz came a great
dissatisfaction with natural language, and the reasoning waged within such
a form. Logicians after Leibniz opined that natural language could not ade-
quately mirror the world, and so they tried to construct artificial languages that
could perform that most difficult task. With the Handbook, The Development
agrees that at one point logic took a “mathematical turn.” Modern philosophi-
cal logic thereby became a “science” that studies valid reasoning “in artificial
languages.”15 How The Development of Modern Logic connects the belief that
logic is about how people actually reason, to the belief that logic is about an
artificial language that deems the way people reason fundamentally flawed is
neither explained nor justified.
The chapter “The Philosophy of Alternative Logics” in The Development is
one that, in our view, rightly propounds the need to acknowledge the rather
fluid and plural nature of logic. To some extent, it represents the line of inquiry
philosophers such as Susan Haack have pursued.16 Thus, the genus “logic” is
extended to entail the species “intuitionistic logic,” “quantum logic,” “relevance
13 Haaparanta, “Introduction,” 4–5.

14 Haaparanta, “Introduction,” 5. Cf. Gillian Russell, “One True Logic?,” Journal of
Philosophical Logic 37 (2008), 593–611.
15 Haaparanta, “Introduction,” 3.
16 Susan Haack, Deviant Logic, Fuzzy Logic: Beyond the Formalism (Chicago: University of
Chicago Press, 1996); Philosophy of Logics (Cambridge and New York: Cambridge Univer-
sity Press, 1978).
6 skodo
logic,” “paraconsistent logic,” albeit all as formalized systems.17 The chapter

argues that the “truth-functional propositional calculus” that Frege devised
“rapidly came to be applied to the assessment of natural argumentation, even-
tually achieving a near hegemony in this role.” The problem with this argument
is that it suggests that there have never been any alternatives to this hegemony,
which we argue is false. But to be fair, the article shows that such alternatives
did in fact exist, e.g. in the British logician and political economist Jevons’s
critique of formal logic from 1876. It therefore comes as a surprise that the
authors of this chapter conclude that “Jevon’s appraisal of the state of tradi-
tional logic a century and a quarter ago might as readily be applied to classical
logic today,” especially his dictum that “ ‘we shall do well to learn the old rules,
which are certainly both ingenious and useful.’ ”18
From reviewing these works, which arguably stand for widespread (and
often paradoxical) attitudes and beliefs about logic, we find purist, perfection-
ist, and universalist dispositions structuring ideas of what logic is, with the
occasional forays into the territory of idealism and alternative logics. The con-
clusion we draw from this survey is that logics other than formal logics (in the
three senses described above) tend to be either marginalized, or else not treated
as “real” logics, that is, they are not treated as accurate descriptions of thought
and reasoning. A major implication ensues from this marginalization and
intellectual ostracism, and it is that if prima facie instances of thought do not,
upon philosophical scrutiny, bear formalization, they are by default not worthy
of being labelled thought (or reasoning, or arguing). We situate other logics at
center stage, and in doing so question their marginalization by formal logics.
Before we move on to give an overview of the structure of the book, and
its individual chapters, we wish briefly bring to the fore the fact that formal
logics have been questioned from a variety of disciplines in the human sci-
ences. There are, thus, proponents of alternative logics that do not fit at all in
the picture painted above, even under discussions of “alternative logics.” For
instance, we find the attempt to construct a “narrative logic,” which exhorts us
to realize that historical narratives are semantically constituted in such a way
as to warrant a new logic.19 The leading philosopher of narrative logic, Frank
Ankersmit, extends in his chapter in this volume, entitled “Representationalist
17 Andrew Aberdein and Stephen Reid, “The Philosophy of Alternative Logics,” in The
Development of Modern Logic, ed. Leila Haaparanta (Oxford: Oxford University Press,
2009), 613–724.
18 Aberdein and Reid, “The Philosophy,” 613.
19 Frank Ankersmit, Narrative Logic: A Semantic Analysis of the Historian’s Language
(The Hague: Martinus Nijhoff, 1983).
Introduction 7
Logic,” his views on logic in the sciences. Another example is the attempt to
construct “the logic of the history of ideas” that starts out from the way his-
torian’s actually reason and use language in order to, based on Wittgenstein,
draw out the procedures appropriate to such reasoning. Such a logic crucially
differs from formal logic in appealing to concrete human traditions and beliefs
formed within those traditions as the fundamental schemas in which reason-
ing takes place.20
Another important contemporary alternative logician is Jaakko Hintikka
who, drawing, among others, on the British idealist R.G. Collingwood, argues
that “we have to make a new start in practically all branches of philosophical
studies including logic, foundations of mathematics, language theory, episte-
mology, and philosophical methodology.”21 Hintikka sees the major problem of
prevalent formal logics as being too narrow to incorporate a logic of knowledge
acquisition. Moreover, formal logics cannot cope with the fact that thinking is
a multi-dimensional activity, involving social and various cognitive faculties,
which must be taken into account if logic is to be able to explicate a logic of
knowledge acquisition.22 As it stands now, the purism, perfectionism, and uni-
versalism of logic can only be defended with a view of logic that is meagre, one
Hintikka calls a logic of definitory rules.
Formal logics have also been questioned from the field of contemporary
political and social philosophy.23 Anglo-American political philosophers have
argued that if we are to understand how politics and political thought proceeds
and constitutes political and social reality, then we must acknowledge that for-
mal logic and dialectical logic are ill equipped on their own to enable such an
understanding. Michael Freeden, for example, has tied logic to rationality in
seeing how it operates in the study of ideology. He arrives at a starting point we
share with him: “In the course of the history of human thought rationality has
signified rightness, or moderation, or self-willing and autonomy, or calculated
means-end purposiveness, or obedience to the law of God—terms which,
20 Mark Bevir, The Logic of the History of Ideas (Cambridge: Cambridge University Press,
1999).
21 Jaakko Hintikka, “Who is About to Kill Analytic Philosophy,” in The Story of Analytic
Philosophy: Plot and Heroes, ed. Anat Biletzki and Anat Matar (London and New York:
Routledge, 1998), 253–269, 260. Another prominent philosopher who was criticized for-
mal logic is P.F. Strawson in his Introduction to Logical Theory (London: Methuen, 1960).
22 Jaako Hintikka, Inquiry as Inquiry: A Logic of Scientific Discovery (Dordrecht: Kluwer,
1999).
23 Michael Freeden, Ideologies and Political Theory: A Conceptual Approach (Oxford: Oxford
University Press, 1998); Ernesto Laclau & Chantal Mouffe, Hegemony and Socialist Strat-
egy: Towards a Radical Democratic Politics (London, New York: Verso, 1984).
8 skodo
though all different, are not necessarily mutually exclusive.”24 Indeed, we must
start out from this kind of plurality and make sense of it, and not start out from
a purist, perfectionist, and universalist ideas of logic and then impose them on
this plurality. Though acknowledging that “Logic unquestionably plays a key
role both in the philosophical formation of an argument and in the evaluation
of its validity and persuasive power,” Freeden draws out the consequence logi-
cal perfectionism and purism would have for political thought if perfection-
ism and purism were to dictate the boundaries of arguing: “In the first case,
political theories as well as ideologies simply cannot bear the full weight of the
meticulous logical analysis directed at them by some philosophers.” Freeden
then continues: “Logic and consistency must remain important, but not over-
whelming, criteria for the assessment of arguments. Logical perfectionism can
be detrimental to the optimalization of analytical insight,” and so “We will
need to readmit the role of the emotional as well as the intellectual attrac-
tiveness of arguments, and we will have to examine cultural as well as logical
validations of political thinking.”25 Many more scholars can be adduced that
indirectly have such a critical view of logic, within fields such philosophy of
religion, and ontology.26
This book situates itself in the literature that attempts to overcome the limi-
tations of formal logics. However, it differs from the works discussed in being
more comprehensive and wide-ranging than they are. This range is reflected in
the two parts of the book (entitled “Perspectives from the History of Thought”
and “Perspectives from Contemporary Philosophy”), in that they show how
the meaning, scope, and application of logic has been tackled, and contested,
from a variety of both past and present approaches. What is noteworthy is that
both continental and analytical perspectives are represented in this collection,
thus showing that these two traditions decidedly share similar concerns, and
are not worlds apart. Another noteworthy feature of this book is the variety of
philosophers and philosophical topics it discusses in relation to logic. Among
these are Martin Heidegger, F.C.S. Schiller, Quentin Meillassoux, Jean Luc-
Nancy, Jacques Lacan, love, the unconscious, historical representation, logical
aliens, the concept of God, normative judgment internalism, and m etaphysical
24 Freeden, Ideologies, 149.

25 Freeden, Ideologies, 37.
26 See e.g. Peter Dews, Logics of Disintegration: Post-structuralist Thought and the Claims
of Critical Theory (London and New York: Verso, 1987); Pamela Sue Anderson, A Feminist
Philosophy of Religion: The Rationality and Myths of Religious Belief (Oxford: Blackwell,
1998); Alvin Plantinga, Does God Have a Nature? (Milwaukee: Marquette University Press,
1980); Alain Badiou, Being and Event (London: Continuum, 2006).
Introduction 9
realism. By explicitly attending to the inherent bond between logic and extra-
logical human aspects, yet another valuable contribution arises from the book
in that it shows that logic too can be tied to social, religious, and political beliefs
and purposes. Many valuable studies in the late twentieth century have shown
how ideas of “reason”, “thinking”, “rationality,” and “God” are intimately tied
to such contexts, indeed justify them by means of masking them into seem-
ingly objective and universal traits of human thought or logic.27 But most of
these studies have neglected to take into account how logic has been part of
such purposes too, which implies that logic is an entity that possesses a nature
apart from normative forms of thought, such as political ideology or religious
theology. All chapters in this book, by contrast, acknowledge the normativity
inherent to logic. However, the chapters also show that despite the fact that
any logic is amenable to change or even wholesale rejection (see especially
the chapter by Tanesini), different logics do have philosophical value, they do
further or make clear the presuppositions of thinking, which means that we
should embrace the fact that logic exists in the plural, as necessarily meshed
with other aspects and dimension of the mind, society, politics, and culture.
The studies in this volume together challenge the purist, perfectionist, and
universalist perspectives on logic by disclosing two features of logic. First, that
the formal logics of the twentieth century have been thoroughly criticized by
competent and astute thinkers.28 Thus, based on these historical inquiries we
can safely assert that no formal logic has ever been intellectually hegemonic.
In other words, the book counters the universalistic ambitions of, for exam-
ple, the philosophies of Bertrand Russell, Gottlob Frege, and logical positivism.
Second, the chapters resolutely argue, either historically or philosophically, for
“other” logics or “counter-logics,” or more broadly still, conceptions of reason
and rationality that transgress the conceptual boundaries of formal logic.29
The chapters in the first part reveal the conceptual breadth of other log-
ics in the twentieth century. Christopher Watkin—in “Proving the Principle of
Logic: Quentin Meillassoux, Jean-Luc Nancy, and the Anhypothetical”—deftly
analyzes Quentin Meillassoux’s attempt to revive the Aristotelian principle
of non-contradiction, or the “anhypothetical” principle. As an alternative to
Meillassoux, Watkin posits Jean-Luc Nancy’s reading of Plato through which
Nancy inserts love as the simultaneous origin and impossible object of logic,
27 John Passmore, “Some Critics of Formal Logic,” in A Hundred Years of Philosophy (London:
Penguin Books, 1966), 156–173.
28 See e.g. the chapters by Wendland, Skodo, Watkin, Cejvan, and Ankersmit.
29 See e.g. the chapters by Ankersmit, Tanessini, Svensson, and Kraal, described in more
detail below.
10 skodo
leading to a highly original destabilizing interpretation of the principle of

non-contradiction. Watkin offers us a critical and important insight into the
way in which two leading French philosophers think about logic. The chapter
by Admir Skodo—“The Self, Ideology, and Logic: F.C.S. Schiller’s Pragmatist
Critique of and Alternative to Formal Logic”—gives an intellectual-historical
account of F.C.S. Schiller’s pragmatist-humanist attack on formal logic and his
attempt to construct a “humanist logic,” emphasizing the strange consequences
of Schiller’s commitments to historicism and advances in applied natural sci-
ence. Aaron Wendland’s chapter—“Language, Truth and Logic: Heidegger on
the Practical and Historical Grounds of Abstract Thought”—gives an interpre-
tation, inspired by Hubert Dreyfus, of Martin Heidegger’s critique of formal
logic as a pragmatist-existentialist one. Ervik Cejvan’s chapter—“The Obstacle:
Jacques Lacan’s Critique of the Formal Logical Representations of the Real”—
discusses Lacan’s critique of formal logic in the context of his psychoanalytical
theory. In so doing, Cejvan attempts to clarify certain aspects of Lacan’s thought
that have proven to be difficult to render comprehensible, such as his concept
of the “real” and his views on sexual difference. In “Collingwood’s Logic of
Question and Answer Against the Relativization of Reason,” Christopher Fear
starts out from the British idealist R.G. Collingwood’s claim that “all logic is
concerned with discussions,” where the aim is to establish a correct answer to
shared questions—hence Collingwood’s “logic of question and answer.” Fear
shows how Collingwood’s logic was a forceful “revolt” against and alternative
to various formal and abstract logics, whether stemming from the realist, ide-
alist, or analytical traditions in philosophy. He also explores certain neglected
aspects of Collingwood’s logic and argues that they might harbor more prom-
ise than is usually acknowledged.
The chapters in the second part by explore the role of logic in contexts for-
mal logic tends to leave out, with some fascinating insights about the nature
of “folk ontology,” historical writing, concepts, moral action, and indeed logi-
cal principles, that stand in stark contrast to the principles of various formal
logics. In his chapter entitled “Representationalist Logic,” Frank Ankersmit
lays down the outline of a logic which he argues best captures the presup-
positions of historical representation, and which challenges what Ankersmit
calls the “Whig interpretation of logic,” which rests on Aristotelian and mod-
ern formal logic. Representationalist logic, by contrast, captures the best of
both Aristotelian and modern formal logic, but cannot be reduced to either
of them. Alessandra Tanesini—in “On Logical Aliens”—argues that there can
be logical aliens, that is, creatures who think in none of the laws, principles, or
rules of deductive logic deemed to be necessary for thought by thinkers such as
Frege and Putnam (such as the principle of non-contradiction, also discussed
Introduction 11
by Watkin) whom she calls “logical absolutists.” Her argument goes one step
further than the “logical pluralism” of J.C. Beall and Greg Restall in defend-
ing the view that there can be incompatible and rivalrous logics, which are
nonetheless all admissible as the normative laws governing thought. Thord
Svensson—in “The Heart of Metaphysical Pluralism and the Consistency
Dilemma: A Critical Analysis of the Possibility of Incompatible Truths”—
discusses through a critique of Michael Lynch (whom Tanesini also invokes),
Terry Horgan and Mark Timmons, what implications metaphysical pluralism
has for absolutist and perfectionist accounts of logic. He arrives at the conclu-
sion that even though these conceptions of logic do not hold, we still have not
arrived at a position which robustly establishes that at least two incompatible
yet both true accounts of reality are possible. In “The Logic of ‘Oughts’ and the
Bindingness of Past Practice: A Critique of Normative Judgment Internalism
through a Reading of King Lear’s Act I,” Karim Dharamsi takes issue with Ralf
Wedgwood’s internalist treatment of the moral self as a logical operator “ampu-
tated,” in Dharamsi’s words, from history and custom. Contra Wedgwood—
and logical purism and absolutism—Dharamsi provocatively argues, based on
a reading of Shakespeare’s King Lear, that any logic of “oughts” must incorpo-
rate the constraints of historical tradition and socio-political context. Anders
Kraal—in “First-Order Logic, Incongruism, and Anti-Formalism”—argues that
classical theistic doctrines are on the one hand, incongruous with formal logic;
and on the other hand, that they are best analyzed by means of informal logic.
Kraal thus suggests that the concept of God warrants the critique of formal
logic. Johan Modée—in “Zombies, Selves, and the Possibility of Afterlife”—
challenges the applicability of a logical argument (the conceivability argu-
ment or, as Modée colorfully calls it, “zombie argument”) to arguments about
the existence of divine beings, and in so doing can fruitfully be read as a riposte
to Kraal. Modée and Kraal present us with two conflicting views on what con-
tent logic can rest on, empirical (Modée) or non-empirical (Kraal), and reveal
how debates about logic in the philosophy of religion cannot be settled by an
agreed upon foundation. In “The Logocentric Predicament and the Logic of
Question and Answer,” Giuseppina D’Oro uses Collingwood’s logic of question
and answer and his arguments about the nature of metaphysical propositions
in scientific inquiry (addressed historically by Fear in the first part), to argue
that in doing away with a certain non-question begging kind of circularity as
a counter-argument, Collingwood’s logic is able to meet the “logocentric pre-
dicament,” that is, the problem that “there is no non-circular justification of
deductive inference.” The logic of question and answer can do so because it
identifies the predicament as being generated by the charge of circularity
itself, which renders that charge open to critique. The type of Collingwoodian
12 skodo
logic that D’Oro defends becomes a philosophical method by which the meth-
odologically, but not historically, fundamentally different modes of inquiry can
be discerned.
Bibliography
Andrew Aberdein and Stephen Reid, “The Philosophy of Alternative Logics,” in The
Development of Modern Logic, ed. Leila Haaparanta (Oxford: Oxford University
Press, 2009), 613–724.
Pamela Sue Anderson, A Feminist Philosophy of Religion: The Rationality and Myths of
Religious Belief (Oxford: Blackwell, 1998).
Frank Ankersmit, Narrative Logic: A Semantic Analysis of the Historian’s Language
(The Hague: Martinus Nijhoff, 1983).
Alain Badiou, Being and Event (London: Continuum, 2006).
Jon Barwise and John Etchemendy, Language, Proof, and Logic (Chicago: The University
of Chicago Press, 2002).
Mark Bevir, The Logic of the History of Ideas (Cambridge: Cambridge University Press,
1999).
John W. Burbidge, “Hegel’s Logic,” in Handbook of the History of Logic: Volume 3: The
Rise of Modern Logic from Leibniz to Frege, ed. Dov M. Gabbay & John Woods
(Amsterdam: Elsevier, 2004), 131–177.
John P. Burgess, Philosophical Logic (Princeton: Princeton University Press, 2009).
———, The Development of Modern Logic, ed. Leila Haaaparanta (Oxford: Oxford
University Press, 2009).
Peter Dews, Logics of Disintegration: Post-structuralist Thought and the Claims of
Critical Theory (London and New York: Verso, 1987).
Michael Freeden, Ideologies and Political Theory: A Conceptual Approach (Oxford:
Oxford University Press, 1998).
Handbook of the History of Logic: Volume 3: The Rise of Modern Logic from Leibniz to
Frege, ed. Dov M. Gabbay & John Woods (Amsterdam: Elsevier, 2004).
Susan Haack, Deviant Logic, Fuzzy Logic: Beyond the Formalism (Chicago: University of
Chicago Press, 1996).
———, Philosophy of Logics (Cambridge and New York: Cambridge University Press,
1978).
Leila Haaparanta, “Introduction,” in The Development of Modern Logic, ed. Leila
Haaaparanta (Oxford: Oxford University Press, 2009), 3–11.
Errol E. Harris, An Interpretation of the Logic of Hegel (Lanham: University Press of
America, 1983).
Introduction 13
Jaako Hintikka, Inquiry as Inquiry: A Logic of Scientific Discovery (Dordrecht: Kluwer,

1999).
———, “Who is About to Kill Analytic Philosophy,” in The Story of Analytic Philosophy:
Plot and Heroes, ed. Anat Biletzki and Anat Matar (London and New York: Routledge,
1998), 253–269.
Anssi Korhonen, Logic as a Universal Science: Russell’s Early Logicism and its
Philosophical Context (Basingstoke and New York: Palgrave Macmillan, 2013).
William Kneale and Martha Kneale, The Development of Logic (Oxford: Clarendon
Press, 1962). See also
Ernesto Laclau & Chantal Mouffe, Hegemony and Socialist Strategy: Towards a Radical
Democratic Politics (London, New York: Verso, 1984).
W.J. Mander, British Idealism: A History (Oxford: Oxford University Press, 2011),
551–552.
Ernst Mayr, “When is Historiography Whiggish?,” Journal of the History of Ideas 51
(1990), 301–309.
G.R.G. Mure, A Study of Hegel’s Logic (Oxford: Clarendon, 1950).
John Passmore, “Some Critics of Formal Logic,” in A Hundred Years of Philosophy
(London: Penguin Books, 1966), 156–173.
Alvin Plantinga, Does God Have a Nature? (Milwaukee: Marquette University Press,
1980).
Gillian Russell, “One True Logic?,” Journal of Philosophical Logic 37 (2008), 593–611.
P.F. Strawson, Introduction to Logical Theory (London: Methuen, 1960).
Mary Tiles, “Kant: From General to Transcendental Logic,” in Handbook of the History
of Logic: Volume 3: The Rise of Modern Logic from Leibniz to Frege, ed. Dov M. Gabbay
& John Woods (Amsterdam: Elsevier, 2004), 85–130.
part one
Perspectives from the History of Thought
∵
chapter 1
Proving the Principle of Logic

Quentin Meillassoux, Jean-Luc Nancy, and the Anhypothetical
Christopher Watkin
1 The Anhypothetical Principle in Plato and Aristotle
Since the birth of philosophy itself, thinkers have been searching for the prin-
ciple of all logic: not a logical proof but the proof of the validity of logic itself.
The search is for a justification of logic that is not waiting for us at the end of
a chain of reasoning but that validates all reasoning in the first place. Both
Plato and Aristotle call this principle of logic the “anhypothetical principle”:
the principle that does not need to be hypothesized.1
The anhypothetical for both Plato and Aristotle is not simply an object
of dialectic, but at the origin of dialectic. For the Plato of Republic book 6,
hypotheses and conclusions belong to διάνοια, the faculty of logical rea-
soning, whereas the anhypothetical principle is apprehended with νους, a
direct intellectual intuition of the truth.
In the lower half of Plato’s division of the intelligible, the part exercising
διάνοια,
the soul, using what was imitated earlier as images, is forced to investi-
gate from hypotheses, not to a first principle, but to a conclusion. But in
the other half it proceeds from hypotheses to an anhypothetical first
principle and without the images of the former, making methodical prog-
ress through the Ideas to the Ideas themselves.2
In the case of διάνοια, its “hypotheses” include “the odd and the even, and the
figures, and the three kinds of angles and the other cogeners of these in each
1 See principally Plato, The Republic, trans. Paul Shorey (Cambridge, ma: Harvard University
Press, 1937), 510b4–9, 510c3–d2, 511b3–c2; and Aristotle, The Metaphysics: Books i–ix, trans.
Hugh Tredennick (Cambridge, MA: Harvard University Press, 1980), 1005b.
2 Plato, Republic 6, 510b4–9.

18 watkin
field of inquiry.”3 They are regarded as obvious to everyone, and so the math-
ematician does not think it worthwhile giving an account of them. As for
the “other half” of the intelligible, the half exercising νους, it attains knowledge
not in the complete absence of hypotheses, but in a way that uses hypotheses
as a Wittgensteinian ladder, in order to go beyond them:
Understand that by the other segment of the intelligible I mean that

which logos itself grasps by the power of dialectic, not using the hypoth-
esis as the first principle, but literally as an hypothesis in order that it may
go as far as the anhypothetical to the first principle of all things and
grasping it, again clinging to the things that cling to it, may descend to a
conclusion, making no use at all of the objects of sense, but of Ideas
themselves, proceeding through Ideas, to Ideas and ending in Ideas.4
And of course for Plato the first principle of all things is the Form of the Good,
the Good beyond being.
Turning to Aristotle, we find an anhypothetical somewhat differently for-
mulated. For Aristotle the anhypothetical is not the highest Idea or the Good
beyond being but rather the assumption behind all reasoning, an assump-
tion which neither is nor can be directly established by any hypothesis. For
Aristotle, insofar as the philosopher is “the student of the whole of reality in its
essential nature,”5 it is the philosopher’s task to investigate also the principles
of syllogistic reasoning about what is. And what is the most certain principle
of Being qua Being?
the most certain principle of all is that about which one cannot be mis-
taken; for such a principle must be both the most familiar (for it is about
the unfamiliar that errors are always made), and anhypothetical. For the
principle which the student of any form of Being must grasp is no hypoth-
esis; and that which a man must know if he knows anything he must
bring with him to his task. Clearly, then, it is a principle of this kind that
is the most certain of all principles.6
3 Plato, Republic 6, 510c3–d2.

4 Plato, Republic, 6, 511b3–c2.
5 Aristotle, Metaphysics 4, 1005b.
Proving the Principle of Logic 19
This principle, then, is not the result of thinking but it is brought to the task of
thinking in the first place. It is not thinking’s product but its principle, and for
Aristotle, the principle in question is non-contradiction:
Let us next state what this principle is. ‘It is impossible for the same attri-
bute at once to belong and not to belong to the same thing and in the
same relation’; and we must add any further qualifications that may be
necessary to meet logical objections. This is the most certain of all prin-
ciples, since it possesses the required definition; for it is impossible for
anyone to suppose that the same thing is and is not, [. . .] for the man who
made this error would entertain two contrary opinions at the same time.
Hence all men who are demonstrating anything refer back to this as an
ultimate belief; for it is by nature the starting-point of all the other axi-
oms as well.7
So for Aristotle the anhypothetical is 1) brought by the philosopher to any

analysis she undertakes, and 2) something about which it is impossible
to be mistaken. Aristotle refers us to Heraclitus as one who professes to say
that the same thing is and is not, but dismisses the contention on the grounds
that, in his very attempt to contradict the law of non-contradiction, Heraclitus
affirms it. In Aristotle’s eyes, Heraclitus affirms in the form of what he says the
very principle he seeks to deny by the content of what he says.
This, for Aristotle, is an indirect proof of the principle of non-contradic-
tion. The principle is not itself demonstrated to be originary (how could it
be, other than by the very principle which is the object of the demonstra-
tion?), but it can be demonstrated that it is impossible to deny it, for any
denial will always already have affirmed in the form of what it says the very
principle it seeks to deny by the content of what it says.
2 Quentin Meillassoux’s Anhypothetical
In one contemporary re-working of this theme of the anhypothetical,

Quentin Meillassoux seeks to strengthen the Aristotelian form of the anhy-
pothetical and offers an anhypothetical demonstration the principle of
non-contradiction which, he claims, is of unprecedented rigour. Both in his
published After Finitude and his eagerly anticipated The Divine Inexistence,

20 watkin
Meillassoux claims to establish philosophy on the basis of a demonstrated

anhypothetical principle.
Rejecting Plato’s anhypothetical idea of the Good in favour of Aristotle’s
indirect proof, Meillassoux seeks to arrive at a primary proposition which
cannot be deduced from any other propositions, and yet which can be dem-
onstrated. He sensibly begins by conceding that the demonstration of the
anhypothetical must necessarily be an indirect proof, or the anhypothetical
would no longer be primary:
This proof, which could be called ‘indirect’ or ‘refutational,’ proceeds not

by deducing the principle from some other proposition—in which case it
would no longer count as a principle—but by pointing out the inevitable
inconsistency into which anyone contesting the truth of the principle is
bound to fall. One establishes the principle without deducing it, by dem-
onstrating that anyone who contests it can do so only by presupposing it
to be true, thereby refuting him or herself.8
Meillassoux rejects all attempts to understand the beginning of reason as an

axiom, postulate, thesis or, in Plato’s sense, a hypothesis. Instead, he claims to
have a founding principle of rational thought which is not itself posited irratio-
nally or contingently but is founded in reason. He seeks to go beyond Aristotle,
for whereas Aristotle only proves that contradiction is unthinkable, not impos-
sible, Meillassoux wants to go one further and demonstrate the impossibility
of contradiction. In other words, he claims not simply that his principle is
anhypothetical, but that it is also absolute. In this, he readily admits that he is
proceeding against the unanimous consensus of contemporary thought that
dismisses any anhypothetical principle of thought as an illusion.
The name he gives to this principle for which we have been waiting since
Plato is the principle of factiality. The principle is elaborated not in terms of
contradiction and noncontradiction, but in terms of necessity and contin-
gency. There is no metaphysical law, Meillassoux asserts, to guarantee any nec-
essary being or necessary law. If there is no necessary being, then it follows that
facticity (or contingency, the possibility of a thing being otherwise than it is) is
not itself a contingent fact, because if facticity were itself a fact there could be
a necessary being. So there is no necessity, except the necessity of contingency
itself:
8 Quentin Meillassoux, After Finitude: An Essay on the Necessity of Contingency, trans. Ray
Brassier (London and New York : Continuum, 2008), 61.
what is, is a fact [est factuel], but that what is is a fact, this itself cannot be
a fact. Only the facticity of what is cannot be a fact. Or again, in other
words: it cannot be a fact that what is is a fact [. . .] The contingency of
beings, and it alone, cannot be a contingent property of that being.9
The principle of factiality is precisely this: that facticity cannot be a fact. An

example: We may say that an object is de facto red, but not that it is de facto de
facto.10 Factiality is the non-facticity of facticity,11 or the non-contingency of
contingency, and contingency is itself therefore necessary in order to avoid
a necessary being which, Meillassoux argues, after the death of God, we
have no grounds to admit into our thinking. In order to avoid falling back
into metaphysics, Meillassoux stresses that the principle of factiality does
not merely maintain that contingency is necessary, but that only contin-
gency is necessary,12 as a direct correlate of the absence of any necessary
being, event or law.
Meillassoux is at great pains to stress that the necessity of contingency does
not replace the “laws of nature” with a meta-law of contingency itself: “there
is no law of becoming, because there is the becoming of laws.”13 So although
the necessity to which Meillassoux appeals is eternal, this eternity does
not signify the eternity of the laws of becoming, but rather the eternity of
the becoming of laws.14 In an address entitled “Temps et surgissement ex
nihilo” (“Time and appearance ex nihilo”) Meillassoux describes this state
of affairs as an inverted Platonism: there is an illusory fixity of objects but
a real contingency “behind” that fixity; the intelligible is on the side of the
most radical becoming, the sensible on the side of fixity.
To prove that the principle of factiality is both anhypothetical and abso-
lute, Meillassoux summons forth an imaginary interlocutor, whom he calls
a correlationist. The correlationist holds that the world as it is perceived is
always already correlated to our faculties of apprehension, and while the
correlationist is ready to admit that contradiction is unthinkable, he denies
that he must conclude from this that it is also impossible. Meillassoux
responds to the correlationist with an indirect proof of the impossibility of
9 Quentin Meillassoux, “L’Inexistence divine” (PhD Diss., Université de Paris I, 1997), 44.
CW’s translation.
10 Meillassoux, “L’Inexistence divine,” 46. CW’s translation.
11 Meillassoux, After Finitude, 79.
13 Meillassoux, L’Inexistence divine, 5. CW’s translation.
14 Meillassoux, L’Inexistence divine, 158. CW’s translation.
22 watkin
contradiction, showing that the correlationist can only contest the principle of
factiality if he already presupposes that it is absolute:
The sceptic is only able to conceive of the difference between the ‘in-
itself’ and the ‘for-us’ by submitting the ‘for-us’ to an absence of reason
which presupposes the absoluteness of the latter. It is because we can
conceive of the absolute possibility that the ‘in-itself’ could be other than
the ‘for-us’ that the correlationist argument can have any efficacy.
Accordingly, the anhypotheticity of the principle of unreason pertains to
the ‘in-itself’ as well as to the ‘for-us,’ and thus to contest the principle is
already to have presupposed it.15
In other words, the correlationist says that something may be unthinkable but
not impossible, but (says Meillassoux) to say that something is unthinkable
and not impossible is to admit that the “in-itself” could be radically different
from the “for-us” or, put another way, it is to admit that there is an absolute
possibility that the “for-itself” is contingent, without a necessary law of rea-
son . . . thus proving the principle of factiality. Meillassoux is establishing the
absolute contingency of the in-itself with an indirect proof: factiality is always
presupposed by any attempt to deny it.
This position differs only in one very small point from hypothetical rea-
son, Meillassoux concludes. Whereas hypothetical reason establishes itself
as contingent, without sufficient reason, the principle of factiality goes
one step further in order to show the necessity of that contingency, its very
essence. What hypothetical reason merely states, Meillassoux’s anhypo-
thetical claims to demonstrate.
We have argued elsewhere that Meillassoux’s proof that unreason is
absolute is doubtful.16 It is our burden here to show that Meillassoux’s proof
of the principle of factiality is itself vulnerable to disproof by the same indi-
rect means through which he seeks to establish it. This argument will take
us on a detour through the thought of Jean-Luc Nancy.

16 See chapters 4–6 of Christopher Watkin, Difficult Atheism: Post-Theological Thinking in
Alain Badiou, Jean-Luc Nancy and Quentin Meillassoux (Edinburgh: Edinburgh University
Press, 2011).
3 Jean-Luc Nancy, Thought, and Love
Whereas Meillassoux seeks to develop Aristotle’s anhypothetical, Nancy

meditates on Plato and specifically on Plato’s great work on love, the
Symposium. In his scattered discussions of Plato’s text, Nancy makes six
observations in relation to love. The first relates to the structure of Plato’s
text. In his early work The Literary Absolute, Nancy makes the point that it
is inexact to call the Symposium a dialogue, for the text is in fact a narra-
tive incorporating (or recalling) a dialogue which is, in turn, interspersed
with speeches.17 In other words, the dialectic, thought or reasoning in the
text does not appear ex nihilo but is presented to the reader by a supporting
narrative context, a supporting gesture which carries and presents the dia-
lectic. This point about the necessity of the supporting gesture will become
crucial in our argument later on.
Secondly, in later works Nancy widens his comments on the Symposium to
consider it’s subject matter: love. Nancy rejects the simple view that love is
merely the topic for which Plato has reserved this particular treatise, which
would make of the Symposium as a sort of Platonic De Amore. Rather, the text
touches on the essence of thought itself:
The Symposium signifies first that for Plato the exposition of philosophy,
as such, is not possible without the presentation of philosophical love.
The commentary on the text gives innumerable confirmations of this,
from the portrait of Eros to the role of Socrates and to the figure—who
appeared here once and for all on the philosophical scene—of Diotima.
Although the Symposium speaks of love, it also does more than that; it
opens thought to love as to its own essence.18
One may philosophise about any number of subjects, but when it comes to
philosophizing about philosophy itself, it is (with a certain etymological appro-
priateness) to love that we must turn. In the mise en scène of the Symposium, it
is love that burdens the assembled guests to speak about love, a desire for truth
that leads them to speak of desire. Their dialectic, the very logic of their argu-
ments, is impelled by love. If we call the Symposium a treatise of love, then we
17 Philippe Lacoue-Labarthe and Jean-Luc Nancy, The Literary Absolute: The Theory of
Literature in German Romanticism, trans. Philip Barnard and Cheryl Lester (Albany: State
University of New York Press, 1988), 87.
18 Jean-Luc Nancy, A Finite Thinking, ed. Simon Sparks (Palo Alto: Stanford University Press,
2003), 248.
24 watkin
must intend both a subjective and an objective genitive. Nevertheless, as much

as it is love that propels the dialectic, love itself never takes the limelight: let us
not forget that Diotima, the decisive “speaker,” is absent, her words conveyed
to us by Socrates.
In the Symposium, Nancy argues, Plato’s thought touches its limits, not
in the sense that there are certain mystical topics about which he (and we)
cannot think philosophically, but in the sense of touching its own source,
its own truth; it thinks its own birth and its erasure, handing over its own
task to love.19 Philosophy and love, it seems, have momentarily changed
places, for “Instead of philosophy here being occupied with gathering and
interpreting the experiences of love, in the final analysis it is love that receives
and deploys the experience of thinking.”20
Thought, then, is not its own source. It is welcomed and deployed by
love, and when philosophy would take love as its object, in the very act of so
doing it reveals itself as love’s object.
Thirdly, the relation of love to thought is not a theme that Nancy develops
only in relation to the Symposium. More broadly, he seeks to understand
love as the movement of thought, where “ ‘love’ thus employed would be, so
to speak, existential rather than categorical, or again it would be the name of
the act of thinking as much or more than it would be its nature.”21
Love, here, is not a phenomenon or a property, nor an affect that thought
can isolate or contemplate. It is the act by and through which contempla-
tion itself arrives. It is love that invites the philosopher to think, that pro-
vokes her to contemplation, and without love there would therefore be no
dialectic, no logic:
The thinking of love—if it is necessary to solicit it, or if it is necessary

that it be proposed anew, as a theme to be discussed or as a question
to be posed—does not therefore lay claim to a particular register of
thinking: it invites us to thinking as such. Love does not call for a cer-
tain kind of thinking, or for a thinking of love, but for thinking in
essence and in its totality. And this is because thinking, most prop-
erly speaking, is love. It is the love for that which reaches experience:
that is to say, for that aspect of being that gives itself to be welcomed;
in the movement across discourse, proof, and concept, nothing but
19 Nancy, A Finite Thinking, 248.

21 Nancy, A Finite Thinking, 247–8.
this love is at stake for thought. Without this love, the exercise of the
intellect or of reason would be utterly worthless.22
What is at stake here is not a question for philosophy—“why is there some-

thing rather than nothing?,” “what is the good life?,” “what is the relation
between parts and wholes?”—but rather the question of philosophy—“why
is there philosophy rather than no philosophy?” or “why are the questions of
existence, or the good life, or parts and wholes, worth asking?” And the answer
to the questions “why dialectic?,” “why logic?,” “why philosophy?,” for Nancy,
is, quite simply, love. Love is the welcoming of the world, or the welcoming of
that which reaches experience, a welcoming which is necessary for thought
to bother thinking anything at all. Just as the Symposium has its own intradi-
egetical narrative gesture, the “framing” narrative which supports and presents
the Platonic dialogue, so also dialectic is borne along on its own ineradicable
gesture: when thought presents arguments and conclusions, the movement of
presentation itself is a movement of love.
The complicity of love and thought, Nancy continues, is evident in
the joining of φιλíα and σοφός in the word “philosophy” itself: the love of
wisdom. Love, therefore, is not the same as thought, but neither can it be
separated from thought, such that on the one hand we would have cold,
rational thought and on the other hand we have warm, sentimental and
faintly irrational love. The relation between thought and love is more
nuanced than that.
In his early text entitled The Categorical Imperative Nancy explores the
relation between love and truth. Love, he says, is that which, in the truth,
does not give itself to be known.
The reason it does not give itself to be known is not that it is mystical or
divine in some undefined way, but that it is the giving of all that is given to
be known, the movement by which anything that comes to be known does
so. And there is one thing that cannot be given, transacted, handed over,
presented, and that thing is the giving itself, the presentation itself.
In an attempt to gesture towards this loving moment of presentation or
giving, Nancy refers to “the imperative truth,”23 truth as a movement, not as
a product. Like the narrative gesture of the Symposium through which the
dialectic on the theme of love is delivered to us, love is the giving, the pre-
sentation of thought. So love cannot be reduced exhaustively to an object

23 Nancy, L’Impératif catégorique : La Philosophie En Effet (Paris: Flammarion, 1983), 90. CW’s
translation.
26 watkin
for philosophy’s contemplation, because it is also the movement through

which that object arrives.
Fourthly, it follows that the name of love cannot simply be the noun “love,”
for this would reduce love to a substance or a faculty, an object of logical
analysis. The name of love is “I love you,” or rather “the ‘I-love-you,’ in the
same way we talk about ‘the cogito.’ ”24
Love does not fully arrive, like a thought, rather it traverses and is always
missed by philosophy.25 Love is the essence of thought, in all its irreducible
multiplicity from erotic love through brotherly love to Christian neighbour-
love and beyond, but thought always misses its essence, always arrives just
too late to find love exposed. When philosophy does try to make an object
of love it is always, Nancy insists, a missed appointment, leaving philoso-
phy with one of love’s dismembered parts (sex, sentiment, self-sacrifice . . .)
or one of its sublimations (friendship, charity . . .).26 Philosophy cannot
address love directly, because love is philosophy’s address, its invitation, its
source.
Fifthly, if we widen our perspective a little we find that, although philoso-
phy cannot contemplate love as such, it is nevertheless always already in
love. Although philosophy always arrives too late for its meeting with love,
it nevertheless arrives in love and for the sake of love. In other words, while
philosophy may want to make an object of love to interrogate and exam-
ine by means of its own dialectic, it constantly finds itself always already
in love, desiring the interrogation and examination by virtue of which it
would examine desire:
This is why love is always missed by philosophy, which nevertheless does

not cease to designate and assign it. Perhaps it cannot help but be missed:
one would not know how to seize or catch up with that which exposes. If
thinking is love, that would mean (insofar as thinking is confused with
philosophy) that thinking misses its own essence—that it misses by
essence its own essence. In philosophy (and in mysticism, in poetics, etc.)
thinking would thus have said all that it could and all that it should have
said about love—by missing love itself. Loving, and loving love, it will
have lost love. It is thence that Saint Augustine’s amare amabam draws its
exemplary force of confession.27

26 Nancy, A Finite Thinking, 245–6.
27 Jean-Luc Nancy, The Inoperative Community, trans. Lisa Garbus, Peter Connor, Michael
Holland, and Simona Sawhney (Minneapolis: University of Minnesota Press, 1991), 90–91.
So love, the essence of thought, cannot be exhausted as the object of the

thought of which it is always already the impulse or desire. But this is not a
council of despair, as Nancy’s delicate and complicated allusion to Augustine
indicates. The amare amabam to which Nancy refers in this quotation is from
Confessions 1:3. It is a beautiful sentence in Latin: “Nondum amabam, et amare
amabam, quaerebam quid amarem, amans amare,” which can, somewhat less
poetically, be translated: “I was not yet in love, and I loved to be in love, I sought
what I might love, in love with loving.” To seek to treat love as an object, to
desire love, is always already to be in love, always already to be in love with
love. So the dialectic that seeks to interrogate love, such as we find in Plato’s
Symposium, already has its object, but not as a possession, indeed not as an
object. It possesses love—rather, it is possessed by love—in its own move-
ment towards love, in and as the dialectic itself, as a Diotima whose absence
can be evoked but who is not thereby made present.
Sixthly and finally, this relationship between love and thought does not
allow love to be exiled outside thought, as its mystical or incalculable
“Other.” It would be hasty and mistaken indeed to suppose Nancy to be say-
ing that love is the other of thought, that logic is on the side of precision,
care and calculation, and love is an excess that overflows the calculative.
This relation of calculation and excess would traduce love’s intimate rela-
tion to thought, and would set them, falsely, in opposition to each other.
And yet, Nancy does talk about love in terms of an “in excess”: not in excess
of calculation, but in excess of the logic of calculation and excess: Love frus-
trates the simple opposition between economy and noneconomy. Love is pre-
cisely—when it is, when it is the act of a singular being, of a body, of a heart,
of a thinking—that which brings an end to the dichotomy between the love in
which I lose myself without reserve and the love in which I recuperate myself,
to the opposition between gift and property.28
In other words, love suspends the mastery of self that maintains the bound-
ary between what is mine and what is yours, between giver and receiver,
indeed between the “me” and the “you.”29
Love does not leave all logic and thinking behind; far from it. Rather, love
goes before all thinking, and carries it along. It is both in excess of the care-
fully demarcated categories of the logical and the illogical, the economic and
the non-economic, calculation and the incalculable, and at the same time it is
the condition of possibility of those very categories; it is the movement or the
imperative of the thought that establishes such categories.

29 Jean-Luc Nancy, L’Adoration, déconstruction du christianisme, 2 (Paris: Galilée, 2010), 88.
28 watkin
Another way of saying the same thing is that love is not the contradiction of
thought, as if love left all logic and thinking behind, but love is the contradic-
tion of contradiction and non-contradiction:
According to this schema, the nature of love is shown to be double and

contradictory, even though it also contains the infinite resolution of its
own contradiction. This nature is thus neither simple nor contradictory:
it is the contradiction of contradiction and noncontradiction. It operates
in an identical manner between all the terms in play: the access and the
end, the incomplete being and the completed being, the self and the
beyond the self, the one and the other, the identical and the different.
The contradiction of contradiction and noncontradiction organizes love
infinitely in each of its meanings.30
For Western thought, Nancy continues, love is double, ambivalent and in con-
flict. For Baudelaire, Wagner and Strauss, for St John of the Cross to Strindberg
via Racine, Kleist and Marivaux or via Maturin, Monteverdi and Freud, love is
both necessary and impossible, sweet and bitter, free and a imprisoned, spiri-
tual and sensual, enlivening and deathly, clear-sighted and blind, altruistic and
egoistic:
For all, these oppositional couples constitute the very structure and life
of love, while at the same time love carries out the resolution of these
very oppositions, or surpasses them: in the realisation of love, the subject
of love is dead and alive, free and imprisoned, restored to the self and
outside of the self.31
This is the sense in which love is not the contradiction of logic but, more fun-
damentally, the contradiction of the system of logic and excess, or in other
words the contradiction of contradiction and non-contradiction.
4 Conclusion
Our Nancean detour complete, we can now return to Meillassoux to recon-

sider his anhypothetical in the light of Nancy’s unfolding of the relation of
love to thought. Meillassoux, it will be remembered, argues for the indirect
30 Nancy, A Finite Thinking, 250. Translation altered.

31 Nancy, The Inoperative Community, 87.
proof of the principle of factiality on the basis that maintaining the neces-
sary possibility of a difference between the “in-itself” and the “for-us” indi-
rectly validates the principle that only contingency is necessary. And it will
be remembered that the principle of factiality relies on the principle of non-
contradiction: that a thing cannot be both necessary and contingent at the
same time and in the same way. But we have now also seen that, for Nancy,
love does not contradict logic at all. Love does not contradict the principle of
non-contradiction but rather it contradicts the contradiction of contradic-
tion and non-contradiction, or the contradiction between what is contradic-
tory and what is not.
Yet someone might argue: this is still by Nancy’s own admission a con-
tradiction, so non-contradiction is being indirectly affirmed after all. To
argue in this way would be inexact, but Nancy’s language serves to muddy
the waters a little. The phrase “the contradiction of contradiction and non-
contradiction” conveys a certain reflexivity at the expense of clarity, and it
needs to be unravelled. When we take time to explore Nancy’s neat phrase
“love contradicts the contradiction of contradiction and non-contradiction”
we see that the contradiction operated by love is not a mere gain-saying or
adopting a contrary position.
“Contradiction” itself means something new here, something altered, not
something that adopts a contrary position to contradiction but that cuts
across its categories, knocking them out of joint. After all, how else could
one “contradict” the contradiction of contradiction and non-contradiction.
When Nancy writes of love contradicting, it is not to be understood as gain-
saying but as both making possible and, at the same time, as a destabilizing.
Love makes contradiction possible insofar as the philosophical gesture of
contradicting is carried along by love; love makes contradiction contingent
in that, without this carrying along there would be no philosophical prin-
ciple of non-contradiction as such at all. The principle relies on love like a
gift relies on the movement of giving. So in terms of logic, it is not the case
that to love is to contradict logic, and thereby to have assumed the prin-
ciple of non-contradiction in the process of denying it, as Meillassoux’s and
Aristotle’s indirect proof would have it. Rather, to love is to have shown the
insufficiency of the logic of contradiction to capture the relation of love to
logic. This does not amount to saying that that contradiction is thinkable, but
that the principle of noncontradiction itself is only thinkable on the basis of a
love that cannot be reduced to the principle whose thinking it impels. Love is
the origin of the opposition between contradiction and non-contradiction
in thought, but it cannot itself be reduced either to contradiction (viz. “love
and logic are contradictory”) or to non-contradiction (viz. “love and logic
30 watkin
are not contradictory”). And the impossibility of this reduction neutralises

Meillassoux’s and Aristotle’s indirect proof.
We must, however, go one step further, because love does not merely
neutralise Meillassoux’s indirect proof but it shows how Meillassoux’s posi-
tion itself provides an indirect proof of that very neutralisation. It does this
because presenting the principle of non-contradiction as a principle already
relies on the supporting philosophical gesture which is, in Nancy’s terms, a
gesture of love. In other words, any attempt to assert the principle of non-
contradiction will always already have cut across the distinction between
contradiction and non-contradiction.
This does not—and this is an important point—invalidate or devalue the
principle in any way. Quite to the contrary: it makes the principle possible.
Meillassoux’s indirect proof is that in the form, or the gesture of doubting
that the principle of factiality is absolute, one affirms that which one seeks
to deny by the content of such doubting.
The indirect proof at which we have arrived after the detour through
Nancy is that in the action of affirming the law of non-contradiction one cuts
across the logic one would seek to secure by that affirmation. Put another
way, Meillassoux’s indirect proof can only work after one strategic act of for-
getting. Namely, Meillassoux must forget his own desire, the desire to estab-
lish the indirect proof in the first place. Like Diotima, the final and decisive
“speaker” in Plato’s Symposium, it is Meillassoux himself, Meillassoux’s own
desire, that is absent from his account of his own indirect proof. The detour
seeks merely to put him back in the picture, where in fact he has been
all along.
With this indirect disproof of the indirect proof we arrive at a new for-
mulation of the anhypothetical principle, different to Plato’s, Aristotle’s and
Meillassoux’s. It is consonant with Plato’s evocation of love in relation to
the Idea of the Good, it follows the pattern of Aristotle’s and Meillassoux’s
respective indirect proofs, and is presupposed by all three philosophers. The
anhypothetical is the “I-love-you,” the “amabam” of Augustine’s “amare ama-
bam.” Like Aristotle’s anhypothetical it is not established by a hypothesis,
for the reason that all hypotheses rely on it and are borne by it. And like
Aristotle’s anhypothetical it is something that the philosopher “brings with
him to his task,” and it acknowledges philosophy as, precisely, the task of a
desiring and loving philosopher. But unlike Aristotle’s anhypothetical it is
not simply the case that contradicting the “I-love-you” affirms it. It is rather
the case that either contradicting it or not contradicting it repeats its own
gesture or movement. Like Plato’s anhypothetical, the “amare amabam” it is
that from which all thinking flows, and like Plato’s anhypothetical it does not
jettison dialectic altogether but uses dialectic to gesture towards that which,
in dialectic, cannot be accounted for by dialectic. But unlike Plato’s anhypo-
thetical it is in no way mystical. It is no threat to logic to show that its prin-
ciple is love, any more than it is a threat to the dialectic of the Symposium to
note that it is offered to us in a movement of that which it seeks to define.
Bibliography
Aristotle. The Metaphysics: Books i–ix, trans. Hugh Tredennick, Loeb Classical Library.
(Cambridge, MA: Harvard University Press, 1980).
Philippe Lacoue-Labarthe and Jean-Luc Nancy, The Literary Absolute: The Theory of
Literature in German Romanticism, trans. Philip Barnard and Cheryl Lester (Albany,
NY: State University of New York Press, 1988).
Quentin Meillassoux, After Finitude: An Essay on the Necessity of Contingency, ed. and
trans. Ray Brassier (London: Continuum, 2008).
———, “L’Inexistence divine” (PhD Diss., Université de Paris I, 1997).
———, “Temps Et Surgissement Ex Nihilo,” Paper presented at the Conference Autour
de Logiques des mondes, ENS Paris, 2006. Available at http://www.diffusion.ens.fr/
index.php?res=conf&idconf=701. Last accessed June 2012.
Jean-Luc Nancy, L’Adoration, déconstruction du christianisme, 2 (Paris: Galilée, 2010).
———, A Finite Thinking, ed. Simon Sparks (Stanford, CA: Stanford University Press,
2003).
———, L’impératif Catégorique, La Philosophie En Effet (Paris: Flammarion, 1983).
———, The Inoperative Community, trans. Lisa Garbus, Peter Connor, Michael Holland,
and Simona Sawhney, ed. Peter Connor (Minneapolis, MN: University of Minnesota
Press, 1991).
Plato, Lysis, Symposium, Gorgias, trans. W.R.M. Lamb, Loeb Classical Library (Cambridge,
MA: Harvard University Press, 1961).
———, The Republic, trans. Paul Shorey, Loeb Classical Library (Cambridge, MA:
Harvard University Press, 1937).
Christopher Watkin, Difficult Atheism: Post-Theological Thinking in Alain Badiou, Jean
Luc Nancy and Quentin Meillassoux (Edinburgh: Edinburgh University Press, 2011).
chapter 2
The Self, Ideology, and Logic

F.C.S. Schiller’s Pragmatist Critique of and Alternative to Formal Logic
Admir Skodo
1 Introduction1
Classic philosophies have to be revised because they have to square

themselves up with the many intellectual and social tendencies that have
revealed themselves since those philosophies matured. The conquest of
the sciences by the experimental method of inquiry; the injection of evo-
lutionary ideas into the study of life and society; the application of the
historic method to religions and morals as well as to institutions; the cre-
ation of the sciences of ‘origins’ and of the cultural development of
mankind—how can such intellectual changes occur and leave philoso-
phy what it was and where it was?2
The eminent British historian Herbert Butterfield viewed the scientific revolu-
tion (from Copernicus to Darwin) and the historical revolution (perhaps best
exemplified by Ranke) as the two decisive intellectual transformations that
created modern Western consciousness.3 They instated two, at times compet-
ing and at times complementing,4 perspectives by which modern Western
man sees, conceives, and constructs himself and the world in which he dwells.5
Both revolutions shook and crumbled pre-modern conventions governing the
1 I would like to thank Helge Ax:son Johnson’s Stiftelse for a generous research grant that has
enabled the research and writing of this chapter.
2 John Dewey, The Influence of Darwin on Philosophy: And Other Essays in Contemporary
Thought (New York: Henry Holt and Company, 1910), v.
3 A point later made by Donald R. Kelley, Foundations of Modern Historical Scholarship:
Language, Law, and History in the French Renaissance (New York: Columbia University
Press, 1970), 6.
4 For a relevant contemporary statement on how the sciences have become historical,
see W.R. Sorley, “The Historical Method,” in Essays in Philosophical Criticism, ed. Andrew Seth
and R.B. Haldane (London: Longmans, Green, and Co., 1883), 102–126, 102.
5 Kenneth B. MacIntyre, Herbert Butterfield: History, Providence, and Skeptical Politics
(Wilmington: ISI Books, 2011), 49–99.

The Self, Ideology, and Logic 33
legitimacy and boundaries of thought, and in their stead erected new ones.
Both revolutions, for example, demolished, by means of superior empiri-
cal methods, the dating of the world and the human species propagated by
Christian churches and the Bible, and so undermined Christian claims to true
knowledge of nature. The scientific revolution, furthermore, queried Christian
assumptions about morality, while the historical revolution challenged the
Enlightenment belief in the uniformity and unshakeable foundations of man’s
mental and behavioral faculties, such as man’s rationality, consciousness, and
historical progress.
In spite of crucial differences, both science and history characterized human
life in an immanent frame, that is, by recourse to the earthly origins, changes,
developments, and in immediate contexts of things and persons. By the end of
the nineteenth and the beginning of the twentieth century, such an immanent
frame had substantially gained in purchase, which is readily observable in the
higher education, literature, and politics of the time once we recognize that
Darwinism, probability theory, statistics, historicism, idealism, pragmatism,
and philology belong to the thicket of the two revolutions, as do nationalism
and the industrial organization of society.6
Broadly speaking, there were two ways in which philosophers of the
nineteenth and early twentieth centuries could adapt to these two revolu-
tions. They could either joins ranks with one, and go to intellectual war with
the other, or they could combine the two in a single systematic scientific-
historical-philosophical worldview. More often than not, historicism, positiv-
ism, and analytical philosophy became traditions aligned around the first type
of response, while idealism and pragmatism followed the second trail: Dewey’s
passage above attests to precisely this fact.7 This second type of response is
also evident in some thinkers of the nineteenth and early twentieth centuries
whose thought resolutely refuses to sit easily in any disciplinary category or
intellectual tradition, such as Nietzsche.8
The latter attitude (combining the two revolutions), however, was only pos-
sible during a time when the natural and human sciences were in principle
6 See e.g. the studies in European Intellectual History Since Darwin and Marx: Selected Essays,
ed. W. Warren Wagar (New York and London: Harper Torchbooks, 1966).
7 Cf. Richard Rorty, Consequences of Pragmatism (Minneapolis: University of Minnesota Press,
1982), especially 139–160.
8 Friedrich Nietzsche, The Gay Science: With a Prelude in German Rhymes and an Appendix of
Songs, trans. Josefine Nauckhoff (Cambridge: Cambridge University Press, 2001); Nietzsche
and Science, ed. Gregory Moore and Thomas H. Brobjer (Aldershot: Ashgate, 2004).
34 skodo
open to contributions from scholars trained in one form of science (human

or natural), but aspired to utilize insights from and even make contributions
to the other. But from roughly the second quarter of the twentieth century,
the human and natural sciences were beginning to comprise specialized and
technical research disciplines, and could be practised legitimately mainly by
those that had undergone rigorous specialized training in a controlled and
controlling institutional setting such as a university or research institution,
and usually including a doctoral program, publications in specialized techni-
cal journals, and participation in seminars and conferences.9 Dewey’s passage
had, within the span of a few decades after its publication, begun to sound
quaint among philosophers, and the question he asked at the end of it no lon-
ger acted as a legitimate philosophical question.10 Yet, roughly between the last
three decades of the nineteenth and the first three decades of the twentieth
century it was fully possible to evince a holistic philosophy fuelled both by the
scientific and the historical revolution.
These six decades not only witnessed the pinnacle of the scientific and
historical imaginations, but saw too the birth of the modernist artistic imagi-
nation. It is commonplace to separate both science and history from art, and
argue that the former denotes traditions in philosophy and social science (for
example, logical positivism, analytical philosophy, philology, and psychoanaly-
sis), while the latter term captures the literature and art of the time (for exam-
ple, Hugo von Hofmannsthal, Marcel Duchamp, James Joyce, Franz Kafka, and
André Gide), with little or no overlap between them. And while the former
imaginations look for certainties in morals, order and perfection in nature,
and clearly patterned change in history (whether cumulative or disjunctive,
reformist or revolutionary), the latter finds morals to be fragile and duplici-
tous constructions, erratically revolving around a fundamentally fractured and
indelibly aesthetic nature of the self, whose psychic volatility and social malle-
ability render a vision of reality as essentially fluid. This chapter will question
this separation, the reason for which will be explained shortly.
Among the pragmatist philosophers, there was one in particular who
embodied a philosophical attitude similar to that of Dewey, and to some extent
Nietzsche (it is no coincidence that he has been analyzed as a thoroughgoing
9 William H. Sewell, Logics of History: Social Theory and Social Transformation (Chicago:
Chicago University Press, 2005), 2–4.
10 The contemporary revival of classical pragmatism, not least the thought of Dewey, did
not mean the revival of the historical Dewey. It was rather geared to find new languages
for contemporary philosophical perspectives in analytical and continental philosophy.
Nietzschean).11 But unlike most other pragmatists, that philosopher was atten-
tive to the anti-foundationalism of the modernist artistic imagination. Perhaps
that is the reason why Bertrand Russell dubbed him the “literary” wing of
pragmatism.12 The philosopher in question is the Briton F.C.S. Schiller (1864–
1937), who held positions at the universities of Oxford, Columbia, Cornell and
Southern California. This chapter examines Schiller’s philosophy of logic, in
which, arguably, his greatest achievements lay.13
The foregoing narrative sets the stage for Schiller’s philosophy of logic, for
Schiller believed that logic was the only science that had remained impervi-
ous to the innovations wrought by the two revolutions. He thus exclaimed that
“[logicians] have trusted that their traditional scheme of instruction would
weather this storm [Schiller’s and others’ critiques of formal logic], as it has
survived the revolt of renascent literature against Medieval Scholasticism and
the nineteenth-century revolt of science against dogma and tradition [emphasis
added].”14 Schiller also recognized that the modernist scientific imagination
was influencing the modernist artistic imagination: “A new science, moreover,
has slowly risen into prominence in the shape of Psychology, which has already
exercised some influence on literature [emphasis added].”15 Schiller went so far
as to opine that “aesthetics can perform the functions of ethics.”16 Psychology’s
influence on literature, Schiller believed, was for the better, and logic too
would be better off if it made room for the insights of psychology, and its fur-
ther exploration in modernist literature.
This chapter argues that the conceptual resources provided by the two
revolutions, and the modernist artistic imagination, allowed Schiller to chal-
lenge what he perceived as the orthodoxies of formal logic. The argument of
this chapter is a historical one: it seeks to unearth a specific historical mode
of thinking about logic. In emphasizing the difference of Schiller’s logic from
the way we today think about logic, the chapter aims to show that Schiller’s
logic, though foreign from the perspective of today’s philosophical culture,
was meaningful and legitimate in its own historical setting. The specificity
11 George J. Stack, “Nietzsche’s Influence on Pragmatic Humanism,” Journal of the History of

Philosophy 20 (1982), 369–406.
12 Bertrand Russell, “Dr Schiller’s Analysis of The Analysis of Mind,” The Journal of Philosophy
18 (1922), 645–651.
13 Mark J. Porrovecchio, F.C.S. Schiller and the Dawn of Pragmatism: The Rhetoric of a
Philosophical Rebel (Plymouth: Lexington Books, 2011), 145–162.
14 F.C.S. Schiller, Formal Logic: A Scientific and Social Problem (London: Macmillan and Co.,
Limited, 1912), viii.
15 Schiller, Formal Logic, 396.
16 F.C.S. Schiller, Social Decay and Eugenical Reform (London: Constable and Co., 1932), 66.
36 skodo
of this mode lies in the following conceptual themes, to be explored below:

(1) Schiller analyzed logic in the context of a holistic philosophy of life that
takes for granted the primacy of the concrete, historical, and social self;
(2) human thought, including logic, has no fixed foundations, and the presup-
positions of logic acquiesce to radical change; (3) A true logic, therefore, must
own up to (1) and (2), which means that logic must be humanized, acknowl-
edging that human thought is radically plastic, and infused with non-logical
elements that influence logical thinking, such as irreducibly conflicting ideolo-
gies, values, and a variety of practical purposes; (4) formal logic denies (1), (2),
and (3), and constructs the tenets of logic based on empirically false, histori-
cally meaningless, and practically useless assumptions, and is therefore struc-
turally flawed as both a descriptive and normative science of thought.
2 The Self and Logic
Schiller’s philosophy of logic was pitted against three types of formal logic:
the logic of text-books used in higher education, resting on logical principles
such as the law of identity, the law of contradiction, and syllogistic deduction;
the logic of British absolute idealism, especially that of F.H. Bradley; and the
symbolic logic of a rising group of philosophers including Bertrand Russell and
G.E. Moore. Schiller believed that, notwithstanding their differences, these log-
ics shared an aloofness from engaging with concrete or actual thinking, and so
their formal nature was sufficient to label them all as formal logic.
For Schiller, but for other pragmatists of his time as well, philosophical
inquiry could never be, and should never aspire to be, purely philosophical (in
the academic sense prevalent in Schiller’s time). The subtitle of Schiller’s book
Formal Logic is telling as to his own view on the scope of Logic: A Scientific
and Social Problem. A constant line of attack from pragmatists in the early
twentieth century was that formal logicians did not recognize the historical,
social, and practical aspects of logic. This line of attack carried illocutionary
force in the early twentieth century because it came from internal criticism,
that is, academic logicians themselves had since the early twentieth century
begun to see insoluble problems to formal logic, and most of these Schiller
knew personally, e.g. Alfred Sidgwick, Bernard Bosanquet, C.S. Peirce, and John
Dewey. Schiller himself was a teacher of formal logic at the prestigious Oxford
and Cornell Universities. That every truth, every logical operation, every logi-
cal principle, rested on some concrete practice, conceivable or actual, was a
fairly conventional presupposition in this period. It was a socially recognized
way of dealing with new dilemmas.17 As such, it allows us to grasp a historically

distinctive mode of thinking about logic.18
As is the case with the other pragmatists of his time, such as Dewey, under-
standing Schiller’s logic properly requires embedding it in a broader philo-
sophical project. Schiller was an iconoclast, and as behoved an iconoclast, he
followed the way of pragmatism to one extreme, which he called “humanism.”
The core of this humanism can be unearthed in Schiller’s first book, Riddles
of the Sphinx, published in 1892, where Schiller was committed to viewing the
person or the “Self” as
[. . .] the most indispensable of all postulates, it is the Alpha, the starting-
point, and it would not be surprising if it turned out also the Omega, the
goal of philosophy. [. . .] all acts of knowledge are performed by selves, the
whole of our cognitive machinery, principles, axioms, postulates and cat-
egories, are invented by and modelled upon selves.19
As this passage shows, Schiller’s humanism was philosophically radical in that

it rendered the study of any form of thought essentially anthropological. The
self, according to Schiller, comes not just with the ability to reason logically but
also with the capacity to develop, change, imagine, feel, dream, desire, con-
trol, deliberate, imagine, choose, co-operate, invent, force, joke, die, deceive,
act, and much more, all of which are connected, all of which have grown and
developed through evolution, and from the particular chains of experiences a
self has undergone since its birth within a specific historical culture. Such is
the process human selves, including logicians, call their lives. Philosophy, for
Schiller, was ultimately the study of life in its actual and conceivable totality.
17 John Passmore, “Some Critics of Formal Logic,” in A Hundred Years of Philosophy (London:
Penguin Books, 1966), 156–173. This is an explanatory historical context in which
Wittgenstein’s later philosophy ought to be embedded.
18 For a good historical approach to American pragmatism see David Hollinger, “William
James and the Culture of Inquiry,” and “The Problem of Pragmatism in American
History,” in In the American Province: Studies in the History and Historiography of Ideas
(Bloomington: Indiana University Press, 1985), 3–23, 23–44.
19 F.C.S. Schiller, Riddles of the Sphinx: A Study in the Philosophy of Humanism (London: Swan
Sonnenschein & Co, 1910), 142. See also F.C.S. Schiller, “Axioms as Postulates,” in Personal
Idealism: Philosophical Essays by Eight Members of the University of Oxford, ed. Henry Sturt
(London: Macmillan, 1902), 47–134; and F.C.S. Schiller, Logic for Use: An Introduction to the
Voluntarist Theory of Knowledge (London: G. Bell & Sons Ltd, 1929).
38 skodo
Issuing from this conception of philosophy is the view that the philosophi-
cal study of logic must take into account everything that falls within the pur-
view of the concrete self, because it is the thinking of such a self that simply
is the content of logic. But because the logician can only conduct such a study
from the vantage point of a concrete self (the logican is inescapably a concrete
self), logic can never arrogate formal or other perfection, completeness, uni-
versality, and uniformity.
According to Schiller, there is no ultimate reality or ideal beyond human
experience, and human experience is not in need of it: “The intellectual cos-
mos also neither has nor needs fixed foundations whose fixity is an illusion.”20
Schiller draws a methodological consequence from this belief that he intends
to act as the searchlight of pragmatism: “it is a methodological necessity to
assume that the world is wholly plastic, i.e. to act as though we believed this,
and will yield us what we want, if we persevere in wanting it.”21 This conse-
quence, in turn, has profound consequences for the nature of “axioms,” or the
most fundamental principles of thought that guide human thinking: “We con-
ceive the axioms as arising out of man’s needs as an agent, as prompted by his
desires, as affirmed by his will, in a word, as nourished and sustained by his
emotional and volitional nature.”22 In more elaborate terms, the self:
[. . .] thinks with his whole heart and personality, that his feelings enter
constantly and copiously into his reasonings, that his nature selects the
objects of his thought, and determines his aims and his motives and his
methods and the values he assigns to his objects, while his education and
history determine the meanings and associations of the instruments of
his thinking, viz. the words he uses.23
If the foundation of logic is the self, then that foundation is radically mallea-
ble, or “plastic,” since everything about the nature of the concrete self is provi-
sional. It is with these presuppositions that Schiller both criticizes formal logic
20 Schiller, “Axioms as Postulates,” 57. Though I do not mention it in this chapter other than
in a cursory manner, the advances in the physics and biology were important in persuad-
ing Schiller into taking this stance. Concepts such as “matter,” “force,” “causality,” “origin,”
and “substance” took on whole new meanings in the late nineteenth and early twentieth
century and these meanings allowed for indetermination or worse. See especially Schiller,
Riddles.
21 Schiller, “Axioms as Postulates,” 61.
22 Schiller, “Axioms as Postulates,” 86.
23 Schiller, Logic for Use, 101.
and professes an alternative humanist logic. If there are no fixed and universal
standards for logic, what is there? What, if anything, ensures the validity and
objectivity of truth, deduction, induction, and the like? What should stay, if
anything, and what should go, of formal logic?
One example of Schiller’s method at work suffices to tease out some answers
to these questions, since it branches out to most of his key concepts and argu-
ments. I have in mind Schiller’s ingenious criticism of the syllogism: if it is true
that “all men are mortal” and that “Socrates is a man,” then it necessarily fol-
lows that “Socrates is mortal.”24 Both its truth and the necessity of its validity
have been taken for granted since Aristotle, and so if it can be shown that it
is a bad form of reasoning, an important step will have been taken toward its
reform, and the reform of formal logic in general.
Schiller’s first line of attack is conventional for that time—namely, that the
conclusion begs the question, for the truth of the major premise depends on
the conclusion: for the logician, or any other thinker for that matter, to prove
that “all men are mortal,” he must know that Socrates is mortal prior to the
proof, because Socrates falls within the premise “all men are mortal.” This of
course, uproots the concept of proof from its force in the syllogism. The best
line of defence the logician can put up, according to Schiller, is to postulate
that the premise is not based on empirical observations. It is rather a universal,
or a law of nature, the conclusion of which is a particular instance. But this
retort does not convince Schiller, for it assumes that the universal is absolute,
applicable to any particular, for any purpose, by anyone and in any context,
while Schiller’s point is precisely that a particular, a specific purpose, a particu-
lar person, and a specific context, constitute thinking.
The absurdity of this postulate is brought out in the alternative, highly
value-laden, and socially circumscribed premise “all negro slaves are men,”
the truth of which is necessarily relative to concrete selves.25 Thus, Schiller
wrote, “No one in his senses, we shall say, will argue about ‘Socrates,’ whether a
defunct philosopher or negro slave, a tomcat or a character in fiction, and with-
out knowing what the problem is that has arisen about him.”26 This example
shows well the rationale of the pragmatist dictum that truths are species of
values and judgements with practical origins, desires and consequences; and
that treating truths in mere propositional form is therefore nothing more than
a pastime of philosophers in ivory towers.
24 F.C.S. Schiller, “Are All Men Mortal?,” Mind 44 (1935), 204–210.

25 Which he might have owed to Bradley: Passmore, A Hundred Years of Philosophy, 158.
26 Schiller, “Are All Men Mortal,” 207.
40 skodo
Schiller’s counter-premise exemplifies his firm belief that that differences

in value “seem to be ultimate and irreconcilable.” And this allows for rhetoric
and power to enter the scene as procedures by which truth and validity can
be made.27 Moreover, because logicians too are selves who assign values and
meanings to concepts and practices, “the logicians themselves continue to dif-
fer widely as to the nature, the function, the value, and even the existence, of
their science.”28 Hence, Schiller argues for a comprehensive value-pluralism,
and the next section will discuss how the recognition of that pluralism leads
Schiller to adopt a principle of toleration.
For the moment, is important to take note of the fact that Schiller arrives at
his pluralism and principle of toleration for reasons that can only be under-
stood historically. Schiller believed, much like the other British and American
pragmatists of his time, that he was living in an age that had just come out of a
historical period in which “monism” or “absolutism” of values reigned supreme
(the Victorian era), but which had to give way to an acceptance of pluralism,
since social, political, and scientific reality had thoroughly discredited that
reign. In Schiller’s own words: “History has declared against intolerance, and in
practice we have all to confess nowadays that there is truth beyond the limits
of the beliefs we hold, because they seem to us the truest.”29
These discussions of the Socrates syllogism, and a host of other discussions
on the elements of logic, pose, according to Schiller, grave problems for formal
logic due to a generic fact—namely, “the abstraction from meaning” in formal
logic.30 But this abstraction, which formal logic took to be a virtue, proved fatal
to formal logic because it left the meaning of its words and sentences, its, as it
were, raw material, hopelessly ambiguous or indeterminate. The reason why
formal logic leaves every word and sentence indeterminate or ambiguous is
that they have a wide range of potential meanings and contexts, and mere “ver-
bal,” or formal, meaning (such as “Socrates,” “man,” and “all men are mortal,” in
the classical syllogism) does not entail a particular meaning. Determination
and disambiguation of meaning can only be achieved by the contextualized
linguistic acts performed by selves. In Schiller’s words, “real” meaning “always
arises in a particular situation, and it is always personal; i.e. it is what men
mean when they use words to express and convey their meaning.”31

28 Schiller, Formal Logic, vii; F.C.S. Schiller, Humanism: Philosophical Essays (London:
Macmillan and Co., Ltd., 1912), 49.
29 Schiller, Formal Logic, 406.
However, in expressing meaning persons are always communicating their

experience of the world to other persons, which explains why the “the mean-
ing of words then becomes social, without ceasing to be personal.”32 This char-
acter of meaning affects the character of truth, for truth, just like meaning, is a
value continuous with the living self:
Like the other values also the career of a truth is profoundly influenced
by man’s social nature; it has not merely to commend itself to its maker
for the nonce, but to continue to give him satisfaction and to continue to
seem the right remark for the occasion. Now this it will hardly do, unless
it succeeds in winning recognition also from others, and is judged valu-
able, ‘good’ and ‘true’ by them. Should it fail to do so, the penalty is in
every case the same, viz. condemnation as ‘false,’ rejection and superses-
sion by a better ‘truth.’ Hence so long as it lasts it is being tested and, it
may be, contested.33
The validity and objectivity of concepts and principles, at bottom, stand and
fall with social edifices, which in turn stand and fall depending on the inter-
actions and communications between different selves. “Axioms,” for instance,
are at bottom practical postulates regimented by working social practices. For
Schiller, the history of thought and the study of evolution in the early twenti-
eth century provided ample evidence that this was the case.34
Schiller’s most daring suggestion for a reform of the syllogism was to treat
it as a mode of inquiry for solving concrete problems for concrete selves. The
practical use of the Socrates syllogism could, for instance, arise if it addressed
a problem concerning Socrates. Schiller offered one: “a ‘problematic’ Socrates
has turned up and there are doubts about him. He is under grave suspicion. Is
he a man or a ghost?”35 This question might not make sense to philosophers
today, but if we understand it historically, we will find it be yet another example
of the historical specificity of the philosophical problems deemed legitimate
by leading philosophers in the early twentieth century. The reason why Schiller
felt he could pose such a question in the context of a discussion of logic was
the fact that he was a life-long devotee to psychical research, and the fact that

33 Schiller, Riddles, 132–133.
34 See especially Schiller, “Axioms.” For a contemporary questioning, akin to Schiller’s, of the
idea of progress see A.S. Pringle-Pattison, The Philosophy of History (London and Oxford:
35 Schiller, “Are All Men Mortal?,” 207.
42 skodo
in the early twentieth century psychical research was as contested as it was

hailed by prominent psychologists and philosophers. Schiller succeeded Henri
Bergson as the President of the Society for Psychical Research in 1914, of which
William James had been the President before Bergson.36
Questions like this one thus preoccupied some of the most fascinating phi-
losopher of the early twentieth century. And they were accepted as admissible
in philosophical discussions.37 Such questions could arise when some people,
called “mediums,” claimed to somehow be in contact with people long since
deceased, and could act as channels through which the spirits or ghosts of
the deceased could communicate with the living. A philosophical question,
then, would be, can we trust that this or that medium is telling the truth
when he or she claims Socrates is communicating with us, is and if so how?
According to Schiller, in order for this to even count as a problem, the meaning
of words such as “personality,” “self-identity,” “mortal,” “truth,” and “validity”
must be radically different than the one implicit in the standard form of the
syllogism, or in any formal logical system.
Whether the syllogism can actually be revised to meet these requirements
and how, was, unfortunately, something Schiller never explored. His own
humanist logic was far from systematic. Still, apart from exemplifying a contin-
gent mode of thinking, he did point to some aspects of human thinking which
lucidly evince that formal logic is sorely lacking in describing and regulating
them; and he did point the way to directions conducive to adequate solutions
of those problems.
So far we have seen Schiller radically humanize truth, meaning, and inquiry.
We need not inquire further into the other logical concepts and theories he
found wanting and attempted to revise, such as the law of identity, induction,
and the correspondence theory of truth.38 This section has already established
just to what extent Schiller was ready to go in analyzing logic in its concreteness.
36 See e.g. Renée Haynes, The Society for Psychical Research, 1882–1982: A History (London:
Macdonald, 1982); Philosophy and Psychical Research, ed. Shivesh C. Thakur (London:
Allen & Unwin, 1976); William James, Essays in Psychical Research (Cambridge, Mass.:
Harvard University Press, 1986).
37 Mark. K. Porrovecchio, “The Curious Case of F.C.S Schiller,” Society for Psychical Research,
[http://www.spr.ac.uk/main/article/curious-case-f-c-s-schiller, accessed Jan. 4 2012].
38 For a pithy discussion on these matters, see Reuben Abel, The Pragmatic Humanism of
F.C.S. Schiller (New York: Columbia University Press, 1955).
3 Logic and Ideology
Schiller’s logic requires historical understanding, since it might present itself,

from the perspective of our dominant philosophical cultures, not least con-
temporary pragmatism, solely as a response to a widely disseminated purely
philosophical “Formal Logic.”39 A philosopher may focus solely on the
“purely” philosophical aspects of Schiller’s logic, but it comes at the cost of
judging irrelevant what Schiller and his fellow pragmatists deemed essential.
Contemporary pragmatism often neglects this insight from historical research,
and treats historical instances of pragmatism as contributions to today’s philo-
sophical problems, such as metaphysical realism, the nature of truth, scientific
explanation, logical formalism, and multiethnic democracy.40
In doing so, contemporary pragmatism loses sight of the fact that some of
the crucial divergences in Anglo-American pragmatism in the first half of the
twentieth century can be attributed to the competing political ideologies to
which its proponents were committed. In debates between pragmatists, a prag-
matist could take a radically different stance on philosophical issues depend-
ing on the particular intersections between his philosophy, his political beliefs,
and other non-logical forms of thought.
Bertrand Russell’s response to Schiller’s criticism of Russell’s logic, and Max
Eastman’s review of Schiller’s own logic accent these dissimilarities. Russell,
the major British intellectual and logician, called Schiller the literary wing of
pragmatism (Dewey being the scientific and James the religious). Schiller was
a major critic of Russell’s formal logic. But in one response to Schiller, Russell
avowed that rhetoric rather than logic often serves to steer the nature of a
debate on logic, thus partly accepting one of Schiller’s main points about the
nature of logic: “He [Schiller] and I are agreed, I think, that it is impossible to
produce logical arguments on either side of the questions which divide us;”
39 Abel, The Pragmatic Humanism; Douglas McDermid, The Varieties of Pragmatism: Truth,
Realism, and Knowledge from James to Rorty (London: Continuum, 2006); John R. Shook,
“F.C.S. Schiller and European Pragmatism,” in A Companion to Pragmatism, ed. John R.
Shook and Joseph Margolis (Oxford: Blackwell, 2006), 44–54; Ahti-Veikko Pietarinen,
“Remarks on the Peirce-Schiller Correspondence,” in Transatlantic Encounters: Philosophy,
Media, Politics, ed. E.H. Oleksy and W. Oleksy (Frankfurt am Main: Peter Lang, 2011), 61–70.
40 See e.g. Hilary Putnam, Pragmatism: An Open Question (Oxford: Blackwell, 1995); Richard
J. Bernstein, The Pragmatic Turn (Cambridge: Polity Press, 2010); James Johnson, The
Priority of Democracy: Political Consequences of Pragmatism (Princeton: Princeton
44 skodo
since they rest on “different logics,” the only effective retort must therefore be
“of the nature of rhetoric rather than logic.”41
Eastman was, along with Sidney Hook, one of Dewey’s best pupils and in
the first four decades of the twentieth century a Marxist, indeed one of the
most prominent Marxists in America of that time. It was by no accident, there-
fore, that Eastman inculpated Schiller on account of the conservative princi-
ple underscoring his revision of formal logic. Such a principle, according to
Eastman, will render it “impossible for anyone to build up a logic of science
and of practical life [emphasis added],” since it acknowledges that formal
logic works practically, and so fulfils the pragmatist criterion of truth. Instead,
according to Eastman, one must take a revolutionary, “democratic,” “system-
wrecking,” and transformative leap in logic.42 True logic, for Eastman, had to be
revolutionary, whereas for Schiller, according to Eastman, it must be conserva-
tive. What logic was and what it was supposed to do differed between Eastman
and Schiller on account of their political ideologies. Schiller was indeed a polit-
ical conservative, but of a very idiosyncratic kind; and he did attempt to justify
that conservatism with his philosophy. Yet, it is far from clear, as will be shown
below, that Schiller’s philosophy is straightforwardly conservative.
In any case, such ideological disputes as that between Eastman and Schiller
are not relative to the 1920s and the 1930s. It is important to recognize that
contemporary pragmatism harbors ideological content as well. And perhaps
we may learn something from pragmatist logic of the interwar years in recog-
nizing this fact: the difference between early twentieth century pragmatism
and contemporary pragmatism is that the former publically avowed the ines-
capability of ideology, while the latter does not. For instance, the ideological
content of Morton White’s pragmatist philosophy of science is inscribed as a
formal feature of philosophy that, moreover, attempts to appropriate histori-
cal authority in the name of ideological neutrality: “According to holistic prag-
matism, scientists’ warpings are carried out with concern for the elegance or
simplicity of the theory they adopt and with the intention to warp the heritage
conservatively—that is, by engaging in what James calls minimum modifica-
tion of it and what Quine calls minimum mutilation of it.”43
How was Schiller’s logic connected to his ideological beliefs? Schiller, as
already mentioned was a conservative. He was also part of the British eugenics
41 Russell, “Dr Schiller’s Analysis,” 651.

42 Max Eastman, “Mr. Schiller’s Logic,” The Journal of Philosophy, Psychology and Scientific
Methods 9 (1912), 463–468, 464, 465.
43 Morton White, A Philosophy of Culture: The Scope of Holistic Pragmatism (Princeton:
Princeton University Press, 2002), 2.
movement, and co-founded the Eugenics Education Society in 1907. These

two commitments will help us situate Schiller’s logic in his ideology, and vice
versa.44 According to Schiller, at the turn of the nineteenth century, two work-
ing “truths” had been shown to have reverted into “truth-claims,” in desperate
need of new concepts, postulates, experiments, verifications, adjusted desires,
and changed purposes. First, the social-Darwinist doctrine of the “survival of
the fit” had been seriously questioned by the success of democracy and social
reform in Britain: the “unfit,” in brief, were surviving, and that fact seriously
challenged the postulate that the “unfit” are doomed to perish. Second, the
so-called higher races, the fit stock, which Schiller identified with the English
nobility, were seen as degenerating, and so losing power in society, which
Schiller took to mean an oncoming racial suicide.45
Now, in philosophical terms, according to Schiller’s humanist logic, we
attain knowledge by an ongoing process of inquiry whereby “truth-claims are
professed and put to testing and experiment,” which at some point “verifies”
the truth-claim, and gains “social recognition” as a “truth,” which implies the
sidelining of other truth-claims as “falsities.” However, it is always the case
that the “truth-claim character persists into the ‘truth.’ ”46 This fundamentally
unstable foundation of truth had revealed itself in interwar Britain, instilling
in Jeremiah-type conservatives and eugenicists like Schiller a prophetic sense
of looming crisis.
In Schiller’s view, then, the truth-claims about man professed by left-wing
progressives had proved themselves as possessing efficient causality, and so
were becoming truths, while the truths of the aristocratic superiority were
turning back into truth-claims. There had arisen a contest between these
incompatible truths, in which logic, if it was to have any role to play in the
pressing problems of the day, had to take sides, for it was implicitly already on
some side.
Logical principles, for Schiller, were tied to other extra-logical spheres
of thought, such as biological principles. For instance, on occasion Schiller
argued that heredity marked the boundaries of our thinking: “Heredity, which
seems to render our moral, intellectual and physical characteristics more or
less dependent on the action of our parents and ancestors, limits, if it does not
44 See e.g. G.R. Searle, Eugenics and Politics 1900–1914 (Leyden: Noordhoof International
Publishing, 1976).
45 F.C.S. Schiller, Tantalus or the Future of Man (London: Kegan Paul, Trench, Trubner & Co.,
Ltd., 1924), 47–48.
46 Schiller, Logic for Use, 105–106.
46 skodo
destroy, our freedom and our responsibility.”47 Thus, the quality of thinking,
health, beauty, and strength which nature indifferently and callously distrib-
uted across the human stocks coincided with social strata.
However, at the same time, Schiller was committed to viewing thought as
essentially historical and voluntary. Those commitments qualified Schiller’s
biological determinism and he argued that hereditary qualities are not abso-
lute or fixed. The very fact that biological facts and laws have meanings and
truths proves that they are inescapably value-laden, and so ongoingly deter-
mined by concrete selves. In the end, for Schiller, it is not some abstract con-
servative or eugenical principle or fact that is decisive for practice. All such
principles and facts “are meant for the guidance of moral agents, with whom
the decision must remain.”48 Biological truths, therefore, were no less exempt
from agency, irreconcilable differences, debates, inquiries, and contests than
were ideological ones.
This complicates Schiller’s logic, and his commitment to conservatism and
eugenics, for it shows that Schiller’s logic supervened on his ideological views:
both eugenics and conservatism were in the first place forms of social inquiry
and therefore ought to be regulated by humanist logic.49 Thus, the pressing
social issues of the day had convinced Schiller that conservatism, ironically,
must change, even take on a new, revived, character, radically different than
the conservatism of the Tories and the House of Lords.50 And eugenics must
be experimental and progressive, for that is the mode in which truth-claims
are applied and put to practical use. Moreover, because every truth-claim and
truth is social in nature, eugenics too “would have to be backed by a powerful,
enthusiastic, and intelligent public sentiment.”51 And because eugenics would
be a trial-and-error experimental practice, just one out of many competing
practices, and not internally coherent at that, according to Schiller, it “will
47 Schiller, Riddles, 231–232.

48 Schiller, Social Decay, 34–35.
49 Schiller, Social Decay, 27.
50 This was a common theme to the “aristocratic revivalism” of the late nineteenth and early
twentieth centuries, but it was justified in different, sometimes radically different, ways.
One of these ways was that of Oscar Levy, a Jewish-German intellectual who successfully
introduced Nietzsche into England. See e.g. Dan Stone, Breeding Superman: Nietzsche,
Race and Eugenics in Edwardian and Interwar Britain (Liverpool: Liverpool University
Press, 2002).
51 Schiller, Tantalus, 59.
regard the toleration of differences of opinion as among the cardinal principles

of a sanely progressive social order.”52
All these provisions notwithstanding, Schiller did believe that eugenics had
become necessary in mass society: the “higher” races should be encouraged to
breed more (positive eugenics) while the lower ought to abstain from breeding
(negative eugenics). Moreover, he seemed to be disgusted by the “lower” races,
calling them weeds, hordes, masses, imbeciles, and the like. Still, for the rea-
sons just stated, Schiller felt that the criticism from the prominent intellectu-
als J.B.S. Haldane and Bertrand Russell, saying that eventually political dissent
would be labelled degenerate and eugenics perverted into arbitrary exercise of
power, was incorrect. He was quite ready to accept the possibility of the failure
of eugenics, if the British public should will it. This acceptance signals a fairly
strong commitment to the strictures of Schiller’s humanist logic.53
There could, for Schiller, be no starker contrast to his own humanist logic
than formal logic. Schiller modelled the structure of his logic on the value-
pluralism of the society of his time together with the historically situated con-
crete self. He intended for that logic to accurately describe the social function
of thinking for such selves, on the one hand, and aid in refining such think-
ing for social uses, on the other. Formal logic, in contrast, had or would have,
devastating social effects, according to Schiller. First, since the “ideal of formal
perfection is Fixity,” formal logic postulates the existence of fixed and perma-
nent truth, and since such a vision of truth entails impermeability to change,
formal logic conceives of change as sign of imperfection, a symptom of falsity.
The practical effect of this postulate serves “to commend Formal Logic to the
blindest and most intractable sort of conservatism.” Second, since the ideal
proof of a formal logical operation is meant to arrive at certainty, the effect
of this ideal is that it debars thought that is risky, and outlaws thought that
is probable. This is pernicious, according to Schiller, to the best scientific and
everyday life practice, where decisions are made, problems solved, questions
answered, leaps taken, through endless series of concrete situations riddled
with uncertainty, or at best, probability. Third, since the concept of truth in for-
mal logic is absolute, it is conceived to be true regardless of any, and in every,
actual and possible circumstance; a truth is necessarily true in formal logic.
This means that “ ‘Necessity’ is as evidently the tyrant’s plea in logical as in
political absolutism and neither has any use for the freedom of human activ-
ity.” The fourth, and final, reason for why formal logic has deplorable social
52 F.C.S. Schiller, Eugenics and Politics: Essays by Ferdinand Canning Scott Schiller (London:
Constable & Co., 1926), 60.
53 Schiller, Eugenics, 29.
48 skodo
effects owes to the fact that formal logic not only postulates absolute truths,
but one and only one system of thought that is able to carry that truth, and that
is formal logic. For this reason, formal logic cannot abide by the plurality of
views containing truth: “The absolute system of immutable Truth is one. Not
more than one view, therefore, can be true.”54
4 Conclusion
This chapter has underscored the historical contingency of the philosophy of

logic. Focusing on F.C.S. Schiller’s pragmatist logic has allowed history to weigh
in on the nature of logic, for Schiller’s logic clearly reveals, in more ways than
one, that extra-logical contexts are crucial for understanding the function and
place of logic in concrete human life. The encounter with Schiller serves to
remind us that philosophy in history can appear as very familiar and yet very
foreign from the perspective of the present. On the one hand, Schiller’s value-
pluralism and anti-foundationalism certainly chime well with many of today’s
leading philosophical and historical perspectives, and can easily act as a source
of conceptual inspiration to them. On the other hand, what must surely baffle
these perspectives is Schiller’s commitment to eugenics, conservatism, and
psychical research. It is clear that Schiller’s logic does not make much sense
unless it is understood in these various contexts. Another way of expressing
this conclusion is to say that philosophy too leads a life in history.
Bibliography
Reuben Abel, The Pragmatic Humanism of F.C.S. Schiller (New York: Columbia
John Dewey, The Influence of Darwin on Philosophy: And Other Essays in Contemporary
Thought (New York: Henry Holt and Company, 1910).
Max Eastman, “Mr. Schiller’s Logic,” The Journal of Philosophy, Psychology and Scientific
Methods 9 (1912), 463–468.
European Intellectual History Since Darwin and Marx: Selected Essays, ed.
W. Warren Wagar (New York and London: Harper Torchbooks, 1966).
Renée Haynes, The Society for Psychical Research, 1882–1982: A History (London:
Macdonald, 1982).
54 Schiller, Formal Logic, all quotations from 397–398.

David Hollinger, “William James and the Culture of Inquiry,” in In the American
Province: Studies in the History and Historiography of Ideas (Bloomington: Indiana
University Press, 1985), 3–23.
———, “The Problem of Pragmatism in American History,” in In the American Province:
Studies in the History and Historiography of Ideas (Bloomington: Indiana University
Press, 1985), 23–44.
William James, Essays in Psychical Research (Cambridge, Mass.: Harvard University
Press, 1986).
James Johnson, The Priority of Democracy: Political Consequences of Pragmatism
(Princeton: Princeton University Press, 2011).
Donald R. Kelley, Foundations of Modern Historical Scholarship: Language, Law, and
History in the French Renaissance (New York: Columbia University Press, 1970).
Kenneth B. MacIntyre, Herbert Butterfield: History, Providence, and Skeptical Politics
(Wilmington: ISI Books, 2011).
Douglas McDermid, The Varieties of Pragmatism: Truth, Realism, and Knowledge from
James to Rorty (London: Continuum, 2006).
Nietzsche and Science, ed. Gregory Moore and Thomas H. Brobjer (Aldershot: Ashgate,
2004).
Friedrich Nietzsche, The Gay Science: With a Prelude in German Rhymes and an
Appendix of Songs, trans. Josefine Nauckhoff (Cambridge: Cambridge University
Press, 2001).
John Passmore, “Some Critics of Formal Logic,” in A Hundred Years of Philosophy
(London: Penguin Books, 1966), 156–173.
Philosophy and Psychical Research, ed. Shivesh C. Thakur (London: Allen & Unwin,
1976).
Ahti-Veikko Pietarinen, “Remarks on the Peirce-Schiller Correspondence,” in
Transatlantic Encounters: Philosophy, Media, Politics, ed. E.H. Oleksy and W. Oleksy
(Frankfurt am Main: Peter Lang, 2011), 61–70.
Mark J. Porrovecchio, F.C.S. Schiller and the Dawn of Pragmatism: The Rhetoric of a
Philosophical Rebel (Plymouth: Lexington Books, 2011), 145–162.
Mark. K. Porrovecchio, “The Curious Case of F.C.S Schiller,” Society for Psychical
Research, [http://www.spr.ac.uk/main/article/curious-case-f-c-s-schiller, accessed
4 January 2012].
A.S. Pringle-Pattison, The Philosophy of History (London and Oxford: Oxford University
Press, 1923).
Hilary Putnam, Pragmatism: An Open Question (Oxford: Blackwell, 1995); Richard J.
Bernstein, The Pragmatic Turn (Cambridge: Polity Press, 2010).
Richard Rorty, Consequences of Pragmatism (Minneapolis: University of Minnesota
Press, 1982).
Bertrand Russell, “Dr Schiller’s Analysis of The Analysis of Mind,” The Journal of
Philosophy 18 (1922), 645–651.
50 skodo
G.R. Searle, Eugenics and Politics 1900–1914 (Leyden: Noordhoof International

Publishing, 1976).
William H. Sewell, Logics of History: Social Theory and Social Transformation (Chicago:
Chicago University Press, 2005).
John R. Shook, “F.C.S. Schiller and European Pragmatism,” in A Companion to
Pragmatism, ed. John R. Shook and Joseph Margolis (Oxford: Blackwell, 2006),
44–54.
W.R. Sorley, “The Historical Method,” in Essays in Philosophical Criticism, ed. Andrew
Seth and R.B. Haldane (London: Longmans, Green, and Co., 1883), 102–126.
George J. Stack, “Nietzsche’s Influence on Pragmatic Humanism,” Journal of the History
of Philosophy 20 (1982), 369–406.
F.C.S. Schiller, “Are All Men Mortal?,” Mind 44 (1935), 204–210.
———, Social Decay and Eugenical Reform (London: Constable and Co., 1932), 66.
———, Logic for Use: An Introduction to the Voluntarist Theory of Knowledge (London:
G. Bell & Sons Ltd, 1929).
———, Eugenics and Politics: Essays by Ferdinand Canning Scott Schiller (London:
Constable & Co., 1926).
———, Tantalus or the Future of Man (London: Kegan Paul, Trench, Trubner & Co.,
Ltd., 1924).
———, Formal Logic: A Scientific and Social Problem (London: Macmillan and Co.,
Limited, 1912), viii.
———, Humanism: Philosophical Essays (London: Macmillan and Co., Ltd., 1912).
———, Riddles of the Sphinx: A Study in the Philosophy of Humanism (London: Swan
Sonnenschein & Co., 1910).
———, “Axioms as Postulates,” in Personal Idealism: Philosophical Essays by Eight
Members of the University of Oxford, ed. Henry Sturt (London: Macmillan, 1902),
47–134.
Dan Stone, Breeding Superman: Nietzsche, Race and Eugenics in Edwardian and Interwar
Britain (Liverpool: Liverpool University Press, 2002).
Morton White, A Philosophy of Culture: The Scope of Holistic Pragmatism (Princeton:
Princeton University Press, 2002).
chapter 3
Language, Truth, and Logic

Heidegger on the Practical and Historical Grounds of Abstract Thought
1 Introduction
The idea of ‘logic’ itself disintegrates in the turbulence of a more original

questioning.
—martin heidegger, “What is Metaphysics?”
In an essay entitled “The Elimination of Metaphysics Through the Logical

Analysis of Language,” Rudolf Carnap takes Heidegger to task for the produc-
tion of “philosophical nonsense.”1 Specifically, Carnap examines several ques-
tions and assertions from Heidegger’s “What is Metaphysics?”—including,
“What about this Nothing?” “Anxiety reveals the Nothing” “The Nothing noth-
ings” “The Nothing exists”—and argues that they amount to “meaningless
metaphysical pseudo-statements.” Carnap’s criterion for classifying Heidegger’s
assertions as nonsense is rooted in what his friend and colleague, A.J. Ayer,
calls the “principle of verification.” According to this principle, “a sentence has
literal meaning if and only if the proposition it expresses is either analytic or
empirically verifiable.”2 The tautology “All bachelors are unmarried men” is
analytically true if its predicate, ‘unmarried men’, is contained within its sub-
ject, ‘bachelors’. Similarly, the contradiction, “All bachelors are married men” is
analytically false (or absurd) since its predicate, “married men,” is negated by
its subject, “bachelors.” For Carnap and Ayer, all analytic judgments (whether
they be tautologies or contradictions) are true or false in virtue of their form.
And whilst they tell us nothing about the world, analytic judgments are mean-
ingful insofar as we can verify the various logical relations between the subject
and predicate of a specific sentence. On Carnap and Ayer’s account, the only
1 Rudolf Carnap, “The Elimination of Metaphysics through the Logical Analysis of Language,”
in Heidegger and Modern Philosophy: Critical Essays, ed. Michael Murray (New Haven: Yale
2 A.J. Ayer, Language, Truth and Logic (London: Penguin, 2001), 171.

52 wendland
other meaningful propositions apart from analytic judgments are those with
empirical content. Following the early Wittgenstein, Carnap and Ayer thought
that all complex states of affairs or empirical occurrences could be reduced
to and thus captured by elementary propositions that reflected a given set of
facts. And if simple propositions were able to reflect certain facts, then their
truth or falsity could be verified by the facts. For example, observing drops of
water falling from the sky confirms the truth of the proposition “It’s raining.”
And insofar as empirical propositions are verifiable they too are meaning-
ful. When, however, the principle of verification is applied to assertions like
“Anxiety reveals the Nothing,” we see that this statement is neither logically
verifiable, because it is neither a tautology nor a contradiction, nor is it empiri-
cally verifiable, since “the Nothing” is not a fact like rain. And therefore Carnap
dismisses Heidegger’s writing as metaphysical nonsense that needs to be elim-
inated from our philosophical vocabulary.
Perhaps the most obvious (or ironic) criticism of Carnap’s and Ayer’s prin-
ciple of verification is the extent to which it is meaningless on its own terms,3
but from a Heideggerian point of view the most fruitful critique of the veri-
fication principle comes from W.V.O. Quine and Wilfird Sellars: namely, that
Carnap and Ayer assume words and sentences have a direct relation to a given
reality without explaining how that reality is given.4 Like Quine and Sellars,
Heidegger is concerned with the conditions through which reality is presented
to human beings such that our signs can correspond to it. And when he says
“the idea of ‘logic’ itself disintegrates in the turbulence of a more original
questioning,”5 Heidegger’s point is not, as Carnap would have it, that human
inquiry should violate the laws of logic; but rather that the intelligibility of logic
is itself grounded in the essence of human beings: i.e., the being whose being
3 As noted in the text, propositions are meaningful if and only if they are empirically observ-
able or tautologies. But the verification principle doesn’t seem to be observable in the same
way that rain is. And insofar as the verification principle is that through which empirical
statements acquire meaning, it cannot itself be an empirical statement, since that through
which observation is made possible cannot itself be observed without the introduction of an
infinite regress or some sort of circular argument. Yet the verification principle cannot be
tautology either, for although it is not a fact it has a relation to the facts insofar as it is the
criterion through which facts are judged. And if the verification principle is neither a fact nor
a tautology, it is a performative self-contradiction or nonsense on its own terms.
4 See W.V.O. Quine, From a Logical Point of View (Cambridge: Harvard University Press, 1980)
and Wilfrid Sellars, Empiricism and the Philosophy of Mind (Cambridge: Harvard University
Press, 1997).
5 Martin Heidegger, “What is Metaphysics?,” in Basic Writings, ed. Krell (New York: Harper
Collins, 1993) 105.
Language, Truth, and Logic 53
it is to question the meaning of (its) being.6 Put otherwise, Heidegger’s aim is

not to cast doubt on the law of non-contradiction or the principle of identity,
but rather to show how the functioning of these laws or principles presuppose
the existence of human beings who are capable of questioning their own exis-
tence, adopting one practical activity over another, and thereby presenting a
world that can then be articulated in a predicate calculus. With that said, this
essay examines the conditions of propositional truth and our predicate calcu-
lus through a study of Heidegger’s work on Aletheia or unconcealment.
2 Heidegger on Propositional Truth and Unconcealment
Aletheia and unconcealment, for Heidegger, refer to a certain “truth of being”

whereby a series of practices create a context or clearing that enables entities
and aspects of the world to show up as the subject matter of our discourse. In
Being and Time, Heidegger illustrates the extent to which our propositional
engagement with reality presupposes a practical context through a descrip-
tion of our average everyday working world. Briefly, the idea is that human
beings are thrown into a specific historical community where particular
potentialities-for-being, say, that of becoming a butcher, a baker or a can-
dlestick maker, are presented to them. When we adopt one possibility over
another we define who we are, but we also disclose certain salient entities and
aspects of the world. In the case of a butcher, knives and cutting boards show
up as discrete entities designed for a specific task. And once these entities are
made manifest in the context of a given activity, we are able to assign sub-
jects and predicates to objects, such as, “This knife is sharp,” and ultimately
symbolize these assertions in the notation of formal logic. Whilst detailing the
derivation of logic from practical activity, this essay alludes to the temporal
horizon that characterizes intelligible human activity. Heidegger discusses this
phenomenon by highlighting the fact that our present practical engagement
with the world presupposes an appropriation of the potentialities-for-being
available in our cultural inheritance as well as the possibility of projecting that
potentiality into the future. So to continue with our earlier example, a human
being can only be a butcher if the activity of being a butcher is available in the
community into which she is born. But actively being a butcher also means
that it is a potentiality-for-being that her community is likely to support for
the foreseeable future. This suggests that the meaningful use of our predicate
6 Martin Heidegger, Being and Time, tr. Macquarrie & Robinson (New York: Harper & Row,
1962) p. 32 [12] (Hereafter BT).
54 wendland
calculus presupposes the temporal horizon built into the human activities
that present entities and aspects of the world as the entities and aspect that
they are. Finally, and insofar as human activity entails a temporal horizon,
Heidegger construes humans as historical beings: that is, as a being whose
being is bound by the possibilities available in a given socio-historical situa-
tion. And through the derivation of logic from practical activity we see that the
truth of any valid deduction is determined by, grounded in, or “disintegrates
into” the particular entities and aspects of the world that show up in certain
historical circumstances.
In §44 of Being and Time, Heidegger tells us that his “analysis takes its depar-
ture from the traditional conception of truth, and attempts to lay bare the onto-
logical foundations of that conception.”7 The traditional conception of truth,
as Heidegger characterizes it, treats assertions or judgments as the site of truth
and claims that the essence of truth lies in the correspondence or agreement
of an assertion with its object. Heidegger appeals to Aristotle who, “as the father
of logic, not only has assigned truth to the judgment as its primordial locus but
set going the definition of ‘truth’ as ‘agreement.’ ’’8 Yet Heidegger wonders what
enables assertions to correspond with objects and what is presupposed in this
particular definition of truth. “What,” he writes, “is tacitly posited in this rela-
tional totality of the adaequatio intellectus et rei? And what ontological charac-
ter does that which is thus posited have itself?”9 Heidegger’s answer to the first
question is Aletheia or unconcealment: if aspects of the world were concealed,
then our assertions would never be able to indicate those aspects or corre-
spond to them in an intelligible way. Hence, Heidegger calls unconcealment
the “primordial phenomenon of truth,” and he sees the agreement between
assertions and entities or aspects of the world as parasitic upon it.10 As for
the second question, Heidegger identifies the ontological character of Aletheia
with our disclosedness—“only with Dasein’s disclosedness is the most primor-
dial phenomenon of truth attained”—and he goes so far as to say: “There is’
truth only insofar as Dasein is and so long as Dasein is. Entities are uncovered
only when Dasein is; and only as long as Dasein is, are they disclosed.”11 And in
order to illustrate the extent to which propositional truth and our predicate
7 BT, 257 [214].

8 BT, 257 [214].
9 BT, 258 [215].
10 BT, 262 [219].
11 BT, 269 [226] “Dasein” is Heidegger’s term of art for human beings. The literal translation
of Dasein into English is “There-being.” Heidegger’s uses the term to capture the fact that
human beings are the world they find themselves in.
calculus is grounded in the disclosure of the world achieved in our practical

activities, it is worth analyzing Heidegger’s conception of correspondence and
unconcealment in greater detail.
Heidegger begins his account of propositional truth in Being and Time by
wondering: “With regard to what do intellectus and res agree?”12 He immedi-
ately rejects correspondence theories of truth that characterize truth as an
agreement between some sort of ideal, mental or linguistic representation
and reality itself. The problem with these theories, as Heidegger sees it, is their
inability to explain the way that intellectus and res agree. Alluding to Husserl’s
account of ideal intentional contents converging with real objects, Heidegger
writes:
This relationship [between intellectus and res] thus pertains to a connec-

tion between an ideal content of judgment and the Real Thing as that
which is judged about. Is this agreement Real or ideal in its kind of Being,
or neither of these? How are we to take ontologically the relation between
an ideal entity and something that is Real and present-at-hand? . . .
Or is the ontological meaning of the relation between Real and ideal
something about which we must not inquire? Yet the relation is to be one
which subsists. What does such ‘subsisting’ mean ontologically?
Why should this not be a legitimate question? Is it accidental that no
headway has been made with this problem in over two thousand years?
Has the question already been perverted in the very way it has been
approached—in the ontologically unclarified separation of the Real and
ideal?13
Elucidating truth in terms of a relation between ideal content and real objects
doesn’t amount to an account of the truth relation itself. And by positing a
distinction between the ideal and the real, Heidegger doubts whether “in their
kind of being and their essential content they give us anything at all with regard
to which they can agree.”14 In short, for a correspondence theory of truth to do
any work, it cannot simply assert that intellectus and res correspond; rather it
must specify the content of that so-called “correspondence.”
Although Heidegger rejects traditional correspondence theories of truth as
“very general and empty,” he is nevertheless concerned with offering a phe-
nomenological account of our experience with the ways in which assertions
12 BT, 258–9 [216].

13 BT, 259 [216–217].
14 BT, 258–9 [216].
56 wendland
correspond to the world.15 Correspondence, for Heidegger, is a characteristic

of our engagement with reality—“Asserting,” Heidegger writes, “is a way of
Being towards the Thing itself that is.”16—and far from identifying assertions
as a kind of ideal, mental or linguistic representation, Heidegger suggests we
experience assertions as directing us towards specific entities and aspects of
the world:
Suppose someone here in the classroom states the proposition ‘the board
is black’ and does so in an immediately given context of question and
answer. To what do we then attend in understanding the statement? To
the phonetic articulation? Or to the representation that performs the
making of the statement and for which then the sounds uttered are
‘signs’? No, rather we direct ourselves to the blackboard itself, here on
the wall! In the perception of this board, in making present and thinking
about the blackboard and nothing else, we participate in the perfor-
mance of the statement. What the statement immediately presents is
that about which it states something.17
In this example, the assertion “the board is black” corresponds not by repre-
senting a fact, say, the board being black, but by providing us with a certain
direction or orientation to the board that presents a fact about it: namely, that
it is black. Heidegger characterizes the orientation that enables an assertion
to indicate its object as a kind of “uncovering.” Through assertions “the entity
itself which one has in mind shows itself just as it is in itself; that is to say, it
shows that it, in its selfsameness, is just as it gets pointed out in the assertion
as being—just as it gets uncovered as being.”18 From here Heidegger writes:
To say that an assertion ‘is true’ signifies that it uncovers the entity as it is
in itself. Such an assertion asserts, points out, ‘lets’ the entity ‘be seen’ in
its uncoveredness. The Being-true (truth) of the assertion must be under-
stood as Being-uncovering.19
15 Heidegger’s rejection of traditional representational accounts of truth as “very general

and empty” (BT, 258 [215]) does not in principle rule out the possibility of a meaning-
ful description of the correspondence relation between the Real and ideal. But for an
account of the seemly insurmountable obstacles any such description faces, see Hubert
Dreyfus, Being-in-the-World (Cambridge: MIT Press, 1991), 121–127.
16 BT, 260 [218].
17 Martin Heidegger, Metaphysical Foundations of Logic, tr. Heim (Bloomington: Indiana
University Press, 1984) 125–6.
18 BT, 261 [218].
19 BT, 261 [217].
The being-true of an assertion as being-uncovering is, however, “ontologi-

cally possibly only on the basis of Being-in-the-world.”20 And Heidegger says
“this latter phenomenon, which we have known as a basic state of Dasein, is
the foundation for the primordial phenomenon of truth”: namely, Aletheia or
unconcealment.21
Aletheia, on Heidegger’s account, is the expression Ancient Greeks used to
capture the phenomenon of uncovering, unhiddenness, or unconcealment.22
Unconcealment is a privative notion that consists in “taking entities out of their
hiddenness and letting them be seen in their unhiddenness.”23 Hiddenness,
for Heidegger, “has a dual sense: 1. having no awareness of; and 2. having no
possible context.”24 Sense 1) refers to a kind of ontic or inner-worldly conceal-
ment in which something is manifest but we lack an explicit understanding
of it, whereas sense 2) refers to a kind of ontological or worldly concealment
whereby the context through which something can manifest itself is lacking.
The dual sense of concealment suggests that unconcealment comes in two
forms: the unconcealment involved in making what is manifest explicit, and
the unconcealment that creates the context by which entities can be manifest
in the first place.25 The being-true as being-uncovering of assertions appears to
be the kind of unconcealment that makes what is explicit. Through assertions,
Heidegger writes, “that which is manifest may be made explicitly manifest in
its definite character.”26 The disclosure bound up with the human way of being
seems to be the kind of unconcealment that creates the context by which enti-
ties can be manifest in the first place. “The uncoveredness of entities within
the world is grounded in the world’s disclosedness. But disclosedness is that
basic character of Dasein according to which it is its ‘there’.”27 Breaking Aletheia
down into unconcealment as making explicit and unconcealment as making
manifest highlights the extent to which the truth of assertions presupposes our
ability to disclose the world—for again if no entities or aspects of the world
were manifest, then there would be nothing for our assertions to correspond
20 BT, 261 [219].

21 BT, 261 [219].
22 In Being and Time, Heidegger’s uses “uncovering” and “unhiddenness” to translate
Aletheia, but in subsequent texts “unconcealment” is his word of choice.
23 BT, 262 [212].
24 Martin Heidegger, Sein und Wahrheit (Frankfurt am Main: Klostermann, 1998) GA 36/37,
188.
25 For a detailed account of the relationship between unconcealment and concealment,
see Mark Wrathall, Heidegger and Unconcealment (Cambridge: Cambridge University
Press, 2011) 1–2, 17–18, 57–71.
26 BT, 197 [155].
27 BT, 263 [220].
58 wendland
with—and it suggests that Heidegger was committed to what Charles Guignon

calls an instrumentalist view of language whereby
our ability to use language is grounded in some prior grasp of the nonse-
mantic significance of the context in which we find ourselves. It is only
because we have first understood the nature of reality that we can then
come to comprehend the meanings of words. Language is seen as a tool
for communicating and ordering this prior grasp of reality. Although lan-
guage may play a very important role in making the world intelligible, it
is itself possible only against the background of an understanding that is
nonlinguistic.28
Briefly, the truth of assertions is parasitic on the truth of our practices that
enable entities and aspects of the world to show up as the subject of our talk.
In order to come to terms with the non-semantic significance that under-
girds propositional truth and our predicate calculus, we need to look at
Heidegger’s account of the worldliness of our work-environment and the prac-
tical understanding that that environment entails.29 The worldliness of the
world is, for Heidegger, the referential totality that determines the meaningful
presentation of its component parts and helps define the identity of existent
human beings. Hammers, in other words, are the entities that they are in rela-
tion to wood, nails, tape measures, other such tools and ultimately our ability
to engage in the activity of carpentry. Similarly, a carpenter is a carpenter by
engaging in the act of carpentry and immersing herself in the equipmental
totality that makes that activity possible. Of course, carpentry is carpentry if
and only if it is directed to some sort of end, say that of building a house, and
Hubert Dreyfus defines the significance that grounds our linguistic activity
as “the relational whole of in-order-tos and for-the-sake-of-whichs in which
entities and activities that involve equipment have a point.”30 On Dreyfus’s
reading, we disclose the component parts that make up a particular activity by
engaging in that activity. He writes, “the whole current situation is articulated
by coping,” and he goes on to say, “when I pick up a hammer and hammer with
it, I articulate one of its significations, i.e. the fact that it is used to pound in
28 Charles Guignon, Heidegger and the Problem of Knowledge, (Indianapolis: Hackett, 1983)
117–118.
29 Heidegger’s account of the worldliness of the world is found in §18 of Being and Time
and his discussion of understanding occurs in §31–32. See BT, 114–122 [83–88], 182–203
[143–160].
30 Dreyfus, Being, 223.
nails; and if I use it to pull nails, I articulate another.”31 Put otherwise, our pri-
mary disclosure of entities, aspects and dimensions of the world to which our
propositions correspond is a function of our practical engagement with that
world in a specific, holistic context.
Taking our cue from Heidegger’s statement “to significations, words accrue”
and building on Heidegger’s claim that “language is a totality of words . . . and
as an entity within the world, this totality becomes something which we may
come across as ready-to-hand,”32 language can bee seen as a series of signs that
we use to indicate aspects of our world. As a tool, language is an integral part
of our practical activity (and not the distinct substance that it is thought to be
by traditional correspondence theorists) and it does its job in the context of
our activities by pointing out characteristics of the work in progress, providing
predicates for subjects involved in the referential whole, and allowing com-
munication to take place in the course of communal activity. As an example of
this phenomenon, take the assertion “The hammer is heavy”. Clearly, a ham-
mer cannot be heavy in isolation, but it can be too heavy for a specific task
in a given situation. And through an assertion, I can point all of this out to
someone on the job with me. In this example, the assertion “The hammer is
heavy” picks out a particular aspect of the our work-environment, ‘hammer’,
provides it with a predicate, “heavy,” and in doing so is capable of indicating
to others that this hammer is too heavy for the task at hand.33 Yet the point of
this illustration is not to show us what language is capable of doing, but rather
to highlight the non-semantic significance that enables it to do the things that
is does. As Dreyfus writes: “Language is used in a shared context that is already
meaningful, and it gets its meaning by fitting into and contributing to a mean-
ingful whole.”34
In his influential commentary on Being and Time, Dreyfus details the deriva-
tion of linguistic meaning from an inherited historical context, through practi-
cal understanding, interpretation and assertion, and finally to the predicate
calculus of modern logic. Ultimately, the meaning of a sign is grounded in a
broader, socio-historical background understanding into which all human
beings are thrown, and which Heidegger calls a “truth of being.” Heidegger
defines our “thrownness” in Being and Time as follows:
31 Dreyfus, Being, 215, 215.

32 BT, 204 [161].
33 I’ve borrowed this example from Dreyfus. See Dreyfus, Being, 209.
60 wendland
To Dasein’s state of Being belongs thrownness; indeed it is constitutive for

Dasein’s disclosedness. In thrownness is revealed that in each case
Dasein, as my Dasein and this Dasein, is already in a definite world and
alongside a definite range of definite entities within-the-world.35
Though truths of being are difficult to define precisely because they are the
background upon which we are able to define anything at all, they can be
crudely characterized as the worldviews of different epochs that allow differ-
ent entities and aspects of the world to show up as the entities and aspects that
they are. Heidegger captures the part truths of being play in our lives when
he writes:
The everyday way in which things have been disclosed is one into which
Dasein has grown in the first instance . . . In it, out of it, and against it,
all genuine understanding, interpreting, and communicating, all re-
discovering and appropriating anew, are preformed.36
And Dreyfus attempts to exemplify these truths of being when he asserts: “The
Greeks encountered things in their beauty and power, and people as poets,
statesmen and heroes; the Christians encountered creatures to be catalogued
and used appropriately, and people as saints and sinners; and we moderns
encounter objects to be controlled and organized by subjects in order to satisfy
their desires.”37 And whilst it may be difficult to tease out the concrete content
of the Greek, Christian and Modern worldviews, formally these truths of being
are simply the set of norms and relations on the basis of which all activities
and entities in the Greek, Christian and Modern worlds make sense.
Although a truth of being opens up the world in countless ways and presents
us with numerous possibilities, the actual significance of our inherited possi-
bilities is initially articulated when we takes a stand on our existence. Taking
a stand on our existence is achieved when we adopt a particular potentiality-
for-being, exhibit a certain practical understanding of the world, and express
that possibility’s importance to us by projecting it into the future. Adopting a
particular possibility also involves revealing certain entities and aspects of the
world and concealing others. In the act of hammering a carpenter discloses the
hammer as a tool designed for pounding nails into a piece of wood, yet ham-
mering itself fails to reveal the chemical composition or physical properties
35 BT, 264 [221].

36 BT, 213 [169].
that enable a hammer to do the job that it does. Perhaps more to the point,
successfully engaging in the act of carpentry precludes the carpenter from
pausing her activity in order to analyze the chemical composition or physical
properties of a hammer in the way a chemist or physicist would. And by taking
a stand on our existence through, say, becoming a carpenter, or a chemist, or a
physicist, we hone in on and articulate one way of being-in-the-world that our
socio-historical background makes available.
The specific significance that we articulate in our active engagement with
the world is capable of being explicitly expressed in interpretation. “That which
is disclosed in understanding,” Heidegger writes, “is already accessible in such
a way that its ‘as which’ can be made to stand out explicitly. The ‘as’ makes up
the structure of the explicitness of something that is understood; it constitutes
the interpretation.”38 Whereas understanding discloses entities and aspects of
our world through a certain use, interpretation consists in abstracting from
our practical engagement with the world and using signs to explicitly see what
we understand as the entities or aspects that they are. Heidegger elucidates
the transition from tacit practical understanding to explicit linguistic inter-
pretation by breaking down the “fore-structure” of interpretation into its basic
parts—“fore-having,” “fore-sight,” and “fore-conception”—and highlighting
the extent to which the latter concepts presuppose yet explicitly articulate the
former. “Fore-having” refers to the truth of being or background intelligibility
we have prior to any practical understanding or interpretation:
In every case this interpretation is grounded in something we have in

advance—in a fore-having. As the appropriation of understanding, the
interpretation operates in being towards a totality of involvements which
is already understood.39
As we’ve seen, the socio-historical background into which we have been

thrown serves as the foundation for our practical engagement with the world.
But taking up a practical activity amounts to adopting a certain perspective on
our cultural inheritance. And Heidegger calls the perspective we take on our
cultural inheritance “fore-sight”:
When something is understood but is still veiled, it becomes unveiled by

an act of appropriation, and this is always done under the guidance of a
point-of-view, which fixes that with regard to which what is understood is
38 BT, 189 [149].

39 BT, 191 [150].
62 wendland
to be interpreted. In every case interpretation is grounded in something

we see in advance—in a fore-sight. This fore-sight ‘takes the first cut’ out
of what has been taken into our fore-having, and it does so with a view to
a definite way in which this can be interpreted.40
Our background intelligibility, fore-having, and practical understanding, fore-

sight, facilitate our understanding of the world before any explicit interpre-
tation has occurred. And once certain entities and aspects of the world are
presented through our cultural inheritance and practical activity, we reach the
final feature of Heidegger’s fore-structure of interpretation, “fore-conception”:
Anything which is held in our fore-having and towards which we set our
sights fore-sightedly’, becomes conceptualizable through the interpreta-
tion. In such an interpretation, the way in which the entity we are inter-
preting is to be conceived can be drawn from the entity itself, or the
interpretation can force the entity into concepts which it is opposed in its
manner of Being. In either case, the interpretation has already decided
for a definite way of conceiving it, either with finality or with reservations;
it is grounded in something we grasp in advance—in a fore-conception.41
Heidegger’s appeal to “conceptualization” and his claim that “interpretation

has already decided for a definite way of conceiving” suggests that the third
stage in his interpretive fore-structure entails seeing something as something.
Once entities and aspects of our world have been disclosed in a discrete way,
it is then possible to assign predicates to subjects and make judgments about
them. As Dreyfus puts it: “Once the subject and predicate are isolated the tra-
dition since Aristotle assumes they must be bound together in a judgment.”42
This, for Dreyfus, “leads to modern logic and makes possible computational
formalization.”43 And in Heidegger’s words:
Binding and separating may be formalized still further to a ‘relating’.

The judgment gets dissolved logistically into a system in which things are
‘coordinated’ with one another; it becomes the object of a ‘calculus’.44
40 BT, 191 [150].

41 BT, 191 [150].
44 BT, 202 [159].
Heidegger, however, is quick to point out that “this calculus cannot disown
its ontological origin from an interpretation which understands.”45 And this
means that our most abstract languages “depend on a series of background
practices on the basis of which all activities and objects are intelligible or make
sense.”46
With the derivation of linguistic meaning from a socio-historical truth
of being in place, we are finally in a position to see how the unconcealment
achieved in our practical activities is a condition of possibility for proposi-
tional truth and our predicate calculus. Propositional truth presupposes the
unconcealment brought about through our practical activities to the extent
that assertions are about something in the world, and that an assertion’s truth-
value is determined by that about which the assertion asserts. To continue
with our earlier example of a hammer, the assertion “The hammer is heavy”
is true or false in virtue of being about something: namely, a hammer. But
for this assertion to be about a hammer, the hammer must be disclosed in a
certain context. As Heidegger puts it: “in order for something to be a possible
about-which for an assertion,” the about-which of the assertion “must already
be somehow given for the assertion as unveiled and accessible.”47 Assertions,
however, are not simply about something that has been disclosed, since they
also make specific claims about that which is disclosed. To make an assertion
about a hammer, for instance, the assertion must focus on a particular aspect
of the hammer: namely, its heaviness. Yet in order for an assertion to focus
on a definite aspect of a hammer, say, its heaviness, that aspect must itself be
disclosed in the context of our practical activities: that is, the hammer must be
too heavy for the task at hand. Finally, the ability of true or false assertions to
assert something meaningful about the world depends on the context in which
entities and aspects of the world are disclosed. In Mark Wrathall’s words:
A sculptor and a carpenter might mean very different things in asserting

‘the hammer is heavy’ as a result of the different practices, goals, equip-
mental contexts, etc., within which they each use a hammer. Likewise,
whether the assertion ‘the hammer is heavy’ is true will depend on the
background which is ‘communicated’ by the speaker and hearer.48
45 BT, 201 [158].

47 Martin Heidegger, Basic Problems of Phenomenology, tr. Hofstadter (Bloomington: Indiana
University Press, 1985), 208.
48 Mark Wrathall, “Heidegger and Truth as Correspondence,” International Journal of
Philosophical Studies 7 (1999) 69–88, 81–82.
64 wendland
Wrathall’s appeal to the distinct background of sculptors and carpenters high-

lights the fact that the meaning and truth of assertions are determined by the
context in which they are used and the entities and aspects of the world dis-
closed therein. And insofar as the meaning and truth of assertions presuppose
the disclosure of entities and aspects of our world in the context of our practi-
cal activities, the verification of empirical propositions and their formalization
into a predicate calculus presupposes the practical activities of human beings
who are capable of questioning their being by choosing between the various
possibilities that their culture inheritance makes available and then project-
ing one of those possibility into the future. In a word, the idea of “logic” itself
disintegrates in the turbulence of a more original questioning: the questioning
of human beings.
3 Conclusion
By illustrating the extent to which the meaning and truth of our most abstract
languages depend on our practical activities, Heidegger corrects for a lacunae
in Carnap’s and Ayer’s work by explaining how the reality that is said to verify
our empirical propositions is actually given to us. Briefly, our practical activi-
ties present certain salient entities and aspects of the world such that they
can be the subject matter of our discourse or symbolized in our predicate cal-
culus. This presentation is temporal insofar as our current practical engage-
ment with the world presupposes an appropriation of a potentiality-for-being
available our cultural inheritance along with the possibility of projecting that
potentiality into the future. And it is historical to the extent that human beings
are bound by the possibilities available in a certain socio-historical situation.
What, of course, remains to be show is whether or not the language Heidegger
uses to describe the practical activities that make propositional truth possible
is itself meaningful or mere “metaphysical nonsense.” But this task must be
deferred to another occasion.
Bibliography
A.J. Ayer, Language, Truth and Logic (London: Penguin, 2001).

William Blattner, Heidegger’s Temporal Idealism (Cambridge: Cambridge University
Press, 1999).
Rudolf Carnap, “The Elimination of Metaphysics through the Logical Analysis of
Language,” in Heidegger and Modern Philosophy: Critical Essays, ed. Michael Murray
(New Haven: Yale University Press, 1978).
Hubert L. Dreyfus, Being-in-the-World: A Commentary on Heidegger’s Being and Time,

Division I (Cambridge: MIT Press, 1991).
Charles Guignon, Heidegger and the Problem of Knowledge (Indianapolis: Hackett,
1983).
Martin Heidegger, Being and Time, tr. Macquarrie and Robinson (New York: Harper
and Row, 1962).
———, Basic Writings (New York: HarperCollins, 1993).
———, Basic Problems of Phenomenology, tr. Hofstadter (Bloomington: Indiana
———, Sein und Wahrheit (Frankfurt am Main: Klostermann, 1998).
Cristina Lafont, Heidegger, Language and World-disclosure, tr. Harman (Cambridge:
Cambridge University Press, 2000).
Mark Okrent, Heidegger’s Pragmatism: Understanding, Being and the Critique of
Metaphysics (Ithaca: Cornell University Press, 1992).
W.V.O. Quine, From a Logical Point of View (Cambridge: Harvard University Press, 1980).
Wilfrid Sellars, Empiricism and the Philosophy of Mind (Cambridge: Harvard University
Press, 1997).
Wrathall, Mark Heidegger and Unconcealment (Cambridge: Cambridge University
Press, 2011).
Mark Wrathall, “Heidegger and Truth as Correspondence,” International Journal of
Philosophical Studies, 7 (1999), 69–88.
chapter 4
The Obstacle
Jacques Lacan’s Critique of the Formal Logical Representation of the
Real
Ervik Cejvan
1 Introduction
The writings of the french psychoanalyst Jacques Lacan (1901–1981) come to

the point of being impossible to read. Lacan actually emphasized that his
discourse cannot be treated academically, for example in a thesis.1 It takes an
almost inhuman effort to really grasp all the puns and references which popu-
late virtually every one of Lacan’s sentences. However, an anchoring point in
my reading of Lacan, which I suggest is conducive to understanding the spe-
cific character of Lacan’s thought, is to treat his psychoanalytic theory in rela-
tion to his thought on psychoanalytical practice, especially as that practice
was informed by Freudian practice. In doing so, I aim to show that Lacan has
sought to insert the great Freudian discovery of the unconscious to the vocab-
ulary of philosophers, anthropologists, theologians, and linguists. For Lacan,
the Freudian “discovery” of the unconscious could enlighten scientific thought
after Freud.
Within this interpretative frame, I argue in this chapter that Lacan’s psycho-
analytical concept of the obstacle (l’obstacle) is nothing short of a critique of
the formal logical representation of the question fundamentally concerning
ontology, namely what is. With this argument, we will be able to better under-
stand Lacan’s elusive concept of “the real” (le réel), since the argument will
provide a theoretical context within which Lacan used it.2 The “real” in Lacan’s
1 Anika Lemaire, Jacques Lacan, trans. David Macey (London and Boston: Routledge & Kegan
Paul, 1977) vii; Jacques Lacan, The Other Side of Psychoanalysis, trans. Russell Grigg (New York:
Norton, 2007), 33.
2 Lacan proposes this critique in formal terms in his so-called “formula of sexuation.” Lacan
does not aim to demonstrate how things are—as he is particularly critical of metaphysical
approaches to the real—but rather to falsify the idea that things outside language are as they
are signified (by language). In particular, in the “formula of sexuation” Lacan aims to falsify,
first, the common idea that human relations are determined by sex and gender; and second,
the heteronormative idea that in terms of sexual difference, an exceptionally female (“clito-

The Obstacle 67
theory indicates an X outside language, which “resist symbolization.” Drawing

on this assumption, Lacan’s concept of the obstacle is inserted into the “real” as
that which resists symbolization, and by extension, logical formalization.3 This
ontological condition, i.e. that the real cannot be symbolized and hence for-
malized, I argue, effectively enables a Lacanian critique of the possibility of a
formal logical representation of metaphysical concepts.
2 Background
Psychoanalysis is based on the Freudian theory of the unconscious. Lacan

sought to preserve the Freudian way of explaining psychoanalytic theory
through discussing and comparing it with concepts from contemporary sci-
ence, particularly linguistics, neurology, and thermodynamics. However, the
concept of the unconscious, upon which Freud outlined psychoanalytical
theory, was based on observations derived from analytic practice. The uncon-
scious is, and was already in Freud’s time, a scientifically obsolete or even
impossible concept (This is perhaps why Freud called psychoanalysis an
impossible profession).
In the 1960s, Lacan was concerned with the question of the requirements
for psychoanalysis to become be a science. Freud hoped that this could be
done in the course of future developments in neurology. Lacan, however, bases
psychoanalytical theory in linguistics, and later in formal logic, consequently
refuting the biological reductivist approach to psychoanalysis. This refutation
underlies Lacan’s famous statement that “the unconscious is structured like
a language.”4 In the appropriation of linguistics for the sake of psychoanaly-
sis, Lacan eventually embarks on the question concerning the formal logical
inscription of the unconscious. This move creates a dilemma for Lacan: how
can one show or represent the unconscious formal-logically, since in Freud,
whom Lacan follows, the unconscious knows no contradiction? Thus, prior to
rial”) enjoyment is different to male (“phallic”) enjoyment. Lacan does not, however, deny the
fact that people are burdened by the problems directly related to sex and gender. See Jacques
Lacan, Encore: On Feminine Sexuality, the Limits of Love and Knowledge, 1972–1973, ed. Jacques-
Alain Miller, trans. Bruce Fink (New York: W.W. Norton and Company, Inc., 1999).
3 “le réel est en somme ce quelque chose qui résiste absolument à la symbolisation.” See Jacques
Lacan, Freud’s Papers on Technique, 1953–1954, trans. John Forrester (New York: W.W. Norton,
1998), lecture 17 February 1954.
4 Jacques Lacan, Écrits, trans. Bruce Fink (New York: W.W. Norton and Co., 2006), 736–737;
222–223; 391.
68 cejvan
any formal designation of the logic of the unconscious Lacan has to respond to
the law of non-contradiction, which effectively poses as an obstacle for logical
formalization of the unconscious. This, then, was a key dilemma for Lacan, and
it enables us to better grasp the problematic character of his psychoanalysis.
Lacan embarks on the concept of the obstacle particularly in his twenti-
eth seminar from 1972–1973, published as Encore, in reference to Aristotle’s
analytic concept of enstasis (ενστασις).5 Lacan’s obstacle entails the double
meaning of Aristotle’s enstasis, namely as an “instance” and an “objection.”
I argue that such a rendition is decisive for grasping the Lacanian idea of the
obstacle, more precisely as an obstacle to a logical formalization of the real,
which for Lacan is identical to the unconscious. Such an idea traditional logic
does not take into account simply because, as Lacan maintains, it could not do
so prior to its Freudian discovery.6 The unconscious is nevertheless an underly-
ing condition of both language and logic, indeed any symbolic system, and so
underlies even principles such as the logical operation of negation, separation,
exclusion, along with the pillars of Aristotelian logic. The unconscious under-
lies logic on the level of the structure of language, since, according to Lacan,
the unconscious is structured like a language.
In the later stages of his career, Lacan attempts to show this formally by
inscribing psychoanalytical theory in terms of an idiosyncratic version that
fuses Aristotelian logic, modern symbolic logic, and Lacan’s own formal opera-
5 In Latin, the Greek enstasis is translated as “instantia” and in English as “objection.” Aristotle
defines enstasis: “An objection [enstasis] is a premise contrary to a premise.” See Aristotle’s
Prior Analytics, Book II, Chapter XXVI: Aristotle, Analytica Priora (Prior Analytics), trans.
A.J. Jenkinson, in The Works of Aristotle, Vol I, ed. W.D. Ross (Oxford, Clarendon Press, 1928).
Also see Bruce Fink, “Preface,” in Jacques Lacan, Encore: On Feminine, the Limits of Love and
Knowledge, 1972–1973, ed. Jacques-Alain Miller, trans. Bruce Fink (New York: W.W. Norton and
Company, Inc., 1999), vii–1: “enstasis is the obstacle one raises to an adversary’s argument; it
is also the exception to a universal predicate, hence an instance or counterinstance that
refutes a general claim. This is but one example of the inappropriateness of translating
Lacan’s ‘Instance de la lettre’ as ‘Agency of the Letter.’ ” Lacan also refers to Aristotle’s Rhetoric
and Topics in the Encore: “Consult Aristotle and you will know everything when I at last come
to this business of the enstasis. You can read, one after the other, the passage in the Rhetoric
and the two sections of the Topics that will allow you to truly know what I mean when I try to
integrate my four formulas, $x–Φx and the rest, into Aristotle’s work:” Lacan, Encore, 70,
70 n16. Also, see the useful information in Fink’s “Introduction,” regarding Lacan’s exact refer-
ences to Aristotle in “The Title of the Letter” in Écrits.
6 The philosophical literature on the concept of obstacle in Lacan is rather sparse. One excep-
tion is Jean-Michel Rabaté, who discusses this concept in relation to French phenomenology:
Jean-Michel Rabaté, “Strasbourg: The Nancy School?”, url: http://www.lineofbeauty.org/
index.php/s/article/view//62/135, accessed September 1 2013.
The Obstacle 69
tors. However, notwithstanding these oddities, Lacan does not denounce

logic as such. Rather, Lacan’s critique of logic is related to his own endeavor
to formalize psychoanalytic theory through his so-called mathèmes, a sys-
tem of signs especially designed to communicate psychoanalytical theory to
non-psychoanalytical scientists. Formalization was thus undertaken in the
course of the Lacanian development and dissemination of psychoanalyti-
cal theory. Lacan’s formalizations in particular signal his efforts to commu-
nicate the Freudian discovery of the unconscious to linguistics, philosophy
and even theology, beyond the tendency of its confinement within the walls
of the p sychiatric clinic and the Lacanian seminar. This proved to be a chal-
lenge since, according to Lacan, the pillar of psychoanalysis is not theory but
practice. Psychoanalysis is, so to speak, a talking cure and an art of the inter-
pretation of speech. It is precisely through the analytic experience of work-
ing on the interpretation of desire, unconsciously articulated in the analyzed
subject’s speech, that the analyst learns something about psychoanalysis, and
not primarily in the application of a theory. For this reason, along with Lacan’s
eccentricities, Lacan’s attempts to convey the lessons of practice to outsiders
was bound to fail perhaps from the start.
However, there is point to this aspect in Lacan that holds more promise for
non-psychoanalysts. For Lacan, psychoanalysis is not based on theory, because
it is not based, as all theory is, on the Cartesian cogito, which is “the subject of
science” according to Lacan. The subject of psychoanalysis is the subject of the
unconscious, and so unlike the subject of any theory or of the Cartesian cogito,
the unconscious does not exist where it thinks.7
As I just mentioned, Lacan’s project of the trancscription of psycho-
analytical concepts in something resembling formal logical terms culmi-
nates—and fails as a coherent exposition—in the Encore. At least two,
cryptic, principles are central to Lacan’s Encore: “There is no sexual relation”
(Il n’y a pas de rapport sexuel; which first appears in Lacan’s seminar L’envers
de la psychanalise, from 1969–1970) and “the Woman does not exist” (La femme
n’existe pas; first introduced in the Encore). There are Lacanian analysts and
theorists, like Catherine Clément, who acknowledge their confusion regard-
ing Lacan’s Encore, one of the last seminars which actually concerns psycho-
analysis, especially its practice. However, the purported monsters of the Encore
such as “There is no sexual relation” and “the Woman does not exist,” I believe,
could not have been said by Lacan other than onto-logically, especially as
7 Lacan, Écrits, 129; 171; 727; Jacques Lacan, The Four Fundamental Concepts of Psychoanalysis,
trans. Alan Sheridan (New York: W.W. Norton and Co., 1998), 35–37; 44; 222; 224.
70 cejvan
they are strictly confined to the “inscription” of Lacan’s so-called “formula of

sexuation.”8
Lacan’s “There is no sexual relation” and “the Woman does not exist,” I sug-
gest, should be understood in the context of Lacan’s dealings with the ques-
tion of sexual difference in psychoanalysis. They go back to the late sixties and
Lacan’s elaboration of the mathèmes, which, as mentioned earlier, constitute
a formalized system of signs designed to exlusively represent the concepts of
Lacan’s psychoanalytical theory. In the late 1970s Lacan eventually seems to
abandon the idea of the representation by the mathèmes as he, for the pur-
pose of communicating psychoanalytic theory to non-specialists, finds a more
effective way to demonstrate psychoanalytical concepts visually, in three
dimensions, by using topological models and knots.
3 The Real
According to Lacan, the real cannot be represented in language since it is

outside of language. The correspondence between the real and language is
structurally determined to fail. In other words, as Lacan explains, relying on
Saussurean linguistic: while signifying the signified the signifier does not sig-
nify the signified.9 Reaching the real would require a rather literal version of a
Cartesian way of seeing things clearly and distinctly.
In what respect are we most certain of what is? People have an immediate
ontological sense of their being in the world, perceiving one’s body for exam-
ple. What nevertheless, according to Lacan, poses the obstacle to this immedi-
ate knowledge is discourse, e.g. explanations which seek to reveal how this,
whatever this is, is happening. In discoursing, according to Lacan, humans are
displacing their immediate access to the real for a (failed) symbolic represen-
tation, or knowledge based on, for instance, a formal proof. The logical con-
sistency of that proof may refer to the consistency of logic, however, the real
escapes that consistency.
Up until the 1970s, the way to render the obstacle posed to the real, Lacan
suggests, is by mathematization, which for Lacan is properly meaningless, and
8 See Lacan, Encore, 64–90 (the “formula” is presented on page 73). I will not explicate Lacan’s
“formula of sexuation” here as it would cause a discussion which violates the delimitation of
this chapter. For an important discussion of the the logic of sexuation see Ellie Ragland-
Sullivan, The Logic of Sexuation: From Aristotle to Lacan (Albany: State University of New York
Press, 2004).
9 Lacan, Écrits, 414–418.
The Obstacle 71
as such touches the real.10 Thus, the real is that which is meaningless, and since
formalizations are meaningless, according to Lacan, they best capture the
real.11 The mathematical sign is precisely what it is, it enounces itself precisely
as the real, as an immediate existence, here and now, beyond representation
in discourse (the notion of representation that I employ in this Lacanian con-
text does not concern the representation (or meaning) on a cognitive and neu-
rological level). Representation here means a representation in and through
language, and language is already a representation, since it structurally does
not signify anything real but yet another signifier (the signifier does not sig-
nify the signified, as we have it in Lacan’s appropriation of Sussurean linguis-
tic). The immediacy of the real is beyond signification and hence beyond
representation.
A discourse that desires to ascertain the “status” of the real fails to do so and
effectively poses an obstacle to the immediacy of the real. This means that not
mathematics, but Lacan’s mathèmes are designed to represent any notion in
Lacanian psychoanalytic theory with a mathème unique to a particular notion.
For example, the mathème “J” signifies “enjoyment” ( jouissance), such as it
is designated as a theoretical concept in Lacanian psychoanalysis. The “a” is
the object-cause of the desire (objet petit a), the “A” is the Other (l’Autre), and
so on.12 The purpose of this formalization was to provide a system which
could be exact in the sense of being properly meaningless to any discursive
approach, including the psychoanalytic discourse which is inscribed in it and
which it represents.
But from the Encore and onward, Lacan begins to question the utility of the
mathèmes, since he comes to realize that they actually still work on the level
of symbolic representation, whereas the real resists symbolization, and it pre-
cisely to suggest what is at stake in knowledge and experience of the real that
Lacan devised the mathèmes. Thus, they fail to convey what Lacan is getting at
10 “Mathematization alone reaches a real—and it is in that respect that it is compatible with

our discourse, analytic discourse—a real that has nothing to do with what traditional
knowledge has served as a basis for, which is not what the latter believes it to be—namely,
reality—but rather fantasy. The real, I will say, is the mystery of the speaking body, the
mystery of the unconscious”: Lacan, Encore, 131.
11 Lacan, Encore, 131.
12 For a Lacanian discussion on the mathème see the papers by the participants of the
“cartel” at École de la Cause Freudienne, Guy Briole, Hervé Castanet, Françoise Fonte-
neau, Victoria Horne, Pierre Streliski, “Le mathème lacanien: l’écriture de la psychanalyse,”
url: http://www.causefreudienne.net/etudier/essential/le-matheme-lacanien-l-ecriture-
de-la-psychanalyse.html?symfony=823046cb461701a780c5d4b2d12fd1f, accessed Septem-
ber 1, 2013.
72 cejvan
on the one hand, and they fail to convey the real (even in their failures). Lacan’s
formulas of sexuation, especially, illustrate the failure of the Lacanian math-
ematization of psychoanalytic discourse. To find a better way of discussing the
real, Lacan eventually embarks on the project of showing it by the means of
topological models and knots, hoping that his new formalization, which is still
very much like the old one, would enable a communication of psychoanalyti-
cal discourse to the empirical sciences.13 Whether Lacan’s mathematizations,
in any form, manage to reach the real or not is a discussion which must be
omitted here. What is of concern in this chapter, to repeat, is how they enable
us to understand Lacan’s thought.
4 Lacan on Logic
According to Lacan, what logic proves, and what ontology instigates through
a discourse on what is, is merely what we already know or believe there is: it
seems nevertheless that the problem is that we are never certain of what is.14
Thus, neither metaphysical speculation nor logic, which since Aristotle and
especially among the Scholastics have been producing arguments to fit meta-
physical speculations into a coherent and seriously argued discourse, seem to
ground our knowledge of what there really is, beyond what poses as an obsta-
cle to this knowledge, namely what we already only suppose to know.
In Encore Lacan embarks on a task to approximate the psychoanalytic idea
of the unconscious, its most fundamental idea, in terms of another analytics
which have proved to be effective throughout the history of Western thought,
including modern science—namely the analytics of Aristotle. His purpose is
to communicate psychoanalytical concepts to empirical sciences and to the
13 For Lacan’s elaborations on science, mathème, linguistics and topology see Dany Nobus,
“Lacan’s Science of the Subject: Between Linguistics and Topology,” in The Cambridge
Companion to Lacan, ed. Jean-Michel Rabaté (Cambridge: Cambridge University Press,
2003), 50–69; Bernard Burgoyne, “From the Letter to the Matheme: Lacan’s Scientific
Methods,” in The Cambridge Companion to Lacan, ed. Jean-Michel Rabaté (Cambridge:
Cambridge University Press, 2003), 69–86.
14 Lacan articulates this particularly in reference to death as a domain of faith, the dilemma
of knowing that we will die, and yet, that we are unable to ascertain this as a fact until it
actually happens, and hence, only then, can be ascertained as such. See a video record-
ing of the conference at the University of Louvain October 13, 1972, which is available
in the movie Lacan Parle (1972) by François Wolff. A transcript (in French) is available
at url: http://www.psychasoc.com/Textes/La-mort-est-du-domaine-de-la-foi, accessed
September 1, 2013.
The Obstacle 73
study of logic. On his path of a radical interpretation of Freud, displacing the

Freudian field of psychoanalysis from neurology to linguistics, and then mov-
ing beyond a structuralist approach, Lacan embarks on a question concerning
the formal logical writing of psychoanalytic discourse.
The problem that Lacan identifies is that Aristotelian logic does not allow
an articulation of the logic of the unconscious, since, according to Freud, the
unconscious knows no contradiction. In the seminar Le logique du fantasme
from 1966 Lacan deals explicitly with the question of the inscription of psycho-
analytical theory in the writing of formal logic. It is this investigation that cul-
minates in the Encore. There is thus a crucial connection between Le logique
and Encore regarding the status of Aristotelian logic in Lacanian psychoana-
lytic discourse. Lacan’s aim is not to renounce the Aristotelian foundation of
logic in toto. Aristotle actually brings logic on the level of a philosophical and
scientific enquiry into the “substance of being” (ousia), which after all is what
Lacan seems to do for psychoanalysis. Psychoanalysis, however, is a discourse
of the unconscious, and as such is unaccountable in terms of Aristotelian, and
also predicate logic.15
As Lacan informs us in Le logique, idiosyncratically with reference to the
Vietnam War and Bertrand Russell, his interest in logic has to do with the ques-
tion of a relation between the formal writing of the logic and the truth of an
event. This has a concrete implication for psychoanalysis:
The relation of the truth to the signifier, the detour through which ana-
lytic experience rejoins the most modern process of logic, consists pre-
cisely in the fact that this relation of the signifier to the truth can
short-circuit all the thinking which supports it. And just as a sort of aim
is outlined at the horizon of modern logic—one which reduces logic to a
correct handling of what is simply writing—in the same way for us, the
question of verification, concerning what we have to deal with, passes
along the direct line of the operation of the signifier, in so far as on it
alone the question of the truth remains suspended.16
Hence, Lacan’s involvement with formal logic is to establish that psychoanal-

ysis shares the same questions with logic regarding the questions of formal
writing, truth, verification, and so on, but that it can say something about
these issues which formal logic never can. Lacan thus embarks on the task of
15 Lacan, Encore, 78 n1.

16 Jacques Lacan, Le Séminaire XIV: La logique du fantasme, 1966–1967 (unpublished),
lecture dated December 7, 1966 (in author’s possession).
74 cejvan
d iscerning a logic of the unconscious, since both traditional (Aristotelian) and

modern formal logic is unable to deal with the contradictions which animate
the unconscious, and also since psychoanalysis deals with concepts such as
fantasy, desire and enjoyment, which cannot be verified by way of traditional
and modern logic.
Apart from being Lacan’s critique of the formal logical representation of
the real, Lacan’s reflections on logic are at the same time conceptual tools for
scientifically communicating psychoanalytical theory. The reader will recall
that science for Lacan is based on the Cartesian cogito, which he designates
as the subject of science. Psychoanalysis, on the other hand, is a discourse on
the subject of the unconscious. Contrary to the subject of the Cartesian cogito,
Lacan maintains, the ontological status of the subject of the unconscious is not
located in thinking, where Descartes inserts the consciousness of existence.
The unconscious, psychoanalytically speaking, does not know that is exists.
Psychoanalysis is aware of its existence merely due to the fact that it speaks
(the unconscious), according to Lacan.
5 The Obstacle
Today I must break new ground—with the notion of the obstacle, with
what in Aristotle’s work—whatever else may be said, I prefer Aristotle to
Jaufré Rudel—is precisely called the obstacle, ενσταις [enstasis].17
The obstacle is a rather unstudied, but significant Lacanian concept. As far as

I can see, it first appears (as a concept) in Encore, in reference to Aristotle’s ana-
lytical concept of enstasis. Moreover, the concept of the obstacle is operative
in a seminar entitled Les non-dupes errent (1973–1974), where it is developed
further in terms of “phallic enjoyment,” defined as “what creates an obstacle to
the sexual relation.”18 Later still, the obstacle is depicted visually in Lacan’s use
of topological models. For example in the seminar RSI the obstacle is instanced
simply as an obstacle inside a torus, structurally determined by the shape of
the torus itself, which offers a clear and distinct depiction of the unconscious
17 Lacan, Encore, 69. Jaufré Rudel was a French 12th century troubadour famous for his
poetry of the “love from afar.”
18 Jacques Lacan, Le Séminaire XXI: Les non-dupes errent (unpublished), lecture dated
May 21, 1974 (in author’s possession).
The Obstacle 75
instanced in a language, which perhaps can qualify as a rather accurate varia-

tion on Lacan’s “the unconscious is structured like a language.”19
Lacan’s interpretation of Aristotle’s enstasis as an obstacle designates, I sug-
gest, an inherent non-relation instanced between any real object and its sym-
bolic representation. Plainly speaking, two supposedly separate modes of the
symbolic representation of the real, one logical and the other metaphysical,
basically operate within the same domain of interpretation, namely within the
discourse of language. Neither logic nor metaphysics can thus offer an accu-
rate representation of the real. Hence, the obstacle is the instance of language,
which is an obstacle to the metaphysical interpretation of the real.
By introducing the concept of the obstacle in the Encore Lacan targets
the problem concerning the formal representation of the real. In reference
to Aristotle’s enstasis Lacan discerns a problem concerning particularly the
Freudian idea of sexual difference, according to which the sexed being of
man is supposed as distinct from that of woman, based on the obvious fact
that woman lacks the male sexual organ. Freud does not stop there and even
puts the woman in a position where she envies man’s triumphant possession.
However, for Lacan this is not a reason for rejecting Freud. Instead, Lacan
embarks on a task of updating the Freudian idea of sexual difference by put-
ting its logic to the test. In the process of doing so, Lacan yet again reveals, now
from another angle from the one previously discussed, the problem concern-
ing the accuracy of the formal logical representation of the real.
Lacan thus introduces the concept of the obstacle in his critique of sexual
difference in the Encore, and he renders this critique in logical terms. But as
this logic is thoroughly meshed with Lacan’s own concepts, which are clearly
idiosyncratic and incoherent, it paradoxically prevents what Lacan was, per-
haps self-consciously paradoxically, trying to do: communicate the findings of
his psychoanalytic discourse to any other (scientific) discourse based on the
principles of traditional and modern logic, such as the law of identity, the prin-
ciple of non-contradiction, and the principle of the excluded middle. However,
Lacan’s obstacle concerns an obstacle (non-relation) instanced between the
real and its symbolic representation.
6 An Obstacle to the Sexual Relation
Lacan’s discussion on the obstacle also appears in the context of Lacan’s cri-
tique of the logic of sexual difference, which according to Lacan’s critique is
19 Jacques Lacan, Le Séminaire XXI: RSI, 1974–1975 (unpublished), lecture dated March 18,
1975 (in author’s possession).
76 cejvan
supposedly the logic of a binary relation. Prior to a formal inscription of the

logic of sexual difference in his so-called “formula of sexuation” in the Encore,
in a seminar held a few years before The Other Side of Psychoanalysis, Lacan
articulated his famous or infamous principle that “there is no sexual relation.”
I interpret this principle as meaning that there is no rapport between the desig-
nation of gender and biological sex. What we are dealing with in this principle
is the non-relation based in psychoanalytic experience according to Lacan,
more precisely of an impossibility to logically verify the proposition of a rapport
between gender and biological sex. Put in “Lacanese,” the sexual relation is sup-
posed to be true, particularly on the level of the logical structure of language,
which supposedly captures a biological difference.
We will remember that according to Lacan, the human being is a “speaking
being” (parlêtre), and through which the unconscious, in turn “structured like a
language,” speaks. Sexuality is inscribed in this human condition, but the ques-
tion of the sexed being concerns the body and not language. The latter, not the
former, is the sole domain of psychoanalysis. Hence, Lacanian psychoanalysis
does not deal with biological question of the sex or the sexes. Particularly in
Encore, Lacan is critical of a psychoanalysis approaching the analyzed sub-
ject not primarily as a speaking being but as a sexed being. This is dangerous
because it subdues the analyzed subject, the subject of the unconscious, to its
“right” place on the chart of a heternormative biological-reductivist logic of
sexual difference. What Lacan designates as the obstacle is the idea of sexual
difference posed as an obstacle between the speaking being and the psycho-
analytic discourse of the unconscious.
In the “formula of sexuation” presented in the Encore, I suggest, Lacan
intends to demonstrate formally what he previously proposed in The Other
Side of Psychoanalysis as a doctrine of psychoanalytic discourse, namely that
“there is no sexual relation.” In this formal logical inscription of sexual differ-
ence, Lacan deals with the idea of a relation between the male and the female
sex, determined logically in terms of a binary relation.
However, the inscription of Lacanian symbols in the formula of sexuation
intends to emphasize the fact that psychoanalytic practice does not operate on
the level of sexual difference, not even when it deals with love and sexuality,
but ultimately at the level of the unconscious, which is not sexed but speaking.
Lacan’s “there is no sexual relation” can thus be interpreted as there is no sexual
relation between speaking beings.20
20 The Lacanian supposition of the human being as a speaking being, is rendered as fun-
damental to the psychoanalytic practice of a “talking cure.” Through speech a person
articulates unconscious desire, which extends beyond mere sexual desire. Desire from a
The Obstacle 77
Considering the ontological status of “the sexual relation” (and “the

woman”), Lacan articulates the failure of logic to ascertain the real, which it is
supposed to do in a formal logical (assumed) representation of metaphysical
concepts. It is interesting to note that various elaborations especially on these
two Lacanian notions fail to provide an interpretation in relation to the con-
text of the Lacan’s enquiry, in which Lacan, I argue, does not intend to debate
the issue of sexuality and femininity in any other sense but through a critique
of the assumed logical conditions on which such issues are based on. This is
where, in my view, Lacan introduces the concept of the obstacle, to explain
why “there is no sexual relation.” Thus “there is no sexual relation” designates
the fact that the human beings are primarily related to each other through lan-
guage. More precisely, what makes each human being unique and what brings
us together is not the fact that each human being has a sex but the fact that
the human being is a speaking being. Hence, due to the human condition of
a being entirely shaped by the language, consciously and unconsciously, the
human being lacks an immediate access to the real through language.21
7 The Lacanian Critique of the Formal Logical Representation

of the Real
Logical coherence, Lacan seems to hold, is a consequence of the logical order

inherent in language. An objection, such as Aristotle’s instantia, “a premise
contrary to a premise,” guarantees the completeness/coherence of language
structure in itself but does not establish a rapport with the real. This is basically
a Fregeian objection. Lacan believed this can be seen clearly and distinctly, in
the topological model such as a torus. An intersection of the torus produces
the so-called Mobius strip, which according to Lacan is a model structurally
equivalent to the unconscious “structured like a language”. The unconcsious,
seen as a structure is moreover located on the flip side of the language struc-
ture, which is clearly shown in the Mobius strip, as Lacan maintains.
However, inscribing the logic of sexual difference in predicate logic, Lacan
effectively reduces Aristotelian syllogistic logic, which is conceived and com-
prehended on the basis of the syntactic logic of spoken language, to the
Lacanian point of view, I suggest, is a fundamental condition of interpersonal relations

between the speaking beings.
21 The Lacanian real is an element of psychoanalytic theory and it differs from the question
of what-is of the ontological enquiry in traditional metaphysics, and particularly in what
Heidegger designate as onto-theology.
78 cejvan
s ymbols which require exact knowledge of the meaning of each symbol, which
as such means virtually nothing on the level of spoken language. Hence, Lacan
wants to show that without this precise knowledge any attempt to interpret
these symbols will fail to render their accurate meaning. Consequently, there
cannot be established any relation between the symbol and meaning, unless of
course one knows the meaning of this symbol.
Lacan’s critique of the logic of the binary relation, which has governed the
idea of the sexual difference in psychoanalysis since Freud, is directly con-
nected to Lacan’s ambition to communicate the Freudian “discovery” of the
unconscious to linguistics and logic, which he considered as sciences through
which psychoanalysis can be actualized particularly by way of a formalization
of psychoanalytic discourse. Science, according to Lacan, is a formalized sys-
tem of knowledge, whereas certain properties and conditions of a particular
object of scientific enquiry are outlined in for example observation and calcu-
lation, operations which require a formalization of a scientific discourse. This
rather traditional idea of science goes back to Aristotle’s method of a scientific
analysis, known today as the Aristotelian syllogistic logic—which is precisely
where Lacan anchors his discussion on the matter of formalization as a way
of a scientific communication of psychoanalytic discourse. The difference
between Aristotle’s method of analysis and modern scientific analysis, is that
the latter, especially since Frege and Wittgenstein, abandons metaphysical
speculation.
In displacing ontology by means of an analysis of meaning at the level of syn-
tax, and posing the question of truth as a purely formal operation undertaken
by means of propositional logic, one can perhaps say that modern science in a
way still deals with a formal or symbolic representation of metaphysical con-
cepts. But does this actually have anything to do with the clear and distinct
seeing, which Descartes poses as a still highly relevant criterion of scientific
certainty? Lacan’s answer is negative. The problem Lacan thus points out in
Encore, more precisely in his critique of the inscription of the logic of sexual
difference in terms of a binary relation, is the idea of a formal logical and sym-
bolical representation of the real. The Lacanian real is the domain of truth
outside language, which cannot be represented and hence symbolized, since
it is properly speaking a matter outside of a clear and distinct seeing. “Seeing”
in this case is not a metaphor but a way of immediate understanding, of reach-
ing the real, which does not depend on interpretation, and thus never ends in
merely speculative knowledge. Seeing in this sense Lacan calls “mis/recogni-
tion” (reconnaisance/méconnaissance) which combines the idea of anamnesis
in Plato and a Cartesian scientific ideal of a certainty beyond doubt.
The Obstacle 79
8 Conclusion
One important lesson we can learn from Lacan’s discussion of the concept
of the obstacle in the context of his critique of formal logic or more gener-
ally symbolization is the following: any logical operation is an obstacle to a
clear and distinct seeing of the real. In other words, nothing of the real is actu-
ally revealed by a logical operation. Hence, according to Lacan, formal logic
ultimately does not represent anything but the coherence of its own chain of
signification. Today the idea that logic does not represent the real conditions
of concern in ontological investigations of what-is is rather widely accepted
among logicians. Still, in certain cultures of philosophical analysis, such as the
analytical tradition, the problem of what is is approached by appropriating the
formal logical derivation to prove or reject certain propositions of the ontolog-
ical status of things, assuming that the (formal) logical propositions actually
represent what is. Lacan’s critique of the formal logical approach to the real
warns against such approaches, and it does so by discerning the problem of the
obstacle as posed by the very formalized approach to the real.
A final remark regarding Lacan’s goal of overcoming logical formalization in
psychoanalysis by means of mathematization, or of creating a formalized sys-
tem which would assure the exactness to psychoanalytical theory. It remains
an open question whether Lacan’s elaboration of the mathèmes reached the
deadlock of the symbolic-formal representation it so desperately wanted to
overcome (one can certainly argue that it did). It remains an open question
whether the strict formalism of modern logic tends to create another self-
sufficient metaphysics, as for example this seems to be the problem of logical
positivsm in philosophy (one can, again, argue that it did). Is remains an open
question whether Lacan’s mathèmes are only intelligible in Lacanian theory.
However it may be with the last question, I claim that they overcome the fail-
ure of formal logic to account for the real.
Bibliography
Aristotle, Analytica Priora (Prior Analytics), trans. A.J. Jenkinson, in The Works of
Aristotle, Vol I, ed. W.D. Ross, (Oxford: Clarendon Press, 1928).
Bernard Burgoyne, “From the Letter to the Matheme: Lacan’s Scientific Methods,”
in The Cambridge Companion to Lacan, ed. Jean-Michel Rabaté (Cambridge:
Cambridge University Press, 2003), 69–86.
80 cejvan
Bruce Fink, “Preface,” in Jacques Lacan, Encore: On Feminine, the Limits of Love and
Knowledge, 1972–1973, ed. Jacques-Alain Miller, trans. Bruce Fink (New York:
W.W. Norton and Company, Inc., 1999), vii–1.
Jacques Lacan, The Other Side of Psychoanalysis, trans. Russell Grigg (New York: Norton,
2007).
———, Écrits, trans. Bruce Fink (New York: W.W. Norton and Co., 2006).
———, Encore: On Feminine Sexuality, the Limits of Love and Knowledge, 1972–1973, ed.
Jacques-Alain Miller, trans. Bruce Fink (New York: W.W. Norton and Company, Inc.,
1999).
———, Freud’s Papers on Technique, 1953–1954, trans. John Forrester (New York:
W.W. Norton, 1998).
———, The Four Fundamental Concepts of Psychoanalysis, trans. Alan Sheridan (New
York: W.W. Norton and Co., 1998).
———, Le Séminaire XXI: RSI, 1974–1975 (unpublished), lecture dated March 18, 1975
(in author’s possession).
———, Le Séminaire XIV: La logique du fantasme, 1966–1967 (unpublished), lecture
dated December 7, 1966 (in author’s possession).
———, Le Séminaire XXI: Les non-dupes errent (unpublished), lecture dated May 21,
1974 (in author’s possession).
Anika Lemaire, Jacques Lacan (London: Routledge, 1978).
Dany Nobus, “Lacan’s Science of the Subject: Between Linguistics and Topology,” in The
Cambridge Companion to Lacan, ed. Jean-Michel Rabaté (Cambridge: Cambridge
Jean-Michel Rabaté, “Strasbourg: The Nancy School?,” url: http://www.lineofbeauty.
org/index.php/s/article/view//62/135, accessed September 1, 2013.
Ellie Ragland-Sullivan, The Logic of Sexuation: From Aristotle to Lacan (Albany: State
University of New York Press, 2004).
chapter 5
Collingwood’s Logic of Question and Answer

against the Relativization of Reason
Christopher Fear
1 Introduction
Fearing that his health would not permit him time to leave to the world the
full range and depth of his offerings in philosophy, R.G. Collingwood sat down
in the late summer of 1938 with the intention of getting as much of his think-
ing down on paper and out in public as he could in the time left to him.1 The
result of this flurry of activity was An Autobiography: a short, bold, and stylisti-
cally informal book that is not only immensely readable as the life-story of an
unceasingly active man, but also philosophically rich, in places intriguing and,
above all, accessible to any reader. After four chapters of what are mostly mem-
oirs of his youth, chapter five of this Autobiography describes one of the funda-
mental concepts of Collingwood’s thinking on various subjects: namely, what
he calls his “logic” (or “theory”) of question and answer.2 It is an extremely sug-
gestive and deceptively complex chapter—which is probably part of the rea-
son why Collingwood scholars have offered so many different accounts of the
significance or “place” of “question and answer” in Collingwood’s philosophy.3
In less that fourteen pages Collingwood explains the importance of “the
questioning activity” to human knowledge, and how it offered a theory of truth
and contradiction distinct from both the “correspondence” and “coherence”
theories of truth.4 Most provocatively as far as intellectual historians are con-
cerned, Collingwood also begins to explain here how “question and answer”
showed that in order to know what a man means by a certain proposition
you must reconstruct the question to which it is intended as an answer. And,
1 See Fred Inglis, History Man: The Life of R. G. Collingwood (Oxford: Princeton University Press,
2011), 217.
2 R.G. Collingwood, An Autobiography (Oxford: Oxford University Press, [1939] 2002), 29–43.
3 See Stein Helgeby’s thorough and admirable survey: Stein Helgeby, Action as History: The
Historical Thought of R.G. Collingwood (Exeter: Imprint Academic, 2004), 84–88. Helgeby’s
own reading of “question and answer” is presented on pages 77–100.
4 Collingwood, Autobiography, 36.

82 fear
most relevant to the discussion at hand, Collingwood explains how something

he had learnt in his work as an archaeologist brought about in him a “revolt
against the current logical theories of the time.”5 “I know that what I am going
to say is very controversial,” he warns, “and that almost any reader who is
already something of a logician will violently disagree with it.”6
Although it might have been offensive to his peers, much of what
Collingwood says about his “logic of question and answer,” especially concern-
ing the “conspiracy” of modern logicians7 and the nature of his “revolt” against
them, is likely to find a receptive audience among many of today’s historians
of philosophy. Either because we are more historically sensitive to the chang-
ing questions of the past than our predecessors typically were, or because we
are familiar with certain “pluralistic” tropes of later philosophy, some of us
are more receptive than Collingwood’s contemporaries were to his claim that
“Meaning, agreement and contradiction, truth and falsehood, none of these
belonged to propositions in their own right [. . .] [but] only to propositions as
the answers to questions: each proposition answering a question strictly cor-
relative to itself.”8 In the first half of this chapter I have outlined the reasoning
behind Collingwood’s critique of his contemporary logicians, comprising some
of what he says explicitly about it, and much of what he does not. But there are
also certain warnings in Collingwood’s philosophy concerning what is lost by
what we might call the retrospective relativizing of reason, and these warnings
have implications which I think will be met with rather more hostility today.
In the second half of this chapter I have explained what I think those warnings
are, and closed by offering reasons for thinking those implications might be
worth paying attention to.
2 Logic and Knowledge
As well as an historian of Roman Britain9 and the Waynflete Professor of

Metaphysical Philosophy from 1935 until 1941, Collingwood was also an ama-
teur archaeologist. It was this experience of his work in archaeology, he says in
his Autobiography, that brought about in his mind “a revolt against the current

7 R.G. Collingwood, The Idea of History: With lectures 1926–1928, ed. Jan van der Dussen, (Oxford:
Oxford University Press, [1946]: 1993), 273.
9 See Collingwood, Autobiography, 120–121.
Collingwood’s Logic of Question and Answer 83
logical theories of the time, a good deal like that revolt against the scholastic
logic which was produced in the minds of Bacon and Descartes by reflection
on the experience of scientific research, as that was taking shape in the late
sixteenth and early seventeenth centuries.”10 Bacon’s Novum Organum and
Descartes Discours de la Méthode took on a new significance, he says, because
they expressed a principle of logic which (Collingwood then adds) needs
restating: namely
the principle that a body of knowledge consists not of ‘propositions,’

‘statements,’ ‘judgements,’ or whatever name logicians use in order to
designate assertive acts of thought (or what in those acts is asserted: for
‘knowledge’ means both the activity of knowing and what is known), but
of these together with the questions they are meant to answer; and that a
logic in which the answers are attended to and the questions neglected is
a false logic.11
The implications for historians of philosophy and historians of political the-

ory have been abundantly discussed since the mid-twentieth century. If the
assumption is false that philosophers offer competing answers to a gamut of
eternal questions,12 then (Collingwood says) it is extremely important if you
want to understand your author’s spoken or written statements to reconstruct
the actual question to which his words are proposed as an answer. Without
knowing the question in your author’s mind, you cannot know what he means
by his statements, “even though he has spoken or written with perfect com-
mand of language and perfectly truthful intention.”13 It is in this sense, and in
this sense only, that historical context for Collingwood is not only relevant but
a necessary condition if you wish to follow your subject’s reasoning14—which,
as an historian, Collingwood thinks you must.15

11 Collingwood, Autobiography, 30–31. He says that he first developed his logic of question
and answer in 1917 in a book called Truth and Contradiction. (Autobiography, 42.) Only
chapter two of Truth and Contradiction survives, unpublished. (Bodleian Library, Dep. 16.)
12 Collingwood, Autobiography, 60–62.
14 This is why Collingwood’s logic of question and answer continues to offer a defence of
Quentin Skinner’s contention that the historian must study the context—and not merely
may, as Mark Bevir insists. For more on the value of deploying Collingwood’s logic in
that debate, see my recent article: Christopher Fear, “The Question-and-Answer Logic of
Historical Context,” History of the Human Sciences 26 (2013), 68–81.
15 Collingwood, The Idea of History, 215.
84 fear
But the implications for logicians are no less far-reaching. What Collingwood
describes as a “revolt” in logic is actually more of an iconoclasm. It is a destruc-
tion of the false idols of contemporary logicians by way of a return to the
origins of logic—origins historical and, if I understand Collingwood cor-
rectly, practical. Formal logic is actually, Collingwood says, an abstraction
and like all abstractions it derives from concrete instances of something. In
the Autobiography his demonstration appeals to the early history of logic’s
entwinement with grammar:
The logician’s proposition seemed to me a kind of ghostly double of the

grammarian’s sentence, just as in primitive speculation about the mind
people imagine minds as ghostly doubles of bodies. Grammar recognizes
a form of discourse called the sentence, and among sentences, as well as
other kinds which serve as the verbal expressions of questions, com-
mands, &c., one kind which express statements. In grammatical phrase-
ology, these are indicative sentences; and logicians have almost always
tried to conceive the ‘unit of thought,’ or that which is either true or
false, as a kind of logical ‘soul’ whose linguistic ‘body’ is the indicative
sentence.16
Collingwood goes on to explain that anyone who presupposes the “indica-

tive” nature of propositions finds himself saddled with a theory of truth and
knowledge that bears no resemblance to first-hand experience of constructing
knowledge. Here he has a very good point. Indicative sentences (such as “This
is your room” and “Dinner is at nine”) indicate things that are distinct from
the indication of them. Indications are not exactly propositions, in the sense
of propositions as statements of a proposed truth. “This is your room” is not a
contention: it is simply an indication. The consequence of modelling proposi-
tions on indications, Collingwood thinks, is that checking whether they are
true means “apprehending” reality in order to check the correspondence with
what is indicated. This is the mistake, Collingwood thinks, made by the “realist”
theory of knowledge.17 Anyone with experience of offering and testing conten-
tions knows that we do not and cannot simply check them by “apprehend-
ing” reality. What we actually do is clarify the question to which an answer is
being proposed, and use evidence to falsify, permit, or reinforce a provisional
conclusion. It is not some correspondence with reality that is checked: it is
16 Collingwood, Autobiography, 34 (emphasis added).

17 And by “realism” he means Cook Wilson, H.A. Prichard, H.W.B. Joseph and others. See
Collingwood, Autobiography, 20.
the evidence itself, and the reasoning by which that evidence is connected to
the conclusion. To use a blunt analogy,18 the “realists” supposed (Collingwood
thinks) that knowledge was like the finale of a game of Cluedo: the accusation
is correct if it corresponds to the suspect, room, and weapon found in the enve-
lope. Collingwood’s objection is that, in a case of real-world knowledge, no-one
ever gets to look inside the envelope. We only have our inquiries, our evidence,
and our reasoning to go on, and the game never ends. So when someone claims
to know who murdered Dr. Black, where, and with what weapon, it is his evi-
dence and reasoning he offers as proof—never the contents of the envelope.
The “realists’” false account of knowledge, Collingwood seems to claim, owes
to a fundamental mistake about the nature of propositions.
Collingwood then offers an outline of his own logic of question and answer.
He does not explicate in the Autobiography exactly what his logic of question
and answer derives from, if not from the indicative sentence. But in view of
passages in some of his other works, it seems to me that Collingwood’s logic
of question and answer derives what I will call demonstrative narrative—that
is, from the narrative of scientific investigation and inductive demonstration
itself. Now this seems to me to be a very clever move. Where Collingwood’s
predecessors shoe-horned their theories of truth and knowledge into proposi-
tional logic as it already existed, Collingwood makes logic follow an account of
truth and knowledge that describes what it is actually like to construct knowl-
edge inductively and to decide that things are true. None of this is especially
clear in the Autobiography, where Collingwood’s attack on propositional logic
and “realist” theories of truth are a little rushed and intertwined with each
other, as well as tangled with his explanations of how “question and answer”
rescues historically-remote philosophers from the misunderstandings of their
“realist” readers. But Collingwood’s other works offer evidence that this is his
agenda for abstract logic.
Chapter five of An Autobiography was certainly not the first appearance of
“question and answer” in Collingwood’s work, and it would not be the last. His
ambitious second book, Speculum Mentis (1924), contains the following state-
ment of what “question and answer” is all about. This passage illustrates that,
for Collingwood, the question of what knowledge is ought not to be posed
separately from the question of how knowledge is achieved. The theory of
18 Collingwood offers his own analogies. I have tried to capture the important points in my
own. See Collingwood, The Principles of History and other writings in philosophy of history,
ed. W.H. Dray & W. J. van der Dussen (eds) (Oxford: Oxford University Press, [1999] 2001),
21–9; The Idea of History, 266–73; R.G. Collingwood, Essays in Philosophy of History, ed.
William Debbins (Austin: University of Texas Press, 1965), 98–99.
86 fear
knowledge, epistemology, is again an abstraction from the concrete instances

of people attempting to know things. In future Collingwood would find ways
of expressing the principles of “question and answer” more concise—though
rarely, I think, more elegantly:
A crude empiricism imagines that knowledge is composed wholly of

assertion: that to know and to assert are identical. But it is only when the
knower looks back over his shoulder at the road he has travelled, that he
identifies knowledge with assertion. Knowledge as a past fact, as some-
thing dead and done with—knowledge by the time it gets into encyclo-
paedias and text-books—does consist of assertion, and those who treat it
as an affair of encyclopaedias and text-books may be forgiven for think-
ing that it is assertion and nothing else. But those who look upon it as an
affair of discovery and exploration have never fallen into that error.
People who are acquainted with knowledge at first hand have always
known that assertions are only answers to questions.19
None of this, Collingwood thinks, should be news to anyone. Plato, after all,
“described true knowledge as ‘dialectic,’ the interplay of question and answer
in the soul’s dialogue with itself [. . .] so Kant mildly remarked that the test of
an intelligent man was to know what questions to ask; and the same truth has
lately dawned on the astonished gaze of the pragmatists.”20
The same point is made—characteristically, more concisely—near the end
of Collingwood’s life, in The New Leviathan (1942): “Knowing a thing is more
than merely being conscious of it,” he says; “Knowing involves asking questions
and answering them.”21 And later: “Knowledge is the conviction or assurance
with which a man reaffirms a proposition he has already made after reflect-
ing on the process of making it and satisfying himself that it is well and truly
made.”22 In the years between these two books—approximately the twenty
years of his “mature” career—Collingwood announces several times that
19 Collingwood, Speculum Mentis, or The Map of Knowledge (Oxford: Clarendon, [1924]

1963), 77.
20 Collingwood, Speculum Mentis, 77–78.
21 Collingwood, The New Leviathan: or Man, Society, Civilization and Barbarism, ed. David
Boucher (ed.) (Oxford: Clarendon, [1942] 1992), 11.11. References to the main text of The
New Leviathan are given by paragraph number as standard (identical in all editions).
22 Collingwood, The New Leviathan, 14.22.
this is a basic principle of science, frequently identifying it with Bacon and

Descartes.23
When Collingwood says “science” he does not mean only natural science.
There is nothing “positivistic” about Collingwood.24 He means “science” in the
continental sense: an organized body of knowledge, une science; una scienza;
eine Wissenschaft. (They are always feminine nouns, though this is probably
not what Nietzsche meant when he supposed that “truth is a woman.”25) The
essence of any science is, Collingwood says, “systematic thinking.”26 And this
systematic thinking always begins when “one becomes aware of a mental
hunger that is no longer satisfied by what swims into one’s mouth. One wants
what is not there and will not come of itself. One swims about hunting for it.
This ranging of the mind in search of its prey is called asking questions.”27 So
“science in general,” Collingwood writes in the introduction to The Idea of
History
does not consist in collecting what we already know and arranging it in

this or that pattern. It consists in fastening upon something we do not
know, and trying to discover it [. . .]. That is why all science begins from
the knowledge of our own ignorance: not our ignorance of everything,
but our ignorance of some definite thing—the origin of parliament, the
cause of cancer, the chemical composition of the sun, the way to make a
pump work [. . .]. Science is finding things out: and in that sense history is
a science.28
The starting point of systematic, scientific thinking is not information, prop-

ositions, or “objects” of a specified kind.29 It is what I will call “the specific
23 Collingwood, Autobiography, 30. See also The Idea of History, 269, and The New Leviathan,
31.27.
24 As Quentin Skinner has recently emphasized. See Quentin Skinner, “Some Problems in
the Analysis of Political Thought and Action”, Political Theory 2 (1974), 283–4. See also
Collingwood, The Idea of History, 126–33.
25 Friedrich Nietzsche, Beyond Good and Evil, trans. Helen Zimmern (Mineola: Dover, [1886]
1997), ix.
26 Collingwood, Autobiography, 25–26, 30–31; The Idea of History, 269, 273; An Essay on
Metaphysics (Oxford: Clarendon, 1940), 4; The New Leviathan, 6.47, 6.43, 6.59, 22.8.
27 Collingwood, An Essay on Metaphysics, 37.
28 Collingwood, The Idea of History, 9 (emphasis added).
29 Contradictions of this last point are frequently to be encountered today. See for instance
most of the preface to Mark Bevir, The Logic of the History of Ideas, especially 8–10, 16, and
also 31–7. Compare with Collingwood, The Principles of History, 179.
88 fear
unknown.”30 The specific unknown is the thing that the scientist, whatever his
precise discipline, wants to replace with a specific “known.” The answer is what
he “desires;”31 it is what replaces emptiness with repletion;32 it is the thing that
his investigative hunger presupposes to be the “good;”33 it is what is pursued
in what Collingwood calls the “hunt.”34 And desire, Collingwood writes in The
New Leviathan, involves propositional thinking: “and a proposition is an answer
to a question; and a question offers alternatives; so desire asks and answers the
question ‘What do I want?’”35 Finding things out is, in short, the movement
from the here-and-now of ignorance to the there-and-then of knowledge.36
The basic method of induction, proceeding from the unknown to the
known—from here-and-now emptiness to there-and-then repletion,37 from
confusion to conclusion—is by employing evidence (in a very broad sense).
Whatever typical variations there might be between disciplines, all forms
of investigation aimed at establishing a conclusion have this in common.
“Anything is evidence which can be used as evidence,” Collingwood writes,
“and no one can tell what is going to serve him as evidence for answering a cer-
tain question until he has formulated the question.”38 In English we are accus-
tomed to using the word “evidence” in conjunction with historians’ disputes
and especially with criminal investigations.39 What Collingwood really means
should be more broadly conceived so that there can be overlap with what we
call “considerations.” Evidence is what helps the questioner to construct his
conclusion, of course, but it is also what he offers to other people in order to
demonstrate that the conclusion he has constructed is the right one. The rela-
tionship between evidence and conclusion is called “reason,” and it is this rela-
tionship in particular—this element of demonstrative narrative—from which
abstract logic is derived. The definition of “reason” that Collingwood offers in
The New Leviathan is wonderfully iconoclastic:
30 That the unknown is “specific” is important, according to what Collingwood calls the
“principle of the limited objective.” See The New Leviathan, 31.61–64.
31 Collingwood, The New Leviathan, 11.1–11.
36 Collingwood, The New Leviathan, 7.44–45, 7.69, 8.13.
38 Collingwood, The Principles of History, 38.
39 Indeed Collingwood discusses at length the similarities between a scientific historical
inquiry and a criminal investigation. See The Idea of History, 266–73; The Principles of
History, 21–29.
‘REASON’ as the name of a mental function or form of consciousness,

rational thinking, is thinking one thing, x, because you think another
thing, y; where y is your ‘reason’ or, as it is sometimes called, your ‘ground’
for thinking x.40
When I say that “evidence” ought to be broadly conceived, I mean broad enough
to stand for what is referred to in this passage as “y” and the relations between
y and x. In real-life demonstrative narrative, of course, there are usually lots of
“y”s—a “constellation”41 of reasons. Thus, in view of y and z (the demonstrator
says), I conclude that x is the case. The response might be offered that, in view
of counter-evidence a and b, we cannot conclude x. “All logic,” Collingwood
says, “is concerned with discussions.”42 What he has in mind (I am claiming) is
any “discussion” in which the aim is to establish the correct answer to a shared
question.
The rules of logic should be understood as abstractions from these specific
instances of individuals offering evidence for their conclusions to others who
may or may not be convinced by the demonstration. These others—other his-
torians, other astrophysicists, other philosophers etc.—are convinced by the
proponent’s reasoning when they accept that y is (probably) the case, and
accept that it demonstrates x. They fail to be convinced by it either when they
are not convinced that y is (probably) the case, or when they accept y but are
not convinced that it provides sufficient reason for concluding x. The impor-
tant point is that logic collects instances of y demonstrating x, and instances
of y failing to demonstrate x, and formulates them as rules abstracted from
the concrete cases in which those successes and failures actually arise. For
Collingwood, it is because of “modern logicians”’ lack of experience in concrete
inquiries and demonstrations of this sort that they are “in a conspiracy to pre-
tend that [instead of answering specific questions] a scientist’s business is to
‘make judgements,’ or ‘assert propositions,’ or ‘apprehend facts.’ ” This reveals,
for Collingwood, that modern logicians “have no experience whatever of sci-
entific thinking, and wish to palm off, as an account of science, an account of
their own haphazard, unsystematic, unscientific consciousness.”43
Now these are strong words, and there are several messages here that will
please those who already have certain grievances with any surviving logicians
of the kind Collingwood attacks. Firstly it is clear that, since it is really derived

43 Collingwood, The Idea of History, 273–274.
90 fear
from concrete instances of people posing questions, investigating them, and

offering answers to them by scientific demonstration, abstract logic is rather
beholden to the kinds of questions that people actually pose, and to the kinds
of evidence they actually offer to one another. Since man’s questions about
himself and the world around him are constantly changing and their solu-
tions changing with them44 (as Collingwood puts it), abstract logic is revealed
as the supposedly timeless product of conspicuously historical (and often
historically-specific) demonstrative narratives. Its pretentions to finality are mis-
taken. Compiling a full list of logical relations is not possible, Collingwood says
(before adding, rather audaciously, “though Germans have thought it so”)45—
and this parasitism of the abstract upon the concrete is part of the reason.
Secondly, anyone who has ever been party to a demonstration in their disci-
pline knows that it is standard practice to by-pass some of the rules enshrined
in abstract logic rather often. We can and do point to long-identified falla-
cies, of course, but we do so selectively, adapting to the demands and stakes
of the demonstration at hand. The effect of this adaptation of logic among
working “scientists,” especially among historians, is that where the formal logi-
cian might complain that the historian’s two pieces of evidence, y and z do not
necessitate the conclusion, x, other historians might quite rightly point out
that y and z nevertheless demonstrate enough, given the circumstances and
what is at stake, for the conclusion to be accepted—especially in the absence
of a more plausible, better-evidenced alternative. The logician leaves such a
dispute thinking that historians’ conclusions are based on such insecure rea-
soning that the knowledge they claim to produce is no real knowledge at all;
while historians are pleased to see the back of a man who appears to have
handed over his capacity for circumstanced judgement to a falsely-eternal
compilation of rules that demand from proof something that in practice it can
never offer, and never even pretends to: certainty and finality. Even if some
accommodation could be made with the language of probability in place of
the demand for certainty, the mathematics of probability offer no assistance
where evidence, which is infinitely variable in type, use, and significance, can-
not profitably be given a numerical value, and where evidence that might still
turn up in future is impossible to estimate fully even qualitatively, let alone
quantitatively.
The effect of Collingwood’s “revolt” against the “conspiracy” of modern
logicians is that it rectifies the oddities he identifies in the “realist” theory
of knowledge. The philosophical accounts of what knowing means, what

knowing demands, and what knowing is like, are neatly aligned. And Colling
wood’s logic of question and answer is good news for historians who gain, in
practice, logic as a servant to their discussions, rather than as the master of
them; and gain, as a bonus prize of precisely the sort they like, the satisfac-
tion of showing logicians that their supposedly timeless science is a product
of history.
3 Reason and Relativism
It is my view that Collingwood’s logic of question and answer offers strong

defences of several of the contentions for which he is best known and most
criticized as a philosopher of history, including the long-embattled claim that
the historian must “re-enact” past thoughts in his own mind.46 Here I will dis-
cuss only the contention that touches most significantly on the way the idea of
different, historically-contextual “logics” (in the plural) might be employed by
historians of philosophy today.
It is perhaps surprising in view of what we have just heard that Collingwood
insists, again in his Autobiography, that historians of philosophy must not
only ask what their subject concluded, but must also ask whether what he
decided was right or true. “History did not mean knowing what events fol-
lowed what,” Collingwood writes in his Autobiography, “it meant getting inside
other people’s heads, looking at their situation through their eyes, and think-
ing for yourself whether the way in which they tackled it was the right way.”47
Withholding these assessments in the history of philosophy is not only unnec-
essary, Collingwood says, it is “crippling.”48 And he adds, by way of rhetorical
force, that “everybody who has learnt to think historically knows it already;
and no amount of argument could teach it to a person who had not learnt
to think historically.”49 Examples of such assessments are easy to come by in
Collingwood’s own treatment of past philosophy. The Idea of History provides
not only descriptive accounts, but also critiques of each of the thinkers dis-
cussed. The New Leviathan contains more colorful examples: an especially
amusing one is: “Plato is the man who planted on the European world the

47 Collingwood, Autobiography, 58–9. See also The Idea of History, 215–216.
92 fear
crazy idea that education ought to be professionalized.”50 In many cases, this

included, Collingwood’s wording seems intended as a deliberate affront to the
orthodox conceit of the intellectual historian’s impartiality.
This contention that historians must ask the “rightness” question is, I think,
likely to find considerably less favor among many of today’s intellectual histo-
rians than the critique of propositional logic we have just seen. Today Skinner
preaches the exact opposite of this,51 and seems to have invited the rather
gothic charge that he has reduced the discipline to a “conducted tour of a
graveyard.”52
If, as Collingwood says they do, the statements of past authors in the his-
tory of philosophy provide answers to questions which are not ours, which are
not eternal or perennial, but which are in fact limited to the problems of their
own time; and if historians of philosophy are to follow past reasoning in light
of those historical contexts; then can asking “Was he right?” still be legitimate,
let along obligatory? Collingwood’s insistence that historians “must” ask this is
aimed at what he identifies as a symptom of the “realist” assumption that phi-
losophers deal with eternal questions. Because of this mistaken belief, he says,
the question, ‘what was Aristotle’s theory of duty?’ would be an ‘histori-

cal’ question. And it would be wholly separate from the philosophical
question, ‘was it true?’ Thus the ‘history’ of philosophy [for ‘realists’] was
an inquiry which had nothing to do with the question whether Plato’s
theory of Ideas (for example) was true or false, but only with the question
what it was.53
This for Collingwood is tantamount to an emasculation of the history of phi-

losophy.54 Historians of philosophy can and should describe past thinkers’ the-
ories and say and explain why they, and the supportive reasoning they offered,
were wrong or right. And he is quite unapologetic for the apparent anachro-
nism: “What is required, if I am to know Plato’s philosophy,” he says in The Idea
50 Collingwood, The New Leviathan, 37.4. See also Collingwood’s declaration that Hobbes’s
Leviathan is “the world’s greatest store of political wisdom”. (The New Leviathan, lx.)
For other examples see also The Idea of History, 76, 173.
51 See Skinner, “A reply to my critics”, 256–257; Quentin Skinner, Visions of Politics, vol. I,
Regarding Method (Cambridge: Cambridge University Press, 2002), 53.
52 Skinner, Regarding Method, 125.
54 See Collingwood, Autobiography, 72–76.
of History, “is both to re-think it in my own mind and also to think other things
in the light of which I can judge it.”55
Two questions ought to arise in response to what Collingwood is arguing
here. The first and most important is “Why?”: Why “must” historians ask not
only what past authors’ reasoning was, but also ask whether they were right?
The second question, which is more relevant to the subject at hand, is “How?”:
How, if abstract logic is as historically contingent as Collingwood seems to say
it is, can we consciously unleash later logic upon unwitting past thinkers? What
counts as a logical proof undoubtedly varies between cultures. Even under-
graduates encountering old philosophy for the first time notice that people
in the past were easily convinced of a claim by what appear to be some quite
insignificant considerations in conjunction with some very weak reasoning.
The more old philosophy you read, the clearer it becomes that what counts
as good reasoning is in a state of constant change and development. The mes-
sage that many historians of philosophy have taken from this is that the logics
of the past are distinct from our own, and that imposing upon past thinkers
logical standards which would have been unknown to them is anachronistic,
unfair, and inimical to understanding. The effect of this is that some authors
seek to defend the arguments of past thinkers against any new critiques that
other authors might want to advance. It is a defence that operates by sealing
off old arguments behind a protective barrier designed to filter out unwelcome
anachronistic logic. To be clear, these are not defences of the original argu-
ment: they are rather notices that trespassers will be disqualified.
Collingwood’s answer to the first question is not easy to find. His
Autobiography offers the claim that historians “must” ask “Was he right?”, but
not the reasoning—that is, the reader gets his “x” several times, but never his
“y”s. His Essay on Philosophical Method is also unenlightening. The answer is to
be constructed, I think, from important passages found in The New Leviathan.
What Collingwood refers to in his writings as “historical thinking”—for all
intents and purposes simply the reasoning historians use to answer their ques-
tions properly—is a type of “theoretical reason.” What Collingwood says about
theoretical reason is fundamental to the logic of his argument that the histo-
rian of philosophy “must” ask the truth question, so I will quote it in full:
Reason is distinguished into theoretical reason and practical reason: i.e.

reason for ‘making up your mind that’ (reason for what logicians call a
proposition) and reason for making up your mind to’ (reason for what
moralists call an intention). We shall see that, of these two, practical

94 fear
reason is the prior: it is the original form of reason, theoretical reason

being a modification of it; and by the Law of Primitive Survivals a practi-
cal element is always present in a case of theoretical reason.56
Collingwood’s argument, it seems to me, is that historians of philosophy “must”

ask whether past authors’ claims were right if (and perhaps only if) they want
their historical work to have any implications for present practice. And if they
have come to their particular line of theoretical reason by modifying a practical
question, they must want this—so long as the practical question has not been
forgotten along the way. We can take “practice” as broadly as we like. For pro-
fessional philosophers like Collingwood the discovery that there is an error of
reasoning behind a past and now generally-accepted answer to a question has
implications for the later “practice” of philosophy that presupposes it.57 This is
what he means, I think, when he writes in The Idea of History that history pro-
vides for us more than “a mere inventory of our intellectual possessions at the
present time” by answering the question “by what right we enjoy them.”58 The
history of philosophy, then, is not only for answering questions about what
ideas were held by someone, or what arguments were dominant during a given
period, or what logic was commonly accepted as demonstrative of truth. It is
also for establishing whether those ideas and logics were right. If those ideas
were wrong, and we have among us a vicious inheritance, the historian might
next inquire whether the error was due to inherited presuppositions that were
allowed somehow to go unchallenged.59 Historical thinking and philosophical
thinking therefore throw light upon each other, in the sense that in history
we come to appreciate the legitimacies and identify the illegitimacies of the
settled answers to past questions and problems as we have inherited them. The
history of philosophy is, then, (or should be) critical in the full sense, and initi-
ated by questions of present importance—at least, according to Collingwood.
It is clear that Collingwood’s insistence that historians of philosophy must
ask “Was he right?” presupposes that the questions to which historians provide
answers arise out of practical questions, questions of what to do.60 The “Was
he right?” question does not follow from conceptual analysis of what historical
56 Collingwood, The New Leviathan, 14.3–38, 1.66–68. See also The Idea of History, 406–407.
57 See Collingwood, An Essay on Metaphysics, 21–33.
60 For a detailed account of the rapprochement between theory and practice see David
Boucher, The Social and Political Thought of R. G. Collingwood (Cambridge: Cambridge
reasoning is, and much less does it follow from an analysis of the kinds of
actions, utterances, or “objects” that historians take as their subject-matter. It
follows instead from a presupposition about why historical questions are asked
in the first place. In short, the “Was he right?” question is the “primitive” ele-
ment of a practical question that survives into the kind of theoretical question
to which it is the job of historians to provide well-reasoned, well-evidenced
answers.
That is Collingwood’s answer to our first question of why historians of phi-
losophy must ask “Was he right?”. Our second question was “How?”: How, if
abstract logic is as historically contingent as Collingwood seems to say it is,
can we consciously unleash our own logic upon unwitting past thinkers? In
order to illustrate what is at stake I took the why question first. It also answers
charges of illegitimacy, since if something “must” be done then it is (at least)
sometimes a legitimate thing to do. But logically the second question comes
first, because “must” implies “can.” And it is here, where the possibility of apply-
ing contemporary logic to past reasoning, that some of today’s intellectual
historians would be inclined to ignore Collingwood. The challenge, then, is to
show that it can be done by explaining how.
Collingwood does not answer this question: the need never arose, because
his presupposition that it is possible to assess the reasoning of the past was
never challenged. So I am now departing from what Collingwood might have
argued himself. (And if the reader does not like it he can at least take comfort
from seeing that I have been quite explicit about it.) But again, Collingwood’s
logic of question and answer has wonderfully iconoclastic potential. The icon
I have in my sights is the “relativist” and/or “pluralist” principle that different
compilations of abstract logic, or different standards of what counts as a dem-
onstration among different cultures, geographical, historical, etc., enjoy pri-
vate validity that make them immune to later or “foreign” logic. It is, in short,
an iconoclastic demonstration that we can assess the rightness, logic, and even
truth of past arguments, which operates by clearing this polytheistic pantheon
of relative plural “logics.”
Collingwood’s definition of reason is simple enough to accommodate every
one of what are called different “logics.” Here it is again:
‘REASON’ as the name of a mental function or form of consciousness,

rational thinking, is thinking one thing, x, because you think another
thing, y; where y is your ‘reason’ or, as it is sometimes called, your ‘ground’
for thinking x.61

96 fear
The flexibility of Collingwood’s definition owes to the infinite possible ways in

which conclusion relates to reason. We understand, or follow, someone else’s
reasoning when we can see that he thinks y is a good reason to conclude x, and
perhaps we understand it even better when we can see why he thinks it is a
good reason. Neither means, of course, that we therefore commit ourselves to
thinking it a good reason. Understanding and criticism are not mutually hostile.
What varies historically, culturally, and according to the demands and
stakes of a demonstration, is not the “form” of reason itself (it is always what
Collingwood says it is), but “what counts” as a satisfactory reason in what is
therefore a satisfactorily-reasoned demonstration. We have seen that the
abstract rules of logic do not govern “what counts”: the abstract rules of logic
are derived from what counts—derived, that is, from concrete instances of
demonstrations succeeding and failing. When, in discussions of this kind,
a challenger refers to an abstract logical fallacy to show why y is not a good
ground for x, he is pointing out that a logical error of this kind has been seen
before, and has been shown to be a fallacy. But anyone with experience of this
kind of discussion knows that the discussion does not therefore end. Abstract
logic does not “govern” discussion: we say, perhaps a little ambiguously, that
logic “informs” discussion. It remains up to the discussants to decide whether
the alleged fallacy has been employed and, even if it has, whether that dam-
ages the conclusion enough to warrant rejecting it.
This shows that the place of logic in rational (i.e. reasoned) discussion is
that of a servant, rather than that of a master. Abstract logic is a very useful ser-
vant, which really does help to remind us of the ways in which demonstrations
can fail or succeed. But the false idea that the rules of logic govern rational
discussions and demonstrations, rather than serve them, inspires among other
things (1) the resentment of abstract logic, and/or (2) the hypostatization of
abstract logic as a “logical framework.” Meanwhile, the realization that logic is
after all only a servant to rational discussion inspires (3) a revanchist denigra-
tion of logic as such, and/or (4) the suspicious belief that our own logic is a
servant conspiring against us.
Some of these positions (1, 3, and 4) are conducive to what Collingwood
calls “irrationalism.”62 (Position 2 alone is not so conducive to “irrationalism”).
What these positions have in common is that they permit and perhaps even
celebrate the co-existence of different “forms” or “frameworks” of logic. Our
interest is not in irrationalism here: it is only the hypostatization of abstract
logic as a “logical framework” which needs to be challenged. The historical
fact that different things are counted as demonstrative in different historical

and cultural contexts is thought to show that “what counts” is governed by

the attendant “logical framework.” This “government” consists in the supposed
unavailability of yet-unencountered considerations: novel concepts, the iden-
tification of certain fallacies, the interrogation of certain undemonstrated pre-
suppositions, etc. It is this “unavailability” that is transformed in the name of
anti-anachronism into a retrospective prohibition.
But if logic does not govern criticism but rather, as I have said, informs it—
that is, if rational thinking uses logic—then what falsely hypostatized “logi-
cal frameworks” apparently made impossible is actually nothing more than
what was missing from past discussions. We do not have to argue that past
discussants could or should have identified a fallacy or an undemonstrated
presupposition. It is enough to say that they simply did not see it—and we can
add that this might have been quite normal. Far from protecting past reason-
ing from present critique, it is precisely those unseen faulty presuppositions
and lingering fallacies that are now identifiable as errors. Of course, if today’s
interrogations of evidence show that past presuppositions were true then, but
are untrue now, such as the presupposition that the Germans intend to invade
Britain, we say that “His answer was right” without committing ourselves to the
belief that the presupposition remains true.
Perhaps Collingwood is wrong about historians being obliged to ask whether
past philosophers were right. But the possibility of doing so is clear, so long
as abstract logic is understood to inform discussion as a servant, rather than
thought to govern it as a master. It follows that an historian who wishes to
comment on, say, the errors of mediaeval historiography63 is by no means dis-
qualifying himself as an historian who understands past thinking. In fact, he
understands it so well that he is not afraid to explain to his own peers what was
wrong with it.
4 Closing Reflections
For reasons the reader might already have noted, Collingwood is commonly
identified as operating in a “dialectical” tradition. Louis Mink has always
emphasized this: “the key to understanding Collingwood’s thought is an appre-
ciation of how fundamentally and pervasively dialectical it is,” he says.64 I have
avoided the term “dialectical” here, since it commonly invites more confusion
63 See for example Collingwood, The Idea of History, 52–56.

64 Louis Mink, “Collingwood’s Historicism: A Dialectic of Process’, in Critical Essays on
the Philosophy of R. G. Collingwood, ed. Michael Krausz, 168, see also 170–3. See also
98 fear
than it does clarity. It does not mean that Collingwood’s philosophy is only rel-
evant if one presupposes “dialectics,”65 especially not if that is intended to indi-
cate a form, framework, or tradition of logic. Neither does it indicate “Hegelian”
dialectic, which Collingwood says involves “at least one very bad mistake,”66
or Marx’s ostensibly inverted version, which still “pretends to extract a concrete
rabbit from an abstract hat.”67 Neither does it secretly dispense with the free-
dom of the will,68 nor does it indicate a belief in the inevitability of progress.
Progress properly defined—i.e. defined as solving new problems without los-
ing hold of solutions to old ones “so that there is gain without any correspond-
ing loss”—is an ideal,69 a presupposed aim of rigorous inquiry, not necessarily
a reality,70 and still less an inevitability.71
What “dialectic” actually does mean in Collingwood’s philosophy under-
lines what I have tried to say here about the priority of concrete discussion
to abstract logic. I will close with an abridged passage from Collingwood’s last
book, The New Leviathan, which epitomizes the true meaning of “dialectic” in
his philosophy: namely, that it indicates a kind of concrete discussion, or an
attitude towards concrete discussion. I am afraid I have been unable to para-
phrase more concisely or anywhere near as gracefully as the original, so I will
leave to your scrutiny Collingwood’s own (abridged) formulation:
All logic is concerned with discussions; but Plato distinguished two kinds
of discussions, ‘eristical’ and ‘dialectical’ (Meno, 75 c–d). What Plato calls
an eristic discussion is one in which each party tries to prove that he was
right and the other wrong. In a dialectical discussion you aim at showing
that your own view is one with which your opponent really agrees, even
if at one time he denied it . . . The essence of dialectical discussion is to
discuss in the hope of finding that both parties to the discussion are right,
and that this discovery puts an end to the debate. Where they ‘agree to
differ’, as the saying is, there is nothing on which they have really
agreed . . . Ever since the world began, if anybody has become intolerable
Louis O. Mink, Mind, History, and Dialectic: The Philosophy of R. G. Collingwood

(Bloomington/London: Indiana University Press, 1969).
65 Mink, “Collingwood’s Historicism,” 177.
70 See Collingwood, The New Leviathan, 34.5–58, 35.11.
71 There is no denial that progress can be halted or suffer setbacks. That is part of the threat
Collingwood thinks is posed by “barbarism. See The New Leviathan, 497–8.
to his neighbours, what has made him intolerable has been something
maniacal or fanatical about his demeanour. The lesson of making oneself
tolerable, the lesson of cultivating a type of demeanour which other peo-
ple find they can stand, often depends on distinctions which in them-
selves are slight; a little more here, and a little less there, may suffice to
convert what no man can stand into something tolerable. It is distinc-
tions of this kind which are systematically pursued by the man who
makes a point of coming to an agreement with everyone from whom he
might have differed; that is, what Plato calls the dialectical man.72
Bibliography
Mark Bevir, The Logic of the History of Ideas (Cambridge University Press, 1999).
David Boucher, The Social and Political Thought of R. G. Collingwood (Cambridge:
R.G. Collingwood, Speculum Mentis, or The Map of Knowledge (Oxford: Clarendon,
[1924] 1963).
———, An Essay on Philosophical Method (Oxford: Clarendon, 1933).
———, An Autobiography (Oxford: Oxford University Press, [1939] 2002).
———, An Essay on Metaphysics (Oxford: Clarendon, 1940).
———, The New Leviathan: or Man, Society, Civilization and Barbarism, ed. David
Boucher (Oxford: Clarendon, [1942] 1992).
———, The Idea of History: With lectures 1926–1928, ed. Jan van der Dussen (Oxford:
Oxford University Press, [1946] 1993).
———, Essays in the Philosophy of History, ed. William Debbins (Austin: University of
Texas Press, 1965)
———, The Principles of History and other writings in philosophy of history, ed. W.H.
Dray & W. J. van der Dussen (Oxford: Oxford University Press, [1999] 2001).
Alan Donagan, The Later Philosophy of R. G. Collingwood (Oxford: Clarendon Press,
1962).
Christopher Fear, “The question-and-answer logic of historical context,” History of the
Human Sciences 26 (2013), 68–81.
Stein Helgeby, Action as History: The Historical Thought of R. G. Collingwood (Exeter:
Imprint Academic, 2004).
Fred Inglis, History Man: The Life of R. G. Collingwood (Oxford: Princeton University
Press, 2011).
72 Collingwood, The New Leviathan, 24.57–24.6, 45.9–91. See also 29.61.

100 fear
Louis O. Mink, Mind, History, and Dialectic: The Philosophy of R. G. Collingwood

(Bloomington/London: Indiana University Press, 1969).
———, “Collingwood’s Historicism: A Dialectic of Process”, in Critical Essays on the
Philosophy of R. G. Collingwood, ed. Michael Krausz (Oxford: Clarendon Press, 1972),
154–78.
Friedrich Nietzsche, Beyond Good and Evil [1886] trans. Helen Zimmern (Mineola:
Dover, 1997).
Quentin Skinner, “Some Problems in the Analysis of Political Thought and Action,”
Political Theory 2 (1974), 277–303.
———, “A Reply to my Critics,” in Meaning & Context: Quentin Skinner and His Critics,
ed. James Tully (Princeton: Princeton University Press, 1988).
———, Visions of Politics, vol. I, Regarding Method (Cambridge: Cambridge University
Press, 2002).
part two
Perspectives from Contemporary Philosophy
∵
chapter 6
Representationalist Logic
Frank Ankersmit
1 Introduction
In 1925 Dietrich Mahnke published his Leibnizens Synthese von Universal

mathematik und Individualmetaphysik.1 Since the book did not go beyond a
careful discussion of what was written on Leibniz in the three to four decades
before its publication one could hardly claim that the book changed the face of
philosophy. Nor even that of Leibniz-scholarship only. Nevertheless, the book
remains to be read in Germany down to the present day.2 Not only because
Mahnke gave a very well informed and even-handed account of the work on
Leibniz by philosophers such as Erdmann, Wundt, Russell, Couturat, Cassirer,
Kabitz, Pichler, and many others, but also because of what the very title of
Mahnke’s book seemed to promise.
Self-evidently, “Universalmathematik” should be related to the univer-
salism of the sciences and their reliance on mathematics, whereas the term
“Individualmetaphysik” is suggestive of a world of individual entities as stud-
ied in the humanities, and especially in history. Had Friedrich Meinecke not
used Goethe’s individuum est ineffabile as the epigraph for his magisterial study
of the historicist discovery of the historical world in Western thought?3 So did
the title of Mahnke’s book not suggest that we should turn to Leibniz’s philoso-
phy of the substance or the monad if we wish to get down to the logical bottom
of the relationship between the sciences and the humanities, of where they
differ and where they agree from a logical point of view?
In this chapter I shall prepare my way for such a reading of Leibniz, but
without actually presenting it. I shall restrict myself here to a preliminary,
systematic rather than historical analysis of several aspects of the logic of
1 Dietrich Mahnke, Leibnizens Synthese von Universalmathematik und Individualmetaphysik

(Halle: M. Niemayer, 1925).
2 The book was reprinted in 1965 and often used in German universities as a textbook for
undergraduate courses in the history of philosophy.
3 Friedrich Meinecke, Die Entstehung des Historismus (Munich and Berlin: R. Oldenbourg,
1936).

104 ankersmit
(historical) representation allowing us to get a grasp of the main similarities

and differences between the sciences and historical representation. Though
the analysis presented here did have its source of inspiration in Mahnke’s book,
it will develop independently of what Mahnke had to tell his readers about
Leibniz and his interpreters at the turn of the nineteenth and the twentieth
century. My conclusion will be that whereas modern formal or symbolic logic
replaced traditional Aristotelian scholastic logic by what Cassirer calls “rela-
tionalist logic,” representation contains elements of both while, at the same
time, not being reducible to either of them. So the thesis I wish to defend in
this chapter is that (historical) representation has a logic of its own, apart from
Aristotelian and modern formal or symbolic logic. In another essay I outline
how all of this can be found already in Leibniz’s philosophy of the substance
or the monad.4
2 A Minimal Characteristic of Historical Representation
I shall assume the historical text to consist of a finite set of true singular state-
ments about the past. Such sets I shall call historical representations of the
past5 (to be abbreviated as HR’s). I shall assume, furthermore, that the main
philosophical problems occasioned by historical writing and the historical text
can be investigated by means of an analysis of HR’s, either directly or indi-
rectly. In case the historical text possesses features going beyond the definition
of historical representation given a moment ago, but that can nevertheless be
explained on the basis of a theory of historical representation, I shall say that
the features in question can be explained indirectly (i.e. by moving backwards
from that theory of historical representation to how the historical text presents
itself to its readers). The foregoing implies that the relationship of true sin-
gular statements to what they are about falls outside the scope of a theory of
HR since any such theory deals exclusively with the relationship between such
statements and the HR of which they are part.
Though HR’s consist of statements typically having the form of propositions
either truthfully or falsely attributing a certain predicate to a subject, this it
not the case with HR’s themselves. So they cannot possibly be said to be true or
false. Nevertheless, statements can be formulated about HR’s. Such statements
4 F.R. Ankersmit, “History as the Science of the Individual,” Journal of the Philosophy of History 7
(2013), 396–425.
5 For an explanation of this terminology see my Meaning, Truth and Reference in Historical
Representation (Ithaca and London: Cornell University Press, 2012), 59–63.
representationalist Logic 105
may be of different types and express different truths or falsehoods. But one
type deserves special attention. Suppose a HR consists of the statements s1,
s2, s3. . . . sn (and where n is a finite number), then any statement of the form
“HR = sa” (and where 1 ≤ a ≤ n) will be analytically true, whereas any statement
other than those mentioned just now will be analytically false. Statements of
the form “HR = sa” should always be read as statements about HR’s (giving us
their representational meaning) and never as statements about historical real-
ity, such as for example sa itself. This leaves room for the possibility that state-
ments like “HR is sa” and “HR is sb” are (analytically) true, whereas statements
such as sa and sb themselves are (empirically) false. It follows that one and the
same statement such as sa or sb can be analytically true if interpreted repre-
sentationally and empirically false if interpreted referentially—and where the
representational interpretation is to be associated with “historical writing”
(“Geschichtsschreibung” in German) and the referential interpretation with
“historical research” (“Geschichtsforschung”).6 What is representationally true
may be empirically false, and vice versa. This, then, gives us the logical basis for
the distinction between “historical research” and “historical writing”—a dis-
tinction corresponding to that between the statement and representation—
and that may function as a permanent warning against any attempt to project
any variant of the thesis of the theory-ladeness of empirical facts on the writing
of history.7 Whereas that thesis emphasizes the link between fact and theory,
the preceding argument radically severs it between fact and representation.
Normally it makes sense to distinguish between identification and individu-
ation. A member of a set S can be said to be identified by operation O and
where O is not constitutive of that member. In case of individuation, however,
in the operation O the selection of the member in question and its constitu-
tion are identical. Statements of the form “HR x is sy” do not identify HR in the
sense of fixing the reference of that statement’s subject-term; they cannot be
read as saying: “sy is uniquely true of HR x” for there exists an infinity of HR’s
possessing the property sy. For the same reason sy cannot be said to individuate
HR x. But a HR consisting of the statements s1 to sn can properly be said to be
identified by s1 to sn and to be individuated by “HR is s1,” “HR is s2,” . . . “HR is sn.”
The statements in a historical text have a double function: to describe the past
and to individuate a HR—and from the perspective of representational logic
only the latter function demands our attention. Of this latter function it can
be said that a historical text’s statements 1 to n recursively, or self-referentially
6 For further details on this distinction, Ankersmit, Meaning, x, 60–62.

7 This is not meant, of course, to be a denial of the plausibility of that thesis for the relation-
ship between facts and theories.
106 ankersmit
individuate (or define) a HR, namely that HR of which all statements “HR is
s1,” . . . “HR is sn” are analytically true.
I will now introduce the notion of the representationalist universe.8 The
representationalist universe contains the totality of all HR’s. This totality—and
hence the representationalist universe is infinite. Since all we presently have
is a finite number of HR’s, independent proof is needed to support the claim
of the infinity of the representationalist universe. We can think of two strat-
egies for substantiating it. The first one goes like this. Suppose we have two
HR’s, HR1 and HR2 having no properties in common. We can then think of new
candidates to be added to the representationalist universe, each of them con-
sisting of some permutation of the properties of HR1 and HR2. However, such
combinations will not necessarily result in new HR’s. Obviously, the attempt to
derive in this way new HR’s, one say from Hellenism and another from the Cold
War, will not result in a new HR. Admittedly, in other cases this may be differ-
ent, but our example forces us to recognize that this is not necessarily the case.
It follows that the multiplication of HR’s along these lines will not necessarily
move us from a finite to an infinite representationalist universe. So this effort
to argue for an infinite representationalist universe leaves the case undecided.
Let us now address the problem from the other side, so to say. Let us begin
again with the finite representationalist universe familiar to us. Suppose it to
contain n HR’s (and where n is a finite number) and suppose there to be a
historical representation HR1. We may consider the possibility of adding one
true statement about the past to HR1 and then ask ourselves whether the pro-
cedure results into a new historical representation, say HR2. It will be here as
in the previous paragraph: it depends. Sometimes HR2 will be a new HR, but
not always so. Think of adding “Beethoven died in 1827” to a HR on medieval
theology. It would be somewhat like trying to put a tennis ball on a moving car:
the ball will blow off right away. But in other cases the extra statement may
effortlessly be absorbed by the HR to which it is added.
We might try to circumvent this sadly inconclusive state of affairs in the
following way. Suppose there to be a HR1 that is completely identical with HR2
with the exception that HR1 contains p whereas HR2 contains q. Then we can
think of a potential HR3 being identical with HR1 and HR2 with the exception
of containing both p and q. This seems a safe way to multiplicate HR’s, since
HR3 1) does not state anything not yet stated by H1 and H2, 2) is nevertheless
different from H1 and H2, and 3) thus not yet part of the (finite) set of HR’s we
8 For a formal description of the representationalist universe, see my Narrative Logic:

A Semantic Analysis of the Historian’s Language (Boston and The Hague: Martinus Nijhoff,
1983), 143–147 (where the term “narrativist universe” is used).
already had. So is this not a first step into the direction of an infinite represen-
tationalist universe?
However, we have two possibilities here: if HR3 is not a representation we
shall be left with the finite set of HR’s we already had, but if it is one, we shall,
indeed, have only just one more HR and no more than that. So the prolifera-
tion of representations always stops right after it has begun, unlike numerical
series such as 1, 2, 3, 4, . . . going on infinitely. Of course, we can think of a HR4
having the properties p, q and also r and repeat the same trick with HR3 and
HR4. And, indeed, so can we go on in principle. However, not indefinitely so,
for we began with a finite representationalist universe of HR’s each of them
having a finite set of properties. Hence, even if we assume that each step in this
procedure will actually result in a new HR, even this will not move us from a
finite to an infinite representationalist universe. Finally, suppose we have two
HR’s not being identical except for just one property (as is in the case discussed
just now) but for x properties (x ≥ 2). Assuming again that each individual dif-
ference results in a new HR, their total will be x!; and even though x! may be a
very large number, it will never be infinite. In sum, even if we are as generous as
we could reasonably be with the admission of new HR’s to an already existing
finite set, this will never give us an infinite set.
The second strategy requires us to move from the representationalist uni-
verse to the “real” universe. It is a trivial fact that an infinity of statements can
be made on the real world; in the real world obtains the dispensation of infin-
ity. For example, for any finite length in the real universe we can think of a
smaller one, and so on ad infinitum. Inversely, the proportion of any item on
this series of ever decreasing lengths to any fixed length will then be enlarged
beyond any finite number. If there is no smallest number, than there is no
largest number either. Obviously, the representationalist universe will have to
respect this feature of the real universe, for no statement we can make on the
real world can be ruled out apriori as a candidate for being part of some HR.
Consequently, infinity will be predicated on the representationalist universe
by the real universe.
Hence, the representationalist universe is not defined or determined by any
finite set of HR’s but by the infinite set of true singular statements we can make
about the real universe. Which does not exclude, of course, that if, for either
practical or theoretical reasons, some subset of this set of this infinite set of
potential statements about the real world does not qualify as part of any pos-
sible HR, the set of possible HR’s should become finite. For however much one
subtracts from the infinite, the leftover will nevertheless remain infinite.
Two conclusions can be inferred from the foregoing. In the first place, we
adopted above two strategies for proving the claim that the representational-
108 ankersmit
ist universe is infinite. The first one failed, whereas the other was successful. It
follows from this that although the representationalist universe is infinite, it is
not a continuous one in the sense that one and the same operator (whatever
that operator may be) will allow us to infer all components of that universe
from any other in it, as is the case with the series 1, 2, 3, . . . or in Newtonian
space. For as we saw when following the second strategy, it will often be neces-
sary to appeal to the real universe in order to move from one component to an
other. We shall return to this when discussing the striking similarities between
the representationalist universe and Leibniz’s monadology.
The other conclusion is that the representationalist universe precedes any
individual HR or any set of them. The representationalist universe is not a
product of existing HR’s, however large their number may be. We should there-
fore conceive of the representationalist universe as being logically prior to its
components, viz. individual HR’s. “At the beginning” there was the representa-
tionalist universe and not individual HR’s. The relationship between the rep-
resentationalist universe and its components is like that between a whole and
its parts, and where the latter presupposes the existence of the former. In this
picture each historical text can be said to give us one of those parts that had
always been, and will always remain part of that universe. Each such text is
like stating (x1,y1,z1) and where (x1,y1,z1) corresponds to some specific point in
three-dimensional space whose existence is not dependent on stating (x1,y1,z1).
If the representationalist universe were finite, some subset of the total set of
statements contained by a HR would already suffice for giving us that specific
HR; identification would then be all we need. Now that the representationalist
universe has been found to be infinite we shall need for this nothing less than
all of the statements contained by a specific HR—as is achieved in the opera-
tion of individuation. In this way the claim of the representationalist universe’s
infinity hangs together necessarily with the thesis that HR’s can be defined
only be an enumeration of all their properties (and vice versa): they are both
sides of one and the same coin.
3 “The” as in “the HR . . .”
Now, suppose we have a statement on a HR such as “HR is sx.” Obviously, the

statement is meaningless, as it stands in need of further specification. For its
truth can only be established if we know to which HR the phrase “HR” refers to.
And now a problem announces itself. Recall that we saw in the previous sec-
tion that the representationalist universe precedes all HR’s contained by it. We
will then recognize that there will be more than one HR having the property
of containing sx. In fact, there is a whole class of them—the class of all HR’s
containing sx. The statement “this HR is sx” is true of any member of that set.
So in case we are dealing with HR’s no room seems to be left for statements
like “this HR is sx” and where “this HR” refers to one unique HR. Put differently,
statements about HR’s of the form “this HR is sx” are always statements about
classes of HR’s and never about individual HR’s.
But, fortunately, this need not be the end of the story. For we can move from
classes to individuals. However, doing so will require us to be no longer con-
tent with just one—or any subset of all the properties of a HR—for only the
complete list of all of a HR’s properties may yield statements that are true of
one and only one specific and unique HR. Only statements summing up all the
properties of some HR can be said to be true of only this HR and not of any
other member of some class of HR’s having sx as one of its properties. But in
that case we shall have one and the same list of properties on both sides of the
copula—as we shall recognize when remembering that HR’s are nothing more
and nothing less than the sum of all their properties. Truth coincides here with
identity. For only in such cases can we say e.g. “the HR having the properties
s1, . . . sx, . . . sn is sx.” In case we would for whatever reason not happen to know
that the HR in question has these properties s1 . . . sx . . . sn, the statement will be
an aposteriori truth; in case we do, it will be an analytical truth. This, then, is
how things are like with statements on individuals HR’s, hence with contexts in
which we are entitled to speak of “the HR is . . .” And where the main lesson of
the foregoing is that we can only speak of “the HR is . . .” in case all of the prop-
erties of the HR in question are enumerated on the place of the dots.
4 Russell on “the”
In the first paragraph of chapter 16 of his Introduction to Mathematical

Philosophy of 1919 Russell writes: “in this chapter we shall consider the word
the in the singular, and in the next chapter we shall consider the word the in the
plural.”9 It may be thought excessive to devote two chapters to one word, but
to the philosophical mathematician it is a word of very great importance; like
Browning’s Grammarian with the enclitic δε, I would give the doctrine of this
9 Note that Russell does not deal here with “the” as it occurs in phrases like “the average tax-
payer,” or “the man in the street.” This is of interest in the present context since “the φ” as used
in such phrases does not refer. Not because there exist no φ’s—as was the case in Russell’s
argument—but because φ’s should be seen here as aspects of the world. See Ankersmit,
Meaning, Truth, 92.
110 ankersmit
word if I were “dead from the waist down and not merely in a prison.” Russell
gave in this chapter his mature views on description and which is generally
regarded—both because of its argument and its philosophical style—as not
only one of the major contributions to contemporary philosophy, but to all
of Western philosophical thought. Russell’s theory of description was mainly
an attempt to deal with the problem of language (more specifically of denot-
ing expressions) about unreal objects. But this compelled Russell to develop
a theory of the proposition whose implications went far beyond his original
problem and that are still decisive for the philosophical analysis of proposi-
tions and of what is expressed by them.
Russell first stated his theory on description and denotation in his essay
entitled “On Denoting” that was published in Mind in 1905.10 Since this essay
provides us with a better background for both getting a grasp of Russell’s
own argument and for how it relates to historical representation, I shall dis-
cuss below this essay and not the chapter from Introduction to Mathematical
Philosophy mentioned above. Russell’s point of departure here are proposi-
tions like “C(x)” and where x is an undetermined variable allowing, however,
for quantification, as in “all x,” “no x,” or “some x.” “All,” “no”, or “some” are for
Russell the most primitive of denoting phrases; they “denote” by saying that all,
no, or some objects in the world are C. Clearly, these denoting phrases have no
meaning themselves, but they will ensure that the phrases in which they occur
have one. Russell then goes on to consider phrases like “the round square” or
“the present King of France,” hence phrases suggesting the existence of objects
that clearly do no exist. There are no round squares and France is no longer a
Kingdom. So how to understand such phrases? Both Meinong and Frege came
up with an answer to this question and both answers are condemned as unsat-
isfactory by Russell.
Meinong had argued that such phrases denoted existing objects even
though there are no round squares or Kings of France. But this sins against
Ockham’s razor and invites contradictions of the kind that round squares both
exist and do not exist.11 Of more interest is, therefore, how Frege tried to tackle
the problem. Frege’s point of departure was his well-known sense/reference or
meaning/denotation distinction. And then the problem arises what the phrase
“the present King of France” in the proposition “the present King of France is
bald” refers to if there exists no present King of France. Or, more specifically,
10 Bertrand Russell, “On Denoting,” Mind 14 (1905), 479–493.

11 Recently Edward Zalta has developed a theory of abstract objects implying a rehabili-
tation of Meinong’s argument. See: http://mally.stanford.edu/theory.htm#dis. I’d like to
thank Jaap den Hollander for this reference.
what happens if we apply the meaning/denotation to this problem of a non-

existent denotation? Russell has two answers. Firstly, suppose we differentiate
(with Frege) between the denotation and meaning of the phrase “the present
King of France”, this would permit us to say 1) “the present King of France”
2) is a “King of France,” and where 1) gives us the phrase’s denotation and 2) its
meaning. But this sentence is nonsense since 1) has no denotation. Secondly,
Frege himself would have dealt with the problem of this example by saying
that the phrase “the present King of France” denotes the (a) null-class. And,
indeed, as Russell is willing to grant, this idea of the null-class avoids the prob-
lems entailed by Meinong’s solution. On the other hand, however it merely
pushes the problem to a new level by confronting us with the question how
the notion of existence hangs together with this idea of the null-class. Taking
together Meinong and Frege, Russell therefore concludes: “thus we must either
provide a denotation in cases in which it is at first sight absent, or we must
abandon the view that denotation is what is concerned in propositions which
contain denoting phrases.” The first option was embraced by Meinong with his
paradox of non-existing things and by Frege with his null-classes. But Russell
himself chooses the second option.
When clarifying this second option Russell takes as his example “the father
of Charles II was beheaded”, and he begins with stating that this proposition
is asserting that there was an x who was the father of Charles II, and who
was beheaded. First, the proposition “the father of Charles II was beheaded”
implies that there must have been a person of whom this is true. Russell
reduces this claim to the terminology of truth and falsehood by rewriting that
proposition as “it is not always false of x that x was the father of Charles II
and was beheaded.” Meinong’s and Frege’s stumbling block of existence is now
out of the way, and has been sublimated into the so far more manageable cat-
egories of truth and falsehood. Next, Russell insists that the proposition also
involves uniqueness: for it is not only true that there was an x who was father
of Charles II and was beheaded—no, there was also only one x of whom this
is true and not more than just this x. So any analysis of “the father of Charles II
was beheaded” will have to do justice somehow and somewhere to this dimen-
sion of uniqueness. Russell proposes to satisfy this requirement by stating that
apart from the assertion we already had—viz. that there was an x who was the
father of Charles II and was beheaded—we shall also need the assertion “if y is
the father of Charles II y must be identical with this x.” In fact, all phrases con-
taining “the,” as in “the present King of England,” “the center of the mass of the
solar system,” “the revolution of the earth around the sun,” and so on, express
this appeal to uniqueness and will have to be rewritten accordingly. Taking all
this together we can conclude that “the father of Charles II was beheaded” is
112 ankersmit
equivalent to “it is not always false of x that x was the father of Charles II and
that x was executed and that ‘if y was the father of Charles II, y is identical
with x’ is always true of y,” One could rephrase this in terms of symbolic logic
as follows:
(Ex) { FCx . (y) [FCy → (y = x)] . Bx}
and where (Ex) stands for: “there is an x,”, “.” for “and,” “FC” for “father of Charles
II,” and “B” for “beheaded.” Though this reformulation has the disadvantage
that the phrase “there is an x” possesses the unpleasant propensity to resusci-
tate again all the old existential worries that Russell’s procedure had precisely
meant to exorcise. The “it is not always false etc”. variant does not encourage
this kind of misreading.
Two remarks are in order from the perspective of the present argument. In
the first place, Russell’s main worry had been phrases denoting things that do
not exist, such as the present King of France. He solved the problem by trans-
lating existential claims into terms of truth and falsehood. However, in 1950
P.F. Strawson published an article arguing that we should suspend our judg-
ment about propositions about the present King of France as long as we do not
know what a speaker saying this at the beginning of the 21st century may have
in mind with it.12 And then Russell can be said to have created problems where
they do not actually exist. More important is, however, a second point. Russell’s
strategy had been to empty the subject-term of propositions of all content and
to shift this content to the predicate part of the proposition while replacing the
subject terms by quantifiers such as all, some, one, or no. In this way Russell
contributed to what has come to be known since Quine as the “canonical nota-
tion” of the proposition. Strawson once summarized it as follows:
[. . .] the relevant part of Quine’s programme of paraphrase can most sim-
ply be summed up as follows. All terms other than the variables of quan-
tification will be found, in canonical notation, to be general terms in
predicative positions. The position of the singular term is reserved for the
quantifiers and the variables of quantification; and since quantifiers
themselves cannot count as terms, the only singular terms left are the
variables of quantification.13
12 P.F. Strawson, “On Referring,” Mind 59 (1950), 320–344.

13 P.F. Strawson, “Singular Terms and Predicates,” in Philosophical Logic, ed. P.F. Strawson
(Oxford: Oxford University Press, 1967), 79.
Subject-terms thus evaporate into predicate terms and only the variables of
quantification may still remind us of its existence in a previous, logically less
enlightened age.
5 The Logical Basis of the Difference between the Sciences and

(Historical) Representation
Now, the decisive fact is that this will not work for HR’s. Suppose we get into
our hands some book on some part of the past and let us say that this book
presents a historical representation for that specific part of the past. We can
then formulate phrases like “the historical representation presented in this
book” analogous to Russell’s “the present King of France” or “the author of
Waverley” and attribute certain properties to what is denoted by “the histori-
cal representation presented in this book.” But suppose that we go on to apply
Russell’s strategy to such phrases denoting HR’s. We saw that part of that strat-
egy requires us to do justice to the uniqueness apparently being part of denot-
ing phrase. More specifically, what is said about what is denoted by a denoting
phrase is not only true of it, but of only it and of nothing else.
The crucial thing is that this strategy cannot be applied to HR’s, since any
property we might propose as singling out one specific HR from all others can
be attributed to other HR’s as well. Suppose we have a HR1 consisting of the
following set of statement about the past: s1, s2,. . . . sf, . . . sh, . . . sn. Then the set
sf . . . sh can never be the exclusive property of HR since we can think of any
desired amount of HR’s also possessing this set. I recall to mind here that we
found in the previous section that the representationalist universe precedes
the construction of HR’s, that it is not staked, or limited by existing HR’s and
that the construction of HR’s has the character of individuation (on the basis
of all of a HR’s properties) rather than of identification (on the basis of some
subset of properties believed to be a HR’s unique possession). As was argued
in section 3, this is, however, not the end of the story. For what is truly unique
about HR1 is that it consists of s1, s2, . . . sf, . . . sh, . . . sn. Put differently, nothing
short of a delimitive enumeration of all of its properties is needed to give us
HR1’s uniqueness. But this gives us back again the original subject-term; hence,
subject-terms in propositions on HR’s cannot be eliminated as is the case in
Russell’s analysis and in Quine’s canonical notation.
If anywhere, we may discern here the logical ground for the distinction
between the sciences and the humanities (as exemplified by the writing of his-
tory). The sciences have their logical model in Russell’s and Quine’s analysis of
the proposition. The sciences dissolve subjects into their properties, replacing
114 ankersmit
them by the variables of quantification. They presuppose the possibility of this

procedure. For scientific theories are, basically, relations between properties as
characterized by the variables in a scientific theory and therefore have the logi-
cal structure of Quine’s canonical notation. But it is impossible to dissolve HR’s
into their properties, or into any subset thereof: they will successfully resist
even the most strenuous effort to do so. The subject-term truly is unassailable
here. The sciences and the humanities are fundamentally different disciplines;
and these differences have their ultimate roots in subject and predicate logic
and the different roles assigned to subject and predicate in either the sciences
or the humanities. But differences do not exclude the possibility of similari-
ties as well. Below I shall mention two of them. Firstly, relations are basic to
both the sciences and the humanities and, furthermore, in both of them rela-
tions strongly reduce the importance of the role of true statements about the
world and of the subject terms that are mentioned in them. Both the sciences
and the humanities are essentially relationalist. Secondly, I hope to show that
accepted wisdom about models in mathematics and the sciences also holds
for representation.
6 Cassirer’s Substanzbegriff und Funktionsbegriff (1): Relations
Russell’s theory of denotation and description has been decisive for the evolu-
tion of symbolic logic and contemporary philosophy of language. Of course
its details have intensively been discussed—I mentioned already Strawson’s
article in Mind published in 1950. But few people would doubt that it is basi-
cally correct, and, even more importantly, realize themselves that the theory
was the winner in a fierce struggle for survival of alternative attempts to define
the nature and tasks of logic at the beginning of the previous century. Yet
such an awareness is indispensable for an adequate assessment of the conflict
between Russell’s theory the account of historical representation presented in
the previous section. In order to correct this all too easy and unwitting identi-
fication with the winner we’d best turn to Ernst Cassirer’s Substanzbegriff und
Funktionsbegriff of 1910 (to be abbreviated as SF). Cassirer’s book may, there-
fore, help us to circumvent “the Whig interpretation of the history of logic,” to
use Herbert Butterfield’s well-known terminology.
The two notions Substanzbegriff (concepts expressing substance) and
Funktionsbegriff (concepts expressing function) in the title of SF stand for
two kinds of logic: traditional Aristotelian logic and modern formal logic. The
book’s main argument is that only the latter can adequately account for mod-
ern developments in mathematics and the sciences. Aristotelian logic has its
basis in a metaphysics claiming that the world consists of individual objects

or things each having their own essence or substance (hence Cassirer’s notion
of the Substanzbegriff ). Aristotelian logic moves from there into two direc-
tions: in the first place into that of the abstraction of concepts and, secondly,
into that of syllogistic logic. Thus one can subsequently say of Socrates he is
1) a human being, 2) a mammal, 3) an animal, 4) a mortal living organism, and
5) a material object. Properties can be attributed to the elements denoted by
a certain concept, for example when saying that “all human beings are mor-
tal.” And these, in their turn, allow for syllogistic reasoning, as when inferring
from the above-mentioned premises that Socrates must be mortal. In a sense
Aristotelian logic is unexceptionable: it will never lead us astray. And in sci-
ences where classification is basic—as in Linnaeus’s biology—it may even be
quite helpful.
Yet it has one major weakness: “vor allem ist es aber die Kategorie der
Relation, die durch diese Grundlehre des Aristoteles zu einer abhängigen und
untergeordneten Stellung herabgedrückt wirdt.”14 This is undoubtedly correct:
for Aristotle relations can only be relations between things; it is inconceivable
that there should be relations without things or sets of things. Things have a
logical priority to the relations between them; and a relation without the things
that are related by it, would be like a bridge hanging in empty air. Replacing
things by their concepts or substances, or by ideas (to use Berkeleyan terminol-
ogy) will not remedy this limit to the scope of Aristotelian logic. Cassirer even
goes as far as seeing from this perspective no relevant differences between
Aristotelian logic and Mill’s (empiricist) logic. Think, for example, of Mill’s
extensional definition of the number x as being exemplified by any set of x
things in his System of Logic (thus inviting Frege’s scathing comment on Mill’s
“arithmetic of pebblestones and peppercorns” empirically always giving us five
of each of them, when adding two more to the three pebblestones of pepper-
corns that we had already15). And whatever the differences between Aristotle
and Mill may be, it is certainly true that the logic of both presupposes the pos-
sibility of reduction to objects, and thus both resist the independence of rela-
tions with regard to what is related by them. In sum, traditional Aristotelian
logic recognizes only two kinds of relationship: that between a thing and its
properties, and that between part and whole.
Cassirer points out that Aristotelian logic has two decisive and closely
related disadvantages in comparison with modern relational logic as developed
14 Ernst Cassirer, Substanzbegriff und Funktionsbegriff: Untersuchungen über die Grundfragen

der Erkenntniskritik (Hamburg: F. Meiner, 2000), 7.
15 Cassirer, Substanzbegriff, 28, 29.
116 ankersmit
in the course of the second half of the nineteenth century. In the first place,
Aristotle’s “Begriffspyramiden” fail to do justice to mathematics and to the
practice of science. Points, lines, and planes are not the generic properties of
concrete objects in actual reality but the building stones of a purely abstract
construction. Similarly, the hierarchical rigidity of Aristotelian logic prevents
it from adapting to the pliability and multiformity of scientific theory. More
specifically, in Aristotelian logic abstraction inevitably results in the loss of
conceptual content, whereas the effort of the sciences is precisely to achieve
the reverse. The best scientific theory is the theory combining a maximum of
content with the greatest scope. Cassirer’s second point is that the sciences
essentially succeed in attaining this remarkable feat by showing how differ-
ent variables describing natural processes hang together—or in the appropri-
ate terminology, what is the relationship between such variables. And, indeed,
the scientific theory typically has the form of a general statement expressing a
relationship between a given set of variables such as mass, velocity, tempera-
ture, volume, and so on.
This gets us to the second term in the title of Cassirer’s book: the notion
of function. For it will need no elucidation that the two are most intimately
related to the point of even having the same meaning. Think of the simple
second order function f(x) = x2; then the function can be said express the rela
tionship between what is on the left side of the copula to what is on the right
side of it.16 Finally, in his exposition of the philosophy of arithmetic, Cassirer
discerns two strategies for the logical derivation of numbers: Russell’s strategy
for doing so in terms of classes and Gauss’s strategy to achieve the same by
means of the mutual relationship between numbers.17 And in agreement with
his relationalist approach to the philosophy of mathematics and science he
prefers the latter.
7 Cassirer’s Substanzbegriff und Funktionsbegriff (2): Models
The other shortcoming of Aristotelian logic—and, in fact, little more than

an ontological re-statement of what was said a moment ago about Cassirer’s
advocacy of a relationalist logic—concerns the relationship between math-
ematics and physics, on the one hand, and reality, on the other. The main
16 Cassirer, Substanzbegriff, 23.

17 Cassirer, Substanzbegriff, 50–52, 70. See also Cassirer’s remark on 59, 62–64 on Dedekind’s
theory of numbers as developed in the latter’s Was sind und was sollen die Zahlen? of 1893.
This book was the main inspiration for Cassirer’s account of arithmetic.
difference between Aristotelian logic (and its latter-day variants as embraced

by Mill) and relationalist logic is that the latter is not an abstraction from real-
ity, but an entirely “apriorist” construction.18 Think, for example, of geometry.
The notions of point, line, and plane are not arrived at by abstraction from a
stone in one’s garden, the straight road connecting the city in which you live
with some other or from a well-kept lawn, but from the axioms defining them.
This time Hilbert is Cassirer’s main guide and Cassirer summarizes Hilbert’s
views as follows:
[. . .] gegenüber der Euklidische Begriffsbestimmung, die die Begriffe des

Punktes order des Geraden, von denen sie ausgeht, als unmittelbare
Gegebenheiten der Anschauuung nimmt und die ihnen somit von
Anfang an einen bestimmten unabänderlichen Inhalt aufprägt, wird hier
der Bestand der ursprünglichen geometrischen Figuren ausschliesslich
durch die Bedingungen bestimmt, denen sie gehorchen.19
In sum, 1) the definition of axioms need not satisfy our intuitions about space,
and 2) whatever follows from the definition the axioms is true, again, regard-
less of our intuitions about them. But that does not imply that the truth of
geometry should be without consequences for our knowledge of reality. On
the contrary, precisely because of their being emptied of all empirical content
the truths of geometry will also be true of those aspects of reality that are not
at odds with how points, lines, and so on are defined by the relevant set of
axioms in geometry. Or, to out the point more forcefully, this is where models
in mathematics and the sciences are exactly the reverse of what we ordinarily
associate with that notion. One may build, for example, a model of an airplane
in order to find out about its aerodynamic properties. In such cases (as in the
models used in economics) we are first given an empirical reality, such as an
airplane or a national economy and we may, next, build a model of it imitating
its properties and behavior under certain circumstances. But in mathematics
and the sciences it is the other way round. There we have, first, a calculus as
defined by a set of axioms and we may then see, next, what aspects of reality
18 I shall not discuss here whether the term “apriorist” is wholly appropriate. Decisive is here
whether one prefers a Platonist account of mathematics (and of numbers) or a construc-
tivist and strictly nominalist one. The Platonist will have his problems with the term (as
was the case with Russell before he was convinced by Wittgenstein that the truths of logic
and mathematics are tautologies), whereas the constructivist will be ready to accept it.
But a discussion of this age-old issue is of no relevance in the present context.
19 Cassirer, Substanzbegriff, 99.
118 ankersmit
correspond to it, if any. In the case of the airplane and of national economy the
“theory” is a model of reality; in that of mathematics and the sciences reality is
a model of the wholly abstract calculus.
8 Representationalist Logic
Cassirer’s way of opposing traditional Aristotelian logic to modern symbolic

logic is of interest in the present context since it may help us to come to a better
understanding of historical representation. More specifically, I will argue that
historical representation has features in common with both Aristotelian and
symbolic logic while, at the same time, it cannot be reduced to either of them.
To begin with, the affinities between Aristotelian logic and historical represen-
tation will be obvious. We saw in the previous section that the components in
the representationalist universe can only be individuated by an enumeration
of all their properties (viz. of all the statements contained by a HR) and are
recursively defined by all of them. It follows 1) that each true statement on a
HR is analytically true and, hence, 2) the notion of the subject of such state-
ments already contains all the attributes that can truly be predicated of it.20
This is where historical representation agrees with the so-called praedicatum
inest subjecto principle and that is generally seen as one of the main claims of
Aristotelian logic. With Aristotle the claim has a metaphysical basis in that for
Aristotle each object is defined by its substance as expressed by all the state-
ments giving the object’s substance.21 Anyway, this is where Aristotelian (scho-
lastic) logic agrees with what was said in the previous sections about historical
representation.
But historical representation also has its affinities with modern relationalist
logic (as this term is understood by Cassirer). In order to see this we need to
have a closer look at the representationalist universe. Just as in the sciences
where not all variables that one could possibly conceive of are realized, so it is
with the HR’s in the representationalist universe. This observation is not merely
a reflection of the actual state of affairs in the writing of history—and where
the number of HR’s will not exceed the number of books on history that were
20 Ankersmit, Narrative Logic, 134–139.

21 Of course that leaves contingent truths about the object out of the picture. But whatever
one may think of this unclarity in Aristotelian logic and metaphysics, it need not bother
us within the context of a discussion of the components of the representationalist uni-
verse, since here the distinction between essential and contingent properties cannot be
made. Each change in the properties of a HR, however small, will give us a different HR.
written between, say 3,000 BC and the present day—but also has its ground
in another feature of historical representation. Elsewhere I have argued that
the meaning of historical representation is fixed or determined by other rep-
resentations, they do not have a meaning apart from those of other HR’s.22 It
follows from this that HR’s tend to be comments on each other and, next, that
they will tend to cluster together in groups of HR’s, whereas the “space” in the
representationalist universe between such clusters is relatively empty. A HR
without any clear relationship to other HR’s will tend to lose its meaning and
to disintegrate into its constituent statements—and no HR will be willing to
run that risk since it means its death. HR’s will thank their cohesion and unity
to the presence of other near-by HR’s; they are much like social beings prefer-
ring the presence of others to an existence in which they live in isolation from
the rest of society. Nevertheless, these clusters may take different forms. HR’s
on unique individual things—think of HR’s of entities as individual such as
Caesar, Charlemagne or Napoleon—will tend to form clusters around a fixed
center, much like the globular star clusters surrounding our galaxy; those on
large but vague and ill-defined topics such as Hellenism, the Enlightenment, or
the Cold War will rather be like galaxies themselves, having a formal complex-
ity agreeing with that of the topic addressed in them.
Now, this is where relations come in. For if the meaning of a HR in some clus-
ter is fixed by that of other more or less “nearby” HR’s being part of that same
cluster, the cluster can be said to structure meaning. That is to say, the cluster
will determine how one specific HR contributes to the meaning of some other.
If we are given HR1 and HR2 that are both elements from the same cluster,
the cluster defines their relationship and, hence, their meaning—though, of
course, all the other elements of the cluster will also contribute to their mean-
ing. And, inversely, we can only properly speak of the meaning of individual
HR’s on the basis of how the cluster relates it to other members of the cluster.
Assuming then, as I argued elsewhere, that historical representation has no
source outside meaning itself, it follows that in historical representation mean
ing is basically relational. Here, then, we see where historical representation
agrees with modern relationalist logic, and differs from Aristotelian logic.
Next, we saw that Cassirer’s relationalist logic compels us to see reality as
a model of an abstract mathematical calculus instead of the reverse. Or as
Dupuy put it:
[. . .] a model is an abstract form, as it were, that is embodied or instanti-

ated by phenomena. Very different domains of phenomenal reality—
22 Ankersmit, Meaning, chapter 7.

120 ankersmit
hydrodynamics and electricity, for instance, or light and sonic

vibrations—can be represented by identical models, which establish an
equivalence relation among them. A model is the corresponding equiva-
lence class. It therefore enjoys a transcendent position, not unlike that of
a Platonic Idea of which reality is only a pale imitation.23
In a future essay I shall discuss in what way the relationship between HR’s and
(historical) reality can be said to repeat Leibniz’s argument about the relation-
ship between substances or monads and phenomenal reality, and I restrict
myself here to what can be said about the issue of the relationship between HR’
and (past) reality from the perspective of meaning. If representionalist mean-
ing is essentially relationalist, HR’s are not models of past reality, but past real-
ity a model of the relevant HR’s. Relations between HR’s are decisive, and not
how they apply to reality. As long as we remain unaware of the relationalism
of HR’s, individual HR’s could be seen as models of those parts of past reality
that are represented by them. This is how we initially tend to conceive of how
HR’s relate to past reality. But when we recognize the relationality of HR’s the
emphasis shifts away from past reality to the relations between HR’s and the
clusters in which they tend to gather together. And then past reality can be said
to be a model of how historians represent it. Here, again, historical representa-
tion sides with modern relationalist logic rather than with Aristotelian logic.
And this brings me to the conclusion to this section. At the end of the nine-
teenth century Aristotelian logic was abandoned for modern formal, symbolic,
or relationalist logic mainly because only the latter could provide us with a
logical account of the newest developments in mathematics, and in the sci-
ences. As a result, logic was now seen to be the natural ally of mathematics and
the sciences. History and the humanities now found themselves confronted
with having to make the unpalatable choice between either associating them-
selves with futile and outdated Aristotelian logic, or the heroic declaration of
being able to do without any logic at all. But as this section suggests, the case
of the historian and of the practitioner of the humanities is not in the least
as hopeless as that. For we can think of a representationalist logic combining
features of Aristotelian logic (the praedicatum inest subjecto principle) with
the relationalism of symbolic logic and with its capacity to endow reality with
meaning having its origins in the domain of pure abstraction.
23 Jean-Pierre Dupuy, The Mechanization of the Mind: On the Origins of Cognitive Science,
trans. M.B. DeBevoise (Princeton: Princeton University Press, 2000), 29, 30. I owe this ref-
erence to Jaap den Hollander.
9 Conclusion
In this essay I have tried to discover the logical basis for the difference(s)
between historical representation and the sciences. Both are predicated on a
different logic, though saying this does not full justice to the complexity of the
situation. A better characteristic of the relationship between the sciences and
the humanities (insofar as representation is essential to the latter) can be given
if we realize ourselves that the shift from Aristotelian scholastic logic to mod-
ern formal logic left room for a (representationalist) logic combining elements
of both. Representationalist logic shares with Aristotelian logic its respect for
individuality,24 and with formal logic the respect for relationalism and the lat-
ter’s conception of the model.
Insofar as my argument makes sense it follows for two reasons that little is
to be expected from any epistemological approach to the agreements and dif-
ferences between the sciences and the humanities (as exemplified by either
Diltheyian, Collingwoodian, or Gadamerian hermeneutics or by any of their
(logical-)positivist counterparts). In the first place, such approaches have an
ineradicable tendency to take the true statement as their model and thus to
remain blind to the logical dimensions of historical representation. And it is
only from the latter that any good is to be expected. Secondly, and more inter-
estingly, if past reality is a model of a HR (as is the case in the sciences) and
not the reverse, epistemology is helpless. For a clarification of the relationship
between a HR and past reality as a model of that HR will require us to appeal to
logic and not to epistemology. The model, as understood in the philosophy of
the sciences, is a logical and not an epistemological concept. Epistemology can
sui generis not explain the relationship between numbers, points, or lines, nor
their role in mathematics and the sciences—and so it is with HR’s and their
role in the representation of the past.
24 Though with the crucial reservation that the “individual things” meant here are not the
inhabitants of our trustworthy universe of stars, houses, dogs and molecules (as is the
case with Aristotle’s things) but of the representationalist universe. This qualification is
not without its ironies. For it confronts us with the fact that the “individual things” of
representationalism are of a different character than these Kings of France, fathers of
Charles II, Walter Scott, and so on, that Russell had in mind. Here modern formal logic
as exemplified by Russell is still more Aristotelian than representationalist logic, which
does not unwittingly embrace an Aristotelian metaphysics claiming that the world con-
sists of individual “middle-sized dry” things having certain properties. See Ankersmit,
Meaning, 154.
122 ankersmit
Bibliography
F.R. Ankersmit, “History as the Science of the Individual,” Journal of the Philosophy of
History 7 (2013), 396–425.
F.R. Ankersmit, Meaning, Truth and Reference in Historical Representation (Ithaca and
London: Cornell University Press, 2012).
———, Narrative Logic: A Semantic Analysis of the Historian’s Language (Boston and
The Hague: Martinus Nijhoff, 1983).
Ernst Cassirer, Substanzbegriff und Funktionsbegriff: Untersuchungen über die Grund
fragen der Erkenntniskritik (Hamburg: F. Meiner, 2000).
Jean-Pierre Dupuy, The Mechanization of the Mind: On the Origins of Cognitive Science,
trans. M.B. DeBevoise (Princeton: Princeton University Press, 2000).
Dietrich Mahnke, Leibnizens Synthese von Universalmathematik und Individualmeta
physik (Halle: M. Niemayer, 1925).
Friedrich Meinecke, Die Entstehung des Historismus (Munich and Berlin: R. Oldenbourg,
1936).
Bertrand Russell, “On Denoting,” Mind 14 (1905), 479–493.
P.F. Strawson, “Singular Terms and Predicates,” in Philosophical Logic, ed. P.F. Strawson
(Oxford: Oxford University Press, 1967).
———, “On Referring,” Mind 59 (1950), 320–344.
chapter 7
On Logical Aliens
Alessandra Tanesini
1 Introduction
Frege wrote about the laws of logic that “they are the most general laws, which
prescribe universally the way in which one ought to think if one is to think at
all.”1 Hence, for Frege the laws of logic are the laws of thought. They are not
psychological laws which describe the mental processes that occur when one
is thinking. Instead, these laws stipulate how one ought to think. They are con-
stitutive of rationality, and consequently they are what makes thought possible
at all.2 Further, Frege claims that these laws have universal application. Every
thinking being is a being whose thought is governed by the same logical laws.
Frege’s view exemplifies a position which I call logical absolutism. This is
the view that there is only one correct logic whose application is universal.
This position is almost universally accepted, presumably because its denial,
logical pluralism, appears to be a non-starter. Logical pluralism, as I under-
stand it, is the view that there is more than one, mutually incompatible but
equally admissible, logic. Or to put the matter differently, there could be beings
who are capable of thought*, but whose thoughts* are governed by laws of
logic* which are incompatible with our own.3 In other words, logical pluralism
entails the possibility of logical aliens. In this paper I explain and defend
1 Gottlob Frege, The Basic Laws of Arithmetic: Exposition of the System, trans., ed., and intro.
Montgomery Furth (Berkeley and Los Angeles: University of California Press, 1964), 12.
2 The claim that for Frege the laws of logic are necessary conditions for the possibility of
thought has an obvious Kantian flavour. This Kantian aspect of Frege’s account of logic has
been discussed by James Conant, “The Search for Logically Alien Thought: Descartes, Kant,
Frege, and the Tractatus,” Philosophical Topics 20 (1991), 115–180; see especially 134–137. I shall
not address the issue as to whether the laws of logic are the sole necessary conditions for the
possibility of thought.
3 I use the asterisk to indicate that in so far as the laws of logic are constitutive of thought,
aliens cannot think. However, they could think*. That is to say, they could engage in a law-
governed activity which plays in their lives a role that is not dissimilar from the role that
thinking plays in ours. The same considerations apply mutatis mutandis to logic*, belief*,
judgement*, etc. Hereafter, I sometimes use “logic” to refer to logic proper but also logic* and
to thought proper but also thought*. What is meant should be clear from the context.

124 tanesini
logical pluralism and the possibility of aliens. In my view logical absolutism

is incompatible with a proper understanding of the normativity of logic given
the fact of human finitude. For this reason, I conclude one must accept both
pluralism and the possibility of alien thought*.
This chapter consists of four sections. In the second I explain what I mean
by logical pluralism and contrast my view with the kind of logical pluralism
proposed by J.C. Beall and Greg Restall.4 I also argue that their position can-
not do justice to the normativity of logic as the set of laws governing human
thought. In the third section, I discuss in some detail an argument offered
by Hilary Putnam against the view that all logical principles can be revised.
Putnam derives from this argument a further conclusion which is tantamount
to a denial of logical pluralism. I argue that Putnam’s argument that there
are at least some unrevisable laws of logic fails. I also show that even if the
argument were successful, it would not warrant the further conclusion that
wholesale logical disagreement is impossible. In the fourth section I consider
an argument which has been developed by James Conant. Conant extracts this
argument from some of the considerations offered by Frege against the psy-
chologistic logician. The argument is best seen as directed against all forms of
logical pluralism, rather than merely as an argument against psychologism in
logic. I show that, even if the argument might be successful against psycholo-
gism, it fails to refute logical pluralism. Since none of these arguments are in
the least convincing, I conclude that logical pluralism has not been refuted.
2 Logical Pluralism
The position occupied by the logical pluralist should not be confused with that
occupied by the deviant logician. Supporters of deviant logics typically are
logical absolutists. They believe that there is only one correct logic; or, at least,
only one correct logic for any given region of discourse.5 They argue, however,
that such a uniquely correct logic is one which is an alternative to classical
4 Beall and Restall have outlined and defended their position in a series of papers and a book.
See, especially, J.C. Beall and Greg Restall, “Logical Pluralism,” Australasian Journal of
Philosophy 78 (2000), 475–493, Greg Restall, “Carnap’s Tolerance, Meaning, and Logical
Pluralism,” The Journal of Philosophy 99 (2002), 426–443 and J.C. Beall and Greg Restall,
Logical Pluralism (Oxford and New York: Oxford University Press, 2006).
5 Some deviant logicians support a global reform of logic, others are in favour of a local reform.
For a good discussion of these issues, see Susan Haack, Deviant Logic, Fuzzy Logic: Beyond the
Formalism (Chicago and London: The University of Chicago Press, 1996), 42–46.
On Logical Aliens 125
logic. In a word, the deviant logician believes that classical logic is incorrect
and that it must be either globally or locally abandoned in favour of another
logic which is incompatible with it.
I have described logical pluralism as the view that there is more than one,
mutually incompatible but equally admissible, logic. In order to make the posi-
tion clear, I need to say a few things about what I mean by a logic being admis-
sible, and two logics being incompatible. I offer my account by contrasting it
with some features of the position which Beall and Restall have also labelled
“logical pluralism.” In what follows I argue that my account, unlike theirs, does
full justice to the normativity of logic as the domain of the laws of thought of
finite beings.
A logic is characterized by its laws.6 These are laws about logical conse-
quence.7 They tell us which propositions or thought contents follow logically
from other propositions or thought contents. Thus, I take the notion of logi-
cal consequence as the primitive in terms of which the notion of logical truth
is defined.8 There are at least a couple of reasons for this choice. Firstly, as
John Etchemendy remarks, logical consequence is the primitive logical notion
because, “[l]ogic is not the study of a body of trivial truths; it is the study of the
relation that makes deductive reasoning possible.”9 Secondly, Frege’s charac-
terization of the laws of logic as the laws of thought encourages this approach.10
Logic, therefore, is normative. For instance, if Modus Ponens is a logical law
so that Q follows from P and If P then Q, then one goes wrong if one fails to
accept Q whenever one accepts both P and If P then Q. In my view, a view
that I cannot fully defend here, these laws are constitutive of thought. In other
words, nothing can violate these laws and still count as fully fledged thought.
Thus, the laws of logic are akin to the laws of physics, which cannot be violated.
Hence, if I am right, the individual who fails to accept Q, whilst accepting both
P and If P then Q (assuming that Modus Ponens is a law of logic) does not
6 In this paper I am exclusively concerned with deductive logic. Thus, “logic” hereafter
should be read as shorthand for “deductive logic.” Since deductive reasoning does not
constitute the whole of thinking, there will further laws of thought besides those dis-
cussed here.
7 Beall and Restall, Logical Pluralism, 35.
8 For instance, one can define logical truth as follows: a logical truth is a logical conse-
quence of any sets of propositions whatsoever.
9 John Etchemendy, The Concept of Logical Consequence (Stanford: CSLI Publications,
1999), 11.
10 Admittedly Frege wavered between two different conceptions of logic. The one which I
have described in this paper, and another according to which logic is the science of the
most general truths.
126 tanesini
reason in an illogical way. Rather such an individual fails to reason because

there is no such thing as an illogical thought. I shall return to the constitutive
nature of logical laws and the consequent impossibility of illogical thought in
the final section of this paper where I argue that the impossibility of illogical
thought does not, contra Frege and Conant, entail the impossibility of logical
aliens. In this section I aim to provide a fuller characterisation of logic plural-
ism under the assumption that the laws of logic are constitutive of thought.11
I agree with Beall and Restall that the intuitive pre-theoretic notion of
logical consequence as the relation that holds between the premises and the
conclusion of a deductively valid argument provides a useful starting point
in the study of logic. My motivation for taking the intuitive notion as a start-
ing point is that logic is an a priori discipline and we have no real alternative
when we wish to study it. Beall’s and Restall’s motivation seems somewhat
different. They think of logic as the study of the ordinary concept of logical
consequence.12 They aim to reveal that this concept is not precise but vague
or unsettled. Setting aside the possibility that the vagueness might reside in
the conceptual system alone rather than in reality, they derive pluralism about
logical consequence relations from the unsettledness of the ordinary concept.13
I presume, however, that at bottom our concern is the same. We are all inter-
ested in the logical norms that govern thought. Beall and Restall adopt some-
thing like a Ramsey-Carnap-Lewis approach to defining theoretical terms with
regard to logical consequence. They flesh out what is implicit in the ordinary
conception of logical consequence and aim to show that more than one pre-
cise notion equally satisfies the ordinary conception. My approach is different
since I believe that our ordinary folk theory of logical consequence as mani-
fested in the ordinary concept could turn out to be wrong about the logical
laws that actually govern our thought. However, as I said above, we have no
alternative but to start from the ordinary concept of logical consequence.
Logical consequence is what logicians attempt to capture in their formal
deductive systems, they try to prove results about it, and even to mechanize it.
Hence, as Etchemendy remarks, “the question of whether a particular deduc-
tive system for a particular language is sound and complete is always a sensible,
and indeed important, one to ask.”14 Answers to this question are not merely of
technical interest because the proofs of soundness and completeness give us
11 Beall and Restall do not think about the normativity of logic in this way. Rather, they think
that the norms of logic are regulative of thought and can be trumped by other norms.
Beall and Restall, Logical Pluralism, 17.
12 Beall and Restall, “Logical Pluralism,” 491 and Logical Pluralism, ch. 2.
13 Beall and Restall, Logical Pluralism, 27–8.
14 Etchemendy, The Concept of Logical Consequence, p. 157.
results about the relation between the formal notions of syntactic and seman-
tic consequence. Whilst the formal notion of semantic consequence, employed
in these proofs, is not guaranteed to coincide with the pre-theoretical notion of
logical consequence, nevertheless the fact that it makes sense to ask such ques-
tions shows that there is a relation—namely, logical consequence—which is
not generated by the formal deductive system and which the system tries to
capture.
An admissible logic is a deductive formal system which accurately reflects
at least an aspect of the semantic relation of logical consequence. The debate
between classical and deviant logicians is over which deductive system is
admissible, both camps presuppose that only one such system will fit the bill.
Logical pluralists deny this presupposition.
“Logical Pluralism,” as I said above is a label adopted by Beall and Restall to
characterize their view. In what follows I develop further the contrast between
the view I wish to defend here and their position. Beall and Restall propose
the following schematic characterization of validity which they take to capture
our pre-theoretical concept of logical consequence:
Generalised Tarski Thesis (GTT): An argument is validx if and only if, in

every casex, in which the premises are true, so is the conclusion.15
Since logical consequence is the relation that holds between the premises and
conclusion of a valid argument, GTT identifies logical consequence with truth-
preservation in every case. However, as the presence of the subscript “x” in GTT
indicates, this characterization needs to be complemented with a specifica-
tion of the cases. Beall and Restall also point out that there are many different
specifications of such cases, because they reflect some aspect of the ordinary
concept of logical consequence This latter point, they claim, is at the heart of
their conception of logical pluralism.16
Beall’s and Restall’s position can be summarized as follows:
(1) Logical consequence is the relation that holds between the premises and
the conclusion of a deductively valid argument.
(2) A deductively valid argument is one that preserves truth in all cases.
(3) There are at least two equally admissible, but non-equivalent, specifica-
tions of all the cases.17

16 Beall and Restall, “Logical Pluralism,” 478.
128 tanesini
Beall and Restall introduce the notion of an admissible specification of the

cases to refer to a specification of the cases which exhibits “the features required
by the (settled) notion of logical consequence.”18 These settled features are:
necessity; normativity and formality. That is to say, the pre-theoretical notion
of logical consequence has some settled or core components. One such com-
ponent is that the relation is necessary as it applies to all cases; another is that
logical consequence is a normative relation because to endorse a claim is to
commit to what logically follows from it. The final settled component is that
logical consequence is a formal relation in at least two senses: (a) it is indif-
ferent to the identities of the objects referred to in the arguments and (b) it
abstracts from the semantic contents of the thoughts or propositions that form
part of the argument.19
Beall and Restall defend claim (3) in the argument above by way of example.
They point out, for instance, that both Model Theoretic and Possible World
Semantics offer plausible specifications of all the cases, specifications which
can be plugged into the pre-theoretical definition of logical consequence. Yet,
they argue, these specifications lead to different classifications of which argu-
ments are valid. For example, consider the arguments:
This object is red all over. Therefore, this object is not white all over.
This object is red. Therefore, this object is colored.
These arguments are valid, if we adopt the specification of all the cases in
terms of possible worlds, since in every world in which the premises are true,
the conclusions are also true. These same arguments are, however, classified as
not valid if we adopt a model-theoretic semantics. The first argument has true
premises and false conclusion whenever the element in the domain which is
assigned to the name “this object” is a member of the set assigned to “red”
but also a member of the set assigned to “white.” Intuitively, this is possible
because “white” could be interpreted to mean “square.”20
This example, Beall and Restall claim, shows that there are at least two
accounts of logical consequence and validity such that (1) they are not equiv-
alent (they classify different arguments as valid) and (2) they have an equal

19 Beall and Restall, Logical Pluralism, ch. 2.
20 For a discussion of cases understood respectively as possible worlds or Tarskian models
see Beall and Restall, Logical Pluralism, ch. 4.
claim to being a way of fleshing out the pre-theoretical notion because they
are both admissible.
Beall and Restall do not think of these two different specifications of all
the cases as genuine rivals. More surprising is their view that the same can be
said of the relation between, for instance, relevance and classical logic. These,
also, are not rivals. Rather, Beall and Restall think of relevance and classical
logic as different ways of making precise the vague or unsettled pre-theoretical
notion of logical consequence. Different precisifications are all admissible if
they share the settled parts of the role played by logical consequence. There is
no point in asking which of these precisifications is correct as there is no fact of
the matter which could settle this further issue.21 In so far as, for example, both
possible world semantics and model theoretic semantics, in their different
ways, offer precisifications of logical consequence as a necessary, normative
and formal relation, Beall and Restall are happy to accept both as admissible.
Although no further questions about correctness are in their view legitimate,
it makes perfect sense to ask questions about the utility of any precisification
for a given purpose. Thus, rivalry among logics only emerges at the level of
applications.22
Beall and Restall are committed to the view that there is no fact of the mat-
ter as to whether an argument is valid full-stop, since this notion of validity
is vague or unsettled. The only questions that have definite answers concern
whether arguments are valid in a specific precisification of the notion. Thus,
one can say of disjunctive syllogism that it is classically valid but not relevantly
valid. The further question as to whether it is really valid makes no sense
because there is nothing which would corroborate a negative or affirmative
answer.23
Thus formulated their position faces at least two serious objections. First,
it develops a variant of logical pluralism which is somewhat trivial insofar as
their position is compatible with Fregean absolutism about logic. Second, it
cannot do justice to the normativity of logic understood as the collection of
laws governing thought.
21 The view that the pre-theoretical notion of logical consequence is vague is developed in
Beall and Restall, Logical Pluralism, chs 2 and 3. Their earlier view was that the notion was
ambiguous and that each logic was a disambiguation of it. Cf., “Logical Pluralism,” 484.
23 Restall, “Carnap’s Tolerance,” 426.
130 tanesini
In order to see why their account makes logical pluralism seem trivially true,
we need simply to consider that syllogisms are not truth-functionally valid.24
Thus, by their lights propositional logic and first order monadic predicate logic
constitute different but equally admissible logics. To the objection that these
considerations make the view trivial, they respond that not all instances of
pluralism are as uninteresting as that. Further, they add, triviality is no objec-
tion to the truth of a position.25 This response in my view misses the point of
the objection. Critics are not objecting that the view is true in an uninterest-
ing way. Rather, they are indicating that any logical pluralism worth its salt
should be defending a different view, one according to which classical proposi-
tional logic and classical first order monadic predicate logic count as one logic
but which nevertheless admits a plurality of logics. In other words, pluralism
about logics should involve some incompatibility or rivalry among the differ-
ent logics. But as I discuss below, Beall and Restall explicitly deny that this is
the case in their view.
My main objection to their view, however, is that it fails to account for
the normativity of logic as the collections of law constitutive of thought. In
response Beall and Restall would simply deny that logic is normative in this
sense. Instead, they claim that we often have inconsistent collections of beliefs
or commitments. Further, relying on the so-called paradox of the preface
(where one expresses commitment both to every claim in the book and to the
claim that there will inevitably be errors in the book), they conclude that logi-
cal norms can be trumped by other norms.26
I find both claims to be unconvincing if they are intended to refute the view
that logic is constitutive of thought so that illogical thought is impossible. The
paradox of the preface concerns warranted assertibility rather than logical
consequence. It is perfectly rational to be less warranted in asserting the con-
junction of a collection of claims than in asserting each of the conjuncts.27 The
claim that we all hold logically inconsistent commitments is not as obviously
true as Beall and Restall appear to assume it is. It is true that we are all prepared
to voice sentences which, if interpreted in accordance with their common uses,
would express inconsistent thoughts. But it does not follow from this consider-
ation that inconsistent thoughts are genuinely thinkable. Rather, what follows
24 This objection has been formulated by G.C. Goddu in “What Exactly is Logical Pluralism?”,
Australasian Journal of Philosophy 80 (2002), 218–30, see 221.
27 Contra Beall and Restall, I do not see that the point must concern degree of belief (rather
than warrant in assertion), Logical Pluralism, 17–18.
is that one might under the impression that one is thinking thoughts such
that if one did think them, one would think an inconsistency. It is beyond the
scope of this chapter to develop this view. To see how an argument for it can
be developed one would need to explore the connections between possibility
and conceivability. Unless the latter is understood psychologically, only what
is possible can be conceived. Consequently, one cannot genuinely conceive or
think what is not possible. Therefore, one cannot think inconsistent thoughts.
These considerations, or something like them, are at the root of the view that
can found in Wittgenstein’s Tractatus Logico Philosophicus, according to which
logical contradictions do not express genuine propositions.28
Whilst a full defence of a constitutivist account of the normativity of logic is
beyond the scope of this chapter,29 in what follows I want to show that it could
do justice to the intuition that if there is a plurality of logics, then these must in
some sense be incompatible. I begin by showing that Beall and Restall cannot
make sense of the thought that different logics must in some important sense
be rivals. Beall and Restall claim that “pluralism is not a recipe for wholesale
agreement” among logics.30 They think there is scope for disagreement among
logical systems only with regard to how useful they turn out to be when applied
to mathematics or when developing a semantics for a given area of discourse.31
Elsewhere, Beall and Restall present the case by way of an analogy with
Euclidean and non-Euclidean geometries. They point out that, once applied,
these geometries can be seen as competing models for the physical space of
our region. Similarly, they claim, “once applied, there is scope for genuine dis-
agreement between logical systems. However, this disagreement comes about
simply by applying the logic to model the validity of the real argument.”32
This analogy is revealing. A geometry is in itself simply an abstract structure;
we can think of it as a model for space only once we have developed an inter-
pretation for it. The model is then subjected to tests to see how well it models
empirical reality. Thus three elements are in play here. There is: physical space,
28 Ludwig Wittgenstein, Tractatus Logico-Philosophicus, trans. D.F. Pears & B.F. McGuinness
(London: Routledge and Kegan Paul, 1974), §4.461.
29 An inferentialist account of semantic content would be an essential part of this posi-
tion. Offering such an account is beyond the scope of this chapter. For an extremely
sophisticated version of such a view see Robert Brandom, Making It Explicit: Reasoning,
Representing, and Discursive Commitment (Cambridge: Harvard University Press, 1994).
31 Thus, for instance, they remark that relevant logic might be better suited than clas-
sical logic to providing a semantics for fictional discourse. Beall and Restall, Logical
Pluralism, 57.
132 tanesini
an uninterpreted abstract structure and an interpretation that assigns to ele-

ments of the structure a meaning in physical terms. When geometries are con-
ceived solely as abstract structures, different geometries are simply different
from each other. It would make no sense to think of them as incompatible or
rivals. Things, including abstract entities, can only be either the same or differ-
ent. They cannot be rivals. Rivalry or incompatibility only exists among items
which have correctness conditions.
Beall and Restall think of logic in ways that are analogous to the applica-
tion of geometry to space. In their view there exists a plurality of formal sys-
tems governed by syntactical rules. These systems are abstract structures and
as such they cannot be rivals. Each of these systems includes some syntactical
rules which are plausibly interpreted as playing a role which reflects some set-
tled components of the role played by logical consequence given the ordinary
concept of the term. This is what warrants thinking of these formal systems
as logics. Once so interpreted each of these systems can be used to model the
validity of real arguments. Some of these models will be better than others
because they best capture our intuitions about which arguments ought to turn
out as valid and which ought not to. It is only in this sense that logics can be
each other rival.
This picture in my view misconstrues the relation between logic and
thought. To see this consider that, generally speaking, what is modelled in
a model exists independently of the model itself. Hence, one may say that
physical space and abstract structures exist independently of each other. In
particular, physical space exists independently of interpreting some abstract
structure or other as a model for it. The same however cannot be said of logic
and thought. Clearly if the norms of logic are constitutive of thought, thought
does not exist independently of them. But even if logical norms are just regu-
lative, so that good thought is thought that obeys logic (whilst bad thought
violates logical norms), the role of logic is to guide reasoning and not to model
it. Either way the picture of logic as a model reflecting relations that pre-exists
it is based on a false picture of normativity. Norms do not describe (as models
do), they prescribe.33
I have claimed that logical laws are constitutive of thought and that Beall
and Restall fail to do justice to the normativity of logic. It might seem to follow
that I am committed to logical absolutism. If logic is constitutive of thought, so
that thought is impossible without logic, then there can only be one logic or at
33 For an incisive critique of these false conceptions of normativity see, Mark Norris Lance
and John Hawthorne, The Grammar of Meaning: Normativity and Semantic Discourse
(Cambridge: Cambridge University Press, 1997).
least only one for any given domain of discourse. This conclusion is premature
as it neglects the finitude of those (human) beings who are capable of thought.
But in order to make this point I need first to say something about our concept
of logical consequence, and how we can make sense of the idea that there
might be incompatible laws about what logically follows from what.
In my view the pre-theoretic notion of logical consequence is a “family
resemblance” concept, rather than a concept which is vague or unsettled. Two
related aspects of family resemblance concepts are important in this context.
First, there need not be any property which is shared by all the instances of
a family resemblance concept. Second, a distinctive feature of family resem-
blance concepts is that we might disagree as to whether something is an
instance of it, and both be correct. Further, we do not need to say that in this
instance we would be using distinct concepts. Thus, our apparent disagree-
ment is not explained away as mere equivocation.34
For example, two people might disagree as to whether an activity is a sport.
Consider, for instance, the case of ballroom dancing. Some might argue that it
is a sport, and as such deserves to be included in the Olympic Games. In order
to make their case, they might point to the similarities with gymnastics which
is a well-established Olympic discipline. Others might deny that ballroom
dancing is a sport, and point to the similarities with other forms of dancing.
They might claim that since ballet is not a sport, although its practice requires
that one possess highly developed athletic skills, ballroom dancing is best seen
as an art form. In this case, both parties are making incompatible recommen-
dations on how to extend the concept of a sport. Thus, their disagreement is
genuine; it is not an equivocation. Further, since each party’s recommendation
is permitted by the current use, and neither is prescribed by it, there is a sense
in which both parties are right: they are both within their rights to make the
recommendations they do.
I think of some disagreements among supporters of different logics along
similar lines. They would be making incompatible prescriptions about how
one ought to think in novel cases which are not determined by pre-existing
laws.35 But not all disagreements about logic need to be thought of as incom-
patible extensions of a pre-existing common system of logical laws. There
might also be cases of beings which are alien to one another from the start.
34 Michael Lynch has defended metaphysical pluralism by arguing that truth is a family
resemblance concept. The logical pluralist takes this position one step further. Lynch’s
discussion of family-resemblance concepts can be found in Truth in Context: An Essay on
Pluralism and Objectivity (Cambridge, Mass.: The MIT Press, 1998), 62–66.
35 This idea was suggested to me by Alan Weir.
134 tanesini
In these cases each population would be entertaining some propositions that

are beyond the grasp of the other group, and reason about them. In these cases,
their thinking may be governed by laws which are incompatible with the laws
of the other group.
Both possibilities are secured by the finitude of both parties. In the first kind
of case, a group might discover that its current logical laws do not supply an
answer in all cases; this group might split into sub-groups that make different
recommendations about their extension. This case is possible because it is a
mark of finitude that it involves thinkers who have not thought all the thoughts
which are possible for them to think. What makes the second kind of case pos-
sible is the existence of propositions that human beings are not capable of
entertaining either because they are too complex or because they include con-
cepts (including formal ones) we cannot grasp. These might be accessible by
other (alien) beings. It is important that among the concepts* that we cannot
grasp are included logical* constants, since otherwise these logics* would fail
to be formal in the sense of being independent of the semantic contents of
thoughts*. Nothing (logically) rules out this possibility since our best accounts
of what counts as a logical constant are either given by enumeration or are
derived from a primitive notion of logical consequence.
Incidentally, and this is a point to which I return in section 3 below, the
first kind of disagreement illustrates that logical laws might be subject to evo-
lution or rational revision. This occurs when current laws do not fully deter-
mine whether a given argument is valid because the argument introduces a
new vocabulary in the form of novel logical constants. Thus, revisions of the
laws of thought are possible. It is important to keep cases such as this one
distinct from the kind of rational revision which is the topic of Putnam’s argu-
ments (discussed in section 3). Although Putnam appears to suggest that he is
concerned with whether logical laws or principles are rationally revisable, his
arguments lend themselves to be interpreted differently. What often seems to
be actually at stake in Putnam’s arguments is whether our theories about what
the laws of logic might be are revisable. When so understood, it is apparent
that the study of logic even though conducted a priori is no more immune
from error than any other human attempt to gain knowledge.
The way of looking at disagreement over logic that I have proposed here has
definite advantages over Beall’s and Restall’s. First, as I have argued above, it
helps to make plausible the idea that genuine logical disagreement might exist.
Second, it also helps to make sense of the idea that there are genuine logi-
cal laws that prescribe thought and which different formal systems attempt to
capture. Logical consequence is not a vague notion, which needs to be substi-
tuted by several different precise ones. But it is one whose range of application
might not be determined in every instance. Thus, unlike Beall and Restall I
am not committed to the counter-intuitive claim that it makes no sense to ask
whether, for example, disjunctive syllogism which is classical valid but not rel-
evantly valid is really valid. Given my account the question makes sense, and
it might encourage us to explore novel extensions of our reasoning practices
so as to give a definite answer to this question. In order to achieve this, we can
advance a whole host of considerations in debate.
Let me now turn to the issue of triviality. I have claimed that given Beall’s
and Restall’s characterization, even classic propositional and classic first-order
logic turn out to deploy different notions of logical consequence. I have also
claimed that this fact trivializes the position. After all Frege would not have
been concerned by it. Classical first-order predicate logic is an extension of
classical propositional logic.36 In order to obtain a genuine pluralist position
one must argue that there are incompatible, yet equally admissible, kinds of
logical consequence.
The notion of incompatibility I have in mind is more easily expressed in
terms of logical truths, but a formulation in terms of logical consequence can
be derived from it.
There are two ways in which two logics L1 and L2 can be incompatible, and
in order to characterize them both, I need to introduce the notion of a logic
being an extension of another:
A logic L1 is an extension of a logic L2 if and only if the class of the gram-

matical sentences of L1 properly includes the class of the grammatical
sentences of L2, and the class of the logical truths of L1 properly includes
the class of the logical truths of L2, and the additional logical truths of L1
all contain essentially occurrences of L1’s additional vocabulary.37
I am now in a position to characterize a strong and a weak sense in which two

logics are incompatible:
Strong Incompatibility between two logics L1 and L2: There is a sentence

S, such that S is a logical truth in L1 (L2), and the contradictory of S is a
logical truth in L2 (L1).
Weak Incompatibility between two logics L1 and L2: L1 is not an exten-
sion of L2, and L2 is not an extension of L1, and there is no logic L3 such
that L3 is an extension of both L1 and L2.
36 The notion of a logic being an extension of another is defined below.

37 For this definition, see Haack, Deviant Logic, 4.
136 tanesini
The relation of strong incompatibility captures a sense in which two logics

can be straightforward rivals. Supporters of the two logics will contradict one
another, and the contradiction cannot be explained away by invoking meaning
variance. The relation of weak incompatibility is trickier. There need not be
anything which the supporter of one logic asserts, but which the other is pre-
pared to deny. A change of logic in this instance involves changes in concepts
or meaning. Meaning or concept variance, however, is not a sufficient condi-
tion for lack of conflict. What the notion of weak incompatibility is meant to
capture is the possibility that two logics might be such that there is a wide-
spread failure of intertranslatability.38
This weakened notion of incompatibility is central to the case for logical
pluralism, because it shows that we do not need a common metalanguage
within which we can talk about both logics and show that they are in conflict
in order to claim that two logics are incompatible. I return to this issue in the
third section where I discuss an argument, offered by Conant, against the pos-
sibility of logical aliens which smuggles in the presupposition that two logics
are incompatible only if it turns out that one treats as logical truth a sentence
whose contradictory is a logical truth in the other logic.
Having thus characterized logical pluralism, and shown why it is not a non-
starter, I now turn to consider arguments which can be taken to show why it
must be mistaken. I show that these arguments fail.
3 The Revisability of Logical Laws
In a series of papers about the status of logical and mathematical necessity

Hilary Putnam has offered a few arguments against the possibility of what
I have called logical aliens. For instance he claims:
The idea is that the laws of logic are so central to our thinking that they
define what a rational argument is. This may not show that we could
never change our minds about the laws of logic, i.e., that no causal pro-
cess could lead us to vocalize or believe different statements; but it does
show that we could not be brought to change our minds by a rational
argument. [. . .] And indeed, Aristotle remarks that if anyone pretends to
38 I do not address Quine’s arguments that in any case where a translation fails or interprets
the other culture as holding a logic different from our own, we should conclude that there
was a problem with the translation. These arguments have been refuted, at least to my
satisfaction, by Haack, Deviant Logic, 8–24.
disbelieve one of the laws of logic and undertakes to argue with us, we
can easily convince him that his own argument presupposes the very
laws of logic that he is objecting to.39
Like Frege, Putnam takes the laws of logic to be the laws of thought. He explains
this point by saying that logic is definitive of rationality. Since nothing could
count as a thought unless it stood in rational relation to other thoughts, it fol-
lows that the laws of rationality are the laws of thought. They are what make
thinking possible.40
Putnam holds that we can derive almost immediately two conclusions from
this conception of logic: first, at least some laws of logic cannot be revised, and
that no one could offer an argument against our laws of logic without employ-
ing the very laws of logic he is arguing against. I think that neither conclusion is
warranted. But before showing that this is the case, I need to make explicit why
Putnam’s argument is relevant to whether or not logical aliens are possible.
The second conclusion that Putnam claims to derive from the conception
of the laws of logic as laws of thought is almost tantamount to denying logical
pluralism. Putnam’s conclusion states that there could not be an individual
who explicitly disagrees with us about logic, and who does not implicitly rely
on that very logic. There must be only one logic. Putnam’s conclusion requires
that the individual in question possesses a conceptual apparatus that enables
him to talk about logic. This is indicative of a lack of clarity as to whether it is
the laws of thought themselves or our theories about them which are allegedly
immune from rational revision.
In my description of what logical aliens would have to be like, I have been
silent about the breadth of their expressive capacities. Prima facie, at least,
it seems possible to think that there could be creatures whose thoughts are
governed by the laws of a particular logic, but whose conceptual apparatus
does not equip them to think and talk about these laws. I shall not explore
this issue, and assume here, for the sake of argument, that logical aliens would
have to have logical concepts* required to talk about the laws governing their
thoughts*.
In a note to the article I am considering Putnam remarks that there are two
ways of challenging the laws of logic. First, somebody might assert what another
person denies. Second, a person might understand differently from another
39 Hilary Putnam, “There is at least one A Priori Truth,” in Realism and Reason, Philosophical
Papers Vol. 3 (Cambridge: Cambridge University Press, 1983), 98–114, see 109–110.
40 Thus Putnam takes the normativity of logic to be constitutive of thought, along the lines
I have endorsed in the first section of this paper.
138 tanesini
some of the notions that are crucially involved in spelling out these laws of
logic.41 Thus, Putnam distinguishes two different ways of disagreeing about
logic. Such a distinction might not be very sharp, since as Quine suggested, one
could interpret a disagreement about logical truths as a disagreement over the
correct meaning of logical connectives, for example.42 Nevertheless, there is an
intuitive difference which is captured by Putnam’s distinction. It is the differ-
ence between, say, asserting not (P or not P) and proposing an understanding
of negation and disjunction which leads one to assert that P or not P is not a
logical principle. I shall keep this distinction in mind when I consider whether
Putnam’s conclusion follows from his conception of the laws of logic as the
laws of thought.
The first conclusion that Putnam derives from his conception of the laws
of logic as the laws of thought or rationality is that, at least some logical prin-
ciples cannot be rationally revised. Putnam’s argument for this conclusion is
deeply unconvincing. It does not seem impossible that even though the laws
of logic are definitive of what constitutes a rational argument one could revise
some of them at a given time whilst relying on others which are kept fixed.
Subsequently, one might revise those laws which one previously kept fixed in
light of the new revised principles one has already acquired. In other words, as
Sellars claimed about the rationality of science as a self-correcting enterprise,
“any claim [can be put] in jeopardy, though not all at once.”43 This point is so
obvious that it is hard to believe that Putnam could have missed it.
In fact, Putnam’s argument essentially relies on another argument which
he has developed earlier in the same paper. This is an argument that attempts
to establish that there is at least one a priori truth. This a priori truth is the
minimal principle of contradiction: not every statement is both true and false.44
Putnam hopes to show that this principle is an a priori truth by showing that it
is unrevisable. Putnam’s argument for this conclusion is an instance of Modus
Ponens:
1. The minimal principle of contradiction (S) is unrevisable
41 Putnam, “There is at least one A Priori Truth,” 110.

42 W.V.O. Quine, “Carnap and Logical Truth,” in The Ways of Paradox and Other Essays,
Revised and Enlarged Edition (Cambridge, Mass.: Harvard University Press, 1976),
107–132, 112.
43 Wilfrid Sellars, Empiricism and the Philosophy of Mind, intro. Richard Rorty and a Study
Guide by Robert Brandom (Cambridge, Mass.: Harvard University Press, 1997), 79.
2. If S is unrevisable then S is an a priori truth

Hence, S is an a priori truth.
Putnam defends the truth of the first premise by claiming that we cannot
conceive of circumstances under which it would be rational to give up this
principle. In order to defend this claim, he considers what such circumstances
would look like. First, he considers whether it could ever be rational to assert
what the minimal principle of contradiction denies, namely that every state-
ment is both true and false. This, Putnam claims, is inconceivable because it is
tantamount to envisaging circumstances under which it would be rational to
assert every statement and its negation. But, as Putnam himself acknowledges,
this is not the only way in which this logical principle could be revised.
Instead, we must also consider whether we can conceive of circumstances
under which it would be rational to give up simultaneously both the minimal
principle of contradiction and the law of double negation. In such circum-
stances, one would hold that it is not the case that not every statement is both
true and false, without also holding that every statement is both true and false.
Putnam dismisses this case as describing a genuine possibility in a rather odd
fashion:
[. . .] in that case the statement ‘every statement is both true and false’
would still have the status of being a priori false, even if the statement of
which it was the negation isn’t a priori true. And to concede the existence
of such a status as a priori falsity is, I think, as much as to concede the
existence of such a status as a priori truth.45
Both claims are incorrect. To concede the existence of the status of a priori
falsity is not on a par with the concession that there are a priori truths, given
Putnam’s understanding of what the a priori is. Also, it is not correct to claim
that in the circumstances under consideration it is being conceded that the
statement “every statement is both true and false” is a priori false. I take up
these two issues in turn.
For Putnam, as for Quine, apriority is identified with unrevisability.46 Thus
for him, a statement is unrevisable if and only if it is a priori. It should be quite
obvious that this identification is designed for a priori truths, but that it leads to
absurdity in the case of falsity. It makes no sense to say that a priori falsehoods

140 tanesini
are those falsehoods which can never be revised. So, what is an a priori false-
hood? It cannot be defined as the negation of an a priori truth without making
unwarranted assumptions about bivalence. Perhaps, it might be a statement
which is never rational to start believing. If this is characterization is accepted,
there might be a priori falsehoods even though there are no a priori truths. In
other words, there might not be any statement which it would never be ratio-
nal to give up, although there are some statements which it is never rational
to start believing. Thus, if Putnam claims that accepting that some statements
are a priori false is essentially the same as admitting that some statements are
a priori truths, because the existence of a priori falsehoods entails the exis-
tence of a priori truths, he is wrong.
Putnam’s other claim in the passage quoted above is also mistaken. Under
the circumstances Putnam asks us to consider we are not entitled to infer that
the claim that “every statement is both true and false” is a priori false. In this
case the statement that “it is not the case that not every statement is both true
and false” is true; however, since the law of double negation does not hold, we
cannot infer the truth of the claim that “every statement is both true and false”
from the truth of its double negation. From this, it does not follow that under
these circumstances the claim that “every statement is both true and false” is
false. Rather it follows, that such a claim does not take the value true, which is
not to say that it takes the value false. In other words, Putnam holds that the
circumstances he is considering are:
not not (every statement is both true and false) is true

not (every statement is both true and false) is not true
(every statement is both true and false) is false
but, they could be equally described as follows:
not not (every statement is both true and false) is true

not (every statement is both true and false) is not true
(every statement is both true and false) is indeterminate
To see that this is a plausible interpretation of the circumstances envisaged

by Putnam, consider the following re-interpretation of this case. Suppose that
under those circumstances we assign the value indeterminate to any sentence
from which we withhold both assent and dissent. Also, suppose that we would
say of a sentence that it is false only if we dissent from it, and that it is true only
if we assent to it.
Further, we can distinguish two different types of negation: internal and

external. These two notions capture the difference between asserting the nega-
tion of a sentence and not asserting that sentence. The following principles
hold for these two notions of negation:
Not1 (internal): not1 A is true if A is false, false otherwise

Not2 (external): not2 is true if A is false, false if A is true
Suppose that we neither assent to nor deny that every statement is both true
or false (S). Then, we assign to that claim the value indeterminate. In such cir-
cumstances we are not2 prepared to assert it. Similarly, we are not2 prepared
to deny it, by asserting that it is not1 true. But, we are prepared to assert the
negation1 of the negation1 since that would be tantamount to holding the orig-
inal claim to be false, which we do not. In other words, the circumstances that
Putnam asks us to envisage could be explained thus:
not1 not1 (every statement is both true and false) is true

not1 (every statement is both true and false) is not2 true
(every statement is both true and false) is indeterminate
Putnam’s argument against the unrevisability of the minimal principle of

contradiction is that circumstances under which it could be rationally given
up are inconceivable, since under such circumstances one would hold every
sentence and the negation of every sentence to be true. Instead, I have shown
that there are different circumstances which would involve a revision of the
principle, and which do not entail such patent absurdity. Admittedly, it is hard
to understand what kind of situation would move us rationally to withhold
judgement about the claim that every statement is both true and false. But this
is not the point, many things which are possible are hard to imagine. Rather,
the burden of proof is on Putnam to show the patent absurdity of trying to
conceive circumstances under which the minimal principle of contradiction
would be rationally revised.
It might be objected that since my account involves a certain amount of
conceptual revision, it must be seen as involving a change of topic, rather than
a change of logic. That is, one might claim that in the example I have presented
the meanings of words like “not,” “true,” and “false” have changed, so that this
example does not show that the minimal principle of contradiction, as origi-
nally understood, has been revised. This objection is unconvincing. Under the
circumstances I have outlined we would have revised our analyses of truth,
falsity, and negation. And thus, we would have become convinced that the
142 tanesini
new analyses were superior to the old ones, and in light of these new analyses
we would abandon our old logical principles. In other words, a change in the
analyses of logical concepts is not sufficient to show that there is no conflict in
the logics adopted.47
I have argued that Putnam’s argument for the claim that at least some laws
of logic are unrevisable fails. I now turn to the second conclusion which he
draws from the argument. He claims that whoever disagrees with us on mat-
ter of logic must implicitly rely on the same logic that that person apparently
rejects. Of course, Putnam does not rule out the possibility of some disagree-
ment over logical matters. Putnam himself has supported the adoption of
three-valued logic in order to deal with quantum phenomena. What Putnam
thinks is impossible is wholesale disagreement.
If it is granted that Putnam’s argument against the revisability of all logical
laws fails, this second conclusion can also be immediately rejected. There is,
however, a further reason why Putnam’s argument for this second conclusion
fails. The unrevisability of at least some logical laws does not entail that whole-
sale logical disagreement is impossible.
Suppose that some laws of logic are unrevisable, so that under no circum-
stances would we rationally be prepared to give them up. In such a case, if
there were beings who disagree with us about those laws, nothing they could
say would convince us to give up those laws. There might be reasons for giving
up those laws, reasons which are grasped by those beings but which are forever
beyond our limited intellectual capacities, for instance. Thus, even if it were a
fact that there are some logical principles which we can never give up, it would
not follow that there could not be beings whose thought was governed by a
logic* which did not have those principles as laws.
It should not come as a surprise that unrevisability does not entail the
impossibility of logical pluralism. The question of revisability is an epistemic
question concerning the status of logical notions. The issues raised by logi-
cal pluralism are instead in response to questions concerning the metaphysi-
cal status of logical notions. More specifically, logical pluralism raises doubts
about the absolute necessity of logical laws, and these doubts are likely to be
independent of what we might think about our fallibility concerning which
laws are the correct ones.
47 For a good argument in defence of this claim, see Haack, Deviant Logic, 10.
4 Logical Aliens: A Final Defence
I now turn to an argument offered by Conant which is explicitly directed

against the possibility of logical aliens. This argument is presented as an expli-
cation of Frege’s attack against psychological interpretations of logic as offered
in the introduction to The Basic Laws of Arithmetic, where Frege writes:
[W]hat if beings were even found whose laws of thought flatly contra-
dicted ours and therefore frequently led to contrary results even in prac-
tice? The psychological logician could only acknowledge and say simply:
those laws hold for them, these laws for us. I should say: we have here a
hitherto unknown type of madness.48
As is well-known, psychologism in logic is the view that the laws of logic are
psychological laws. Logic, according to this view, is an empirical science.
Frege notes that the psychologistic logician is committed to the intelligibil-
ity of a scenario in which we encounter beings, who are capable of thought,
and who contradicted our laws of logic. Instead, Conant argues on Frege’s
behalf that the scenario is unintelligible. The argument consists of two parts:
in the first Conant appears to argue that there cannot be logical aliens, in the
second he concludes that we cannot understand the use of “cannot” in the
claim that there cannot be logical aliens. I consider these two parts in turn.
First, imagine that there are beings who do not accept a basic law of logic:
for instance, the principle of non-contradiction, not (p and not p). For a psy-
chologistic logician such beings could exist. They would be creatures whose
psychology is very different from ours. Conant disagrees. He offers a dilemma
to the psychologistic logician.
[. . .] either (1) [the psychologistic logician] can claim that his account
reveals that the judgments of the aliens conflict with ours, in which case
his idea of one judgment’s conflicting with another can be shown to tac-
itly rely upon the idea of their logical incompatibility [. . .],
or (2) he can refrain from telling us anything about the logical relation in
which their judgments stand to ours, in which case he can tell us nothing
about their thought whatsoever.49
48 Frege, The Basic Laws of Arithmetic, 14.

49 James Conant, “The Search for Logically Alien Thought,” 146–7.
144 tanesini
If the psychologistic logician chooses the first option, she concedes that
there is a non-psychological notion of incompatibility. Hence, she concedes
that psychologism is false. Also, Conant continues, since “we can only dis-
cern a disagreement between our beliefs and those of others against a shared
background,”50 and “it is the principles of logic [. . .] which make such discern-
ment possible,”51 the alien beings are not logical aliens at all. Their thought is
governed by the same logical principles as ours.
If the psychologistic logician opts for the second option, she also fails to
establish the existence of logical aliens, since she establishes at most that
[. . .] the noises we and the aliens make merely differ from one another,
[. . .], then they are no more in disagreement with one another than the
moos of two different cows.
Hence,
Rather than showing us that they think differently, [the psychologistic

logician] will be unable to show us that they are so much as capable of
thought.52
Hence, Conant concludes that psychologism in logic must be false since logical
aliens are impossible. It is worth noting, perhaps, that it is not clear why Conant
believes that the burden of proof lies with the psychologistic logician. In other
words, it is not clear why such a logician cannot rest content with choosing the
second option, and simply claim that it is not up to her to establish that there
are logical aliens. She might claim that the onus lies with Conant. He needs to
show that logical aliens are impossible, and his argument fails to do this.
It is also worth noting that the first horn of Conant’s dilemma is less than
convincing. Conant at some stages seems to rely on the Kantian idea that since
logic is constitutive of rationality it cannot be rationally disagreed with. But,
as I have already remarked in the context of my discussion of Putnam’s argu-
ment, this point merely shows that it is not possible to revise all the principles
of logic at once. It does not show that any logical principle is immune from
rational revision, and—consequently—rational disagreement.
There is, however, a way of generalizing and strengthening Conant’s
argument so that it is directed against all forms of logical pluralism, and its
50 Conant, “The Search,” 147.

c onclusion states that there cannot be logical aliens. The argument so modi-
fied would read thus:
There cannot be logical aliens because if there were such creatures, there
would be creatures whose logic is incompatible with our own. However, the
notion of incompatibility invoked here is a logical notion. Therefore, at least
some aliens’ statements stand in a logical relation of incompatibility with at
least some of our statements. But, if they stand in these relations, their state-
ments are part of the same logic as our own. Hence, such creatures are not
logical aliens, after all.
The argument, however, even when it is thus construed fails. It basically
assumes that if two logics are incompatible, there must be a metalanguage
within which it is possible to talk about both logics and show that there are
some statements which are logically true in one logic and such that their con-
tradictories are logically true in the other. Conant is aware that his argument
crucially depends on this assumption, since he writes that:
The first horn of the dilemma rests in part on the claim that it is one of
the criteria for whether someone affirms a judgment with which we dis-
agree that he means to deny what we assert.53
As I have already argued in this paper there is a weaker notion of incompatibil-

ity available to the defender of logical pluralism. Once this notion is adopted,
it stops Conant’s argument from going through.
Similar remarks can be made against Donald Davidson’s argument that no
intelligible sense can be made of the notion of incommensurable conceptual
schemes. In a nutshell, Davidson’s argument is that a total failure of intertrans-
latability is a necessary condition for the existence of incommensurable con-
ceptual schemes. Further, “it is essential to this idea that there be something
neutral and common that lies outside all schemes.”54 However, since no sense
can be made of this neutral something, it follows that no sense can be made of
the notion of incommensurable conceptual schemes. Conant’s argument basi-
cally has the same structure as Davidson’s; in both cases it is assumed that there
cannot be conflict (conceptual or logical) unless there is a common framework
against which the two systems can be seen to be in direct opposition. Hence,
if I am correct that we can make sense of a conflict without requiring a neutral
metalanguage, these arguments fail.

54 Donald Davidson, “On the Very Idea of a Conceptual Scheme,” in Inquiries into Truth and
Interpretation (Oxford: Clarendon Press, 1985), 183–198, 190.
146 tanesini
In the second part of his argument Conant claims that strictly speaking we
cannot be said to understand the conclusion of the first part of his argument. In
his view, we cannot make any sense of the “cannot” in the sentence “there can-
not be logical aliens.” This is the consequence of a view that Conant attributes
to Wittgenstein. It is also a view which has been recently endorsed by Putnam.55
For Conant logical truths do not have negations which we understand.
This second part of Conant’s argument seem to take the following form:
We seem to be in a position to assert: there cannot be logical aliens. But in
order to assert this sentence, we must understand what it says. The sentence
seems to say that the possibility of logical aliens must be rejected. However, in
order to understand this, we would have to understand that whose possibility
is rejected. That is to say, we would have to understand something which is
impossible. But, what is impossible is not genuinely conceivable. Hence, what
is impossible cannot be understood. Thus, no sense can be made of the sen-
tence that there cannot be logical aliens.56
Whilst I have much sympathy for these considerations, they are of no use
against the logical pluralist as they smuggle in the Wittgensteinian notion
of illogical thought. Conant merely presumes that to claim that there could
be logical aliens is the same as claiming that there is such a thing as illogical
thought. His argument can therefore be re-cast as follows: There cannot be
illogical thought. But, the thought that there cannot be an illogical thought is
itself an illogical thought. Hence, neither of the following can be coherently
stated: there can be illogical thought there cannot be illogical thought.
Thus, formulated the argument has some power. But, it derives its power
exclusively from the fact that the notion of illogical thought is nonsense, since
there is no thinking without logic. Hence, this argument does not succeed in
refuting logical pluralism unless it can be shown that logically alien thought
would have to be illogical thought. However, as this paper as shown, logical
alien thought* is not illogical. As a result, Conant’s second argument also fails.57
55 Hilary Putnam, “Rethinking Mathematical Necessity,” in Words and Life, ed. James Conant
(Cambridge, Mass.: Harvard University Press, 1995), 245–263, 256.
57 A much earlier version of this paper was presented to the philosophy departments at
Queen’s University Belfast and at the University of Nottingham. A more recent version
was delivered at the Selves and Logic conference held in Höör. I would like to thank the
audiences at these events for helpful comments and suggestions.
Bibliography
J.C. Beall and Greg Restall, Logical Pluralism (Oxford and New York: Oxford University
Press, 2006).
———, “Logical Pluralism,” Australasian Journal of Philosophy 78 (2000), 475–493.
Robert Brandom, Making It Explicit: Reasoning, Representing, and Discursive
Commitment (Cambridge: Harvard University Press, 1994).
James Conant, “The Search for Logically Alien Thought: Descartes, Kant, Frege, and the
Tractatus,” Philosophical Topics 20 (1991), 115–180.
Donald Davidson, “On the Very Idea of a Conceptual Scheme,” in Inquiries into Truth
and Interpretation (Oxford: Clarendon Press, 1985), 183–198.
John Etchemendy, The Concept of Logical Consequence (Stanford: CSLI Publications,
1999).
Gottlob Frege, The Basic Laws of Arithmetic: Exposition of the System, trans., ed., and
intro. Montgomery Furth (Berkeley and Los Angeles: University of California Press,
1964).
G.C. Goddu “What Exactly is Logical Pluralism?”, Australasian Journal of Philosophy 80
(2002), 218–30.
Susan Haack, Deviant Logic, Fuzzy Logic: Beyond the Formalism (Chicago and London:
The University of Chicago Press, 1996).
Mark Norris Lance and John Hawthorne, The Grammar of Meaning: Normativity and
Semantic Discourse (Cambridge: Cambridge University Press, 1997).
Michael Lynch, Truth in Context: An Essay on Pluralism and Objectivity (Cambridge,
Mass.: The MIT Press, 1998).
Hilary Putnam, “Rethinking Mathematical Necessity,” in Words and Life, ed. James
Conant (Cambridge, Mass.: Harvard University Press, 1995), 245–263.
———, “There is at least one A Priori Truth,” in Realism and Reason, Philosophical
Papers Vol. 3 (Cambridge: Cambridge University Press, 1983), 98–114.
W.V.O. Quine, “Carnap and Logical Truth,” in The Ways of Paradox and Other Essays,
Revised and Enlarged Edition (Cambridge, Mass.: Harvard University Press, 1976),
107–132.
Greg Restall, “Carnap’s Tolerance, Meaning, and Logical Pluralism,” The Journal of
Philosophy 99 (2002), 426–443.
Wilfrid Sellars, Empiricism and the Philosophy of Mind, intro. Richard Rorty and a Study
Guide by Robert Brandom (Cambridge, Mass.: Harvard University Press, 1997).
Ludwig Wittgenstein, Tractatus Logico-Philosophicus. Trans. D.F. Pears & B.F. McGuin-
ness (London: Routledge and Kegan Paul, 1974).
chapter 8
The Heart of Metaphysical Pluralism and the

Consistency Dilemma
A Critical Analysis of the Possibility of Incompatible Truths
Thord Svensson
1 Introduction
It is commonly assumed that only one of two incompatible facts can hold. For
instance, either it is a fact that God exists or it is a fact that God does not exist.
It is also commonly assumed that only one of two inconsistent propositions
can be true; for instance, either the proposition “God exists” is true or the prop-
osition “God does not exist” is true. In virtue of being incompatible, both prop-
ositions, it seems, cannot be true at once. However, some thinkers have argued
for the possibility of some exception to this commonly made assumption.1
In this chapter, I critically consider how Michael P. Lynch and Terry Horgan
together with Mark Timmons have sought to make sense of the possibility of
incompatible but true propositions, especially with regard to one key criticism
of it, sometimes and appropriately called the consistency dilemma.2 Although
I find each attempt to deal with the criticism innovative, I will argue that both
attempts face some serious problems.
I proceed as follows: In sections 2 and 3, I describe some basic features of
Lynch’s position, metaphysical pluralism, and his own preferred account of
concepts, which constitutes an essential part of his argument for metaphysi-
cal pluralism. In section 4, I turn more directly to the core idea of this kind of
1 See for instance Hilary Putnam, Representation and Reality (London: mit Press, 1989), 114 and
forward. For some who feel uncertain about the precise content of this idea, see for instance
Simon Blackburn, “Enchanting Views,” in Reading Putnam, ed. Peter Clark et al. (Cambridge:
Blackwell Publishers, 1994), 12–30, 16–18, and Peter Byrne, God and Realism, (Aldershot:
Ashgate Publishing, 2003), 38.
2 See Michael P. Lynch, Truth in Context—An Essay on Pluralism and Objectivity (London: mit
Press, 1998) and Terry Horgan and Mark Timmons, “Conceptual Relativity and Metaphysical
Realism,” in Philosophical Issues 12: Realism and Relativism, ed. Ernest Sosa et al. (Oxford:
Blackwell Publishers, 2002), 74–96.
© koninklijke brill nv, leiden, ��4 | doi ��.��63/9789004270183_�10

the heart of metaphysical pluralism 149
pluralism, the possibility of incompatible and true propositions. I argue that

Lynch’s analysis of this possibility is somewhat unsatisfactory in that it only
makes sense relative to a certain, somewhat questionable, account of concepts
and in what sense two propositions can be true yet incompatible. In section 5,
I consider and object to another proposal presented by Horgan and Timmons.
In section 6, I offer some concluding remarks which more explicitly connects
my criticism to the overall theme of this book on the matter of what does not
necessarily follow from certain warranted critiques of perfectionist and abso-
lutist accounts of logic.
2 Lynch’s Metaphysical Pluralism
In Truth and Context—An Essay on Pluralism and Objectivity, Michael P. Lynch

defends what he calls Metaphysical Pluralism. His advocacy of this position is
partly situated within the context of another aim, that of arguing for what he
calls Relativistic Kantianism.3 According to this latter position, it is possible to
coherently combine metaphysical pluralism with a realist theory of what it
is for a proposition to be true.4 In this chapter I will confine myself to deal-
ing with metaphysical pluralism and examine to what extent Lynch is able to
make sense of this position, especially with regard to a commonly made criti-
cism of it.5 Before considering this key objection we need to get acquainted
with the kind of pluralist position Lynch accepts and his main reason for doing
so. According to metaphysical pluralism, different and incompatible accounts
of the metaphysical nature of various features of the universe can be equally
correct. In explaining this key idea of metaphysical pluralism, Lynch writes the
following:
Metaphysical pluralism is pluralist because it implies that true proposi-

tions and facts are relative to conceptual schemes or worldviews; it is
metaphysical because the facts in question concern the nature of reality—
facts about God, mind, and the universe.6
3 See Lynch, Truth in Context, 3–4.

4 Lynch, Truth in Context, 3–4. In this respect, Relativistic Kantianism is different from for
instance Hilary Putnam’s position “internal realism” which contained an epistemic account
of what it is for a proposition to be true.
5 The objection is for instance part of a very common criticism of Putnam’s notion of “concep-
tual relativity.”
6 Lynch, Truth in Context, 3.
150 svensson
Lynch also explains metaphysical pluralism as
[. . .] the idea that there can be more than one true metaphysic, that there
can be a plurality of incompatible, but equally acceptable, conceptual
schemes. These conceptual schemes are ways of dividing reality into
objects and kinds of objects; they are ways of categorizing the world. The
pluralist intuition is that the world does not dictate to us which of these
ways of categorizing is the best, the most correct, or the way the world
really is in ‘itself.’ The pluralist denies that there are any absolute facts
about ultimate reality; the facts themselves reflect our conceptual point
of view.7
According to Lynch, metaphysical theories are not true in an absolute but only
in a relative sense, because they are only true relative to a certain conceptual
scheme, relative to how we conceptualize reality. Lynch traces part of this idea
back to Kant and his “insight that our knowledge of the world, and hence the
‘world as it is for us’ is in some sense constructed.”8 However, he observes that
Kant was no pluralist in that according to him, all humans shared one common
conceptual apparatus. This is not how Lynch sees it because according to him
our metaphysical theories can be true relative to more than one single scheme.
In working out this idea, he develops what he regards as a Wittgensteinian
account of conceptual schemes.9 According to this account, a scheme “is a
network of general and specific concepts used in the propositions we express
in language and thought.”10 Some of the concepts constitutive of a concep-
tual scheme are basic and foundational in virtue of being presupposed by less
foundational ones in the network.11 However, even what is considered basic
concepts can according to Lynch change and vary between different cultures
which according to him sets him apart from the Kantian account of conceptual
schemes. On an extended use of the notion of logic, taken as a foundational
system of concepts employed to make sense of reality, Lynch can be held to
accept logical pluralism; meaning that people’s basic accounts of what makes
sense may differ. In virtue of accepting metaphysical pluralism Lynch also
7 Lynch, Truth in Context, 10–11

8 See Lynch, Truth in Context, 11.
9 In chapter 2 in Truth and Context, Lynch brings out in detail the difference between his
account of conceptual schemes and the ones that he thinks are attributable to Kant and
Quine. Unfortunately, I do no have the space to go into this.
opposes the idea of there being eternal and absolute facts about the subject-
matter of logic apart from how such facts are conceptualized by certain people
at a certain time. This also entails that metaphysical investigations according
to him have the purpose of trying to describe but also develop the content of
such basic concepts rather than trying to reveal the absolute nature of some
mind-independent entity believed to be referred to by such concepts.12
A strong motivation for metaphysical pluralism, according to Lynch, is the
“apparent intractability of metaphysical disputes.”13 It seems that very few, if
any, of the questions pursued and explored within the context of metaphysics
have been given a solution that all or most metaphysicians agree on. In the
following quotation, Lynch exemplifies this circumstance with regard to the
context of mathematics and logic:
For a fourth type of example, we can turn to mathematics and logic.

Consider numbers. Are they to be identified with the all too perishable
scratches we make on the blackboard or the ideas we have when making
those marks or something altogether different, as Plato thought, existing
outside of space and time? Surely, it seems, they must be something. Or
take points on a Euclidean plane. Points can be taken to be basic abstract
particulars, sets of convergent spheres, composed of intersecting lines, or
logical constructions out of volumes. Every one of these answers would
seem to be conflict with the others [. . .] Similar debates can take place
over the ontological status of sets, functions, propositions, possible
worlds, and properties.14
In the face of our long history of disagreement and uncertainty about how to
respond to questions about the true metaphysical nature of various features of
our universe, the correct attitude to adopt is that of metaphysical pluralism. In
virtue of adopting this stance one should thus stop search for one metaphysi-
cally true analysis of justice, time or knowledge. However, as we have seen, the
reason for being a metaphysical pluralist according to Lynch is not an episte-
mological one. That is, Lynch does not believe that one metaphysical theory
about the nature of logical systems, properties, possible worlds and personal
identity is true, but due to our inability to find out which one it is, we have
to, at least for the time being, be modest and tolerate more than one theory.
According to him the motivation for metaphysical pluralism goes deeper than

13 Lynch, Truth in Context, 11, 16–17.
152 svensson
that and has to do with the very nature of metaphysical concepts and issues,
not just our ability to analyze them correctly. As Lynch claims:
[. . .] it is the nature of the disagreement that fosters the relativist intu-
ition: the metaphysical concepts themselves seem to be responsible for
the suspicion that there is no absolute way to resolve the dispute.15
The reason for metaphysical pluralism is thus directly connected to our meta-
physical, foundational, concepts and not to some inability on our behalf to
access some mind-independent and external domain believed to be target of
those concepts. In the next section, I’ll account for this part of Lynch’s argu-
ment for metaphysical pluralism.16
3 Clay and Crystal—Two Accounts of Concepts
Lynch’s case for metaphysical pluralism is very much based on a certain

account of concepts.17 According to him we should differentiate between two
accounts of concepts.18 According to one traditional account, a concept is a
distinct and transparent entity with a determinate application. Or differently
put, the extension of a concept is determinate and precise, either a certain
object belongs to the extension of it or it does not. According to Lynch, on this
account of concepts, they can be compared to crystals in being “rigid, pure,
and transparent, with sharp edges and definite borders.”19
Lynch contrasts this account of concepts with another one according to
which a concept may have a rather indeterminate and flexible use. On this
latter account, it is not established from the outset how a concept is to be used

16 In opposing metaphysical absolutism, the opposite position to metaphysical pluralism,
Lynch opposes both what he calls content-absolutism and fact-absolutism and argues
for fact- and content-relativism. He also thinks that the former entails the latter, that is,
you cannot be a content-absolutist without being a fact-absolutist. Content-relativism
means that the complete content of a proposition is relative a scheme, see Lynch, Truth
in Context, 15.
17 According to Lynch himself, he is working with a ‘minimal concept of a concept’ accord-
ing to which concepts are constituents of propositions, and also compose our thoughts.
According to the same view, to have a concept is to see something in a certain way, see
Lynch, Truth in Context, 56.
18 Lynch, Truth in Context, 56–57 and forward.
in each and every case; its use may rather vary and change without its ceasing
to be the same concept. On this account, a concept is more like clay that can
be shaped and reshaped into different figures without being destroyed; that is,
without ceasing to be one and the same bit of clay. As Lynch writes, in bringing
out the key idea in this account of concepts:
[. . .] Concepts are not ‘absolutely determinate’ closed circles, but elastic
and flexible. Concepts on this view are more like sculpting clay. Unlike
crystal, which breaks easily, you can stretch and pull a piece of clay in
radically different directions before it tears apart.20
On this account of concepts, one and the same concept may be used differ-
ently without ceasing to be that very concept.
In elaborating on this account of concepts, Lynch naturally once more
refers to Wittgenstein and especially his idea of a family resemblance concept.
In explaining what is so special about this kind of concept, Lynch describes a
common explanation of how people can be held to come to know many of the
concepts they master, not only family resemblance concepts. Usually people
are taught certain concepts in virtue of being presented with a set of paradig-
matic examples of to what kind of objects the concepts apply. Through this
exemplification they are expected to get a grip of to what the concepts apply
to; what kind of objects the concepts designate. Relative a certain a certain set
of such paradigmatic examples the application of a concept is rather deter-
minate. The concept of gold for instance can be held to apply to all objects
sharing the same underlying property as the one possessed by objects ini-
tially presented as paradigmatic examples of gold. However, regarding a fam-
ily resemblance concept this is not the case because such a concept can be
applied in accordance to more than one set of paradigmatic examples, accord-
ing to different paradigms so to speak, and no such set of paradigmatic objects
can be singled out as being more correct than the others.21
In exemplifying this Wittgensteinian oriented account of conceptual flex-
ibility, Lynch describes a fictive scenario involving two anthropologists in an
argument over if a certain activity on a remote island should be conceived of as
a religious activity or as a game. That is, can the concept of religion be applied
to what the people on the island are doing or is the concept of a game more
applicable? According to Lynch, both anthropologists can be correct in virtue
of using the concept of a game along different paradigms. The reason for this

154 svensson
is that the activity on the island is complex and rich in details and has some
features commonly associated with our concept of a game and some features
commonly associated with our concept of religion. The adults are for instance
divided into two groups. Each group starts to dig a hole and tries to make the
hole as deep as possible while at the same time trying to fill up the hole the
other group is digging. When some time has passed each hole is measured and
the group with the deepest hole celebrates and the other group has to bring
food to them. According to Lynch the practice also contains some “sacred ritu-
als . . .; for example, prior to and after the digging, each group prays to their
twin deities, who are generally depicted as bearing shovels and trowels. And
all disputes are settled by the high shaman.”22 According to Lynch, one anthro-
pologist may then argue that they are witnessing a game, pointing to the fact
“there are team [. . .] and offense and defence, and clear winning and losing”
while the other anthropologist may argue that they are rather observing a reli-
gious activity, pointing to the fact that the people in each group are taking
the activity very seriously, it involves sacred rituals and is controlled by the
shamans.23 In drawing upon different paradigms of what constitutes a game
and a religious activity both anthropologists can then according to Lynch be
equally correct. According to him it is also important to resist the inclination, if
it would arise, to think that the different uses of “game” by the anthropologists
should be seen as uses of different and distinct concepts of a game. In his eyes
it is rather the same concept of game used differently.
In explaining the flexible feature of certain concepts, Lynch also introduces
a distinction between minimal and robust concepts. A minimal concept is neu-
tral to many metaphysical theories while the latter kind of concept consists of
a more substantial development of the minimal one and due to this, is not so
neutral from a metaphysical perspective.24 Lynch also talks about the minimal
concept as what people often have in mind when not reflecting too much on
the concept’s content. Lynch writes:
A minimal concept of F is a concept whose ordinary use ‘floats free’ of

metaphysical questions (or most metaphysical questions) surrounding
Fs. It is a way of thinking about something that is neutral with regard to
issues about its ontological nature. In the example of the concept of
mind, our minimal concept of mind is whatever it is that thinks and has
experience. In contrast, what I will call a robust concept of F is a concept

23 Lynch, Truth in Context, 64–65.
whose ordinary use consists of a commitment to some particular onto-

logical view of Fs. Thus, a robust concept is a way of thinking about some-
thing that is not neutral with regard to questions about its ontological
status. The Cartesian concept of mind, with its commitment to a non-
physical substance, is a comparatively more robust concept.25
According to Lynch, we can thus extend and develop a minimal concept of

some subject-matter, like our concept of mind, into a more substantial and
detailed concept of it. Usually such an elaboration of a concept follows from
taking into account metaphysical issues and questions. Since the latter can be
done differently, we can have several robust concepts in connection to one and
the same minimal concept. Lynch thinks that his distinction between mini-
mal and robust concepts is most helpful in making sense of how people can
talk about the same subject-matter very differently and still be held to present
equally correct accounts of it.26
4 Metaphysical Pluralism and the Consistency Dilemma
According to Lynch’s preferred account of concepts, a concept can be applied

differently without ceasing to be one and the same concept. In this section, I
will describe how Lynch uses this account of concepts in explaining how two
incompatible propositions can be true. Lynch is familiar with the problems
that some associate with the possibility of incompatible truths and that many
find it hard to accept this idea without reducing it to something trivial. For
instance, with regard to Putnam’s idea of conceptual relativity, which Lynch
has much sympathy for, many appear to share Susan Haack’s opinion that the
idea of conceptual relativity seems, as she puts it, to be either a “momentous
tautology” or “manifestly false.”27 To appreciate what some people find puzzling
about Putnam’s idea, let us consider one fictive example of it made famous by
25 Ibid.
26 Lynch also points out that the distinction is a relative one in the sense than one concept
can be more robust or minimal than a different one, see Lynch, Truth in Context, 69.
27 See Susan Haack, “Reflections on Relativism: From Momentous Tautology to Seductive
Contradiction,” in Philosophical Perspectives 10: Metaphysics, ed. James E. Tomberlin.
(Oxford: Blackwell Publishers, 1996), 302. See also Byrne, God and Realism, 38 and John
Searle, The Construction of Social Reality (New York: Free Press, 1995), 155 and forward for
similar comments. This does not mean that no one apart from Putnam himself under-
stands it; to many others it seems more evident what it is about. This is also one of the
circumstances that make the discussion so fascinating.
156 svensson
him. Two people, Smith and Johnson, are counting objects in a bag with four
particulars. They present different accounts of how many objects the bag con-
tains because Smith thinks that mereological sums are objects, Johnson does
not. According to Putnam, Smith and Johnson can be held to present incom-
patible but equally correct accounts of the number of objects contained in
the bag. In making sense of this circumstance, Putnam seems to claim that
words like ‘object’ and ‘existence’ in the accounts may have different uses.28
With regard to this explanation more than one critic has wondered in what
sense the two accounts then can be thought to be incompatible? For them to
be incompatible, the key words within them, it seems, must be used with the
same meaning. If not, the accounts would become consistent. Lynch refers to
this worriment as the “consistency dilemma” for metaphysical pluralism and
thinks that it is important to deal with it. Recall, metaphysical pluralism states
that two statements can be incompatible and equally true and accurate. If we
were presented with an explanation of how this is possible which explicitly or
implicitly entailed that the statements were about different things or had dif-
ferent meanings, this explanation would be considered unsuccessful in that it
would seem to remove any hope for incompatibility between the statements.
Obviously, Lynch does not think that the idea of metaphysical pluralism
(or what Putnam calls “conceptual relativity”) should be considered a trivial
(or absurd) thesis.29 According to Lynch, to show that this is not the case, we
need to show how the following four statements can be true at once.
(1) Smith and Johnson are expressing distinct propositions. (2) Smith and
Johnson are expressing incompatible propositions. (3) Smith and Johnson
are expressing true propositions. (4) Smith and Johnson are not employing
28 Or as he puts it more generally “the same situation can be described in many different
ways, depending on how we use the words. The situation does not itself legislate how
words like ‘object,’ ‘entity’ and ‘exist’ must be used. What is wrong with the notion of
objects existing ‘independently’ of conceptual schemes is that there are no standards for
use of even the logical notions apart from conceptual choice.” See Putnam, Representation,
114. Putnam has also stated that: “A corollary of my conceptual relativity—and a contro-
versial one—is the doctrine that two statements which are incompatible at face value
can sometimes both be true and the incompatibility cannot be explained away by saying
that the statements have ‘different meaning’ in the schemes they respectively belong.” See
Hilary Putnam, Realism with a Human Face, (Cambridge, Mass.: Harvard University Press:
1990), x. It is not that clear how one should interpret this remark, and especially not how
it should be related to the previous ones quoted in this foot-note.
29 Of course, nor does Putnam.
completely different concepts of ‘object’ or ‘exist’ or ‘number’; they are

not talking past each one another.30
Lynch’s proposal on how to satisfy all four conditions simultaneously basically

builds on the distinction between robust and minimal concepts presented in
previous section and a contra-factual analysis of how a proposition may be
relative to more than one conceptual scheme. Let us go over this in detail, con-
sidering one condition at the time.
In one initial response to the worriment just considered, that Smith and
Johnson are using the terms ‘object’ and ‘existence’ with different meanings
(which would make us think that they are not talking about precisely the same
matter), Lynch declares that if Smith and Johnson are seen as participating in
a traditional metaphysical debate about the nature of an object, or how many
objects our universe contains, they will think that they are addressing and
responding to precisely the same question. In Lynch’s terminology they are
working with one and the same minimal concept of an object, as Lynch writes:
‘If Smith and Johnson are metaphysicians, then by their own lights they will
be sharing the same minimal and absolute concept of an object, not employ-
ing different concepts!31 Due to this, some critics of Putnam are according to
Lynch wrong to think that Smith and Johnson are using different concepts of
an object, at least if this is taken to mean that they are using completely differ-
ent concepts of an object. That is, they are not talking past each other.
Lynch also however wants us to accept that this minimal concept of an
object is a flexible one and, in virtue of this, can be extended and interpreted
differently. The same is also true of the metaphysical concept of exists. Lynch
writes:
For if existence is not absolute in that there is no common property

shared by all existents, then the door is open to taking ‘exist’ as similar to
‘game’ or ‘art’. To do so is to take our concept of existence as a flexible
concept that is learned initially by reference to certain paradigms and
then extended past those paradigms as experience requires.32
According to Lynch, Smith and Johnson should then be attributed one and
the same minimal concept of an object and exists but different robust con-
cepts of an object and exists. Moreover, in virtue of relying on different
robust concepts of for instance an object, the propositions put forward, partly

158 svensson
constituted of them, are distinct. That is, when Smith states that the bag con-
tains a certain number of objects and Johnson objects to this they are defending
and questioning different propositions. Each such proposition is also accord-
ing to Lynch true and accurate relative to the conceptual scheme it belongs to.
So far Lynch may be taken to have accounted coherently for the majority
of the four conditions which, according to him, is needed to take care of the
consistency dilemma. That is, he has made sense of the claim that Smith and
Johnson are expressing (1) different and (3) true propositions and that (4) they
are not employing completely different concepts of “object” or “exists.” Hence
it remains to account for how Johnson and Smith may be held to offer incom-
patible accounts. In explaining this last part of his solution to the dilemma,
Lynch writes:
In what sense of ‘incompatible’ are Smith’s and Johnson’s assertions

incompatible? According to the pluralist, they are (or could be) extend-
ing their shared minimal concept of an object differently. Thus the prop-
ositions they are expressing are relative to different conceptual schemes
and are therefore logically consistent. At the same time, there is a clear
and important sense in which the pair of propositions are incompatible;
if these propositions were relative to the same scheme, they would be incon-
sistent. This fact is necessarily true of that pair of propositions: in every
possible world where these propositions are relative to the same scheme,
only one is true. And it is in precisely this sense that Johnson and Smith
are rightly said to be expressing consistent but incompatible
propositions.33
This passage is supposed to add the last part of his solution to the consistency
dilemma and account for more precisely in what sense the two propositions
may be incompatible. I should concede that I am a bit puzzled by what he
states. According to Lynch, the propositions in question would become incom-
patible if they are considered relative to one and the same conceptual scheme.
This appears accurate, but then they would no longer both be true. That is, on
Lynch’s approach, the propositions are either true or incompatible, but not
both at the same time. But if not, I fail to appreciate how this can be consid-
ered an adequate and non-trivial response to the consistency dilemma. Recall
that Lynch sets out to demonstrate how two incompatible propositions can be
true. Regarding this I expected it to mean true and incompatible at the same
time. To repeat, if not, I cannot help feeling that his account fails to avoid the
dilemma.

Another problem with Lynch’s proposed solution to the consistency

dilemma has to do with his account of concepts. According to Lynch, a con-
cept can be used differently without ceasing to be one and the same concept.
Even though I have some sympathy with this account, I find it less certain that
we in the kind of examples he describes have different uses of the same con-
cept rather than distinct uses of different concepts (and even if the former
would be the case, I would still dispute its importance for the case he is try-
ing to make). I am not claiming that his account of concepts is mistaken but
rather that it seems somewhat arbitrary if we decide to say that we are using
the same concept differently or using different concepts. I am also not sure
what fact that would settle this kind of questions apart from context and our
interest; to draw upon Lynch’s own pluralist position, one may suggest that this
matter can be conceptualized differently. One may also claim that Lynch’s firm
and resolute position on this matter is not in line with the spirit of metaphysi-
cal pluralism. To some extent I think that his conception of concepts in this
respect is shaped all too much by his more general ambition to make sense of
how incompatible accounts of the metaphysical nature of some fundamental
part of reality can be true.34
Lynch does offer some support for his position that we in many contexts
have one concept used differently rather than different concepts, but I am sim-
ply not convinced by his arguments. Recall for instance Lynch’s example with
the two anthropologists in a dispute over whether a certain type of action on
an island should be seen as game or a religious activity. Lynch realizes that
one may feel inclined to propose that the anthropologists are simply having
distinct concepts of a game which would make the claims they make perfectly
compatible. As previously remarked, Lynch strongly objects to this analysis of
what is going on. In explaining why he writes for instance:
Some might wish to conclude that Smith and Johnson are simply equivo-
cating over the word ‘game’; thus, the Islanders’ activity is a game on
Smith’s concept of a game but not on Johnson’s. Yet this is surely incor-
rect, for Smith and Johnson presumably can understand each other very
well. It would be more true to our actual practice to say that we share one
very fluid concept of game rather than many very distinct concepts.35
34 See for instance Lynch, Truth in Context, 56–57 in which he explains that he is interested
in how concepts must be like to make sense of metaphysical pluralism.
160 svensson
I find this reply somewhat peculiar. The idea that the people cannot have two
different concepts because they can communicate successfully seems uncon-
vincing; it is not like we are considering the hypothesis of radical conceptual
relativism. Sometimes Lynch seems to mean that the different uses between
“game” and “game” by Smith and Johnson is not as big as that between “bank”
(as a sandbank) and “bank” (as financial institution or company). That is, Smith
and Johnson are not discussing completely different things as they would be
doing if they were talking about sandbanks and banks as financial institutions.36
Although Lynch may be correct in this, it does not, I think, demonstrate that
Smith’s and Johnson’s uses of “game” should be identified as different uses of
the same concept rather than uses of different and distinct concepts. Even very
similar concepts may still be distinct and different.
Moreover, even if one would accept that certain concepts are quite flexible
in the sense Lynch thinks (that is, that they can be applied differently while
still remaining the very same concepts), one may still question to what extent
this is helpful for his case for the possibility of incompatible but equally true
accounts of some subject-matter. Lynch thinks that: “Unlike words such as
‘bank,’ ‘rise,’ or ‘beat’ English speakers do not take ‘exist’ or ‘there is’ to com-
monly express more than one concept or meaning.”37 There may be some truth
in this. However, when people use concepts like “exists” we do usually take into
consideration what they mean more specifically by it. It is this contextual and
specific meaning that is important and if one person’s specific use of exists is
different compared to another person’s specific use of it, this may enough to
warrant the conclusion that they are not expressing and defending incompat-
ible accounts of something. The additional fact that each use of exists in some
sense can be held to be part of or connected to some more general concept
of exists does not according to me add anything relevant. Lynch, in contrast,
seems to think that it is precisely this circumstance that points to something
exciting and controversial: “It is this fact that disposes us to think that the
neuroscientist and the Cartesian (and the internalist and externalist in epis-
temology) are using the same concept and using different concepts. Rather
than attempting to ‘explain this away,’ we should trust our intuitions, for this
contradiction points to the flexible nature of our thought.”38 However, once we
contextualize the concepts or distinguish between the general concept and the
36 See for instance Lynch, Truth in Context, 68 for support of this reading of what Lynch is
getting at.
more specific one, or more specific uses of it, I fail to see any reason for talking
about a contradiction here.
In presenting another argument for why we should accept that we in many
contexts have one concept used differently rather than different concepts,
Lynch appeals to how the user of the concepts themselves may rule on the mat-
ter. In emphasizing the idea of a minimal and shared concept and responding
to Searle’s critical comment on Putnam’s position, Lynch writes for instance:
So Searle is simply mistaken if he is claiming that the ‘criterion’ for count-

ing objects has been deliberately set differently by Smith and Johnson. If
Smith and Johnson are metaphysicians, then by their own lights they will
be sharing the same minimal and absolute concept of an object, not
employing different concepts! And so it will be for most other metaphysi-
cal debates, including ones that we are inclined to think matter, e.g.,
debates over the existence of God, or minds, or the fundamental particles
of physics. In our little metaphysical debate between Johnson and Smith,
as well as in all other serious metaphysical debates, it seems wrong to say
that the participants in the debate are simply equivocating over terms39
Lynch’s response to Searle seems a bit unfair because Searle’s comment is only
based on what Putnam himself has stated and with regard to that, Searle’s
remark does no seem to be that misplaced.40 Speaking for himself Searle may
in fact agree with Lynch that people discussing the ultimate nature of objects,
God or mind take themselves to address the same subject-matter, but this was
not in question, what was in question was Putnam’s construal and analysis of
such disputes. And this observation may also apply to Lynch’s position. Searle
or anyone else may suggest that people debating such matters commonly
assume the possibility of being correct about it in some important and sub-
stantial sense, either the people in the debate may be assumed to target some
mind-independent and absolute fact or perhaps, which seems to make more
sense to me, they are trying to make explicit what is implicit in our use of cer-
tain concepts, which may vary between different cultures and people. But this
does not seem to be how Lynch sees it, because he thinks that the different
accounts presented by people in such a debate should be seen as robust devel-
opments of a shared minimal concept. And propositions expressed through
such robust concepts would not, I contend, be addressing precisely the same
matter, nor do they need to be in conflict. Taking this into account it does seem

40 See note 28 for an example of how Putnam has put it.
162 svensson
a bit inaccurate to defend the intelligibility and use of a minimal concept

in this context by drawing upon the fact that people discussing some foun-
dational question commonly take themselves to address the same subject-
matter. On many accounts of what is going on within metaphysical debates
this is true, but not on Lynch’s own account and analysis of them.
5 Horgan and Timmons on Conceptual Relativity and Relativized

Content
Terry Horgan and Mark Timmons are in the business of explaining conceptual
relativity. This is not easily done, they think, because the notion of conceptual
relativity is puzzling and in need of some elaboration.41 They do however share
Lynch’s optimism concerning the possibility and usefulness of presenting a
proper and non-trivial idea of conceptual relativity, even if they are not too
optimistic about Lynch’s own proposal, a matter I will come back to.
According to Horgan and Timmons, an accurate explanation of conceptual
relativity must show how we can apply the principles of affirmatory conflict
and mutual correctness to two propositions.42 To accomplish this is to make
clear how two propositions can be in conflict but true. Horgan’s and Timmons’
approach to this matter basically consists of the elimination of some assump-
tions about concepts and words that, according to them, are responsible for
our failure to appreciate what conceptual relativity is all about. One assump-
tion they wish to get rid of is what they refer to as the invariantist idea about
concepts and terms. About this idea they write:
The idea is that the semantic standards that govern concepts and words
cannot vary from one usage to another, insofar as the same concept of
41 As they put it: “We will argue that conceptual relativity is indeed a genuine phenomenon,
albeit an extremely puzzling one.” See Horgan and Timmons, “Conceptual Relativity,” 74.
42 Just like Lynch they present a set of conditions that a successful analysis of concep-
tual relativity must satisfy: “In order to make sense of conceptual relativity, one needs
to explain how the members of the following list of ideas can be mutually compatible:
1. Persons P1 and P2 are making conflicting claims. [Principle of affirmatory conflict] 2. So
they must be employing the same concept of object and using the term ‘object’ with the
same meaning. 3. But they are also making claims that are mutually correct. [Principle
of mutual correctness] 4. So their claims cannot be flatly inconsistent. The gist of the
puzzle is to explain how all four of these claims can be correct.” See Horgan and Timmons,
“Conceptual Relativity”, 78. Compare to Lynch, Truth in Context, 81–82.
word-meaning is employed in both usages: any difference in governing

semantic standards reflects a distinct concept or word-meaning.43
They wish to replace this idea about concepts and meanings with the variantist
conception, according to which:
[. . .] whatever exactly concepts and meanings are, they are subject to cer-
tain kinds of identity-preserving differences in correct usage. One and
the same concept can be used by two persons (or by the one person, at
different times) in ways that are governed by somewhat different seman-
tic standards, while still being the same concept.44
According to the variantist conception the same concept can be used differ-
ently without ceasing to be that very concept. It can be used in accordance
to different but identity-preserving “governing semantic standards.” With refer-
ence to Derrida, Horgan and Timmons label this kind of identity-preserving
context-sensitivity diffèrance.45 To exemplify it, they refer to a comment made
by David Lewis on Peter Unger’s position concerning the concept of flatness.
Drawing upon Lewis’ comment, they suggest that the concept of flatness can
be used with different precision without ceasing to be the very same concept.
In expressing this idea they write:
The standards of precision that govern a particular use constitute the

specific current setting of what may be called the precision parameter for
flatness. As the passage from Lewis makes clear, this parameter is contex-
tually variable: it can take on different specific settings in particular con-
texts. This contextual variability is semantically built into the single
concept flatness, and into the meaning of the term ‘flat.’ What you mean
when you use ‘flat’ in such a way that the sidewalk counts as flat is some-
what different from what Unger means when he uses ‘flat’ in such a way
that it doesn’t; [. . .] these differences in meaning, and in concept, are
identity-preserving differences. There is a diffèrance in meaning, and in
concept, between yourself and Unger.46
43 Horgan and Timmons, “Conceptual Relativity,” 77–78.

44 Horgan and Timmons, “Conceptual Relativity,” 79.
164 svensson
According to Horgan and Timmons, one and the same concept of flatness can
be applied equally correct on different surfaces once each such use is made
relative a certain setting of the precision parameter for flatness. Like Putnam
and Lynch, Horgan and Timmons also exemplify much of what they hold to
be true of conceptual relativity through a fictive scenario involving two people
counting objects, this time called “Carnap” and the “Polish logician.” Timmons
and Horgan say that they may present different but equally true accounts of
how many objects there is. The reason for this is that Carnap and the Polish
logician are using the notion of an “object” differently, although in an identity-
preserving manner.47
Timmons’ and Horgan’s account may come across as rather similar to
Lynch’s proposal, which they are familiar with. However, Timmons and Horgan
suspect that there is an important difference between his proposal and the one
they are offering because according to them, Lynch’s proposal may contain a
thought they call relativized content which according to them cannot be part
of a successful account of conceptual relativity. The reason for this is that it
violates the principle of affirmatory conflict, which they take to be essential for
conceptual relativity. In explaining what they mean they return to the example
with Carnap and the Polish logician, and claim that if one accepts the prin-
ciple of relativized content, one seems to think that when Carnap states that
there are four objects, he is implicitly conceding that this is just true relative
to how he counts them and the Polish logician is judged to make the same
concession with regard to how many objects he sees.48 According to Horgan
and Timmons, this is a problematic analysis it would make the two accounts
compatible, and for this reason they do not accept it.49 In expressing how they
regard the matter, they write:
On our view, the content of what Carnap is saying is properly expressed

this way: There are exactly three objects on the table, while the content of
what the Polish logician is saying is properly expressed this way: There are
exactly seven objects on the table. Both statements are categorical, and
thus neither of them is an implicit relativity claim. Although the respec-
tive statements are semantically governed by different settings of the
mereology parameter, and although each statement is correct under the

49 In some sense, the concern they have is similar to Lynch’s ambition to avoid getting
caught in the consistency dilemma; even if they, as we have seen, are worried that his
attempt to avoid this kind of worriment fails.
particular parameter-setting that governs it in context, neither statement

is implicitly about its own governing parameter-setting.50
Although I appreciate the reason for why they wish to avoid the outcome of
what they call relativized content, I fail to appreciate what difference it would
make if the “relativeness” of a statement to a certain parameter-setting is
thought to be part of the content of the statement or part of some meta-anal-
ysis of the statement. I mean, Horgan and Timmons must at least accept the
latter and if they do, this would still make the statements compatible; which
would seem to lead to the same problematic outcome they think follow from
the principle of relativized content. That is, even if I would happen to agree
with them that ‘Implicit reference to a setting of the mereology parameter is
not a component of the content of what either of [Carnap and the polish logi-
cian] is saying’ this concession does not change the fact that the statements,
according to Horgan and Timmons, are governed by different parameter-set-
tings, and as far as I can tell, it is this latter circumstance that is relevant as far
as the thesis of conceptual relativity is concerned.51 My inability to make sense
of this step in their reasoning towards an adequate analysis of conceptual rela-
tivity may also be what explains why I find it difficult to agree upon or make
sense of how they more specifically wish to construe the kind of conflict they
think is involved in conceptual relativity. According to Horgan and Timmons,
Carnap and the Polish logician are not making relativizing claims; if they were
we would not have a conflict between them. In trying to make sense then of
precisely how Carnap and the Polish logician are involved in some kind of con-
flict, Horgan and Timmons claim that they are not offering directly contradic-
tory statements, but rather statements that fail to be correctly co-affirmable.
In elaborating on this possibility, Horgan and Timmons write:
Failure of correct co-affirmability is a generic kind of affirmatory conflict.

One species of this genus—a familiar way that two statements can fail to
be correctly co-affirmable—is for them to be directly inconsistent with
one another. But, given our above account of synchronic diffèrance, this
is not the only way. Another species of the genus arises from the fact that
the various permissible settings for contextually variable semantic
parameters are mutually exclusionary; i.e. for a single person P at a single

51 Ibid.
166 svensson
time t, no more than one parameter-setting for a given concept or word

can semantically govern correct usage, by P at t, of that concept or word.52
In expressing the same idea they also write that the “affirmatory conflict at
work in cases of conceptual relativity” is such that:
The two statements fail to be correctly co-affirmable, and this failure

stems from the fact that they are respectively semantically governed by
mutually exclusionary settings of some contextually variable semantic
parameter. The respective claims of Carnap and the Polish logician fail to
be correctly co-affirmable for just that reason; and this feature consti-
tutes their affirmatory conflict with one another. 53
In virtue of the idea expressed in these two quotes Horgan and Timmons take
themselves to have offered an account of how two statements may be incom-
patible, although not incompatible in the sense that they are directly inconsis-
tent, but in the sense that they “fail to be correctly co-affirmable”; or as they put
it, once more drawing upon Derrida, they are “inconsitant.”54 I must concede
that I find it difficult to appreciate to what extent the idea of conflict presented
by them amounts to any sufficiently real conflict. They claim that one and the
same person cannot employ different parameter settings at the same time
because they are mutually exclusionary, but it is not evident why this is the
case. In explaining this they seem to present and rely on a thought rather simi-
lar to the one presented by Lynch. Horgan and Timmons claim for instance
that it is an important part of the analysis of conceptual relativity they offer
that the accounts presented by Carnap and the polish logician cannot be true
relative the same parameter setting, which seems analogous to Lynch’s claim
that relative the same conceptual scheme, Smith’s and Johnson’s account of
how many objects a bag contains cannot both be true. This seems true and
reasonable, but to accept this does not seem to amount to the acceptance of
incompatible and equally correct accounts of the same circumstance. As pre-
viously remarked, the key idea implicit in Horgan’s and Timmons’ proposal
seems similar to Lynch’s, namely that a concept can be taken to be the same
concept (in a certain sense) while still being developed and used differently
(in another sense). I therefore think that the analysis of conceptual relativity

54 Ibid.
presented by them may be susceptible to some of the same problems as Lynch’s

proposal.
6 Concluding Remarks
Lynch as well as Horgan and Timmons are presenting innovative accounts of

how to make sense of incompatible but true propositions. I do however think
that both attempts fail to present an account of such a conflict that does not
come across as rather trivial. To be fair, Horgan and Timmons seem, to some
extent, more mindful of this fact, because they emphasize that Carnap’s and
the Polish logician’s accounts are incompatible in a special sense. Nonetheless,
as they make clear in the beginning of their article, they aim to “propose a con-
strual of conceptual relativity that clarifies it considerably and explains how it
is possible despite its initial air of paradox.”55 As far as I can tell, whatever air of
paradox that surrounded the idea of conceptual relativity from the outset, it is
gone, but so is also the exciting and controversial nature of it. The same verdict
seems true of Lynch’s metaphysical pluralism. He boldly states that “The heart
of metaphysical pluralism is the paradoxical idea that there can be incompat-
ible truths.”56 I agree that the idea in itself seems paradoxical, but not so much
on Lynch’s analysis of it.
In closing and to more explicitly connect my argument to the overall theme
of this collection I wish to make it clear that I have as much sympathy for
Lynch’s criticism of the traditional belief in one absolute and true account
of metaphysical issues, as I have for criticisms concerning absolutist and per-
fectionism accounts of logic. I agree with Lynch that the universe itself does
not dictate how we should use our concepts and how to categorize things into
properties, relations and objects. He is thus correct in opposing the idea of
eternal and absolute facts about, for instance, logical issues, apart from how
such facts are conceptualized by certain people at a certain time. As previously
remarked, on an extended use of the notion of logic, taken as a foundational
system of concepts employed to make sense of reality, our basic accounts of
what makes sense may differ. Even so, this does not demonstrate the possibil-
ity of incompatible and true accounts of such issues in any relevant and strong
sense, at least not for the reason given by Lynch. Different accounts of reality,
logic and our selves, are just that, different.
55 Horgans and Timmons, “Conceptual Relativity,” 74.

168 svensson
Bibliography
Simon Blackburn, “Enchanting Views,” in Reading Putnam, ed. Peter Clark and Bob
Hale (Cambridge: Blackwell Publishers, 1994), 12–30.
Peter Byrne, God and Realism (Aldershot: Ashgate Publishing, 2003).
Susan Haack, “Reflections on Relativism: From Momentous Tautology to Seductive
Contradiction,” in Philosophical Perspectives 10: Metaphysics ed. James E. Tomberlin
(Oxford: Blackwell Publishers, 1996), 297–315.
Terry Horgan and Mark Timmons, “Conceptual Relativity and Metaphysical realism,”
in Philosophical Issues 12: Realism and Relativism, ed. Ernest Sosa and Enrique
Villanueva (Oxford, Blackwell Publishers, 2002), 74–96.
Michael Lynch, Truth in Context: An Essay on Pluralism and Objectivity (London: mit
Press, 1998).
Hilary Putnam, Representation and Reality (London: mit Press, 1989).
———, Realism with a Human Face (Cambridge, Mass.: Harvard University Press, 1990).
John Searle, The Construction of Social Reality (New York: Free Press, 1995).
chapter 9
The Logic of “Oughts” and the Bindingness

of Past Practice
A Critique of Normative Judgment Internalism through a Reading of
King Lear’s Act I
Karim Dharamsi
1 Introduction1
We are invited in the first act of King Lear to witness a ritualized performance
wherein assets are to be divided by a King who is to abdicate his thrown. The
King will give, but he wants in exchange for authority and territory a demon-
stration of filial piety—quid pro quo; a kind of ritual in itself that conforms
to the economy of public display, and civility. At the core of the first scene,
as is my contention, the apparent demands of prudential reason (demands
consistent of what is required of the public ceremony) are intruded upon by
misplaced moral demands of a self-legislative will, untethered from conven-
tion and now working outside its endorsements. Cordelia, the youngest and
best-loved daughter, misunderstands or, indeed, fails to register the deontic
indicators being signalled to her by convention. She decides to tell the truth,
and objects to the rhetorical responsibilities of place and time. But these are
not alien responsibilities. She has been a member of this community and, it
can reasonably be assumed, has enjoyed the benefits of its practices. Indeed,
her very role, as heir, and perhaps her self-understanding is a construct of
Monarchical rule.
Understanding what one ought to do underwrites the very possibility of free-
dom for what Christina Korsgaard calls reflexive beings.2 Such beings exercise
a capacity to reflect on their responsibilities, their acts and how the reciprocal
relations between self-and-world are negotiated. This is as true in King Lear
as it is in life itself. In language consistent with the universe in which the trag-
edy operates, acting outside the boundaries of what is licensable, especially
1 This paper owes its reality to Dr. Admir Skodo’s excellent support and hard work.
2 Christine M. Korsgaard and Onora M. O’Neill, The Sources of Normativity (Cambridge:
Cambridge University Press, 1996), 93.

170 dharamsi
when seemingly unreflective, offends the natural order of things. Cordelia must
die! She has attempted to exit the bindingness of norms, assuming that the
sui generis character of freedom is licence to work outside the deontic limits
of licensed practice. Order must be restored. But this restoration is not merely
cosmetic. It restores social, civic order and thereby restores the socio-historical
geometry in which the ‘self’ itself operates. The self in this respect is coexten-
sive with social order, which in turn is, to shamelessly borrow a phrase, a mir-
ror of nature. For Shakespeare, the integrity of social order is governed by a
rhetorical inheritance, which for him seems entirely consonant with what is
natural. Nature can then be rendered intolerant of our precocious desires to
overcome our historical inheritance—and overcome our obligations.
Shakespeare’s tragedy depends on both Cordelia’s misunderstanding the
operations of ceremony, and her unreflective exit from required etiquette. It
is in Cordelia’s exit that we first begin to enter Shakespeare’s diagnosis of the
possibility of incompatible prudential and moral recommendations for action.
Cordelia’s incompliant response to Lear’s call for aretē or civic virtue in the
space of public reason implies a blurring of the distinction between the moral
and prudential, without eradicating the viability of the practical distinction. In
this way, Shakespeare can be read as offering a view of the self-legislative will
as free only insofar as it is appropriately responsive to the demands of the pub-
lic good. This can minimally require knowing the convention—or the rule one
is to follow—and being able to appreciably judge between different options
within the convention without attempting to operate outside its limits—some-
thing that may be impossible.
Ralph Wedgwood reflects a quietist attitude toward the distinction between
prudential and moral recommendations for action. He argues that the binary
fails to properly illuminate the logic of ‘ought’ as essentially normative, operat-
ing logically prior (if you will) to any narrowly construed moral or prudential
consideration. Instead of collapsing the distinction between the moral and pru-
dential, Wedgwood’s argument depends on accepting a kind of logical-ought
operator that is wedded to practical reasoning: how selves under the disguise
of first person transparency judge what they ought to do. Particularly interest-
ing is Wedgwood’s Platonism about normative facts. He claims “that when a
procedure for answering a normative question reaches the right answer, that
is not because of the intrinsic character of the procedure in question (indeed,
in my view such procedures are fallible), but simply because the answer corre-
sponds to an appropriate normative truth or facts.”3 I am less concerned about
Wedgwood’s Platonism in this paper, as his account is complicated and not
3 Ralph Wedgwood, The Nature of Normativity (Oxford: Oxford University Press, 2009), 7.
The Logic of “ Oughts ” and the Bindingness of Past Practice 171
necessarily material to my concerns. What Wedgwood’s Platonism does pro-

vide him with is a first-person account of action explanation. This account is
divorced from accounts of action sensitive to the role of history and custom,
and Wedgwood implies that “self’” acts as a pure executor of moral judgement.
I understand Shakespeare’s diagnosis in King Lear as recommending against
any view that fails to take history and custom seriously. In other words, under-
stand a “self” requires understanding the socio-historical order within which
that self operates. Contra-Wedgwood, any procedure for reaching the truth
is intrinsically connected to the nature of the truth revealed; and “self” is a
feature, constitutive of its inheritance. In this way, the moral and prudential
oughts are inherently normative but the logic of ought is essentially historical
and self-oriented. It is contingent on matters of custom, how agents under-
stand the ends they wish to secure or the values they endorse. I defend these
latter claims in this paper.
2 Filial Piety and Prudential Reason
“Meantime, we shall express our darker purpose.”4 And with this, Lear’s second
locution and his first substantial one in Shakespeare’s King Lear, begins the
unnatural trajectory of the Ancient King’s decline and death. As noted, the
King has decided to abdicate his thrown, and to relinquish his assets. While
many leaders outstay their welcome—the examples are too numerous, and
often involve a mixing of the pathetic with the horrifying—Lear may be leav-
ing too soon. His reasons for early succession seem sensible; his arguments
seem justified, and so rational. Lear’s primary heirs are his daughters; they are
to be bequeathed an inheritance proportionate to their expression of love. The
“darker purpose” is not an affront to nature, but sadly (perhaps) recognition of
an ordering of things.5 It is a kind of order that comes with a particular form
4 Kenneth Muir, The Arden Shakespeare: King Lear (Cambridge, Mass.: Harvard University
Press, 1972), 5.
5 Cf. Thomas Hobbes, Humane nature, or, The fundamental elements of policie: being a discov-
erie of the faculties, acts, and passions of the soul of man, from their original causes: according
to such philosophical principles as are not commonly known or asserted (London: Matthew
Gilliflower, 1650). Hobbes’s contractarian political ethics marshals a defense of authoritarian
government on the grounds that the sovereign does not enter into contracts with his sub-
jects. Perez Zagorin argues in his Hobbes and the Law of Nature (Princeton: Princeton
University Press, 2009), 64, that: “The sovereign power, as he (Hobbes) expounds its genesis,
is not a party to a contract with its subjects and therefore cannot commit any breach of legal
obligation to them. The people cannot be counterposed to the sovereign as a separate and
172 dharamsi
of participation in civic life—and custom: the social mettle from which selves
emerge. Indeed, the stuff we may detest, the darker purposes we all ultimately
accept, the bureaucratic threads that bind us to process and often to outcome,
and we take as necessary for the well-being of our community and ourselves. In
Shakespeare’s universe, the natural and the civic—physis and nomos—seem to
find ultimate expression when in alignment, when our darker purposes align
with what is best in us. It is courageous, then, to accept our place, to have the
wisdom of etiquette and know our obligations.
The play begins with the Earl of Kent and the Earl of Gloucester anticipat-
ing the division of the Kingdom. An impression is given that discussions prior
to Act 1 have determined the nature the ceremony, even if not the outcome
of its deliberations. Kent opens the play; he thinks that the King has favoured
the Duke of Albany (Goneril’s husband) over the Duke of Cornwall (Regan’s
husband).6 Gloucester’s response confirms Kent’s intuitions about Lear’s affec-
tions, but sobers those intuitions with the rights of ceremony—and the kind
of neutralizing of one’s affection required by affairs of State. Kent reveals here
that his judgement with regard to the requirements of convention may be
impaired, while Gloucester reveals just the opposite. He says:
It did always seem to us (that the King had more affected the Duke of
Albany than Cornwall); but now, in the division of the kingdom, it appears
not which of the Dukes he values most; for equalities are so weigh’d that
curiosity in neither can make choice of either’s moiety.7
What is to come is ceremonial; and ceremony is hardly ever trivial. It has bind-
ing properties, holds communities together, and can be the locus of divisive,
but restricted conflict. Conflict is “restricted” when background assump-
tions apply equally to all parties—even when some or even all of the parties
would be challenged to make explicit any one or more of those assumptions.
The rules of grammar can be said to apply equally to every English language
speaker, but many users of the language who engage in basic forms of endorse-
ment or correction with their children, for instance, might be challenged to
identify systemically the rules they are endorsing or the misapplication they
superior political entity, because without the sovereign in which it is collectively merged, the
people does not even exist and is merely a number of dissociated individuals.” Lear’s request
for an exchange is unusual because it mimics the structure of contract, but it is, rather, an
exchange captured in the way a trade between parties can be.
6 Muir, The Arden Shakespeare: King Lear, 5.
are correcting. Such assumptions, be they about grammar or about social prac-
tice (generally), are binding, and conflicts themselves do not occupy regions
outside constraint of such norms. Here I wish to anchor my remarks to Henry
Jackman’s stipulation with regard to the bindingness of norms. Jackman main-
tains that “[. . .] norms must be in some sufficiently robust sense binding on
us.”8 He adds that “[a]nyone who means anything by their words will not be
able to ‘opt out’ of such norms.”9 I add here a slight but sufficiently provocative
amendment to Jackman’s stipulative definition, namely, that what is intended
by the words anyone uses is also binding; such intentions are responsive to
how conditions under which norms generally apply are appropriately inter-
preted. This applies as much to how words are used as what it is for a meet-
ing to take place. Meetings take place because rational creatures are behaving
in accordance with features of social practice within which expectations for
things called “meetings” are met. When pressed, the same endorsements may
not be met with systematic justifications. It is important to recognize that any
such systematic justification is unnecessary. What is required is behavior (lin-
guistic and otherwise) appropriate to the community’s expectations (broadly
construed).
Kent and Gloucester indicate at the opening of the play that a meeting is
about to take place; their conversation reveals that its purpose and its nature
are known. There may be misunderstanding or legitimate form of misinterpre-
tation with regard to the nature and function of the meeting. Misinterpretation
is normal social fare, but still accommodated by the vastness of binding back-
ground conditions. A member of a community may express confusion in the
following way, “Oh, I didn’t realize you called this meeting for that reason” or
“I’m sorry, I simply cannot accept your claim that-p given your claim that-q” or
“I think it’s strange how he bites his finger nails.” One is still operating within
the space of reasonable expectations and background agreements—and rea-
sonable misunderstanding. However, one (excluding cases of madness) can
occasionally appear to have “opted out” of norms (contra-Jackman) when one
has unintentionally exited or one has never belonged or one has been forcibly
ejected. There can be in either of first two cases dire consequences associated
with such conceptual and practical amputation from community. The last case
is especially egregious and outside the scope of my discussion here.10 In the
8 Hugh Jackman, “Charity and the Normativity of Meaning,” paper presented at the 2004
meeting of the Pacific Division of the American Philosophical Association, 1.
9 Jackman, “Charity,” 1.
10 I have in mind here the most egregious forms of ejecting a person or group from com-
munity. The Jews during the World War II, the Tutsis in Rwanda, the Muslims in Bosnia.
174 dharamsi
instances mentioned, one has been rendered an alien or one has always been
alien. I will refer to these cases of “opting out” as special cases. Cordelia’s sta-
tus unintentionally changes from insider to outsider. And this is the tragedy of
King Lear, if I have it right. To understand the case, I have to examine the main
event of Scene 1 more carefully.
In calling what is to happen in Lear’s court “ceremonial” and a matter of “cus-
tom” I am making a claim about the meanings and intentions associated with
having the meeting in the first place, and the demands being made on those
present—especially those being asked to exchange something for something.
Forgive my quoting Lear’s opening remarks in full, from act, 1, scene 1:
Give me the map there. Know that we have divided

In three our Kingdom; and ‘tis our fast intent
To shake all cares and business from our age,
Conferring them on younger strengths, while we
Unburthen’d crawl toward death. Our son of Cornwall,
And you, our no less loving son of Albany,
We have this hour a constant will to publish
Our daughters’ several dowers, that future strife
May be prevented now. The Princes, France and Burgandy,
Great rivals in our youngest daughter’s love,
Long in our court have made their amorous sojourn,
And here are to be answer’d. Tell me, my daughters,
(Since now we will divest us both of rule,
Interest of territory, and cares of state)
Which of you shall we say doth love us most?
In these cases the victims are able to recognize the operations of normal life outside the
sphere of their amputated status and no doubt wondered why they have been forcibly
ejected. Hanna Arendt’s much discussed Eichmann in Jerusalem: A Report on the Banality
of Evil (London: Penguin, 2006) presents an interesting challenge to those who suggest
that Eichmann was a monster. Arendt: “The trouble with Eichmann was precisely that
so many were like him, and that they many were neither perverted nor sadistic, that they
were, and still are, terribly and terrifyingly normal. From the viewpoint of our legal insti-
tutions, and of our moral standards of judgment, this normality was much more terrify-
ing than all the atrocities put together” Arendt, “Eichmann in Jerusalem,” 276. Arendt’s
remarks continue to be controversial, offending many who think that evil natures express
evil deeds—not Arendt’s terrifyingly normal creatures. But Arendt’s observation speaks
to how profoundly unbelievable an ejection from norms can be; it presents those ejected
with a horrifying realization that the rules of the game they have so naturally endorsed
continue to apply—just not to them. By then it is too late.
What we our largest bounty may extend

Where nature doth merit challenge. Goneril,
Our eldest-born, speak first.11
Lear is asking for something specific. He is King, and it is clear that he has a
“darker purpose,” namely, a form I have already suggested: the bureaucratic
operations of quid pro quo—something for something. Here Lear is operating
as Sovereign, who happens to be a father. This is not merely a matter of simple
emphasis, but an emphasis upon which rests a fundamental understanding of
what is required by the request being made. Lear wants something because
he is giving something. We know from both Lear and later Cordelia that it is
an exchange. And insofar as the request is the request of a Sovereign, what is
required ought to be understood as binding on those who expect to receive
something, but binding, also, because the King has made the request. This is
not binding in any legal sense since Kings do not typically enter into contracts
with their subjects. The binding features run deeper and excavate the sub-
stance of rhetorical interchange between the King and his successors. Each
side will have to deliver.
Following conventional interpretations, Harold Bloom may be right to rec-
ognize the two elder sisters as “transparent vessels of wickedness.”12 While I
agree with Bloom on their transparency, I am not entirely in agreement on
what such access reveals. This is, sadly, outside the scope of this particular
argument. I do, however, think Bloom is right to remind us that Cordelia does
not sound like a victim.13
In response to Lear’s request, Goneril answers, “Sir, I love you more than
word can wield the matter [. . .]” And, Regan, for her turn says, “I am made
of that self mettle as my sister . . .” Adding, “Only she comes too short: that I
profess Myself an enemy to all other joys.”14 Their effusion is theatre; it is, no
doubt, duplicitous. However, for all their moral defects, the sisters are authen-
tic. They are meeting the rhetorical demands of Lear’s request, and keeping
their interests in view. In this sense they are not dishonest but guided by the
recommendations of prudential reason. If we assume with Bloom that the sis-
ters are “transparent vessels” of any kind, then we have to assume too that we
understand them as they understand themselves. In this way the sisters are
not acting against their interests; insofar as this is trivially the case, Goneril
11 Muir, The Arden Shakespeare: King Lear, 5–6.

12 Harold Bloom, The Western Canon (New York: Harcourt Brace, 1995), 67.
13 Bloom, The Western Canon, 67.
176 dharamsi
and Regan are being all that they can be: good prudential reasoners. As readers
there is little mystery associated with the sisters’ character. And Lear is satisfied.
While Goneril and Regan are speaking Cordelia’s mental state is revealed by
way of two asides. The reports of her introspection include, while Goneril is
responding to Lear’s request
Cor. What shall Cordelia speak? Love, and be silent.
And while Regan is responding,
Cor. Then poor Cordelia! And yet not so; since I am sure my love’s
More ponderous than my tongue.15
We learn of Cordelia’s capacity for reflexive self-indulgence, but her muscular

reflections misinterpret the demands of this particular ceremony. She clearly
fails to communicate explicitly (in the public space she occupies) the idea that
Goneril has the “mettle” to disclose, namely, that words underdetermine her
love for her father. Indeed, Goneril’s disclosure resembles Cordelia’s private
thoughts. The problem is that Goneril’s utterance is unconvincing to us, even
if she is convincing Lear. What does this matter? It is, after all, Lear that has
to be pleased. It is evident as we begin to predict Cordelia’s response and eas-
ily estimate that her moral considerations will work against her interests. But
her interests are not merely hers; she is, given her station, answerable to the
interests of State and of the public good. Cordelia is, we also realize, in the
dark with regard to how her moral considerations overplay the requirements
of the situation. She is surprised by Lear’s response. And her surprise should
be a surprise to us. With Goneril and Regan, too, it is too simplistic to impute to
their character a host of negative properties. They are rhetorically attune to the
requirements of context, even if only dispositionally. They may lack Cordelia’s
capacity to reflect, but it is not this capacity’s operations that are being judged.
Cordelia’s turn comes. Lear asks Cordelia to speak. Here is the exchange:
Cor. Nothing, my lord.

Lear. Nothing?
Cor. Nothing.
Lear. Nothing will come of nothing: speak again.
Cor. Unhappy that I am, I cannot heave
My heart into my mouth: I love your Majesty

According to my bond: no more nor less.

Lear. How, how, Cordelia! Mend your speech a little,
Lest you may mar your fortunes.
Cor. Good my Lord,
You have begot me, bred me, lov’d me: I
Return those duties back as are right fit,
Obey you, love you, and most honour you.
Why have my sisters husbands, if they say
They love you all? Happily, when I shall wed,
That lord whose hand must take my plight shall carry
Half my love with him, half my care and duty:
Sure I shall never marry like my sisters,
To love my father all.
Lear. But goes thy heart with this?
Cor. Ay, my good Lord.
Lear. So young, and so untender?
Cor. So young, my Lord, and true.
Lear. Let it be so: thy truth then be thy dower:
For, by the sacred radiance of the sun,
The mysteries of Hecate and the night,
By all the operation of the orbs
From whom we do exist and cease to be,
Here I disclaim all my paternal care,
Propinquity and property of bond,
And as a stranger to my heart and me
Hold thee from this for ever. The barbarous Scythian,
Or he that makes his generation messes
To gorge his appetite, shall to my bosom
Be as well neighbour’d, pitied, and reliev’d,
As thou my sometime daughter.16
In his William Shakespeare’s King Lear, Harold Bloom provides an analysis of

this exchange that is worth considering. Bloom writes:
When Cordelia’s turn comes to bid in Lear’s auction, she voices our con-
tempt for the oily speeches of Goneril and Regan and for the premises
behind the whole charade. We are relieved to hear the bubbles pricked,
but Cordelia’s premises do not present a clear antithesis to the faults in
16 Muir, The Arden Shakespeare: King Lear, 7–8.

178 dharamsi
Lear’s. Her ideas are only a variation on Lear’s; she too thinks of affection
as a quantitative, portionable medium of exchange for goods and ser-
vices (1.1.95–104). Moreover, she sounds priggish. When she parries Lear’s
“So young, and so untender?” with “So young, my lord, and true,” we share
her triumph and her righteousness. We exult with her, but we may well be
put off by the cold competence of our Cinderella. We agree with Kent
when he says that she thinks justly and has “most rightly said” (1.1.183),
but we are probably much more comfortable with his passionate speeches
on her behalf than we were with her own crisp ones.17
In this passage Bloom makes four observations, each of which concern how
readers understand what is to unfold. First, Bloom takes for granted that
Cordelia represents our collective interests in reacting as she does to her sis-
ters’ “oily speeches.” Second, Bloom notes that Cordelia’s “premises” do not
present us with a clear antithesis to the “faults” in Lear’s. He adds that her
ideas are only a variation on Lear’s as she thinks of affection as “quantitative,”
making reference to proportionate distribution in “a medium of exchange for
goods and services.” Third, Bloom recognizes Cordelia’s “priggishness.” This
is, by my lights, a necessary insight in making sense of Cordelia’s role in the
narrative and what her apparent smugness and arrogance represents. Finally,
Bloom rightly notes Kent’s passionate speeches better ally with our expecta-
tions of measured response than does the “competent” righteousness of “our
Cinderella.”
I have suggested, contra-Bloom, that the ceremony is meaningful and not
a charade. Succession is not a light affair. Lear’s request is no doubt bizarre,
but it is not unintelligible when set in the framework of quid pro quo. This is
the “natural” underwriting the civic in Shakespeare’s Ancient Briton. As John
Danby suggests in his Shakespeare’s Doctrine of Nature: A Study of King Lear,
“[i]t is impossible to talk about Nature without talking also about pattern and
ideal form; about Reason as displayed in Nature; about Law as the innermost
expression of Nature; about Custom which is the basis of Law and equally with
Law an expression of Nature’s pattern; about Restraint as the observance of
Law, and the way to discover our richest self-fulfilment.”18
In my view, Lear’s judgement is not impaired. Rather, his office, the locus
of contact between the natural and the civic, is exploiting custom to endorse
a procedural form of just dessert with regard to the sisters’ inheritance. And
custom, following Danby’s remarks, expresses nature’s pattern—or at least can
17 Bloom, The Western Canon, 67.

18 John F. Danby, Shakespeares’s Doctrine of Nature (London: Faber and Faber, 1949), 67.
be understood by those who hold office or are part of the Monarchical class
(in this case) as consistent with the natural course of things. As I have sug-
gested, this appreciation of pattern rendering is not alien to Cordelia; she has
been initiated into the practices and expectations she is rejecting in the first
act. In this way, Shakespeare’s Ancient King is facing off against his daughter
who occupies a liminal realm between the pre-modern and the modern. She
is not one of us, even though she is showing signs, and she’s no longer part of
the community that raised her. Indeed, Cordelia may represent our interests in
distancing herself from Goneril and Regan, but she does not move us into her
corner or motivate us to defend her sisters. Cordelia’s apparent self-legislative
priggishness allows proper distance between our interests and the interests
of those who are to receive Lear’s gifts. While Goneril and Regan overreach
and exaggerate, Cordelia seems amputated from the situation—she renders
herself “outside” the fray and is, remarkably, orphaned. Her very being is now
unnatural in Shakespeare’s creation. She can represent our collective inter-
ests if and only if we understand those interests in very narrow terms that are
qualified simply by our dislike of her sisters. If Bloom is right in suggesting that
Cordelia represents our collective interests, then one cannot help but wonder
how prudential recommendations for the good of the community could not
have animated Cordelia’s response to her sisters’ oily speeches. Shakespeare
has nicely halved the prudential from the moral in presenting us a form of
moral obduracy ignorant of custom, alienated from history and, now, divorced
of future.
Cordelia acts from moral recommendations, but her position transcends
context and she seems to misread the demands of Lear’s rhetorical queues. The
demands may be strictly prudential, but they are also moral when placed in
their respective historical-rhetorical context. When Lear advises she “mend her
speech,” he is sincere; it is not truth he is after—not the kind of truth Cordelia
is motivated to give. Rather, truth here conforms to the requirements of
ceremony—and custom. What is being demanded is “demonstration,” indeed,
a display of filial piety towards the King who happens to be father.
In failing to read the situation Cordelia moves us from what would other-
wise be the typical transitional formality to a dark revolution. In short, and
more to the point of this discussion, Cordelia’s response preserves an ideal of
truth that is procedurally neutral, but it fails to achieve a moral end; she casts
prudential considerations aside. In other words, how she believes she “ought”
to act is attached directly to a purely moral position. And she is the reason for
the tragedy. If Cordelia was sensitive to the rhetorical requirements of custom,
she may have responded differently to Lear’s question. As I have suggested,
Lear’s “dark purposes” are not an affront to the order of things; the order
180 dharamsi
enjoins all parties to play the game properly. By being “orphaned,” Cordelia has
sealed her fate and her father’s. It is her response that is an affront to nature,
and so an affront to the civic order. In short, exiting norms is tantamount to an
exiting of self. And there’s no self-exit. Cordelia must die.
3 The Prudential and the Moral
In Groundwork for the Metaphysics of Morals, Kant famously distinguishes

between moral and prudential reasons for recommending action.19 A state of
schadenfreude is unremarkably immoral; one has moral reasons to reject this
attitude towards others. In contrast, one has prudential reasons to eat well,
rest, and exercise to maintain good health. As Kant points out, reasons that
recommend an action need not be entirely moral or prudential. Kant’s shop-
keeper in Groundwork has both moral and prudential reasons for not cheating
his customers.20
Ralph Wedgwood seems to accept the notional distinction between the
moral and prudential, but defends a “further sense of ‘ought’ that is neither
narrowly moral nor narrowly prudential.”21 On first blush, this seems like a
good idea. Had Cordelia had another option that was not narrowly reduced
to moral considerations, she might have responded differently to Lear. And
Cordelia, under the rubric of this sort of broader moral mandate, might have
asked of herself, following Wedgwood, “Ought I to do what I ought to do?” This
kind of reflexivity might have thwarted the tragic trajectory of the play.
Wedgwood posits a condition under which a conflict arises between rea-
sons which recommend a moral act, x, and a prudential act, y. He suggests a
context by which it is impossible to do both, in which case one might ask, he
argues, (prior to accepting either the moral or prudential recommendations),
“Ought I to do what I ought to do?” Wedgwood concludes that the “ought” of
this query is neither narrowly moral nor prudential; it is normative in a more
19 Immanuel Kant, Groundwork for the Metaphysics of Morals, trans. Arnulf Zweig, ed.
Thomas E. Hill and Arnulf Zweig (Oxford: Oxford University Press, 2002).
20 In Groundwork on the Metaphysics of Morals Kant illustrates the overlapping of the pru-
dential and the moral by way of an example. A shopkeeper has moral reasons to not cheat
his customers—it’s just wrong. He also has prudential reasons; cheating his customers
will likely make for an unsuccessful business. In this way the moral and prudential can
overlap.
21 Wedgwood, The Nature of Normativity, 24.
general sense.22 Cordelia’s asides do seem to demonstrate some confusion

about what ought a response to be to Lear’s question. Her confusion does not,
however, demonstrate a tension between prudential and moral recommenda-
tions. Cordelia interprets Lear’s question and her sisters’ responses as simply
immoral. She interprets the King and her sisters’ as creatures that have been
amputated—not from custom or history—but from the moral realm. Given
what I have maintained, it is here that Cordelia becomes an ironic character.
She is at the end of the first act an outsider, orphaned by her attempt to exit the
normative demands of the quid pro quo. Instead of marriage to a player who
follows the rhetorical rights of inclusion, Cordelia has embraced a truth that
is at once irreducible to history and ceremony. As Lear says, her truth is her
dower. He might well have said, her unnatural truth.
Wedgwood’s defence of normative judgment internalism (nji) is “exclu-
sively concerned with normative judgments that can be expressed by state-
ments of the form ‘A ought to φ,’ when the term ‘ought’ is used in this more
general normative sense.”23 He adds that he is specifically concerned with
first-person form of such judgments, and doing “something of the appropriate
sort.”24 Wedgwood’s discussion is couched in a broader discussion about what
it means to be rational, and his internalism can only be made sense of in terms
of his discussion of rationality. Wedgwood writes:
[. . .] if you are rational, and your answer to the question ‘What ought I to
do?’ is that you ought to go to bed (and going to bed is “of the appropriate
sort”), then you will not just judge that you ought to go to bed; you will
also intend to go to bed.
This is hardly a surprising conclusion. It is widely accepted by philoso-
phers that akrasia is a kind of irrationality; and akrasia consists of will-
ingly failing to do something that one judges one ought to do . . . so long
as φ-ing something of the appropriate sort, if you do not intend to φ, you
will count as willingly failing to φ. So at least as long you φ-ing is some-
thing of the appropriate sort, if you judge that you ought to p, and yet do
not intend to φ, you are being akratic—and so irrational.25

182 dharamsi
Sometimes rational states can be confused for akratic ones, as in cases where
one’s actions seem out of place or when others act in ways that seem patently
outside the scope of normal behavior. One might think here of ss guards during
World War Two as expressing behaviors that to observers would appear akratic.
Wedgwood’s remarks, however, rest on the first-person and so determining
whether one is being irrational seems to also rest with oneself. While I agree
with Wedgwood that what one ought to do can also be what one intends to do
(in the trivial sense), it is not clearly the case that what one intends is what one
ought to—especially in areas of moral judgment where tense competitions
between what ought to be done and one’s intentions arise. Still, Wedgwood’s
internalism puts the emphasis on Cordelia’s own understanding of what she
ought to do—how she ought to respond to Lear. Contra-Shakespeare, the self
is a kind of logical operator for Wedgwood. The logical self operates outside of
the constraints of history or the requirements of a given social order if that self
is moral. Cordelia must live!
As I have suggested, Cordelia’s asides during her sisters’ responses to Lear
suggest reflexive self-understanding. She is struggling to decide on how to
respond to Lear, but her reflections do not challenge her attitude. Her ques-
tions are about how to express her attitude. When she responds, her intentions
are clearly aligned with what she thinks she “ought” to say. Wedgwood main-
tains that one ought to φ when φ-ing is of the appropriate sort; if you intend
to φ, then it is rational to φ and it is what you ought to do. But what makes
something “of the appropriate sort?” Wedgwood writes:
[. . .] nji is only true of judgments of the form ‘I ought to φ’, where φ-ing is
something “of the appropriate sort” . . . for φ-ing to be of the appropriate
sort, it must be the case that one knows that one’s having intention to φ
will make a significant difference to the chances of one’s actually φ-ing.
Let us say that in fact, for φ-ing to be of the appropriate sort, something
stronger must be the case: one must know that one will φ if and only if
one intends to. Let us say that in this case, φ-ing is a course of action that
is “manifestly dependent on intention.”26
Wedgwood and Cordelia share a few things in common. Wedgwood’s formal

claim is that a state of akrasia arises when one simultaneously holds that
“I ought to φ” and then fails to intend to φ. Here he claims that one’s judg-
ment and one’s will are clearly in conflict. Cordelia’s judgment is “manifestly
dependent on intention.” She clearly intends to φ and so φ-ing, as something

that she ought to do (tell Lear what she really thinks), is consistent with her
intention to φ. Lear’s ideal sense of “what is appropriate” is not as narrow as
either Wedgwood’s or Cordelia’s, and this is the crux of the moral and pruden-
tial impediment facing the latter’s imperilled state. Lear is acting on behalf of
history and custom, and Cordelia is acting, in the main, from a contextually
detached first-person vantage. She is not irrational, just out of touch.
Kant associates the moral law with the human will. For some Kantians,
Christina Korsgaard, for instance, the source of normativity resides in the self-
imposition of moral obligations by the autonomous will. If Korsgaard and Kant
are correct, then, contra-Wedgwood, the “ought” of “Ought I to do what I ought
to do?” is not prior to its moral or prudential species, but the very conflict that
self-legislation of the autonomous will requires to assert its freedom to choose
between competing ends. A possible confusion arises with regard to such con-
siderations because Wedgwood ignores questions of what we value in broadly
social or historical terms when we choose x over y or the other way around. He
also ignores Korsgaard’s construal of Kant’s moral theory, and her especially
robust defence of freedom. For Korsgaard, persons are reflexive creatures. We
turn our attention on our own moral activities and unless we have a reason to
act, she says we cannot go forward.27 But questions arise for all of us when we
wonder about our reasons for acting. Sometimes those reasons align our moral
considerations (be they in the first-person) or with our responsibility to the
wellbeing of community, and perhaps its officers.
Cordelia demonstrates her ability to turn her attention to her own moral
activities, but I have maintained that she is insensitive to the rhetorical queues
and certain deontic signals—the stuff that recommends in favor of custom
in much the way Danby suggests that nature and civic order are conjoined
in Shakespeare’s tragedy. For Wedgwood, NJI takes as essential the connec-
tion between normative judgment and practical reasoning or motivation for
action.28 He argues that normative judgments about acts one ought to perform
are essentially connected to motivation and practical reasoning. Wedgwood
argues that such judgments “involve a concept whose essential conceptual
role is its role in practical reasoning.”29 For Wedgwood, an essentially practi-
cal mode of presentation of the ought-relation characterises normative judg-
ments made in practical reasoning. Many concepts are individuated by their
conceptual or inferential role. Wedgwood’s innovative argument suggests that
some concepts serve a special practical purpose. He argues that a concept
27 Korsgaard and O’Neill, The Sources of Normativity, 93.

184 dharamsi
expressed by the term “ought” can be taken to be individuated as appropriate

to how it figures in NJI as true. Wedgwood writes
[. . .] we could partially characterise the essential role of this concept

(ought) as follows: any judgment of the form ‘I ought to φ’, involving this
concept, rationally commits the thinker to intending to φ ( . . . for this
claim to be plausible, φ-ing must be, as I put it, something “of the appro-
priate sort”: that is, φ-ing must be a course of action that is “manifestly
dependent on intention,” in a situation “with no relevant uncertainty”.) to
say that a judgment “rationally commits” the thinker to having a corre-
sponding intention is to say that if this judgment is rational, then that
makes it irrational for the thinker not to have the corresponding
intention.30
Following Jackman, I noted earlier that norms have to be sufficiently binding

on us, and that there are no logical or practical provisions for opting out. I sug-
gested that any case of opting out has to be special. And I added to Jackman’s
claim that what one intends is also binding in some sufficiently robust sense
insofar as the meanings of our words are bound by the practices within which
their correctness conditions are expressed. In the above passage, the “ratio-
nal” corresponds by Wedgwood’s lights to an “information-relative ‘ought.’”31
He claims that a belief’s status as either rational or irrational is relative to a
body of information; in relation to, say, I1 the belief may be rational but in rela-
tion to I2 the belief may be irrational. What one “intends” is normally rational
since what is intended is based on what someone presumably assumes of the
situation. Wedgwood generally resists this kind of information-relative reduc-
tion, favoring instead a relativized conception of rationality that depends on a
thinker’s commitment to a belief at a given time. So in understanding that “I
ought to φ” and the commitments implied, a thinker is rational insofar as her
commitment (explicitly) relativized is rational in relation to the “total” infor-
mation that the thinker possesses at the time in question.32 In conceiving of
a commitment as having a corresponding intention, Wedgwood is suggesting
that another feature of rationality is that standards of rational belief are all ori-
ented towards the goal of having a correct belief.33 In setting up the discussion
in this way, Wedgwood is defending internalism; whether or not it is rational

for a thinker to believe in a proposition’s truth is dependent on “facts” internal

to the thinker at the time she took the proposition to be true. In this way “facts”
about a thinker’s mental states are more relevant to understanding her actions
than are the “facts” external to her thoughts, for those “facts” may not have
been appropriated by her; indeed, they may not have been available. Recall
that Wedgwood is a Platonist about normative facts. He is not concerned about
procedures (how we go about reaching the truth); he believes that the truth is
something awaiting discovery and that the methods are, in the main, fallible.
But there are “facts” internal to the thinker relevant to how she conceives of
her ends.
Accordingly, Wedgwood’s account of the semantics of the practical “ought”
renders an essential conceptual role to its use as efficacious in practical reason-
ing. The practical “ought” is given, he maintains, “by the rule that acceptance of
a first-person judgment involving this sort of ‘ought,’ of the form ‘O<me, t>(p),’
commits the thinker to making the embedded proposition p part of her ideal
plans about what to do at t.”34 Wedgwood suggests that the practical “ought”
has a semantic value such that when “indexed to an agent A at a time t, it is
that property of a proposition p that makes it correct for A to incorporate p
into her ideal plans about what to do at t, and incorrect for A to incorporate
the negation of p into any such ideal plans about what to do at t.”35 In attempt-
ing to give an account of the logic of the practical “ought,” Wedgwood relies
on some basic principles of deontic logic. The source, he argues, of deontic
logic is in making explicit consistency constraints on statements involving the
use of the practical “ought”—what makes a thinker’s plans realizable and con-
straints there may be that prejudice against a plan’s realization. In the end,
Wedgwood’s treatment of the “practical ought” depends on an account of nor-
mative discourse.
The strength of Wedgwood’s argument rests on privileging the first person.
And this seems consistent with the language of self-legislation familiar to
Kantians. However, the first-person in Wedgwood’s account negotiates inter-
ests entirely in indexical terms, divorced from the values that provide context
for “ideal plans” in terms of historical imperatives or custom. In failing to reg-
ister the rhetorical signals of Lear’s succession ceremony, Cordelia is idealizing
outside the context of the practices that inform the kinds of ideals appropriate
to the condition. In this way, Cordelia is operating outside the requirements of
her responsibilities, and yet she is governed by moral considerations. She is not
in an akratic state, but she is an outsider. What is of crucial importance here

186 dharamsi
is that Cordelia is not only a social-outsider; she is an exited self. Her truth is
her dower. Cordelia is Shakespeare’s moral realist, exiled from community and
from self. As is revealed by the narrative of the play, Cordelia’s exile is final and
ultimate.
4 Conclusion
Edmund, Gloucester’s bastard son is introduced to us in the first act—before

Lear’s main bits of the scene. Even at this early stage in the play, Edmund is the
necessary outsider, lacking proper alignment between nature and civic respon-
sibility. His father has “so often blush’d to acknowledge him” and we are asked
with regard to Edmund’s status whether we “smell a fault.”36 It is not uncom-
mon for Shakespeare to associate sin with a bad smell.37 To “smell a fault” is
also to acknowledge Edmund’s uncommon visage, something acutely out of
place but yet always in view. He occupies an important place in King Lear. He
is revealed to the reader early, and his second verbal entry is a response to
Kent: “My services to your Lordship.”38 As Muir suggests in his commentary
of the play, Edmund expresses obligation here in terms of service (or duty).
The utterance conveys a sordid commitment, but it is too early and we remain
in the dark. It is only later that we realize Edmund’s material importance to
the play; he is Cordelia’s double—her self-interested other half whose tragic
arc ends with redemption and re-entry into community. The logic of his self is
restored by the social geometry of Shakespeare’s Ancient Briton.
King Lear is an examination of our philosophical tendency (perhaps) to
divide our prudential and moral considerations. But it is also an examination
of the moral and prudential in terms qualified only in terms of a self divorced
from context, from history and custom. I have tried to show how history and
custom underwrite our considerations for how to act and that our moral and
prudential considerations are answerable to forces outside our narrowly con-
ceived of interests. In the end, the self finds its logic in the socio-historical con-
text which it is and in which it will finally extinguish. At least if my assessment
of Shakespeare—in King Lear—is correct.

37 Following Kenneth Muir’s (1972) recommendation, one might compare Hamlet, Act 3,
Scene 3.
Bibliography
Hannah Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil (London:

Penguin, 2006).
Harold Bloom. The Western Canon (New York: Harcourt and Brace, 1995).
John F. Danby Shakespeares’s Doctrine of Nature (London: Faber and Faber, 1949).
Thomas Hobbes, Humane nature, or, The fundamental elements of policie: being a dis-
coverie of the faculties, acts, and passions of the soul of man, from their original causes:
according to such philosophical principles as are not commonly known or asserted
(London: Matthew Gilliflower, 1650).
Hugh Jackman, “Charity and the Normativity of Meaning,” paper presented at the 2004
meeting of the Pacific Division of the American Philosophical Association.
Immanuel Kant, Groundwork for the Metaphysics of Morals, trans. Arnulf Zweig, ed.
Thomas E. Hill and Arnulf Zweig (Oxford: Oxford University Press, 2002).
Christine M. Korsgaard and Onora O’Neill, The Sources of Normativity (Cambridge:
Kenneth Muir, The Arden Shakespeare: King Lear (Cambridge, Mass.: Harvard
Ralph Wedgwood, The Nature of Normativity (Oxford: Oxford University Press, 2009).
Perez Zagorin, Hobbes and the Law of Nature (Princeton: Princeton University Press,
2009).
chapter 10
First-Order Logic, Incongruism, and

Anti-Formalism
Anders Kraal
1 Introduction
Modern formal logic has oftentimes been said to provide a tool for conceptual
analysis. Gottlob Frege, for example, says in Begriffsschrift that his formal logic
aims at clarifying “the relations of concepts.”1 Similarly, Bertrand Russell and
A.N. Whitehead say in the Principia Mathematica that their formal logic aims
at “the greatest possible analysis of the ideas with which it deals.”2 And Wilfred
Hodges, a distinguished contemporary logician, speaks of formal logic as “an
aid to definition and conceptual analysis.”3
The most widely used version of modern formal logic in contemporary ana-
lytic philosophy is standard first-order predicate logic (henceforth just “first-
order logic”). Indeed, first-order logic, in its Gentzen-style natural deduction
guise, has become “the most universally accepted method (within philosophy)
of ‘doing logic’.”4
As is the case with all formal logics, however, the idea that first-order logic
is an adequate tool for the conceptual analysis of a given proposition is some-
thing that is (or ought to be) open to inquiry, and there is more than one posi-
tion that can be taken in response to it. Let us use the name congruism for the
position that for a given system of formal logic and a given set of propositions,
the propositions of that set can be adequately analysed in terms of the relevant
system of formal logic; and let us use the name incongruism for the opposite
view. Let us moreover use the name formalism for the position that formal logic
1 Gottlob Frege, “Begriffsschrift,” trans. M. Beaney, in The Frege Reader, ed. M. Beaney (Oxford:
Blackwell, 1997), 50–51.
2 Bertrand Russell and A.N. Whitehead, Principia Mathematica, vol. 1 (Cambridge: Cambridge
University Press, 1910), 1.
3 Wilfred Hodges, “Classical Logic I—First-Order Logic,” in The Blackwell Guide to Philosophical
Logic, ed. L. Goble (Oxford: Blackwell Publishing, 2008), 9.
4 Francis Jeffry Pelletier, “A Brief History of Natural Deduction,” History and Philosophy of Logic
20:1 (1999): 2.

First-Order Logic, Incongruism, and Anti-Formalism 189
is a better tool than non-formal logic for the conceptual analysis of a given set
of propositions; and let us use the name anti-formalism for the opposite view.
In this chapter I shall single out a set of propositions with regard to which
I shall defend incongruism (vis-à-vis first-order logic) and anti-formalism.
The set of propositions I shall single out takes as its members what I shall call
“divine nature doctrines,” by which I mean doctrines such as “God exists,” “God
is almighty” and “God is all-good” as these doctrines are understood in classical
theism.5 The selection of this particular set is pertinent in view of a number of
special difficulties that have long been taken to attach to these propositions.6
To the extent that analytic philosophers have made use of modern for-
mal logic in analyzing divine nature doctrines, first-order logic has been the
preferred choice. But exactly how first-order analyses of divine nature doc-
trines are to be carried out in practice is a matter of disagreement.7 Consider
for example the doctrine “God exists.” Some take “God” as an argument and
“exists” as a monadic function,8 and then subject the doctrine to an argument-
function analysis:9
Exists(God)
Others take “God” as a monadic function and subsume “exists” to quantifier

analysis:10
5 My understanding of classical theism accords with that of Brian Leftow in his encyclo-
pedia article “Concepts of God” in The Routledge Encyclopedia of Philosophy, vol. 4, ed.
E. Craig (London and New York: Routledge, 1998), 93–102. Cf. also my encyclopedia article,
“Theism, Classical” in Encyclopedia of Sciences and Religions, vol. 4, eds. A. Runehov and
L. Oviedo (Springer: New York and Dordrecht, 2013), 2239–2240.
6 For more on this, see my papers “The Emergence of Logical Formalization in the
Philosophy of Religion: Genesis, Crisis, and Rehabilitation,” History and Philosophy of
Logic (forthcoming); and “Logic and Divine Simplicity,” Philosophy Compass 6:4 (2011):
282–294.
7 See, once again, my papers “The Emergence of Logical Formalization in the Philosophy of
Religion” and “Logic and Divine Simplicity.”
8 In this paper I follow Fregean terminology and speak of arguments and functions rather
than of singular terms and predicates.
9 See Alvin Plantinga, “Self-Profile,” in Alvin Plantinga, eds. J.E. Tomberlin and P. van
Inwagen (Dordrecht: Reidel, 1985), 38, 95.
10 See Jan Woleński, “Theism, Fideism, Atheism, Agnosticism,” in Logic, Ethics and All That
Jazz: Essays in Honour of Jordan Howard Sobel, eds. L.-G. Johansson, J. Österberg and
R. Sliwinski (Uppsala: Uppsala Philosophical Studies, 2009), 388.
190 Kraal
∃x(God(x))
Others build the argument-function analysis into a quantifier analysis:11
∃x(God(x) ∧ Exists(x))
Others take “God” as an argument but go for an identity analysis:12
∃x(x = God)
Still others take “God” as a monadic function and go for a definite description
analysis:13
∃(x)(God(x) ∧ ∀(y)(God(y) ↔ x=y))
In spite of disagreements as to the correct formal analysis of divine nature doc-

trines, analytic philosophers have widely assumed that divine nature doctrines
are amenable to first-order analysis, whatever that analysis amounts to. In this
paper I contend that this assumption is mistaken, at least insofar as first-order
logic is provided with its standard semantics and the relevant doctrines are
understood within the conceptual framework provided by classical theism’s
doctrine of divine simplicity (about which more later).
In arguing for the relevant versions of incongruism and anti-formalism I
will proceed as follows. In section 2 I offer an overview of standard first-order
logic. In section 3 I defend the relevant version of incongruism by proposing
four criteria for amenability to first-order analysis, and arguing that divine
nature doctrines fail to meet at least one of these criteria. In section 4 I give
two formalistic responses to this version of incongruism. In section 5 I proceed
to defend the relevant version of anti-formalism by arguing that a non-formal-
istic response to incongruism is more adequate than the formalistic responses
of section 4. Section 6 contains some words in conclusion.
11 See Kaj B. Hansen “Formal logic, Models, Reality,” in Neither/Nor, eds. R. Sliwinski and
F. Svensson (Uppsala: Uppsala Philosophical Studies), 86–88. Hansen reads ‘∃x’ as “x
exists in a model,” and ‘Exists(x)’ as “x exists in reality.”
12 See William Mann “Definite Descriptions and the Ontological Argument,” in Philosophical
Applications of Free Logic, ed. K. Lambert (Oxford: Oxford University Press, 1991), 269–270.
13 See W.V.O. Quine, Mathematical Logic (Cambridge, Mass.: Harvard University Press, 1951),
150; Russell and Whitehead, Principia, 32; and Jozef Bocheński, The Logic of Religion (New
York: New York University Press, 1965), 68.
2 An Overview of First-Order Logic
Although the basic ideas of first-order logic go back to Frege’s Begriffsschrift

from 1879, the first explicit first-order system was developed by Wilhelm
Ackermann and David Hilbert in their Grundzüge der theoretischen Logik
from 1928. Frege, Ackermann and Hilbert were all mathematicians, and they
intended their formal logics to function primarily as tools in mathematical
analysis. However, under the influence of philosophers such as Rudolf Carnap
and W.V.O. Quine these tools became popular also in the context of analytic
philosophy more generally.14
First-order logic comprises a propositional level and a quantification level.
The propositional level makes use of propositional parameters (φ, ψ, . . . ),
truth-functional logical constants (¬, ∧, ∨, →, and ↔), and the absurdity sign
(⊥). Propositional parameters are used as substituenda for truth-bearers,
truth-functional logical constants are used to construct complex expressions,
and the absurdity sign denotes an arbitrary contradiction. These symbols are
supplied with a syntax giving the rules for forming formulas: e.g. as follows:
φ and ψ are formulas; ⊥ is a formula; if φ and ψ are formulas, then ‘¬φ’, ‘φ ∧ ψ’,
‘φ ∨ ψ’, ‘φ → ψ’, and ‘φ ↔ ψ’ are formulas. A semantics is provided giving the
truth-conditions of these formulas, e.g. as follows:
Formulas are true or false.

‘⊥’ is false.
‘¬φ’ is true iff φ is false.
‘φ ∧ ψ’ is true iff φ is true and ψ is true.
‘φ ∨ ψ’ is true iff φ is true or ψ is true.
‘φ → ψ’ is true iff it is not the case both that φ is true and ψ is false.
‘φ ↔ ψ’ is true iff φ and ψ are both true or both false.
We also have the usual inference rules:
φ, ψ 𐅂 φ ∧ ψ
φ ∧ ψ 𐅂 φ
φ 𐅂φ ∨ ψ
φ ∨ ψ, [φ] χ, [ψ] χ 𐅂 χ15
[φ] ψ 𐅂 φ → ψ
14 Pelletier, “Natural Deduction,” 2.

15 The notation “[φ] χ” should be understood as meaning that χ can be derived from φ,
although it is not relevant to the inference in question how the derivation is made.
192 Kraal
φ, φ → ψ 𐅂 ψ
[φ] ⊥ 𐅂 ¬φ
[¬φ] ⊥ 𐅂 φ
φ, ¬φ 𐅂 ⊥
The quantificational level comprises argument-function analysis and quan-

tifier analysis. Argument-function analysis analyzes propositions in terms of
arguments and functions, where arguments denote elements and functions
denote sets.16 If ‘α’ and ‘β’ are arguments and ‘Ψ(x)’ and ‘Φ(x)’ are functions,
then ‘Ψ(α)’ and ‘Φ(β)’ say that α is a member of Ψ and that β is a member of Φ.
A truth-conditional semantics is provided for these formulas with the help of
model-theory. Let D be a domain consisting of one item. Let I be an interpreta-
tive function consisting in assigning ‘α’ as a name of that item (‘I(α)’) and in
taking ‘Ψ’ as an extension (‘I(Ψ)’) consisting of all the elements in D satisfying
the function ‘x is Ψ’. Let M be the model consisting of the pair <D, I>. We then
have:
‘Ψ(α)’ is true in M iff I(α) is an element of I(Ψ)
Quantifier analysis presupposes the argument-function analysis but goes

beyond it in making use of the universal quantifier ‘∀x’ (“for all x”) and the
existential quantifier ‘∃(x)’ (“there exists an x”) to analyze truth-bearers involv-
ing universal and existential quantification. ‘∀(x)Ψ(x)’ says that every x is Ψ,
and ‘∃(x)Ψ(x)’ says that there exists an x that is Ψ. A truth-conditional seman-
tics is provided for these formulas with the help of model-theory. Let D be a
domain consisting of two items. Let I be an interpretative function consisting
in assigning ‘α’ as a name of the first of these items and ‘β’ as a name of the
second of these items, and in taking ‘Ψ’ as an extension consisting of all the ele-
ments in D satisfying the function ‘x is Ψ’. Also, let ‘γ’ function as a stand-in for
any name that names an item in D. We then define M as the model consisting
of the ordered pair <D, I>. Truth-conditions are now given as follows:
‘M╞ ∀xΨ(x)’ is true iff M╞ Ψ(γ) for any item I(γ)

‘M╞ ∃(x)Ψ(x)’ is true iff M╞ Ψ(γ) for at least one item I(γ)
‘M╞ Ψ(γ)’ is true iff the item I(γ) is an element of the extension I(Ψ)
16 In this paper I shall only be considering monadic functions, i.e. functions with only one
argument.
The quantificational level makes use of the inference rules of the propositional
level, and four additional inference rules of its own, which can be given as fol-
lows. Let D be a non-empty domain of items over which the variables bound
by quantifiers range, let ‘t’ be a term that is either a variable ranging over one
or several of these items or else a name denoting one of the items, and let ‘Ψ’
be any predicate; then
Ψ(x) 𐅂 ∀xΨ(x)
∀xΨ(x)𐅂 Ψ(t)
Ψ(t)𐅂 ∃xΨ(x)
∃xΨ(x), [Ψ(x)] Φ 𐅂 Φ
(These rules are subject to some well-known restrictions which I omit here).17
Functions are understood set-theoretically, in terms of the extensions of ele-
ments that satisfy the function in question. The standard way of understand-
ing sets accords with these features:18
(F1) A set is a collection of distinguishable objects.

(F2) A set is itself an object.
(F3) If x1 . . . xn are objects, then we can represent the set of these objects
as {x1 . . . xn}.
(F4) If A = {x1 . . . xn}, then A ≠ xi for i = 1, . . . n.
(F4) accords with the foundation axiom of Zermelo-Franckel set-theory, accord-

ing to which every non-empty set A contains an element B which is disjoint
from A.
3 Incongruism
In defending incongruism I shall proceed in two steps. I first propose crite-

ria that must be fulfilled by anything that is amenable to first-order analysis.
I thereafter argue that divine nature doctrines fail to fulfil at least one of these
criteria.
17 Cf. Kaj B. Hansen, Grundläggande logik (Lund: Studentlitteratur, 1997), 286–294.

18 Cf. Robert Stoll, Set Theory and Logic (San Francisco and London: W.H. Freeman and
Company, 1963), 2; Hansen, Grundläggande logik, 168.
194 Kraal
We start with two criteria pertaining to the propositional level of first-order

logic. A first such criterion is that anything that is amenable to propositional
first-order analysis must be a proposition:
(C1) For any x, x is amenable to propositional first-order analysis only if

it is a proposition.
(C1) can be supported as follows:
(i) For x to be amenable to propositional first-order analysis, the infer-

ence rules of propositional logic must apply to it.
(ii) For these inference rules to apply to x, x must be the sort of thing
that can function as a premise or conclusion in an inference.
(iii) For x to be the sort of things that can function as premise or con-
clusion in an inference, x must be a truth-bearer.
(iv) For x to be a truth-bearer, x must be a proposition, i.e. the possible
object of a propositional attitude.
(v) Hence, for x to be amenable to propositional first-order analysis, x
must be a proposition.
A second criterion pertains to bivalence:
(C2) For any x, x is amenable to propositional first-order analysis only if

it is bivalent.
(C2) can be supported as follows:
(i) In order for something to be amenable to propositional first-order

analysis the inference rules of propositional first-order logic must
apply to it, including the rule ‘[φ] ⊥ 𐅂 ¬φ’.
(ii) The rule ‘[¬φ] ⊥ 𐅂 φ’ is valid only on the condition that either φ or
its negation ¬φ is true, for it infers φ from the supposition that ¬φ
entails an absurdity, which would not be valid other than if there
were only one other possible alternative if ¬φ failed to be true,
namely that φ is true.
(iii) Hence in order for something to be amenable to propositional first-
order analysis it must be bivalent.
We turn next to criteria for the quantificational level. A first criterion can be
derived from (C1) and (C2):
(C3) For any x, x is amenable to quantificational first-order analysis only

if it is a bivalent proposition.
(C3) is obvious since first-order logic’s quantificational level presupposes

the propositional level, and amenability to propositional first-order analy-
sis has been seen to require the properties of being a proposition and being
bivalent.
A further criterion is as follows:
(C4) For any x, x is amenable to quantificational first-order analysis only

if it presupposes the existence of more than one item.
(C4) is an outcome of the standard set-theoretical semantics of quantifica-

tional first-order logic. We argue as follows:
(i) The semantics of first-order logic rules out that sets can be mem-
bers of their own sets, or, equivalently, that properties can be prop-
erties of themselves.
(ii) Since a set cannot be a member of itself, a (non-empty) set entails
the existence of both a set and a distinct member of that set; and,
equivalently, a property had by a bearer entails the existence of the
property and a distinct bearer.
(iii) Hence whatever is amenable to first-order analysis will presuppose
the existence of two distinct items.
In what follows I shall take (C1)–(C2) to be necessary and sufficient criteria for
amenability to propositional first-order analysis, and (C3)–(C4) to be necessary
criteria for amenability to quantificational first-order analysis. I now proceed
to argue that even if divine nature doctrines fulfil the requirements specified
in (C1)–(C2) and hence are amenable to propositional first-order analysis, they
don’t fulfil the requirements specified in (C4) and so fail to be amenable to
quantificational first-order analysis.
That classical theistic doctrines fulfil the requirements specified in (C1)–(C2),
might be thought obvious, although there might be room for some doubt. With
regard to the requirement specified in (C1) we reason as follows:
(i) Reality claims function as objects of propositional attitudes, and

thus qualify as propositions.
(ii) Divine nature doctrines involve claims about reality, and thus qual-
ify as propositions.
196 Kraal
(iii) Hence, divine nature doctrines fulfil the requirement specified in

(C1).
With regard to (C2) one could note that many attempts to show that there are
non-bivalent propositions involve controversial assumptions. For example,
attempts to show that propositions about future contingents are non-bivalent
typically presuppose libertarianism,19 and attempts to show that undecid-
able propositions are non-bivalent typically presuppose verificationism.20 An
exception would seem to be propositions expressed by sentences involving
vague predicates, i.e. predicates that seem both to apply and not to apply to
some object. There might perhaps be reason to think that predicates used to
express divine nature doctrines (e.g. “almighty,” “all-good,” “all-knowing”) are
sufficiently vague to induce non-bivalence. But as far as I know there has been
no substantial attempt to show that this is in fact the case, and so the matter
seems rather uncertain.
More serious problems emerge with regard to the quantificational level and
in particular criterion (C4). Divine nature doctrines can be shown to fail to
fulfil the requirement specified in (C4). We reason as follows:
(i) A proposition is amenable to quantificational first-order analysis

only if it fulfils the requirement specified in (C4).
(ii) Divine nature doctrines presuppose the existence of just one item.
(iii) Hence, divine nature doctrines don’t fulfil the requirement speci-
fied in (C4), and so aren’t amenable to quantificational first-order
analysis.
As can be seen, the conclusion, if correct, substantiates incongruism (of the

relevant sort).
A main point that may need some substantiation in the above argument
is the claim in (ii) that divine nature doctrines presuppose the existence of
only one item. This claim is an outcome of the framework significance of clas-
sical theism’s doctrine of divine simplicity, according to which God is identi-
cal to his intrinsic properties.21 Hence in holding e.g. that “God exists,” “God is
19 Cf. Jan Łukasiewicz, Jan Łukasiewicz: Selected Works, ed. L. Borkowski (Amsterdam and
London: North-Holland Publishing Company, 1970), 114–115.
20 Cf. Michael Dummett, Truth and Other Enigmas (Cambridge, Mass.: Harvard University
Press, 1978), 215–247.
21 See e.g. Augustine, The City of God, vol. 3, trans. D. Wiesen (Cambridge, Mass.: Harvard
University Press, 1968), 465–469; Thomas Aquinas, Summa Theologica, vol. 1, trans. the
almighty” and “God is all-good,” classical theism is not presupposing that God
is one thing and his existence or almightiness or all-goodness something else;
rather, it is presupposed that God is identical to his existence, almightiness and
all-goodness. Given this framework, the claim made in (ii) is correct: divine
nature doctrines presuppose the existence of only one item.
I conclude that the incongruity thesis is vindicated: divine nature doctrines
aren’t amenable to first-order analysis.
4 Formalism
A proponent of formalism might concede the incongruism of the preceding

section without abandoning the more basic idea of formalism. For example,
he or she might seek to develop an alternative system of formal logic which is
believed to overcome the problems that have been seen to attach to first-order
logic. In what follows I shall exemplify two formalistic responses to this sort,
both of which aim at accommodating divine nature doctrines within an alter-
native system of logic. I first develop a formal system that doesn’t entail (C4);
I call the resulting system CT-Logic (“Classical Theism Logic”). In a second step
I show that the non-standard logical system developed by the Polish logician
Stanisław Leśniewski in the 1920s doesn’t entail anything like (C4) either, and so
can also be taken to accommodate divine nature doctrines in the relevant way.
We start with CT-Logic. CT-Logic is a restricted and expanded first-order
logic. It is restricted in that it restricts the usual arguments and functions
of first-order logic to items other than God and the divine properties. It is
expanded in that it adds a conceptual apparatus to first-order logic intended
for the treatment of divine nature doctrines.
In developing CT-Logic I develop a formal language governed by CT-Logic.
I first provide a syntax for the symbols of this language. The syntax is similar to
that of first-order logic except that it is supplemented with an additional set of
arguments and an additional set of monadic functions, tailored to the analysis
of classical theistic doctrines.
English Dominican Province (London: Sheed & Ward, 1948), 19; Martin Chemnitz, Loci
Theologici, vol. 1, trans. J.A.O. Preus (St. Louis: Concordia Publishing House, 1989), 51; and
Francis Turretin, Institutes of Elenctic Theology, vol. 1, trans. G.M. Giger (Phillipsburg: P&R
Publishing, 1992), 193. See also Leftow, “Simplicity, divine,” in The Routledge Encyclopedia
of Philosophy, vol. 8, ed. E. Craig (London and New York: Routledge, 1998), 784–788; and
Kraal, First-Order Logic and Classical Theism (Uppsala: Uppsala University, 2010), 167–181.
198 Kraal
The logical constants of propositional first-order logic are adopted. Hence

we have propositional parameters (φ, ψ . . .), truth-functional logical constants
(¬, ∧, ∨, →, and ↔), and the absurdity sign ⊥. The syntax for constructing for-
mulas is also the same as in first-order logic.
Syntactical differences emerge on the quantificational level. We adopt from
first-order logic argument expressions α, β . . ., which we restrict for items other
than God (call them “arguments of type A”); quantifiers ∀ and ∃, whose range
we restrict to items denoted by arguments of type A; functions Ψ(x), Φ(x) . . .,
which we restrict for properties other than divine properties (call them “func-
tions of type A”); and a symbol D used to denote a non-empty domain of items
denotable by arguments of type A and items denotable by arguments of type
B (to be explained below). To these familiar, but restricted, symbols we add
the following: arguments ‘a’, ‘b’ . . ., which we use to denote God (we call them
“arguments of type B”); and function expressions P(x), Q(x) . . ., which we use
to express divine properties (call them “functions of type B”). The symbols of
type B provide us with a means for developing a logic in which rules that gov-
ern propositions about items other than God need not govern propositions
about God, and vice versa.
A syntax comprising arguments and functions of both type A and type B
can be provided as follows. We use the meta-variable ‘δ’ for any singular term
of type A or B if not otherwise specified, and the meta-variable ‘Δ’ for any func-
tion of type A or B if not otherwise specified. We then lay down that: δ is a term;
if Δ(x) is any predicate expression and δ any term, then Δ(δ) is a formula; if Δ(δ)
is a formula in which δ and Δ are of type A, and x is a variable that can take as
values items denoted by singular terms of type A, then ∀xΔ(x) and ∃xΔ(x) are
formulas.
A semantics for the propositional level of our CT-Logic language can be
adopted from first-order logic. A semantics for formulas of the quantificational
level is as follows. Let M be a model <D, I >, where D is a specified domain of
items including God as well as other things, and I is an interpretative function
assigning arguments (“names”) of types A and B to the items in D. The truth-
conditions for formulas of the form ‘Δ(δ)’ can be given thus (we use the meta-
variables ‘δ’ and ‘Δ’ as above):
‘M╞ Δ(δ)’ iff the property-bearer I(δ) has the property I(Δ).
Truth-conditions for quantified formulas can be given recursively:
‘M╞ ∀xΔ(x)’ iff M╞ Δ(δ) for any element I(δ) of type A.

‘M╞ ∃xΔ(x)’ iff M╞ Δ(δ) for at least one element I(δ) of type A.
We have now given a semantic interpretation of the symbols pertaining to the

quantificational level of our CT-Logic language.
We have now arrived at a tool for the formal analysis of divine nature doc-
trines such as “God exists” and “God is almighty” which does not presuppose
an ontological distinction between God and his properties. For let ‘a’ = “God”;
‘P(x)’ = “exists”; and ‘Q(x)’ = “is almighty.” “God exists” can now be analyzed as
P(a)
And “God is almighty” as
Q(a)
Neither the syntax nor the semantics of these respective formulas will force an
ontological distinction between God and his properties.
The inference rules of the propositional level of first-order logic can be
adopted into our CT-Logic language without qualification. The inference rules
of the quantificational level of first-order logic can also be adopted into our
CT-Logic language provided that their variables range only over items that
are denotable by arguments of type A and their functions are similarly all of
type A.
We turn next to Leśniewski. Leśniewski’s system of formal logic aims at
reflecting the structure of reality more adequately than the logic of Frege and
Russell.22 In 1956 Czesław Lejewski gave a classic exposition of this system,
which at the same time was an enrichment of it. I here refer to the resulting
system as “Leśniewski-Lejewski Logic.”23
Leśniewski-Lejewski Logic can be seen to not force an ontological subject/
predicate distinction on the subject-matter of predications, and is in this sense
a more adequate tool for the formal analysis of divine nature doctrines than
first-order logic. This can be seen as follows.24
22 For historical overviews of Leśniewski’s work and its aims, see Peter Simons, “Stanisław
Leśniewski,” in Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/entries/
lesniewski/), accessed Jan. 25, 2010; and Jan Woleński, “Lvov-Warsaw School,” in Stanford
Encyclopedia of Philosophy (http://plato.stanford.edu/entries/lvov-warsaw/), accessed
Jan. 25, 2010.
23 See Lejewski, “On Leśniewski’s Ontology,” in Lesniewski’s Systems: Ontology and Mereology,
ed. J.T.J. Srzednicki (The Hague: Martinus Nijhoff Publishers, [1956] 1984), 123–148.
24 The following exposition is based primarily on Lejewski, “On Leśniewski’s Ontology.”
200 Kraal
We start with the semantic notions of an “unshared name,” defined as a term

that names only one object; a “shared name,” defined as a term that names
more than one object; and a “fictitious name,” defined as a term which as to its
syntax behaves like an unshared or shared name but which fails to name any
object. With these semantic notions we can specify various alternative seman-
tic statuses of names (using ‘α’ and ‘β’ for names):
(a) ‘α’ is an unshared name and denotes an object.

(b) ‘α’ is a shared name and denotes an object.
(c) ‘α’ is a fictitious name and denotes no object.
(d) ‘α’ and ‘β’ are unshared names and denote an object.
(e) ‘α’ and ‘β’ are unshared names and denote distinct objects.
(f) ‘α’ is an unshared name and denotes an object, and ‘β’ is a shared
name denoting several objects, one of which is denoted by ‘α’.
(g) ‘β’ is an unshared name and denotes an object, and ‘α’ is a shared
name denoting several objects, one of which is denoted by ‘β’.
(h) ‘α’ is an unshared name and denotes an object, and ‘β’ is a shared
name denoting several objects, none of which is denoted by ‘α’.
(i) ‘β’ is an unshared name and denotes an object, and ‘α’ is a shared
name denoting several objects, none of which is denoted by ‘β’.
(j) ‘α’ is an unshared name denoting an object, and ‘β’ a fictitious name
denoting no object.
(k) ‘β’ is an unshared name denoting an object, and ‘α’ a fictitious name
denoting no object.
(l) ‘α’ and ‘β’ are shared names denoting the same objects.
(m) ‘α’ is a shared name denoting several objects, and ‘β’ a shared name
denoting all the objects denoted by ‘α’ in addition to at least one
more object.
(n) ‘β’ is a shared name denoting several objects, and ‘α’ is a shared
name denoting all the objects denoted by ‘β’ in addition to at least
one more object.
(o) ‘α’ and ‘β’ are shared names denoting several objects each, some of
which are denoted by both names and some of which are not.
(p) ‘α’ and ‘β’ are shared names denoting several objects each, none of
which is denoted by both.
(q) ‘α’ is a shared name denoting several objects, and ‘β’ a fictitious
name denoting no object.
(r) ‘β’ is a shared name denoting several objects, and ‘α’ a fictitious
name denoting no object.
(s) ‘α’ and ‘β’ are fictitious names both of which denote no object.
A predication, say “God is almighty,” can be analyzed in four different ways,

corresponding to four different explications of “is.” Let ‘α is β’ represent any
predication. The four possible analyses can be given by means of truth-condi-
tion specifications as follows:
‘α ε β’ is true iff (d) or (f) give the status of ‘α’ and ‘β’, otherwise it’s false.
‘α < β’ is true iff (d), (f), (l) or (m) give the status of ‘α’ and ‘β’, otherwise it’s
false.
‘α ⊂ β’ is true iff (d), (f), (k), (l), (m), (r) or (s) give the status of ‘α’ and ‘β’,
otherwise it’s false.
‘α Δ β’ is true iff (d), (f), (g), (l), (m), (n) or (o) give the status of ‘α’ and ‘β’,
otherwise it’s false.
Identity claims can be explicated in three ways:
‘α = β’ is true iff (d) gives the status of ‘α’ and ‘β’, otherwise it’s false.
‘α ™ β’ is true iff (d) or (l) give the status of ‘α’ and ‘β’, otherwise it’s false.
‘α ○ β’ is true iff (d), (l) or (s) give the status of ‘α’ and ‘β’, otherwise it’s false.
Existence claims can likewise be explicated in three ways:
‘ex(α)’ is true iff (a) or (b) gives the status of ‘α’, otherwise it’s false.
‘sol(α)’ is true iff (a) or (c) gives the status of ‘α’, otherwise it’s false.
‘ob(α)’ is true iff (a) gives the status of ‘α’, otherwise it’s false.
Leśniewski-Lejewski Logic can now be seen to not force an ontological subject/

predicate distinction on predications. For let “God is almighty” be formalized
as ‘α ⊂ β’. ‘α ⊂ β’ is consistent with each of the identity claims ‘α = β’, ‘α ™ β’ and
‘α ○ β’, since they are all true if (4) obtains. Hence, Leśniewski-Lejewski Logic
allows for an identity between the objects denoted by ‘God’ and ‘almighty’, and
the same goes mutatis mutandis for other divine attributes.
5 Anti-Formalism
Apart from “formalistic” responses to incongruism of the above sort, it is also

possible to offer an anti-formalistic response by resorting to non-formal logic
when analyzing divine nature doctrines.
By “non-formal logic” I mean (briefly put) the logic that (i) underlies formal
logic, (ii) is presupposed in correct thinking, and (ii) can be expressed in natu-
ral languages such as English.
202 Kraal
Non-formal logic comprises concepts such as implication (“implies”), nega-

tion (“not”), disjunction (“or”), conjunction (“and”) and equivalence (“if and
only if”), which underlie, but are not identical to, first-order logic’s logical con-
stants →, ¬, ∨, ∧, and ↔. Non-formal logic also comprises logical principles such
as the principle that propositions cannot be both true and false (“the law of
non-contradiction”), which underlies the first-order formula ‘¬(φ ∧ ¬φ)’; and
inference rules, such as that if a conjunction of two propositions is true then
each conjunct is individually true (“conjunction elimination”), which under-
lies the first-order inference rule ‘φ ∧ ψ 𐅂 φ’.
In what follows I shall give some examples of non-formal analyses of divine
nature doctrines. In the next section I proceed to argue that these sorts of anal-
yses are more adequate than formal logical analyses.
The examples I shall give correspond to the first-order analyses of the
divine nature doctrine “God exists” surveyed at the beginning of this paper, i.e.
‘Exists(God)’, ‘∃x(God(x))’, ‘∃x(God(x) ∧ Exists(x))’, ‘∃x(x = God)’ and ‘∃(y)∀(x)
(x=y ↔ God(x))’. The examples are as follows:
(i) In the doctrine “God exists” the term “God” is used to denote God
and the term “exists” is used to express the property of existing, and
the doctrine itself asserts that God has the property of existing.
(ii) In the doctrine “God exists” the term “God” is used to express the
property of being God, and the doctrine itself asserts that there
exists an object that has this property.
(iii) In the doctrine “God exists” the term “God” is used to express the
property of being God and the term “exists” is used to express the
property of existing in reality, and the doctrine itself asserts that
there exists an object that has the property of being God and the
property of existing in reality.
(iv) In the doctrine “God exists” the term “God” is used to denote God,
and the doctrine itself asserts that there exists an object that is
identical with God.
(v) In the doctrine “God exists” the term “God” is used to express the
property of being God, and the doctrine itself asserts that there
exists exactly one object that is identical to God.
The above analyses do not themselves involve logical concepts such as impli-
cation, negation, disjunction, and so on. However, on the basis of these analy-
ses we can derive various logical features of the claims, e.g. these ones:
(i*) The doctrine “God exists” implies that existence is a property, and is
inconsistent with the claim that existence is not a property.
(ii*) The doctrine “God exists” is consistent with the claim that there are
properties.
(iii*) The doctrine “God exists” is inconsistent with the claim that there
are no properties.
(iv*) The doctrine “God exists” implies that something is identical to
God, and is inconsistent with the claim that nothing is identical to
God.
(v*) The doctrine “God exists” is consistent with the claim that there is
no more than one God, and is inconsistent with the claim that there
is more than one God.
We see, then, that various logical properties of the doctrine “God exists” can be
derived from the non-formal logic analyses offered in (i)–(v).
It can easily be seen that the non-formal logical analyses (i)–(v) are just
as informative as their formalistic counterparts ‘Exists(God)’, ‘∃x(God(x))’,
‘∃x(God(x) ∧ Exists(x))’, ‘∃x(x = God)’, and ‘∃(y)∀(x)(x=y ↔ God(x))’. But, con-
trary to their formalistic counterparts, they do not presuppose an ontological
distinction between God and the divine properties. The reason why the formal-
istic counterparts presupposed such a distinction was because standard first-
order semantics forces such a distinction. In the case of (i*)–(v*), however,
there are no analogous semantic rules to which the explications are bound,
and hence no equivalent ontological distinction between God and his proper-
ties can be inferred.
I now seek to vindicate anti-formalism by arguing that the above non-for-
malistic response to the incongruity thesis is more adequate than the formalis-
tic responses of section 4. I offer three reasons for thinking this.
First, non-formal logic is epistemically more secure than formal logic. This
can be made clear by examples: the non-formal principle “no proposition
is both true and false” is not certain because it accords with the first-order
formula ‘¬(φ ∧ ¬φ)’, but, rather, the first-order formula ‘¬(φ ∧ ¬φ)’ is certain
because it accords with the non-formal principle “no proposition is both true
and false.” Again, the non-formal principle “every proposition is either true or
false” is not certain because it accords with the first-order formula ‘φ ∨ ¬φ’,
but, rather, the first-order formula ‘φ ∨ ¬φ’ is certain because it accords with
the non-formal principle “every proposition is either true or false.” Granting
that an epistemically more secure logic is more adequate than an epistemically
less secure one, it follows (all else being equal) that non-formal logic is more
adequate than formal logic.
Second, non-formal logic has greater expressive resources than formal logic.
The principles and inference rules of a formal logic can all be expressed also in
non-formal logic, but not all principles and inference rules of non-formal logic
204 Kraal
can be expressed in a formal logic such as first-order logic. For example, the
non-formal logical principle “anything implied by a true proposition is true”
cannot be adequately expressed in formal logic, as e.g. Whitehead and Russell
point out.25 Granting that a logic with greater expressive resources is more
adequate than a logic with more limited expressive resources, it follows (all
else equal) that non-formal logic is more adequate than formal logic.
And third, the real strong-points of formal logic vis-à-vis non-formal logic
are largely irrelevant to the task of getting clear about the conceptual content
of divine nature doctrines. Three well-known strong-points of formal logic are
that it (i) provides a means for compressing long and complicated proposi-
tions into brief formulas that are easy to grasp, (ii) that it can be used in com-
puter programming, and (iii) that it allows for interesting meta-logical results
such as Gödel’s completeness and incompleteness theorems. These are indeed
strong-points; but, as can be seen, they are largely irrelevant to the task of get-
ting clear about the conceptual content of divine nature doctrines. Divine
nature doctrines are not of such a long and complicated sort as to benefit in
any obvious way from being compressed into symbolic formulas, and facility in
computer programming along with meta-logical theorems, important as they
are, are of no direct or obvious relevance to the task of getting clear about the
logic of divine nature doctrines.
For the above three reasons I think that anti-formalism is to be preferred
to formalism when it comes to the conceptual analysis of divine nature doc-
trines. My suggestion, therefore, is that we abandon the use of formal logic in
seeking to get clear about the conceptual content of divine nature doctrines,
and instead go for with non-formal logic.
6 Conclusion
In the foregoing I have defended versions of incongruism and anti-formalism.

According to the relevant version of incongruism, divine nature doctrines
aren’t amenable to standard first-order analysis. According to the relevant ver-
sion of anti-formalism, divine nature doctrines are better analyzed by means
of non-formal logic than by means of formal logic.
The relevant version of incongruism was argued for by developing four
plausible criteria for amenability to first-order analysis, and then arguing that
divine nature doctrines fail to meet at least one of these criteria. The criterion
that divine nature doctrines fail to meet is that of presupposing the existence
25 Whitehead and Russell, Principia, 98–99.

of more than one item. This, I argued, is a presupposition that classical theism
does not meet due to the framework significance of classical theism’s doctrine
of divine simplicity.
The relevant version of anti-formalism was argued for by first proposing
two responses to the incongruity thesis, one formalistic and the other non-
formalistic, and then arguing that the non-formalistic response is the more
adequate response. The non-formalistic response was deemed more adequate
than the formalistic response in that it is epistemically more secure, has greater
expressive resources, and because the strong-points of first-order logic vis-à-vis
non-formal logic are largely irrelevant to the task of getting clear about the
conceptual content of divine nature doctrines.
Should the results of the present paper be considered surprising? Perhaps
not. After all, first-order logic was not developed for the formal analysis of
divine nature doctrines, but for the formal analysis of mathematical proposi-
tions. So if it turns out that first-order logic is conceptually inadequate for the
logical analysis of divine nature doctrines, we shouldn’t be too surprised. On
the other hand, since the use of first-order logic as a tool for the conceptual
analysis of divine nature doctrines has become widespread in contemporary
analytic philosophy, our results, if sound, are not insignificant.26
Bibliography
Wilhelm Ackermann and David Hilbert, Grundzüge der theoretischen Logik. (Berlin/
Göttingen/Heidelberg: Springer, 1949).
Augustine. The City of God, vol. 3, trans. D. Wiesen (Cambridge, Mass.: Harvard
Thomas Aquinas, Summa Theologica, vol. 1, trans. the English Dominican Province
(London: Sheed & Ward, 1948).
Józef Bocheński, The Logic of Religion (New York: New York University Press, 1965).
Martin Chemnitz, Loci Theologici, vol. 1, trans. J.A.O. Preus (St. Louis: Concordia
Publishing House, 1989).
Michael Dummet, Truth and Other Enigmas (Cambridge, Mass.: Harvard University
Press, 1978).
Gottlob Frege, “Begriffsschrift,” trans. Michael Beaney, in The Frege Reader, ed.
M. Beaney. (Oxford: Blackwell, 1997), 47–78.
Kaj B. Hansen, Grundläggande logik (Lund: Studentlitteratur, 1997).
26 Many thanks to Kaj Børge Hansen, of Uppsala University, for helpful remarks on an earlier
version of this paper.
206 Kraal
———, “Formal logic, Models, Reality,” in Neither/Nor, eds. R. Sliwinski and F. Svensson
(Uppsala: Uppsala Philosophical Studies, 2011), 79–99.
Wilfred Hodges, “Classical Logic I—First-Order Logic,” in The Blackwell Guide to
Philosophical Logic, ed. L. Goble (Oxford: Blackwell Publishing, 2008), 9–32.
Anders Kraal, First-Order Logic and Classical Theism: Toward Logical Reorientation.
(Uppsala: Uppsala University, 2010).
———, “Logic and Divine Simplicity,” Philosophy Compass 6:4 (2011): 282–294.
———, “Classical theism,” in Springer Encyclopedia of Sciences and Religions
(forthcoming).
———, “Theism, Classical” in Encyclopedia of Sciences and Religions, vol. 4, eds.
A. Runehov and L. Oviedo (Springer: New York and Dordrecht, 2013), 2239–2240.
Czeław Lejewski, “On Leśniewski’s Ontology,” in Lesniewski’s Systems: Ontology and
Mereology, ed. J.T.J. Srzednicki (The Hague: Martinus Nijhoff Publishers, [1956]
1984), 123–148.
Brian Leftow, “God, concepts of,” in The Routledge Encyclopedia of Philosophy, vol. 4,
ed. E. Craig (London and New York: Routledge, 1998), 93–102.
———, “Simplicity, divine,” in The Routledge Encyclopedia of Philosophy, vol. 8, ed.
E. Craig (London and New York: Routledge, 1998), 784–788.
Jan Łukasiewicz, Jan Łukasiewicz: Selected Works, ed. Borkowski. L. Amsterdam and
(London: North-Holland Publishing Company, 1970), 110–128.
William Mann, “Definite Descriptions and the Ontological Argument,” in Philosophical
Applications of Free Logic, ed. K. Lambert (Oxford: Oxford University Press, 1991),
257–272.
Francis Jeffry Pelletier, “A Brief History of Natural Deduction,” History and Philosophy
of Logic 20:1 (1999): 1–31.
Alvin Plantinga, “Self-Profile,” in Alvin Plantinga, eds. J.E. Tomberlin and Peter van
Inwagen (Dordrecht: Reidel, 1985), 3–97.
W.V.O. Quine, Mathematical Logic (Cambridge, Mass.: Harvard University Press, 1951).
Peter Simons, “Stanisław Leśniewski,” in Stanford Encyclopedia of Philosophy, 2011,
accessed May 21, 2013. http://plato.stanford.edu/entries/lesniewski/.
Robert R. Stoll, Set Theory and Logic (San Francisco and London: W.H. Freeman and
Company, 1963).
Francis Turretin, Institutes of Elenctic Theology, vol. 1, trans. G.M. Giger (Phillipsburg:
P&R Publishing, 1992).
A.N. Whitehead and Bertrand Russell. Principia Mathematica, vol. 1 (Cambridge:
Jan Woleński, “Theism, Fideism, Atheism, Agnosticism,” in Logic, Ethics and All That
Jazz: Essays in Honour of Jordan Howard Sobel, eds. L.-G. Johansson, J. Österberg and
R. Sliwinski (Uppsala: Uppsala Philosophical Studies, 2009), 387–400.
———, “Lvov-Warsaw School,” in Stanford Encyclopedia of Philosophy, 2009, accessed
Jan. 25, 2010. http://plato.stanford.edu/entries/lvov-warsaw/.
chapter 11
Zombies, Selves, and the Possibility of Afterlife

Johan Modée
1 Introduction
The history of human beings is crowded with all kinds of strange, imagined
beings, such as gods, ghosts, souls, witches, fairies, trolls, zombies, gnomes and
so forth. As the archaeological evidence indicates, the consciousness of human
beings achieved the cognitive capability to conceive such beings at least 30,000
years ago.1 After that, the human beings have frequently imagined that their
cognitive imaginations are cases of existent beings. In fact, it has been assumed
that the existence of such beings is more certain than anything.
Another important historic or even prehistoric conception, still prevalent in
many religions, is the idea of afterlife. The basic idea is here that the identity
of an individual is preserved after death. In other words: it is claimed that it is
logically possible for the self to exist post mortem.
Prima facie, we can all, as human beings, conceive that there is a spiritual
substance that can take different forms as soul or divine being. From our cul-
tures, we have historical conceptions of divine beings and human souls. Thus
is may appear to most of us that we actually can grasp the existence of such
beings and things, as representations in our minds. From this, some or even
most of us conclude that they (possibly) also (can) exist. Thus, in this cogni-
tive-historical sense, selves, logic and ontology are strongly connected in the
context of religion. It is natural for human beings to conceive supernatural
things.
Now there is no obvious empirical evidence for the existence of any afterlife,
divine being, or spiritual substance whatsoever. The importance of this fact
cannot be overrated.2 By contrast, the empirical evidence for the existence of
a physical world is overwhelming. But in the face of this evidence and lack of
evidence, at the same time, the absolute majority of people believe that there
are spiritual substances and afterlife. It is a natural, cultural position to see the
1 Steven Mithen, The Prehistory of Mind: A Search for the Origin of Art, Religion and Science
(London: Thames & Hudson 1996).
2 See the discussion in John Searle, Mind, Language and Society: Philosophy in the Real World
(London: Phoenix 1999), 33–37.

208 Modée
world as partly spiritual, in the sense that there is a spiritual substance, and
that human beings are partly spiritual. This is indeed a strange situation. The
Enlightenment and modern science came out very hard on religious ideas. The
many times iterated point was and still is of course that there is no empirical
evidence of spiritual substances—at least not the kind of evidence that the
scientific view requires. The contemporary prevalence of religion is, however,
contradicting this. Religion still has its strongholds, despite the Enlightenment
critique. But I cannot go into any details here, as that would go beyond the
scope of this paper.
Another consideration is my real concern, and that is the status of our cul-
tural discourses. Since the dawn of human history, philosophers and theolo-
gians have tried to demonstrate the possible or even necessary existence of
some divine being, soul or spiritual substance by the use of logical tools. A
related attempt has been to clarify these concepts, showing that they do not
entail any contradiction. The absence of logical contradictions is then taken as
a platform for ontological assumptions.
It is unclear whether or not this discourse has contributed to a situation
where religion has made a comeback or in the contemporary world. One
aspect of that discourse is nonetheless the focus of the present chapter.
2 Zombies and the Philosophical Zombie Argument
One way to classify all these philosophical-theological attempts to demon-

strate the existence of a non-physical reality is to see them as conceivability
arguments, similar to zombie arguments in the philosophical literature.
What are zombies? Zombies are historical-cultural creatures of imagina-
tion, which have their natural place in a religious context. Zombies are corpses
that become alive. Such creatures appear in a prophecy by Ezekiel in the Old
Testament (cf. Ezekiel 37: 1–10). Another context in which they emerge as con-
ceptions is Haiti voodoo religion.
In modern times, zombies have been made world-famous by various zom-
bie movies. The ghouls in George Romero’s’ classic 1968 film The Night of
the Living Dead initiated a movie genre that recently got its latest contribu-
tion in Marc Foster’s blockbuster World War Z (2013), starring Brad Pitt. The
zombie creature in these movies, typically, is a corps that is still alive, in some
mysterious way. In Romero’s 1968 movie corpses became zombies due some
unknown natural or non-natural cause and they move around very slowly in
large crowds, looking for fresh bodies and brains to consume. Contemporary
Zombies, Selves, and the Possibility of Afterlife 209
zombie movies have basically the same conception but here the ghouls often
move incredibly fast.
The zombie of the philosophers, by contrast, is a less dangerous being. It is a
human being without consciousness. The philosophical debate about zombies
is mainly related to the mind-body discussion, issues concerning the debate on
physicalism in general, and—more recently—the debate about conceivability
arguments as such.3
A historical source for these conceivability arguments is Hume’s famous
claim in his Treatise of Human Nature: “[. . .] whatever the mind clearly con-
ceives, includes the idea of possible existence, or in other words, that noth-
ing we imagine is absolutely impossible.”4 Hume claims here that a coherent
conception includes the idea of possible existence, and also that in fact no
imagination of the mind can be without that concept. Descartes famous cogito
is dependent on the same assumption, that conceptual or coherence conceiv-
ability leads to ontological conclusions.5 And, of course, we have the same
assumption behind Anselm’s ontological proof. All these arguments go from
the conceivability of a concept to possible or even necessary existence.
Similarly, the contemporary, basic conceivability argument for zombies has
the following logical form:
1. Zombies are conceivable.

2. Whatever is conceivable is (ontologically) possible.
3. Therefore zombies are (ontologically) possible.
This argument is valid. But the question is if the premises are acceptable. Both
can be challenged. The usefulness of the conceivability argument can also be
disputed, even if Hume claims that this form of argument is “an establish’d
maxim in metaphysics.”6
3 David Chalmers, The Conscious Mind: In Search of a Fundamental Theory (New York: Oxford
University Press 1996); Stephen Yablo, “Is Conceivability a Guide to Possibility?,” Philosophy
and Phenomenological Research 53 (1993), 1–42; Stephen Yablo, “Concepts and Consciousness,”
Philosophy and Phenomenological Research 59 (1999), 455–463; Stephen Yablo, “Textbook
Kripkeanism & The Open Texture of Concepts,” Pacific Philosophical Quarterly 81 (2000),
98–122.
4 David Hume, A Treatise of Human Nature, eds. L.A. Selby-Bigge and P.H. Nidditch (Oxford:
Oxford University Press 1968), 32; cf. Yablo, “Conceivability.”
5 Yablo, “Conceivability.”
6 Hume, Treatise, 32.
210 Modée
3 Gods, Afterlife, and Zombies
The driving idea of this chapter is that arguments for the possible existence
of divine beings basically have the same structure as the basic zombie argu-
ment or that they presuppose a similar kind of argument. A clear example
from the philosophical-theological literature, supporting this idea, is Richard
Swinburne’s different arguments for substance dualism, which he has pre-
sented in several works. But here I want to focus on a particular argument that
he presented in, e.g., Personal Identity7 and in The Evolution of the Soul.8
The first step in that particular argument is a reflection over the “logical pos-
sibility” of afterlife. Swinburne claims that it is generally agreed, at least among
people that are “uninfluenced by philosophical theory,”9 that the notion after-
life is coherent, which means that it is conceptually consistent. There is no
contradiction in the idea that a person may survive his or her own death,
according to Swinburne. Then he says this: “From the mere logical possibility
of my continued existence [i.e., in afterlife, without a body] there follows the
actual fact that there is now more to me than my body; and that more is the
essential part of myself.”10 The argument as such aside,11 the idea behind it is
clearly the philosophical idea that a coherent conception (“logical possibility”)
entails that corresponding existence is metaphysically possible.
I think that this broadly logical move is commonplace in theology and prac-
tical religion, even if the philosophical argument is not at hand. Taking gods as
existent beings, in the absence of direct empirical confirmation, presupposes
that conceivability is a guide to ontological possibility or even to “the truth”
itself. In the absence of evidence, adherents are also quite satisfied with con-
ceivability, which, at best, can be empirically supported by some theological
interpretation of facts (e.g., the “creation,” “afterlife,” etc.). So it could be argued
that religion and philosophical arguments about its rationality are dependent
on some version of the conceivability argument.
Now the metaphysical idea of conceivability arguments, that conceivability
indicates, guide or entails possible existence, has been challenged. I will first
outline the debate, from which we have that attack, and then discuss the impli-
cation of the attack for Swinburne’s argument.
7 Sidney Shoemaker and Richard Swinburne, Personal Identity (Oxford: Blackwell, 1984).
8 Richard Swinburne, The Evolution of the Soul (Oxford: Clarendon Press, 1997).
9 Shoemaker and Swinburne, Personal, 244.
11 See Nicholas Everitt, “Substance Dualism and Disembodied Existence,” Faith & Philosophy
17 (2000), 331–347, for a critical discussion.
4 The Conceivability Debate
Hume’s claim that it was an established maxim in metaphysics that conceiv-

ability entails possibility was not without opponents. For instance, as Yablo has
shown, John Stuart Mill doubted the value of conceivability: “[. . .] our capacity
or incapacity of conceiving a thing has very little to do with the possibility of
the thing in itself; but is in truth very much an affair of accident, and depends
on the past history and habits of our own minds.”12
Even if doubts like Mill’s were well known, the logic behind the conceivabil-
ity argument was first seriously challenged in the early 1970s, with Kripke’s and
Putnam’s work on reference and necessity.13 Already in these works, we find
ideas that later inspired other philosophers to develop new stances, breaking
with the established dogmas. For example, in his 1975 paper, Putman writes:
“we can perfectly well imagine having experiences that would convince us (and
that would make it rational to believe) that water is not H2O. In that sense, it is
conceivable that water isn’t H2O.”14
Thus it is perfectly possibly for us to conceive that water is not H2O, but
something with an entirely different substance. But that is, given the argument
in Putman’s paper, an a posteriori necessary falsehood. Putman’s conclusion,
contrary to Hume’s claim, is therefore that (logical) conceivability is no proof
of (metaphysical) possibility.
Bringing that conclusion to our current case, we can of course conceive
all sorts of things—such as ghosts, zombies, witches, fairies, gnomes, etc.—
but from that only it does not follow that these things are (metaphysically)
possible.
The current debate, which frequently focuses on the conceivability of zom-
bies and what that could entail, is best represented by, one the one, extreme
side, David Chalmers, and, on the other side, a number of philosophers that
dispute Chalmers’ view. I can only summarize the main details of this debate
below.
Chalmers argues that zombies are conceivable, and that, therefore, that
mainstream reductionist materialism or physicalism about consciousness is
false. Chalmers’ alternative physicalist position is his “naturalistic dualism,”
which adds panpsychism to the physicalist doctrines about cognition (e.g.,
12 Mill, quoted in Yablo, “Conceivability,” 2.

13 Saul Kripke, Naming and Necessity (Cambridge, ma: Harvard University Press, 1980);
Hilary Putnam, “The Meaning of Meaning,” in Philosophical Papers, Volume 2: Mind,
Language and Reality (Cambridge: Cambridge University Press, 1975).
14 Putnam, “Meaning,” 233.
212 Modée
functionalism, strong AI). Panpsychism is the doctrine that consciousness is

everywhere where information is processed. And that is basically everywhere
in the strong sense, i.e., in thermostats, rocks, and all.15
The argument behind this view rests entirely on the zombie conceivability
argument.16 Quoting Searle’s presentation of Chalmers’ view:
If it is logically possible, in the sense of not being self-contradictory, to

imagine that there could be zombies that where organized just like we
are and had exactly our behavior patterns, but were totally devoid of con-
sciousness, then it follows that our consciousness cannot logically con-
sist simply in our behavior or functional organization.17
So if it is correct that zombies are logically possible, some version of dualism

must be true, e.g., Chalmers panpsychism, which he sees as a “strangely beauti-
ful” alternative.18
Arguing for the conceivability of zombies, Chalmers uses a theory that he
calls “two-dimensional semantics.”19 In this theory, there are two kinds of con-
ceivability (or “intension”), which correspond to two kinds of possibility. In the
first kind, conceivability entails (logical) possibility, for example that water is
not H2O. Here if water = watery stuff in one world, so it could have been the
a posteriori case that it turned out to be something else than H2O. By contrast,
the second kind does not allow for this unless that that the watery stuff = H2O,
because this is a posteriori true in all possible worlds once the connection is
established.20 Here is a necessary falsehood to claim that water could be some-
thing else than what it is. To claim that something is possible in the second
sense is thus to claim that something is possible across worlds. Another way
of saying this is to say that the first kind of possibility is a conceptual (logical)
possibility, while the second is the metaphysical possibility.
With consciousness, however, Chalmers thinks that the first and the sec-
ond sense of possibility coincide: “[. . .] with consciousness, the primary and
15 Chalmers, Conscious Mind, 297.

16 John Searle, The Mystery of Consciousness (London: Granta Books 1998), 146; Yablo,
“Concepts.”
17 Searle, Mystery, 146.
18 Searle, Mystery, 166.
19 Chalmers, Conscious Mind; David Chalmers, The Character of Consiousness (Oxford:
20 Cf. Putnam, “Meaning.”
secondary intensions coincide [. . .] if something feels like a conscious experi-

ence, even in some counterfactual world, it is a conscious experience.”21
Therefore, we could say that Chalmers thinks that it is enough to have
conceived zombies in the first sense to arrive at second sense in which the
metaphysical doctrine could be that consciousness is everywhere where infor-
mation apparently is causally processed.
Stephen Yablo outlines the argument as follows:
(a) it is conceptually possible that there be zombies, so [. . .] [given Chalm-

ers’ use of two-dimensional semantics]
(a’) there are worlds in the primary intension of “there are zombies,” [i.e., the
first sense of conceivability] so
(a”) there are worlds which if actual make ‘there are zombies’ true, so
(b”) there are conceptually possible zombie worlds [i.e., the second, de re
sense].22
There are many ways to challenge Chalmers’ view, either details of it or as a

whole. One, detailed way is to reject the initial assumption or claim that zom-
bies are conceivable or conceptually possible.
That is, for instance, Daniel Dennett’s approach.23 Dennett thinks that those
who claims that zombies are conceivable haven’t really conceived them as
clear as one could reasonably demand. Consciousness, Dennett says, is a very
complex thing, which is not like some easily, conceivable thing that can be
moved from one place to another. In other words, what is it that we actually
conceive when we conceive a being that is exactly like us except that it lacks
consciousness? Does that proposition make any sense at all, if we really think
hard about what it express?
I think Dennett’s argument is stronger than it seems. It is very, very easy to
say, “look, here’s a being exactly like us but without consciousness,” and then
claim that it is a logically or conceptually possible conception. The easiness in
stating strange things and the fact that there is no obvious contradiction in the
claim give misleadingly the impression that the claim is conceptually clear. But
what the conceivability argument actually and perhaps only can claim, on this
view, is that it is possible to assert that there are zombies in the same way that
21 Chalmers, Conscious Mind, 133.

22 Yablo, “Textbook Kripkeanism,” 110 (my brackets).
23 Daniel Dennet, “The Unimagined Preposterousness of Zombies,” Journal of Consciousness
Studies 2 (1995), 322–326; Daniel Dennet, “The Zombic Hunch: Extinction of an Intuition?,”
Royal Institute of Philosophy Millennial Lecture (1999).
214 Modée
it is possible to assert that there is a square triangle. It has not been shown that
we actually can conceive that sort of being just because it is possible to claim
that it is conceivable when, in fact, it is perhaps just possible assert that there
is a being or thing such as such. For instance, it is very difficult to explain a
zombie being in detail and show how this being actually works (without con-
tradiction). It is easy to submit fuzzy and hasty statements about imagined
beings and reflect very superficially on what we actually say, but—on a closer
examination—it is clear that things are not at all that easy. Fuzzy or vague
concepts can hardly be claimed to be “conceptually possible” without detailed
specification.
Consider the example of Santa Claus, who delivers Christmas gifts to all
children in a few hours. This is easy to say and (in a superficial sense, also con-
ceive, since “all” is a vague word), but Santa’s logistics is of course impossible in
practice. But consciousness is surely an even harder case, compared with Santa
and his impossible logistics. Add a beard, some typical gear including a deer
and you have Santa Claus. And that he can deliver gifts to all children in e few
hours goes of course without saying.
It is far more difficult to conceive the case where you can remove conscious-
ness from a person and still have exactly the same behaviour intact. For exam-
ple, imagine the situation where we experience the smell of cheese with some
verbal approval and the case where the zombie behaves exactly like us. What
sense could be made of the zombie behaviour? It has no consciousness, so how
could it express that it enjoyed the smell of cheese? Is it a lie or what is it? But
how can a zombie lie? Lies seem to presuppose consciousness. If you have no
consciousness you cannot possibly lie.24 Not even the typical zombie behavior
makes sense, if the zombie is entirely devoid of consciousness. A zombie must
be aware of the presence of human brains and be in the position of distinguish
non-zombies from zombies. That awareness seems to entail consciousness.
Another way to reject Chalmers’ position is to challenge the conceivabil-
ity argument as such. This is, for instance, Yablo’s approach.25 For example,
given the physicalist assumption that the world is entirely physical and that
everything apparently non-physical can be reduced to the physical substance,
it is clear that the consistent physicalist must formulate the position so that
zombies are not possible in the metaphysical sense. As Yablo says, it “is not
24 These arguments are developed in Allin Cotrell, “Sniffing the Camembert: On the
Conceivability of Zombies,” Journal of Consciousness Studies 6 (1999), 4–12; and in Nigel
Thomas, “Zombie Killer,” in Toward a Science of Consciousness II, eds. Stuart R. Hameroff,
Alfred W. Kaszniak & A.C. Scott Cambridge (Cambridge, ma: mit Press, 1998), 171–177.
25 Yablo, “Conceivability”; Yablo, “Concepts.”
counterintuitive in the least” to maintain that position for a physicalist, since

physicalism, typically, is an ontological thesis that claims to give a complete
account of consciousness and close the door for any kind of dualism.
5 The Conceivability of Afterlife
The metaphysical possibility of afterlife is very important in many religions.

Christianity and Islam are two cases where the conception of afterlife is
strongly connected to the human self and the identity of the individual. The
claim is that personal identity can be preserved beyond the physical death of
the person. Buddhist and Hindu doctrines of the afterlife are usually seen as
less oriented towards the self and personal identity, even if such matters are
not always clearly articulated in Buddhist and Hindu folk theology.26 Setting
the all the possible differences and local variations aside, it is obvious that the
doctrine of afterlife itself is central in all these world religions. In fact, present
life is of less importance in these contexts. It is the non-physical afterlife that is
the real or basic life. The physical life is a path to the eternal goal.
The human self, according to Christian and Muslim tradition, is composed
of a physical part and a non-physical part. The latter is usually called the soul.
Of greater importance, however, is the divine substance related to the non-
physical part of the self. The divine is a spirit, and the human soul has its origin
in that substance.
One traditional idea seems to connect the individual and human breath to
the eternal life that has its origin or home in or in the presence of the divine
spirit. Quoting Ezekiel’s “zombie prophecy” (37: 1–10, NIV):
The hand of the Lord was on me, and he brought me out by the Spirit of
the Lord and set me in the middle of a valley; it was full of bones. 2 He led
me back and forth among them, and I saw a great many bones on the
floor of the valley, bones that were very dry. 3 He asked me, ‘Son of man,
can these bones live?’ I said, ‘Sovereign Lord, you alone know.’ 4 Then he
said to me, ‘Prophesy to these bones and say to them, “Dry bones, hear
the word of the Lord! 5 This is what the Sovereign Lord says to these bones:
I will make breath enter you, and you will come to life. 6 I will attach ten-
dons to you and make flesh come upon you and cover you with skin; I will
26 John Hick, Death and Eternal Life (Louisville: Westminster 1994); Melford Spiro, “Religion:
Problems of Definition and Explanation,” in Anthropological Approaches to the Study of
Religion, ed. Michael Banton (London: Tavistock 1966).
216 Modée
put breath in you, and you will come to life. Then you will know that I am
the Lord.”’ 7 So I prophesied as I was commanded. And as I was prophesy-
ing, there was a noise, a rattling sound, and the bones came together,
bone to bone. 8 I looked, and tendons and flesh appeared on them and
skin covered them, but there was no breath in them. 9 Then he said to me,
‘Prophesy to the breath; prophesy, son of man, and say to it, “This is what the
Sovereign Lord says: Come, breath, from the four winds and breathe into
these slain, that they may live.”’ 10 So I prophesied as he commanded me,
and breath entered them; they came to life and stood up on their feet—a
vast army [my italics].
Ezekiel’s description of bones that become a zombie army shows us the steps
that make the impression that something actually is conceived and possible.
To bones he adds tendons and flesh and finally breath. Hence: we go from
bones to a zombie conception, and even a “vast army.” All stages can be con-
ceived, thus the whole composition is allegedly conceived. Life is possible and
partly non-physical (breath), given the power and will of God. The human self
must therefore be aware of its divine dependence.
The theological point is mainly moral and not ontological: humans are
dependent on the divine and must honour this relation. The focus is thus on
human accountability. The possibility of afterlife is in the hands of the divine
but also dependent on the moral standard of the individual. Despite all sorts of
differences, it can be argued that Buddhist, Hindu and Muslim ideas of afterlife
are also basically moral arguments along the same line.27
Now this moral point of afterlife is pointless if afterlife itself is metaphysi-
cally impossible. Since there is no clear empirical evidence of afterlife, the
doctrine stands or falls with the feasibility of conceivability arguments. Let us
therefore turn to Swinburne’s afterlife argument again.28 Its first premise is as
follows:
(A) It is logically possible for me to continue to exist without my body.
This premise starts with well-known, easily conceivable features in the real
world: the self and the physical body of the person. Then the move is the
opposite of Ezekiel’s: Swinburne claims that the self, logically speaking, can be
27 Cf. Hick, Death, 46; Jane Smith, “Reflections on Aspects of Immortality in Islam,” The
Harvard Theological Review 70 (1977), 85–98.
28 See Shoemaker and Swinburne, Personal, 30; cf. Everitt, “Substance.”
separated from its body. This idea is seen as logically possible because—prima
facie at least—there is no real contradiction in terms.
Still we must admit that it is an astonishing and startling claim to submit,
that it is possible to exist without a body. Everything in our everyday experi-
ence confirms the opposite.
The anthropologist Pascal Boyer’s theory of how religious concepts are
constructed could highlight what is going on here. In Boyer’s terms, first, reli-
gious concepts are based on familiar ontological categories, such as PERSON,
PLANT, or ANIMAL. For instance, in Christianity and Islam, the post-mortem
souls are persons, and it is clear that the received view also is that the soul’s
person is identical to the person who just died. Secondly, concepts of supernat-
ural agents or entities then “invariably specify information that violates intuitive
expectations associated with the relevant ontological category.”29 For example,
the idea that a PERSON lacks physical body would be such a violation because
we intuitively take for granted that persons are identical to a particular physi-
cal body. Finally, “a supernatural concept also activates the intuitive expecta-
tions that are not violated, among those associated with the relevant ontological
category.”30 An example of this would be that is assumed that souls perceive
things: they can see and feel, without physical eyes, brains and nerve-endings.
Now we can go back to premise (A) again: as already indicated, since it is
a conceivability argument, Swinburne’s premise can be challenged in many
ways. Given Boyer’s analytical tools, here is one additional way of challenging
(A): if a concept now internally violates “intuitive expectations” of the onto-
logical category to which it belong, is it then really possible to claim that it
is logically coherent? Of course, since the dawn of mankind religions have
submitted views of afterlife without having any logical worries. Traditional
communities may have experienced their view as conceptually possible and
unproblematic. As Swinburne says, people “uninfluenced by philosophical
theory” apparently have no trouble with their local conceivability arguments.31
But it is really conceivable that a dead person is alive? For is not that logi-
cal contradiction what the concept afterlife idea really says and what also is
making the whole idea so startling and—using Pascal Boyer’s terminology—
basically “counter-intuitive?”32
29 Pascal Boyer, “Functional Origins of Religious Concepts: Ontological and Strategic

Selection in Evolved Minds,” Journal of the Royal Institute of Anthropology 6 (2000), 197.
30 Boyer, “Functional,” 197.
32 Boyer, “Functional,” 202.
218 Modée
Accepting (A) leads to further trouble: it is difficult to block other “logical

possibilities.” Suppose for instance that the zombie argument also is feasible.
This argument claims that it is conceptually possible that a body can walk
around without consciousness. Hence it presupposes that the consciousness
somehow can be separated from the body. One way is of course the physical
death of the person. But another way would be to go for (A) above. That is, we
could have it like this:
(A) It is logically possible for me to continue to exist without my body,

and
(B) It is logically possible for my body to run around and behave badly
without my consciousness.
In other words: given the doctrine of the soul (A), it is possible that we could
have a self that can be separated from its zombie appearance (B). I—the con-
sciousness separated from the body—could watch my zombie body eating
brains and human flesh in some possible world. To rule out this absurd pos-
sibility Swinburne must argue that (B) is false or absurd. But if (B) is more eas-
ily seen as puzzling or counter-intuitive while (A) should be seen as logically
possible, it is probably for the reason that (A) is not sufficiently conceptually
unpacked and supported by social conventions of acceptance. Because at face
value, compared to (B), it is equally absurd. For there are many aspects of con-
sciousness that clearly are mixed up with bodily sensations. How, then, can the
self be separated from such constitutive bodily aspects of itself? The bodily
environment of consciousness seems to be part of the self.33
Moreover, rejecting (B) leads to trouble for (A), since both rests upon the
conceivability argument. And I have already indicated various difficulties with
that argument above.
6 Final Remarks
This chapter has argued that (1) afterlife arguments and arguments about the
existence of divine beings can be seen as conceivability arguments and that
the feasibility of these religious matters should be considered in such terms
(see above). On the other hand, (2) given that a religious conceivability argu-
ment can be justified, it is difficult to see how different conceptions of afterlife
can be ruled out. And (3), it seems like everyday conceptions of afterlife are not
33 See Everitt, “Substance,” for a discussion.

sufficiently conceived and conceptually unpacked—a criticism also directed

against the conceivability of Zombies.
Moreover, social convention and tradition explains why local religious tra-
ditions may appear so real and essential for those who belong to a particular
social context and also why these traditions strikes outsiders so strange, irra-
tional and unbelievable.34 No one doubts the existence of food and shelter in
any local culture; everyone doubts some or every religious claim about pos-
sible existences—claims that are considered so essential and properly basic in
different local communities.
I leave the reader with this dilemma: either we need to accept that many dif-
ferent ideas of afterlife and the supernatural are conceivable and hence meta-
physically possible, or we need to reject the zombie/afterlife/supernatural
conceivability arguments. The first horn leads to an absurd maximal ontologi-
cal pluralism, the second horn, it could be argued, to some version of athe-
ism or agnosticism. Neither of these two options will be seen as logical for the
absolute majority of selves in our world today.
Bibliography
Pascal Boyer, “Functional Origins of Religious Concepts: Ontological and Strategic

Selection in Evolved Minds,” Journal of the Royal Institute of Anthropology 6 (2000),
195–214.
David Chalmers, The Character of Consiousness (Oxford: Oxford University Press, 2010).
———, The Conscious Mind: In Search of a Fundamental Theory (New York: Oxford
Allin Cotrell, “Sniffing the Camembert: On the Conceivability of Zombies,” Journal of
Consciousness Studies 6 (1999), 4–12.
Daniel Dennet, “The Unimagined Preposterousness of Zombies,” Journal of Con
sciousness Studies 2 (1995), 322–326.
———, “The Zombic Hunch: Extinction of an Intuition?,” Royal Institute of Philosophy
Millennial Lecture (1999).
Nicholas Everitt, “Substance Dualism and Disembodied Existence,” Faith & Philosophy
17 (2000), 331–347.
John Hick, Death and Eternal Life (Louisville: Westminster, 1994).
David Hume, A Treatise of Human Nature, eds. L.A. Selby-Bigge and P.H. Nidditch
(Oxford: Oxford University Press, 1968).
Saul Kripke, Naming and Necessity (Cambridge, ma: Harvard University Press, 1980).
34 For a discussion, see Rationality, ed. Bryan Wilson (Oxford: Blackwell, 1970).
220 Modée
Steven Mithen, The Prehistory of Mind: A Search for the Origin of Art, Religion and
Science (London: Thames & Hudson, 1996).
Hilary Putnam, “The Meaning of Meaning,” in Philosophical Papers, Volume 2: Mind,
Language and Reality (Cambridge: Cambridge University Press, 1975).
John Searle, The Mystery of Consciousness (London: Granta Books, 1998).
———, Mind, Language and Society: Philosophy in the Real World (London: Phoenix,
1999).
Sidney Shoemaker and Richard Swinburne, Personal Identity (Oxford: Blackwell, 1984).
Richard Swinburne, The Evolution of the Soul (Oxford: Clarendon Press, 1997).
Jane Smith, “Reflections on Aspects of Immortality in Islam,” The Harvard Theological
Review 70 (1977), 85–98.
Melford Spiro, “Religion: Problems of Definition and Explanation,” in Anthropological
Approaches to the Study of Religion, ed. Michael Banton (London: Tavistock, 1966).
Rationality, ed. Bryan Wilson (Oxford: Blackwell, 1970).
Nigel Thomas, “Zombie Killer,” in Toward a Science of Consciousness II, eds. Stuart R.
Hameroff, Alfred W. Kaszniak & A. C. Scott Cambridge (Cambridge, ma: mit Press,
1998), 171–177.
Stephen Yablo, “Is Conceivability a Guide to Possibility?,” Philosophy and Pheno
menological Research 53 (1993), 1–42.
———, “Concepts and Consciousness,” Philosophy and Phenomenological Research 59
(1999), 455–463.
———, “Textbook Kripkeanism & The Open Texture of Concepts,” Pacific Philosophical
Quarterly 81 (2000), 98–122.
chapter 12
The Logocentric Predicament and the Logic of

Question and Answer
Giuseppina D’Oro
1 Introduction
It has been argued that all attempts to justify logic presuppose precisely what
ought to be explained (the rules of deductive inference) and thus that no
attempt to justify logic can succeed because there is no non-circular justifica-
tion of deductive inference. This, in a nutshell, is the logocentric predicament.
It is the predicament in which one finds oneself in when one asks for a justifi-
cation of the standards of measurement. The logocentric predicament was the
subject of Lewis Carroll’s “What the Tortoise said to Achilles.”1 Carroll argues
that every valid deductive inference from premises to conclusion presupposes
the principle that deductive inferences are truth-preserving. The principle
that deductive inferences are truth-preserving must therefore be included as
a premise in a deductive argument in order for its conclusion to be justified.
This, however, generates an infinite regress since the new deductive argument
which contains the principle of valid inference amongst its premises will be
justified only by introducing another principle of valid inference, and so on.
Carroll thus concluded that the justification of deductive inference can never
be complete.2
The logocentric predicament has more recently been revisited by Susan
Haack in “The Justification of Deduction.”3 Haack argues that the principle
that deductive inferences are truth-preserving cannot be justified either
inductively or deductively. It cannot be justified inductively because an induc-
tive justification would be too weak. It cannot be justified deductively either,
as a deductive justification would be circular. Although the logocentric pre-
dicament has been discussed primarily in the context of deductive logic, it
applies, mutatis mutandis to inductive logic as well. A deductive justification of
1 Lewis Carroll, “What the Tortoise Said to Achilles,” Mind 14 (1895), 278–280.
2 For a close discussion of Carroll and a book-length analysis of the logocentric predicament
see Robert Hanna’s Rationality and Logic (Cambridge Massachusetts: mit Press, 2006).
3 Susan Haack, “The Justification of Deduction,” Mind 85 (1976), 112–119.

222 D ’ Oro
induction would be too strong, but an inductive one, as Hume pointed out,
would be circular. In this chapter I argue that the logic of question and answer,
as developed in Collingwood’s An Essay on Metaphysics,4 escapes the logocen-
tric predicament and that it does so by rejecting the very assumption which
gives rise to the predicament, namely the view that all circularity is bad and
therefore to be avoided. This is not to say that Collingwood was an advocate
of arguments which are circular in the pejorative sense of being question-
begging. Rather, he sought to distinguish between a form of circularity that
is vicious and one which is not. This chapter tries to illustrate this claim by
considering Collingwood’s logic of question and answer in the context of his
conception of metaphysics as a science of absolute presuppositions.
2 Metaphysics as a Science of Absolute Presuppositions
Collingwood developed the logic of question and answer in the context

of his account of metaphysics as a science of absolute presuppositions. To
understand his defence of an argumentative form which is circular and yet
not question-begging one must first outline his conception of metaphysics.
Collingwood rejected the traditional conception of metaphysics as an onto-
logical investigation into the structures of reality per se on the grounds that
there can be no such thing as a science of pure Being. Reality is always inves-
tigated from a particular point of view. His claim that there is no such thing
as non-perspectival knowledge of pure being is most often glossed as stating
that there is no ahistorical perspective from which reality can be known or
investigated and thus that all knowledge claims are historically relative.5 In the
following I will assume that Collingwood’s claim that there is no presupposi-
tionless knowledge is an attempt to argue not in support of historical relativ-
ism, as is often assumed, but in support of a different claim: the view that any
4 R.G. Collingwood, An Essay on Metaphysics, revised edition with an introduction by Rex

Martin (Oxford: Oxford University Press, 1998 [1940]).
5 For this historicist/relativist reading of Collingwood’s metaphysics see Stephen Toulmin
“Conceptual Change and the Problem of Relativity”, in Critical Essays on the Philosophy of
R. G. Collingwood, ed. Michael Krausz (Oxford: Clarendon Press, 1972) 212–213; Alan Donagan
The Later Philosophy of R. G. Collingwood (Oxford: Clarendon Press, 1962); Nathan
Rotenstreich, “Metaphysics and Historicism,” in Critical Essays on the Philosophy of R. G.
Collingwood (Oxford: Clarendon Press, 1972) 179–200, 199–200. More recently see R. Harrison
“Atemporal Necessities of Thought; or, How Not to Bury Philosophy by History,” in Reading
Kant: New Perspectives on Transcendental Arguments and Critical Philosophy, ed. E. Schaper
and W. Vossenkuhl (Oxford: Basil Blackwell, 1989), 43–54.
The Logocentric Predicament 223
description of reality is relative to the investigative goals of the form of enquiry

from which it is approached. Understood in this way, the task of uncovering
absolute presuppositions is a logical (or as he would put it, a “criteriologi-
cal”) task. It is the role of the metaphysician to make explicit the fundamental
presuppositions which govern forms of enquiries and are normative for their
practitioners, and not merely to describe what certain groups of people believe
at certain points in time. Metaphysics is an attempt to uncover what the prac-
titioners of a science are logically committed to; it is not a purely descriptive
enterprise concerned with what people believe at certain times and places. So
construed, Collingwood’s metaphysics of absolute presuppositions defends a
kind of methodological, rather than historical, relativism. In other words, what
Collingwood defends is not a claim concerning the relativity of presupposi-
tions to time and place, but a claim concerning the relativity of explanation
to subject matter. And in the context of his metaphysics of absolute presup-
positions the claim that “explanation is relative to subject matter” should be
read in the manner of an Hegelian “speculative proposition”: since method
determines subject matter, a particular subject matter requires a certain form
of explanation. To say that explanation is relative to subject matter or that a
particular subject matter requires a particular form of explanation are just two
ways of saying one and the same thing, namely that there is no such thing as a
science of pure Being.
In the following I will be largely presupposing the exegetical claim that
Collingwood is defending a form of methodological rather than historical
relativism,6 and that the task of metaphysics is not to report what people
believe but to uncover the presuppositions which govern certain domains of
enquiry and are normative for their practitioners. Whilst metaphysics must
indeed begin by reflecting on existing explanatory practices (and in this sense
it does contain a descriptive element within it), the metaphysician is not an
experimental philosopher carrying out surveys about practitioners’ beliefs.
The role of the metaphysician is not to conduct opinion polls, but to tease
6 I have defended this exegetical claim in my “The Myth of Collingwood’s Historicism,” Inquiry
53 (2010), 627–641 and in Collingwood and the Metaphysics of Experience (London and New
York: Routledge, 2002). On this see also James Connelly “Metaphysics and Method: A
Necessary Unity in the Philosophy of R. G. Collingwood,” Storia, Antropologia e Scienze del
Linguaggio 5 (1990), 1–2, Tariq Modood “The Later Collingwood’s Alleged Historicism and
Relativism,” Journal of the History of Philosophy 27 (1989), 101–125; A. Oldfield “Metaphysics
and History in Collingwood’s Thought,” and Rex Martin, “Collingwood’s Claim that
Metaphysics is a Historical Discipline,” both in Philosophy, History and Civilization:
Interdisciplinary Perspectives on R. G. Collingwood, ed. David Boucher, James Connelly and
Tariq Modood (Cardiff: University of Wales Press, 1995).
224 D ’ Oro
out practitioners’ implicit commitments by making explicit the norms which

govern their forms of enquiry. In this respect philosophy is both a descriptive
and normative enquiry.7 It is descriptive because it takes existing explanatory
practices as its starting point; it is normative because it shows what must be
presupposed by the practitioners of a science in order for their form of inves-
tigation to be possible.
With these preliminary observations in mind we can now start to explain
how the logic of question and answer is linked to Collingwood’s conception
of metaphysics as a science of absolute presuppositions. Absolute presup-
positions spell out principles which make certain forms of enquiry possible.
The logic of question and answer illustrates the logical regress from a form
of knowledge with its associated conception of reality to the conditions of its
possibility.
3 The Logic of Question and Answer
Every question, Collingwood says, has a presupposition. Without presupposi-

tions, questions would not arise. This claim, as we have seen, lies at the basis of
the view that there can be no knowledge of pure Being. The pursuit of knowl-
edge is conducted in the attempt to answer certain questions. But without pre-
suppositions there would be no questions. Take for example the question “why
did dinosaurs become extinct?” Without the presuppositions that “dinosaurs
at some point existed” and “dinosaurs no longer exist,” that question would
not even arise. Perhaps another question would occur to us, but not that
particular one.
Presuppositions provide necessary rather than sufficient conditions for
questions. The presupposition “Ceri owns a pencil case” is required so that
a question such as “where did you put your pencil case Ceri?” may arise.
Presuppositions do not, however, state sufficient conditions because they do
not entail that a particular question, rather than another one, must arise. On
the basis of the same presupposition I could have asked a different question:
“have you put your pencil case in the school bag Ceri?” Although presuppo-
sitions only spell out necessary conditions for questions to arise—and thus
they do not fully determine what questions could be asked—they also limit
the range of questions to which they can give rise. Thus, for example, the
7 See R.G. Collingwood, An Essay on Philosophical Method, (Oxford: Oxford University Press,
2005 [1933]), reprinted with an introduction by James Connelly and Giuseppina D’Oro,
128–29.
presupposition “Ceri owns a pencil case” could not give rise to the question “do
you need a pencil case Ceri?” unless by the question one meant: “do you need
another pencil case?” So, if we change presuppositions we change the kind of
questions we can intelligibly ask.
The sense in which presuppositions give rise to questions is not causal or
genetic but logical. Whereas there must always be someone to do the presup-
posing (otherwise questions would not be asked), it is not the act of presup-
posing but rather the content of the presupposition and the internal relation
in which that content stands to the question to which the presupposition gives
rise that constitutes the object of the argumentative regress from a form of
knowledge to its presuppositions. The presupposition that dinosaurs existed
on Earth would still have the logical power to give rise to the question “Why did
dinosaurs become extinct?” even if in the absence of anybody assenting to that
presupposition the question would not be asked. It is because Collingwood
is concerned with the way in which presuppositions logically constrain ques-
tions, rather than in the fact of us making presuppositions or asking the ques-
tions, that he speaks of the power of presuppositions to give rise to questions
as to their logical as opposed to causal efficacy.8
A similar relation holds, for example between Dasein and the world for
Heidegger. When Heidegger claims that where there is no Dasein there is no
world,9 he does not mean that if there were no Dasein empirical reality would
disappear. If empirically reality were causally dependent on Dasein for its
existence, then it would indeed be true that where there is no Dasein there is
nothing, as a form of Berkeleyan idealism would claim. What is dependent on
Dasein, for Heidegger, is not empirical reality, but meaningful reality. To say
that only where Dasein is, is there a world, is not like saying that only where
there is water is there life (where water is a causal condition for life). Similarly,
for Collingwood to say that presuppositions give rise to questions is not to say
that they are causal conditions of the asking of questions in the way in which
the presence of water is a causal condition for the existence of life. Since the
dependence of a question on a presupposition is not a dependence of the act
of asking on the act of presupposing, but of the content of the question on the
content of the presupposition, this relation would hold even if certain presup-
positions were no longer made or assented to and consequently the questions
to which they have the power to give rise were no longer being asked. Just as we
would want to distinguish between the text of a play and its performance on
the stage so Collingwood wants to distinguish between the act of presupposing

9 Martin Heidegger, Being and Time (New York: Harper & Row Publishers, 1927), 78 ff.
226 D ’ Oro
and the content of the presupposition even if, in the absence of the act of pre-
supposing no questions could actually be asked. To conflate the logical relation
between an absolute presupposition and the questions to which it gives rise
with the causal relation between the act of presupposing and that of question-
ing, would be tantamount to losing sight of the distinction between the text of
a play and its performance, thereby reducing the content of the dialogue in the
script of a play to the conversation between the actors on the stage. Conflating
a causal conditions for the asking of questions with the power that presup-
positions have to give rise to questions (logical efficacy) would be like arguing
that a producer’s willingness to fund the staging of a play, is a causal condition
for the existence of the play’s text as well as its performance. Just as we would
want to deny that a producer’s willingness to fund the staging of a play is a
causal condition for the existence of the play’s text as well as its performance,
we should be wary of conflating the causal conditions for the asking of ques-
tions with the logical efficacy of presuppositions, or their power to give rise to
questions.
That there is a logical nexus between presuppositions and the questions
to which they give rise is evident from the fact that we fail to understand
questions when we do not share the presuppositions from which they arise.
Suppose that unbeknown to me a dangerous criminal had entered the build-
ing followed by a plain-clothes policeman on his trail, and that the policeman
had seen him run in the direction of my office. Since I was absorbed in reading
a document, I was blissfully unaware of these goings-on. If I were asked: “have
you seen him; where did he go?” I would not understand what the policeman
is asking, not because I do not understand the words he is uttering (for he is
speaking in a foreign tongue), but because I am not sharing the presupposi-
tion that gives rise to that question (that there is a dangerous criminal in the
building). In short: presuppositions have the power to give rise to questions, by
which Collingwood means that they have logical rather than causal efficacy. As
such, they do their work not in so far as they are believed or assented to, but in
virtue of their content.
Often presuppositions are made not for argument’s sake but to in so far as
they are believed to be true. If I did not believe it to be the case that Ceri owns a
pencil case, I would not presuppose it in my reasoning, and if I did not assume
it, I would not be asking questions such as “where is it? Did you put it in your
school bag? Did you forget it at school?” etc. However, while we tend to presup-
pose what we believe to be true, the truth of a presupposition, or even the fact
it is believed to be true, is irrelevant to its logical efficacy or its power to give
rise to questions. Suppose that I falsely presupposed that there are eggs in the
fridge because I bought them yesterday and put them there, but I was mistaken
because Mark had them for lunch while I was out at work. My presupposi-
tion gives rise to the question “Mark, shall we have omelette for dinner?” Mark
can understand my question even if he does not assent to the presupposition
which gave rise to it (“that there are eggs in the fridge”), for he knows he cooked
them for his lunch. Assenting to or believing a presupposition to be true is thus
irrelevant to our ability to understand the questions to which it gives rise. As
Collingwood says: “the logical efficacy of a supposition does not depend upon
the truth of what is supposed, or even upon its being thought to be true, but
only on its being supposed [. . .] the process of thought from question to ques-
tion does not depend on each question being answered truly, but only upon its
being answered: and not upon the questioner’s thinking the answers true, but
only on his accepting the answers given to him, or ‘assuming them for the sake
of argument.’”10 Whilst the logical efficacy of a presupposition is independent
of its truth, grasping the logical entailment between a presupposition and the
question to which it gives rise is a condition for understanding the question.
For if I asked: “Mark shall we have dinosaur eggs for dinner?” he would be justi-
fiably puzzled by my question for he, as in all likelihood the reader, would be in
the dark as to what presupposition one must make in order for such a question
to arise. Questions appear crazy or unintelligible, not when the presupposi-
tions which give rise to them are false—the question “shall we have omelette
for dinner?” is perfectly intelligible to Mark even if it rests on the false pre-
supposition that there are eggs in the fridge—but when the presuppositions
which give rise to them remain unfathomable.
Another important aspect of the logic of question and answer is the claim
that presuppositions may be either relative or absolute. A presupposition is
relative if it could be asserted as a proposition with a definite truth-value in
answer to a question. For example, the presupposition “Ceri has a pencil case”
could be either a claim that, when presupposed, has the power to give rise to
questions (“Where is your pencil case?” etc.), or it can be offered as an answer
to a question such as “What school items does Ceri own?” What determines
whether “Ceri has a pencil case” is a presupposition or a proposition is the role
it plays in the logic of question and answer. If its role is to give rise to questions
then it is a presupposition whose truth or falsity is irrelevant to its logical effi-
cacy; if its role is to answer a question then it is a proposition with a definite
truth-value.
Unlike relative presuppositions, absolute presuppositions have the power
to give rise to questions but can never be asserted in answer to a question as if
they were propositions with a definite truth-value. Absolute presuppositions,

228 D ’ Oro
as Collingwood would say, are never propounded in answer to questions; their

role is exclusively that of giving rise to questions. Consider for example what
Collingwood takes to be the absolute presupposition of the practical sciences
of nature. This is the view that a cause is “an event or state of things by pro-
ducing or preventing which we can produce or prevent that whose cause it
is said to be.”11 This particular sense of causation is a heuristic principle of
the practical sciences of nature because doctors, engineers, mechanics, and
such like, conceive of causes as handles which can be turned to produce the
desired consequences. It is by administering malaria tablets that doctors pre-
vent malaria; it is by tightening a valve that a mechanic gets the car engine to
start once again. Without presupposing that it is possible to alter the course
of nature in this way, medicine, engineering and mechanics would not be
conceivable enterprises. The presupposition “a cause is an event by produc-
ing and preventing which . . .” engenders a particular line of questioning aimed
at discovering the preventable causes of events. Statements such as “malaria
is caused by the bite of a mosquito” or “the engine stopped working because
of a loose cable” are offered in answer to the range of questions that arise for
those investigators who conceive of causes as “handles”. Propositions offered
in answer to such questions, unlike the absolute presupposition which gives
rise to them, have a definite truth-value. The propositions “malaria is caused by
the bite of a mosquito” or “cancer is a virus” can be verified/falsified by showing
respectively that there is a correlation between the sanitization of damps and
a reduction in the incidence of malaria, and that exposure to cancer patients
does not lead to a greater incidence of the disease in the population. But the
presupposition that “a cause is . . .” cannot be verified in the manner in which
we verify specific empirical correlations because it is the pre-condition for ask-
ing the kind of questions—“what is the preventable cause of x?”—which are
answered by propositions such as “malaria is caused by the bite of a mosquito”
or “the engine stopped working because of a loose cable”. To expect the presup-
position that causes are preventable or manipulable to have a definite truth-
value would be tantamount to mistaking it for propositions such as “the engine
stopped working because of a loose cable” or “malaria is caused by the bite of a
mosquito”. This is arguably the trap in which Hume fell when he subjected the
principle of the uniformity of nature to the same test to which he subjected any
ordinary proposition. Is the principle of the uniformity of nature an analytical
proposition concerning relations of ideas? Is it an empirical proposition about
matters of fact? If it is neither an analytical proposition verifiable by detecting
an agreement or disagreement between the ideas expressed, nor an empiri-
11 Collingwood, An Essay on Metaphysics, 296–7.

cal proposition which either corresponds or fails to correspond to the facts

it purports to describe, it must be an unverifiable metaphysical proposition
which should be committed to the flames.12 Having come across an absolute
presupposition of inductive reasoning Hume failed to recognize its role in our
thought as a presupposition (as something which has the power to give rise
to questions) rather than as a proposition and demanded that it should be
proved to be true or false either inductively or deductively. And having found
either attempt at a proof for the principle of the uniformity of nature unsat-
isfactory he endorsed scepticism as the only honest philosophical option.
For Collingwood, what is nonsensical is not the principle of the uniformity
of nature but rather the demand that such principle should be verified in
the manner of a proposition. Hume’s demand for verification is nonsensical
because presuppositions do their logical work not in so far as they are true or
false but in so far as they are presupposed. To say that the notion of truth and
falsity does not apply to absolute presuppositions is not to say that since they
cannot be verified they lack justification and are thus, in some sense, arbitrary,
but rather that the demand that they should be verified is misplaced. To try and
justify the presupposition which governs the practical sciences of nature in the
same way in which one might try to verify the claim that malaria is caused by
the bite of a mosquito would be tantamount to measuring the standard meter
and claiming of it that it either is or is not one meter long. Just as Wittgenstein
denied that one can say of the standard meter either that it is or that it is not
one meter long,13 so Collingwood denies that absolute presuppositions can
be said to be either true or false precisely because in determining the kind of
questions that can be asked within a particular form of enquiry they ipso facto
establish the range of acceptable true/false answers that can be given to them.
12 “If we take in our hands any volume; of divinity, or school of metaphysics, for instance,
let us ask, Does it contain any abstract reasoning concerning quantity or number? No.
Does it contain any experimental reasoning concerning matters of fact and existence?
No. Commit it then to the flames. For it can contain nothing but sophistry and illusion:”
David Hume, An Enquiry Concerning Human Understanding (Oxford, Oxford University
Press, 2007 [1748]), Section 12, “The Sceptical Philosophy,” part 3).
13 Ludwig Wittgenstein, Philosophical Investigations (Oxford: Basil Blackwell, 1953), §50.
230 D ’ Oro
4 The Logocentric Predicament and the Logic of Question and

Answer
But what is the status of the logic of question and answer? And is the kind of
metaphysical knowledge that is obtained through its application something
that one cannot justify unless one justifies the logic of question and answer
itself? Arguably not. Unlike deductive and inductive arguments, the logic
of question and answer is not designed to generate new knowledge by add-
ing to the repertoire of truths that are already known. Metaphysical analysis,
for Collingwood, “does not, like exact or empirical science, bring us to know
things of which we were simply ignorant, but brings us to know in a differ-
ent way things which we already knew in some way.”14 Through deduction
we can extend our repertoire of a priori truths. By means of induction we can
extend our empirical knowledge by moving from observed to unobserved
matters of fact. By contrast, metaphysical analysis makes explicit what is in
some sense already known to the practitioners of first order sciences. What
is already known to them are not the truths at which the deductive logicians
arrive by applying Modus Ponens, or the predictions which are made by means
of inductive inference, but rather the forms of inference through which these
truths are arrived at.
Unlike deductive and inductive arguments, the logical regress from a form of
knowledge to the conditions of its possibility (absolute presuppositions) does
not introduce any new knowledge. Strictly speaking, then, the logic of ques-
tion and answer is not a new kind of logic which generates its own peculiar
kind of knowledge, i.e., metaphysical knowledge. Presuppositional arguments
of the kind that Collingwood deploys simply make explicit the forms of infer-
ence which are presupposed at the first order level. The job of the metaphysi-
cian, as Collingwood says “is not to propound them (absolute presuppositions)
but to propound the proposition that they are presupposed.”15 Thus the meta-
physician does not affirm the claim “causes are events by producing or pre-
venting which [. . .] is true,” or “causes in the sense which they are deployed in
the practical sciences of nature exist;” the metaphysician propounds the claim
that this conception of causality is absolutely presupposed by practical scien-
tists of nature because it is a sine qua non of their form of enquiry. Absolute
presuppositions are always presupposed and never propounded. First order
investigators presuppose them; philosophers propound the proposition that
they are presupposed. To propound them as if they were propositions would
be to mistake their role in the logic of question and answer.
14 Collingwood, An Essay on Philosophical Method, 161.

To clarify why the uncovering of absolute presuppositions through the

logic of question and answer does not involve a viciously circular argument,
I contrast the logical regress from a form of knowledge to the conditions of
its possibility with an argument which is circular in the pejorative sense of
being question-begging. Normally, an argument is deemed to be circular in this
pejorative sense if it begs the question by presupposing in its premise what
it ought to show in its conclusion. Consider for example Berkeley’s continu-
ity argument for the existence of God. It roughly goes like this. Premise 1: we
assume trees, houses, etc., to exist even when not perceived by our own mind;
premise 2: trees, houses etc., do not exist absolutely unperceived (unperceived
by any mind). Conclusion: Therefore there exists a spirit/mind (God) who
continuously perceives trees, houses, and any objects to which we attribute
continuous existence.16 Berkeley’s argument, as stated, is circular because one
of its premises (premise 1: objects exist when unperceived by a finite mind)
relies on the truth of the conclusion (the existence of a continuous perceiver).
Aschenbrenner phrases the objection as follows: “His actual use of it is circular.
He has no reason to believe that remote sensibles exist except on the supposi-
tion that God perceives them, but he has no reason, leaving the other argu-
ment aside, not any need to believe that God or any other perceiver is there
present except that there exist remote sensibles which must be perceived by
someone. The existence of these is begged from the start.”17
Collingwood’s metaphysics of absolute presupposition is not circular in
this question-begging way. Where Berkeley is arguing from the belief in the
continuous existence of unperceived objects to the existence of the presup-
position that would make this belief true (that a continuous perceiver (God)
exists), Collingwood is not inferring from the claim that the practitioners of a
science deploy a certain kind of explanation the conclusion that the absolute
presuppositions which such explanations rely on are true or exist, but only
that they have to be presupposed. In accordance with the logic of question
and answer Berkeley’s argument would have to be restated as demonstrating
not that God exists but only the epistemically much more modest conclusion
that the existence of God is the presupposition on which belief in the con-
tinuous existence of temporarily unperceived objects rests. There is no vicious
circularity in this claim, just an engendering of a form of self-consciousness
about the deeper commitments on which our beliefs rest. Were Collingwood
16 George Berkeley, The Principles of Human Knowledge and Three Dialogues (Oxford: Oxford
University Press, 1996 [1710 and 1713]).
17 Aschenbrenner, quoted in I. C. Tipton, Berkeley: The Philosophy of Immaterialism (London:
Methuen, 1976), 322.
232 D ’ Oro
inferring the existence of a certain kind of cause from the deployment of a

given explanatory practice, he would indeed be making an illegitimate step
from an explanatory premise to an ontological conclusion to which he would
not be entitled without begging the question. But there is no such slippage
from epistemic premises to ontological conclusions in the case of the logic of
question and answer because, as we have seen, the metaphysician does not
propound absolute presuppositions as if they were propositions. Since the
activity of rendering presuppositions explicit does not generate new knowl-
edge (the logic of question and answer is not a knowledge-generating logic),
there is no vicious circularity involved in the process of uncovering presuppo-
sitions and no question-begging of the premises.
Yet whilst the regress from a form of knowledge to the conditions of its pos-
sibility does not involve circularity of a vicious and objectionable kind, meta-
physical analysis does show that a certain form of epistemic circularity is an
unavoidable predicament for first order scientists. For without any presupposi-
tions there would be no questions, without questions no lines of enquiry, and
without any lines of enquiry there would be no science in the Latin sense of
the term scientia, meaning a body of knowledge with a specific method and
subject matter. Thus for Collingwood the outcome of a philosophical reflection
on the nature of the fundamental commitments which govern our explanatory
practices is not that we have to turn to scepticism because no non-circular
justification of absolute presuppositions is forthcoming. His view that the
logocentric predicament is an inescapable predicament of first order scien-
tists arises out of a positive endorsement of presuppositions as enabling con-
ditions of knowledge. The practitioner’s entitlement to presuppose them lies
precisely in the fact that they make knowledge possible. Collingwood would
thus disagree with Hume’s claim that reflection upon our explanatory prac-
tices paves the road to scepticism and “Philosophy woul’d render us entirely
Pyrrhonian, were not nature too strong for it.”18 Rather than pushing us into
scepticism, philosophical reflection renders explicit what the practitioners of
the first order sciences know implicitly and encourages us, as Heidegger puts
it, to “leap into the circle” of knowledge.19
18 Hume, D. (Abstract, 657).

19 Heidegger, Being and Time, 363.
Bibliography
G. Berkeley, The Principles of Human Knowledge and Three Dialogues (Oxford: Oxford
University Press, 1996 [1710 and 1713]).
L. Carroll, “What the Tortoise Said to Achilles,” Mind 14 (1895), 278–280.
R.G. Collingwood, An Essay on Philosophical Method, (Oxford: Oxford University Press,
2005 [1933]), reprinted with an introduction by James Connelly and Giuseppina
D’Oro.
———, An Essay on Metaphysics, revised edition with an introduction by Rex Martin
(Oxford: Oxford University Press, 1998 [1940]).
J. Connelly, “Metaphysics and Method: A Necessary Unity in the Philosophy of R. G.
Collingwood,” Storia, Antropologia e Scienze del Linguaggio 5 (1990), 1–2.
A. Donagan, The Later Philosophy of R. G. Collingwood (Oxford: Clarendon Press,
1962).
G. D’Oro, “The Myth of Collingwood’s Historicism,” Inquiry 53 (2010), 627–641.
———, Collingwood and the Metaphysics of Experience (London and New York:
Routledge, 2002).
S. Haack, “The Justification of Deduction,” Mind 85 (1976), 112–119.
R. Hanna, Rationality and Logic (Cambridge Massachusetts: mit Press, 2006).
R. Harrison, “Atemporal Necessities of Thought; or, How Not to Bury Philosophy by
History,” in Reading Kant: New Perspectives on Transcendental Arguments and
Critical Philosophy, ed. E. Schaper and W. Vossenkuhl (Oxford: Basil Blackwell, 1989),
43–54.
M. Heidegger, Being and Time (Oxford: Blackwell Publishing, 2005 [1927]).
D. Hume, An Enquiry Concerning Human Understanding (Oxford, Oxford University
Press, 2007 [1748]).
R. Martin, “Collingwood’s Claim that Metaphysics is a Historical Discipline,” in
Philosophy, History and Civilization: Interdisciplinary Perspectives on R. G. Colling
wood, ed. David Boucher, James Connelly and Tariq Modood (Cardiff: University of
Wales Press, 1995).
T. Modood, “The Later Collingwood’s Alleged Historicism and Relativism,” Journal of
the History of Philosophy 27 (1989), 101–125.
O. Oldfield, “Metaphysics and History in Collingwood’s Thought,” in Philosophy, History
and Civilization: Interdisciplinary Perspectives on R. G. Collingwood, ed. David
Boucher, James Connelly and Tariq Modood (Cardiff: University of Wales Press,
1995).
N. Rotenstreich, “Metaphysics and Historicism,” in Critical Essays on the Philosophy of
R. G. Collingwood (Oxford: Clarendon Press, 1972), 179–200.
I. C. Tipton, Berkeley: The Philosophy of Immaterialism (London: Methuen, 1976).
234 D ’ Oro
S. Toulmin, “Conceptual Change and the Problem of Relativity”, in Critical Essays on the
Philosophy of R. G. Collingwood, ed. Michael Krausz (Oxford: Clarendon Press, 1972),
212–213.
L. Wittgenstein, Philosophical Investigations (Oxford: Basil Blackwell, 1953).
Name Index
Ackermann, Wilhelm 191 Freud, Sigmund 28, 66–69, 73, 75, 78

Ankersmit, Frank 6, 9n28–29, 10
Anselm of Canterbury 209 Gauss, C.F. 116
Aristotle 9, 10, 17, 18, 19, 20, 23, 29, 30, 39, Goethe, Johann von 103
54, 68, 72–75, 77, 78, 92 Guignon, Charles 58
Augustine 26, 27, 30
Ayer, A.J. 51, 52, 64 Haack, Susan 5, 124n5, 135n37, 136n38,
142n47
Bacon, Francis 83, 87 Hawthorne, John 132n33
Beall, J.C. 11, 124, 125, 126, 127, 128, 129, 130, Hegel, G.F.W. 2, 3, 4n9, 98
131, 132, 134, 135 Heidegger, Martin 8, 10, 51–64, 225, 232
Boole, George 1 Heraclitus 19
Boyer, Pascal 217 Hick, John 215n26, 216n27
Brandom, Robert 131n29, 13n43 Hilbert, David 117, 191
Butterfield, Herbert 32 Hintikka, Jaako 7
Hodges, Wilfrid 187
Carnap, Rudolf 51, 52, 64 Horgan, Terry 11
Cassirer, Ernst 103, 104, 114, 115, 116–119 Hume, David 209, 211
Cejvan, Ervik 10 Husserl, Edmund 55
Chalmers, David 209n3, 211–214
Collingwood, R.G. 7, 10, 11, 81, 82, 84, 86, James, William 42, 43, 44
90, 91–99, 222–224, 224–228
Conant, James 123n2, 124, 126, 136, 143, Kant, Immanuel 2–3, 5, 180, 183, 185
144, 145, 146 Korsgaard, Christine 169, 183
Kraal, Anders 11
Davidson, Donald 145 Kripke, Saul 211
Dennett, Daniel 213
Descartes, René 83, 87, 209 Lacan, Jacques 8, 10, 66–79
Dewey, John 33n2, 33, 34, 36, 37, 43, 44 Lance, Mark Norris 132n33
Dharamsi, Karim 11 Leibniz, G.W. 5, 103, 104, 108, 120
D’Oro, Giuseppina 11 Lejewski, Czeslaw 191, 201
Dreyfus, Hubert 10, 56n15, 58, 59, 60, 62, Lesniewski, Stanislaw 191, 201
63n46 Lynch, Michael 11, 133n34, 148, 149–162,
Dupuy, J.P. 119, 120n22 164, 166–167
Eastman, Max 43–47, 48 Mahnke, Dietrich 103, 104

Etchemendy, John 125, 126 Marx, Karl 98
Eugenics Education Society 45 Meillassoux, Quentin 8, 9, 19–23,
28–30
Fear, Christopher 10 Meinecke, Friedrich 103
Fitch, Frederic 1 Meinong, Alexius 110, 111
Foster, Marc 208 Mill, J.S. 115, 117, 211
Freeden, Michael 7–8 Mithen, Steven 207n1
Frege, Gottlob 1, 9, 10, 110, 111, 115, 123, 124, Modée, Johan 11
125, 126, 129, 135, 137, 143, 188, 189n8, 191, 199
236 Name Index
Nancy, Jean-Luc 8, 9, 22, 23–28, 29–30 Shoemaker, Sidney 210n7–10, 216n28,

Nietzsche, Friedrich 33–35, 46n50 217n31
Skinner, Quentin 83n14, 87n24, 92, 92n51
Ockham, William of 5 Skodo, Admir 10
Smith, Jane 216n27
Peano, Giuseppe 1 Socrates 24
Plato 9, 17, 18, 20, 21, 23–24, 25, 27, 30, 31, Spiro, Melford 215n26
86, 91–93, 98, 99 Svensson, Thord 11
Putnam, Hilary 10, 124, 134, 136, 137, 138, Swinburne, Richard 210, 216–218
139, 140, 141, 142, 144, 146, 148n1, 149n4–5,
155, 156, 157, 161, 162, 211, 212n20 Tanesini, Alessandra 10
Thomas, Nigel 214n24
Quine, W. V. O. 44, 52, 112–114, 136n38,
138, 139 Watkin, Christopher 9
Wedgwood, Ralph 11, 170, 171, 180–185
Restall, Greg 11, 124, 125, 126, 127, 128, 129, Weir, Alan 133n35
130, 131, 132, 134, 135 Wendland, Aaron 10
Russell, Bertrand 1, 9, 35, 36, 43, 44n41, White, Morton 44
47, 103, 109–114, 116, 117, 121, 188, 190n13, Whitehead, A.N. 188, 190n13, 204
199, 204 Wittgenstein, Ludwig 7, 18, 52, 131, 146,
150, 153
Saussure, Ferdinand de 70 Wrathall, Mark 57n25, 63, 64
Schiller, F.C.S. 8, 10, 35–36, 37–42, 43–48
Searle, John 207n2, 212 Yablo, Steven 209n3–5, 211, 212n6, 213, 214
Sellars, Wilfrid 52, 138
Subject Index
A posteriori 211, 212 Empiricism 3, 86

A priori 126, 134, 138, 139, 140 Enstasis 68, 74, 75
Afterlife 207, 210, 215–219 Epistemology 121
Akrasia 181, 182, 185 Eugenics 44, 45
Aletheia (Unconcealment) 53, 54, 57 Evidence
Anamnesis 78 concept of 84, 85, 88, 90, 97
Anhypothetical, the 9, 17–23, 28–30
Archaeology 82 Factiality
Argument-function analysis 190, 192 principle of 21, 22, 29, 30
Assertion 51–59, 63, 64 Filial piety 169, 171, 179
Funktionsbegriff 114, 116
Being 18, 20, 21, 24, 27–29
Geschichtsforschung (Historical
Circularity research) 105
non-vicious 222, 231 Geschichtsschreibung (Historical
vicious 222, 231, 232 writing) 105
Classical syllogism 40 God
Cogito 26, 69, 74 concept of 8, 11, 148, 189, 190, 196–199,
Concept 201–204
family resemblance 133 death of 21
minimal 152n17, 154, 155, 157, 158, 161, Good, the 18, 20, 30
162 Gödel’s completeness theorem 204
religious 217 Gödel’s incompleteness theorem 204
robust 154, 155, 157, 161 Grammar 84
vague 214
Conceptual scheme 149, 150, 156n28, 158 Historical representation 108, 104–108,
Conceivability 210, 211, 213, 214n24–25 110, 113, 114, 118–121
Conceivability argument 208–210, 212, Humanism 37
216, 217–219 Hypothesis 18, 20, 30
Congruism 188 H2O 211, 212
Consciousness 207, 209, 211–215, 218
Conservatism 44, 46, 47, 48 Idealism 7, 33, 36
Contingency 20–22 Identification 105, 108, 113
Contradiction 20, 21, 22, 28 Identity 109
Correlationism 21, 22 Illogical thought 126, 130, 146
Counterfactual 213 Imagination
Custom 171, 172, 178–181, 183, 185, 186 artistic 34, 35
historical 34
Dasein 54, 57, 60, 225 scientific 34, 35
Denotation 110, 111, 114 Incongruism 188, 189, 190, 193–197
Description 109, 110, 114 Individualmetaphysik 103
Individuation 105, 108, 113
Efficacy Inference rules 191, 193, 194, 199, 202, 203
causal 225 Interpretation 59, 61–63
logical 225, 226 Intuitive expectation 217
238 Subject Index
Irrationalism 96 and politics 8, 9, 44

pragmatist critique of 10
Justification propositional 85, 92
deductive 221, 229 purist 1, 2, 5, 8, 9, 11
inductive 221, 222, 229, 230 quantum 5
of question and answer 10, 81, 82,
King Lear 11, 169, 171, 172, 174, 175–179, 181, 83–90, 91–98, 222, 224–228
185, 186 relevance 5–6
representationalist 6, 7, 10, 105, 120,
Language 121
artificial 5 and society 9
instrumentalist view of 58 symbolic 104, 112, 114, 118, 120
natural 5 universalist 1, 2, 5, 7, 8, 9
and practical activity 59, 63, 64 validity of 18
Linguistics whig interpretation of 10, 114
structural 67, 69, 72n13, 73, 78 Logical
Logic alien 8, 10
absolutist 1, 7, 11, 124, 149 consequence 125, 126, 127, 128, 129,
alternative 5, 6 130, 132, 133, 134, 135, 146
anti-formalist 189, 190, 201, 203, 204, framework 96–98
205–206 necessity 128, 142
Aristotelian 104, 114–121 ought 170
and culture 9 Logocentric predicament 11, 221, 232
deductive 10
of definitory rules 7 Mathème 69–72, 79
deviant 124, 125, 127 Mathematization 70–72, 79
dialectical 7, 17, 86, 97–99 Meaning 38, 40, 41, 42, 46, 105, 108, 110, 111,
first-order 188–190, 191, 193, 194–197, 116, 119, 120
197–199, 202, 204, 205 Metalanguage 136, 145
formal 1, 2, 3, 6, 7, 8, 9, 10, 11, 35, 36, 38, Metaphysics 21
39, 40, 42, 43, 44, 47, 48, 53, 81, 82, 84, Model 114, 116, 117, 120
90, 188, 197
and French philosophy 10 Necessity 20
and history 11, 91, 120 Non-contradiction
humanist 10, 39, 42, 45, 46, 47 principle of 10, 19
inductive 88 Normative Judgment Internalism 8, 11,
informal 11 181, 182, 184
intuitionist 5 Normativity 170, 171, 173, 180, 181–185, 187
and love 8, 9, 23, 24, 26, 28, 29, 31 Norms 169, 173, 174n10, 180, 184
narrative 6
normativity of 9 Obstacle, the 66–68, 70, 71, 73–77, 79
other 1, 6, 9 Ontology 207
of oughts 11, 170, 171, 185
paraconsistent 6 Panpsychism 211, 212
perfectionist 1, 2, 7, 8, 9, 11, 149 Personal identity 215
and philosophy of religion 8 Phenomenology 68n6
pluralist 8, 9, 82, 91, 95, 124, 127, 133 Physicalism 209, 211, 215
n34, 135, 146
Subject Index 239
Pluralism Representation 55, 56, 66, 67, 70, 71, 74,

metaphysical 11, 148–152, 155–156, 75, 77, 78, 79
159, 167 Representationalist universe 106–108, 113,
Possible world 212 118, 119, 121n23
Pragmatism 10, 33, 34 n10, 35, 37, 38, 43, Revolution
44, 86 historical 32–35
Presupposition scientific 32–35
absolute 222–224, 226, 227
relative 227, 228 Santa Claus 214
Psychical research 42, 48 Science
Psychoanalysis 10, 66–68, 69–79 as systematic thinking 87, 89
Psychologism 124, 143, 144 history as 87
Positivism 9, 87 natural 87
Possibility Sciences and humanities
logical 207, 210–213, 216, 217 relationship between 103, 113, 114, 121
metaphysical 209, 210, 211–212, 215 Self, the 34, 36, 37–39, 41, 47, 170, 171, 180,
Praedicatum inest subjecto 182, 186
principle of 118, 120 Sexuation 66n2, 70, 72, 76
Public good 170, 176 Sexual difference 66n2, 70, 75–78
Sexual relation 69, 70, 74–78
Quantifier analysis 189, 190, 192 Signifier 70, 71, 73
Quid pro quo Substanzbegriff 114–117
principle of 169, 175, 178, 181
“The” 108, 109
Rational revision 134, 137, 141, 144 Theory of truth
Rationalism 3 correspondence 55, 81
Rationality 9, 20 Truth 36, 39–42, 43, 44, 52–64, 104, 108,
Realism 109, 111, 112, 117, 118n20
epistemological 84n17, 84 Two-dimensional semantics 212, 213
metaphysical 8, 9
Reason Unconscious, the 8, 66–69, 71n10, 72–78
hypothetical 22 Understanding
practical 93, 94 practical 59–62
prudential 169–171, 175, 176, 179–181, Universalmathematik 103
183, 186
theoretical 93, 94 Verification
Réel (the Real) 66–68, 70–72, 74, 75, principle of 51, 52
77–79
Relation 114, 115, 119, 120 Zermelo-Franckel set-theory 193
Relativism 95 Zombie argument 11, 208–219
Relativity
conceptual 149n5, 155, 156, 162–167

Admir Skodo, Ed., Other Logics - Alternatives To Formal Logic in The History of Thought and Contemporary Philosophy-2014

Uploaded by

Copyright:

Available Formats

You might also like

Admir Skodo, Ed., Other Logics - Alternatives To Formal Logic in The History of Thought and Contemporary Philosophy-2014

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Admir Skodo, Ed., Other Logics - Alternatives To Formal Logic in The History of Thought and Contemporary Philosophy-2014

Uploaded by

Copyright:

Available Formats

Other Logics

Michael Krausz, Bryn Mawr College

Annette Baier, University of Pittsburgh

The titles published in this series are listed at brill.com/phc

Copyright 2014 by Koninklijke Brill nv, Leiden, The Netherlands.

This book is printed on acid-free paper.

1 Proving the Principle of Logic: Quentin Meillassoux, Jean-Luc Nancy,

9 The Logic of “Oughts” and the Bindingness of Past Practice:

This book is the final outcome of an international project funded by Stiftelsen

Aaron James Wendland

There is a widespread assumption in today’s western philosophical culture that

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004270183_��2

to a “general logic,” the former of which fruitfully seeks to overcome flaws of

5 Tiles, “Kant,” 102.

metaphysics,” and those who propound a non-metaphysical reading.9 Again,

When we state in everyday language that a person’s logic fails, we nor-

9 Burbidge, “Hegel’s Logic,” 170.

In each period in the history of logic, researchers called logicians have

13 Haaparanta, “Introduction,” 4–5.

logic,” “paraconsistent logic,” albeit all as formalized systems.17 The chapter

24 Freeden, Ideologies, 149.

leading to a highly original destabilizing interpretation of the principle of

Jaako Hintikka, Inquiry as Inquiry: A Logic of Scientific Discovery (Dordrecht: Kluwer,

Proving the Principle of Logic

1 The Anhypothetical Principle in Plato and Aristotle

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004270183_��3

Understand that by the other segment of the intelligible I mean that

3 Plato, Republic 6, 510c3–d2.

So for Aristotle the anhypothetical is 1) brought by the philosopher to any

2 Quentin Meillassoux’s Anhypothetical

In one contemporary re-working of this theme of the anhypothetical,

7 Aristotle, Metaphysics 4, 1005b.

Meillassoux claims to establish philosophy on the basis of a demonstrated

This proof, which could be called ‘indirect’ or ‘refutational,’ proceeds not

Meillassoux rejects all attempts to understand the beginning of reason as an

The principle of factiality is precisely this: that facticity cannot be a fact. An

15 Meillassoux, After Finitude, 61.

3 Jean-Luc Nancy, Thought, and Love

Whereas Meillassoux seeks to develop Aristotle’s anhypothetical, Nancy

must intend both a subjective and an objective genitive. Nevertheless, as much

The thinking of love—if it is necessary to solicit it, or if it is necessary

19 Nancy, A Finite Thinking, 248.

What is at stake here is not a question for philosophy—“why is there some-

22 Nancy, A Finite Thinking, 247.

for philosophy’s contemplation, because it is also the movement through

This is why love is always missed by philosophy, which nevertheless does

24 Nancy, A Finite Thinking, 253.

So love, the essence of thought, cannot be exhausted as the object of the

28 Nancy, A Finite Thinking, 260.

According to this schema, the nature of love is shown to be double and

Our Nancean detour complete, we can now return to Meillassoux to recon-

30 Nancy, A Finite Thinking, 250. Translation altered.

are not contradictory”). And the impossibility of this reduction neutralises

The Self, Ideology, and Logic

Classic philosophies have to be revised because they have to square

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004270183_��4

open to contributions from scholars trained in one form of science (human

11 George J. Stack, “Nietzsche’s Influence on Pragmatic Humanism,” Journal of the History of

© koninklijke brill nv, leiden, ��4 | doi ��.��63/9789004270183_��2

© koninklijke brill nv, leiden, ��4 | doi ��.��63/9789004270183_��3

© koninklijke brill nv, leiden, ��4 | doi ��.��63/9789004270183_��4

© koninklijke brill nv, leiden, ��4 | doi ��.��63/9789004270183_��5

© koninklijke brill nv, leiden, ��4 | doi ��.��63/9789004270183_��6

d iscerning a logic of the unconscious, since both traditional (Aristotelian) and