value and context the nature of moral and political knowledge oct 2006

VALUE AND CONTEXT
This page intentionally left blank
Value and Context
The Nature of Moral and Political Knowledge
ALAN THOMAS
CLARENDON PRESS · OXFORD
1
Great Clarendon Street, Oxford ox2 6dp
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing w orldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
© Alan Thomas 2006
The moral rights of the author have been asserted
Database right Oxford University Press (maker)
First published 2006
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,

or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the a ddress above
You must not circulate this book in any other binding or cover
and you must impose this same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Thomas, Alan, Dr.
Value and context : the nature of moral and political knowledge / Alan Thomas.
p. cm.
Includes bibliographical references and index.
ISBN-13: 978–0–19–825017–3 (alk. paper)
ISBN-10: 0–19–825017–7 (alk. paper)
1. Ethics. 2. Political science—Philosophy. I. Title.
BJ1012.T53 2006
170

.42—dc22 2006017202
Typeset by Laserwords Private Limited, Chennai, India
Printed in Great Britain
on acid-free paper by
Biddles Ltd, King’s Lynn, Norfolk
ISBN 0–19–825017–7 978–0–19–825017–3
13579108642
For Kathryn
This page intentionally left blank
Acknowledgements
This book originated in a doctoral thesis submitted to the University of Oxford in

1995. It has had a very lengthy period of gestation, for various reasons, and little of
the original text of the thesis survives in the present manuscript. I have incurred sub-
stantial debts to others over the course of writing the thesis and this book. The late
Bernard Williams supervised the thesis on which this book is based and it will be clear
from this work how much he shaped my entire conception of philosophy. I take issue
with many of his claims about moral knowledge in the course of this book but my
indebtedness to him will be equally evident throughout. He was the ideal supervisor
of my thesis and became a supportive and wise mentor at the outset of my career. His
passing greatly saddened me.
I was very fortunate that my doctoral thesis was examined by Charles Taylor and
Roger Crisp. Both of them gave me a great deal of valuable feedback that I have incor-
porated into the revisions of my thesis to make it more appropriate for publication as
a book. I am very grateful to them both. For helpful feedback either on the papers
that have been incorporated into this book or on the book itself I am also very grate-
ful to Edward Harcourt, Brad Hooker, Simon Kirchin, Richard Norman, Tom Pink,
Martin Stone, and Philip Stratton-Lake.
For financial support during the writing of my thesis I am grateful to Ian Honey-
man, the former bursar of St Hugh’s College, Oxford; St Hugh’s itself for the award
of a Senior Jubilee Scholarship; the late Jack Campbell-Lamerton, formerly bursar of
Balliol College; and Balliol College itself. Without the support of Ian Honeyman and
Jack Campbell-Lamerton I would have been reluctantly forced, on financial grounds,
to abandon my doctoral studies many years ago. More recently the Arts and Human-
ities Research Council supported work on this book and I am grateful to the Board
for this support.
I owe a particular debt of gratitude to Peter Momtchiloff at Oxford University
Press who has shown a great deal of faith in this project over many years. He was pre-
pared to wait until I was completely happy with the final product. His only comment
during this time was that ethics was an area of philosophy in which people took time
to get things right!
I am also indebted to Peter for his expert choice of the anonymous readers of the

manuscript of this book. The manuscript was reviewed twice, several years apart, by
two readers in each instance but with one person performing this task on two separate
occasions. While I am grateful to all three readers I do have to single out for particu-
lar thanks the individual, whomsoever he or she is, who twice read successive versions
of the manuscript. Each time he or she engaged with the arguments in considerable
detail, suggesting many improvements and often raising deep issues about contextu-
alism that forced me to rethink my own views. Hopefully one day that person will
identify him- or herself so that I can thank them in person. Several footnotes identify
viii Acknowledgements
particular places where I have responded to his or her suggestions in depth; in those
footnotesIrefertohimorher(forwantofabetterphrase)asthe‘firstanonymous
reader’. I would also like to thank Jenni Craig, production editor, and Rowena Anke-
tell, copy editor, for their expert work in the final production stages of this book.
I will conclude with my thanks to three others. Adrian Moore was one of my under-
graduate tutors, briefly my thesis supervisor and later a colleague at St Hugh’s. His
friendship for twenty years has meant a great deal to me and I have learnt more about
philosophy from him than from anyone else, including Bernard. Those who know his
work will see its pervasive influence throughout this book. My mother, Eira Thomas,
was a constant source of moral support throughout my doctoral research; she was
delighted when this book was accepted for publication. I very much regret that she
did not live to see it published. Finally, the reader has reason to be very grateful to my
partner, Kathryn Brown, who has read innumerable drafts of this book over many
years and has immeasurably improved it both substantively and stylistically. My per-
sonal debt to her goes beyond anything that I can put into words, but I dedicate this
book to her. This book would not have existed without her love, support, and prac-
tical advice at every stage of its composition.
PERMISSIONS
I am grateful to the editors of the following journals, or collections of papers,
for giving me permission to reprint material from previously published papers.
I am grateful to Ward E. Jones, editor, for permission to reprint sections from

‘Minimalism and Quasi-Realism’, Philosophical Papers (Nov. 1997), 233–9 which
are incorporated into Chapter 5 and to the editors and Edinburgh University Press
for permission to reprint substantial parts of ‘Consequentialism and the Subversion
of Pluralism’, in Brad Hooker, Elinor Mason, and Dale Miller (eds.), Morality,
Rules and Consequences (Edinburgh University Press, 2000), 179–202 which forms
the basis of Chapter 8. I am grateful to Roger Crisp and to Edinburgh University
Press for permission to reprint material from ‘Internal Reasons and Contractualist
Impartiality’, Utilitas, 14/2 (July 2002), 135–54 which forms the basis of Chapter 4.
I am grateful to Bob Brecher and Kluwer publishers for permission to incorporate
some sentences from ‘Nagel’s Paradox of Equality and Partiality’, published in Res
Publica, 9/3 (2003), 257–284 into Chapter 11. Finally, I am grateful to Peter Burnell
and Peter Calvert, the editors, and to Frank Cass Publishers for the reprinting of
passages from ‘Liberal Republicanism and the Role of Civil Society’, Democratisation
(Aug. 1997), 26–44 which are incorporated into Chapter 12.
Contents
Note to the Reader x
Introduction 1
I. MORAL KNOWLEDGE AND MORAL REASONS
1. The Problem of Moral Knowledge 11
2. The Case for Cognitivism 32
3. Values, Norms, and the Practical 53
4. Internal Reasons and Contractualist Impartiality 67
II. NORM-EXPRESSIVISM AND NON-OBJECTIVISM
5. A Critique of Expressivism 101
6. Non-Objectivism and Internal Realism 130
III. CONTEXTUALIST MORAL JUSTIFICATION
7. Epistemological Contextualism 169
8. A Contextual Model of Moral Justification 198
9. Tradition Based Moral Enquiry 225
10. Moral Belief and the Possibility of Error 246

IV. CONTEXTUALISM IN POLITICAL PHILOSOPHY
11. Political Liberalism and Contextualism 271
12. Political Liberalism and Civic Republicanism 295
Conclusion 321
List of Works Cited 323
Index 341
ANotetotheReader
Part I of this book sets out the basic assumptions of a cognitivist view of ethics that
treats moral properties as anthropocentric but real properties. It explains why there
is a problem of moral knowledge and related issues about moral motivation, such as
the distinction between internalism and externalism about motivation. Those who
are expert in the field and very familiar with these meta-ethical views and the debates
that they have stimulated may choose to omit the first three chapters and pick up the
argument at Chapter 4.
If it is said that Reason itself, rather than any particular statement of its
content, must remain as the final arbiter, then we must wonder what
precisely that is. If not as particular content, then the only sense in
which reason must endure is as an evolving chain of descent. Reason will
endure as whatever evolves or grows out of the current content of reason
by a process of piecemeal change that is justified at each moment by
principles which are accepted at that moment (although not necessarily
later on), provided that each evolving stage seems close enough to the
one immediately preceding it to warrant the continued use of the label
‘reason’ then. (The new stage may not seem very similar, however, to an
earlier, step-wise stage.) That degree of continuity hardly seems to mark
something which is a fixed and eternal intellectual point.
Robert Nozick, Invariances
This page intentionally left blank
Introduction
The aim of this book is to further an ongoing debate between ‘cognitivists’ and ‘non-

cognitivists’ about the possibility and the nature of moral knowledge. The former
assert, and the latter deny, that some of the moral claims we make should be inter-
preted as claims that are often known to be true. I think this debate is deadlocked. It
can only be resolved by re-examining the assumptions underlying our current options.
This book develops a contextualist approach to epistemology that I think offers the
best way forward for cognitivism. Contextualist models of knowledge have recently
received more philosophical attention, even in ethics. However, they have not, so far,
explicitly been placed at the service of the defence of moral cognitivism.
The starting point of my work is the innovative form of cognitivism that David
Wiggins and John McDowell developed in the nineteen eighties: a view that has been
described as ‘the most important contemporary challenge to the terms of the standard,
and perhaps stalemated, dialectic between noncognitivism and naturalistic cognitiv-
ism’.¹ They suggested that we should see moral values as attuned to our particular
sensibilities, but none the less as part of our moral experience and as indispensable to
our best moral explanations.² I sympathize a great deal with the criticisms that they
developed of standard non-cognitivist accounts of morality, particularly those variet-
ies of projectivism which see our moral judgements as the projection of values onto
an evaluatively neutral reality. The origin of such views is Nietzsche’s The Joyful Sci-
ence, although within the analytical tradition this view is particularly associated with
the work of Simon Blackburn and Allan Gibbard.³ Wiggins and McDowell suggested
that the attractions of such views are illusory when we focus our attention on so-called
‘thick’ ethical concepts, such as courage or brutality.⁴ When we examine our use of
such concepts from within our ongoing moral practices, we can come to see that there
is no possibility of dissociating their descriptive and their normative elements. This
seemed to them further to imply that there was no feasible project of separating out
our evaluative ‘projections’ from the non-evaluative reality onto which these judge-
ments were projected.
¹ Stephen Darwall, Allan Gibbard, and Peter Railton, ‘Toward Fin de Si`ecle Ethics: Some
Trends’, Philosophical Review, 101/1 (1992), 115–189 at 164.
² I will refer primarily to the papers collected in David Wiggins, Needs, Values, Truth,3rdedn.

(Oxford: Blackwells, 2000) and to the papers collected in John McDowell, Mind, Value and Reality
(Cambridge, Mass.: Harvard University Press, 2001).
³ Friedrich Nietzsche, The Gay Science, ed. Bernard Williams, trans. Josefine Nauckhoff and
Adrian Del Caro (Cambridge: Cambridge University Press, 2002); Allan Gibbard, Wise Choices,
Apt Feelings (Oxford: Oxford University Press, 1990); Simon Blackburn, Ruling Passions (Oxford:
Oxford University Press, 1999).
⁴ I will adopt the convention of referring to concepts, as opposed to the linguistic terms that
express them, in bold typeface.
2 Introduction
I have always sympathized both with Wiggins’s and McDowell’s overall project
and with the two analogies they used to illuminate their alternative view: between val-
ues and secondary properties, and between moral reasons and mathematical reasons.
Their project was to give an account of moral properties in which such properties
are construed as anthropocentric, but also as real. They are real in so far as they are
indispensable to the explanations we offer of moral phenomena, but not real in the
sense that they are grounded in the universe ‘as it is in itself’ without relation to the
interests and concerns of human beings. To explain this unique kind of metaphysical
status for moral values, they used analogies between the ontological status of values
and that of secondary qualities, such as colours, and between the ‘logical space’ of
moral reasons and the ‘logical space’ of constructivist mathematics.⁵ One of the aims
of this latter analogy was to capture the authority of moral reasons; the way in which
there appears to be a necessary connection between moral knowledge and the will in
an appropriately motivated agent.
However, my sympathy with this theory in its original form only goes so far. My
first, less radical, objection, is that the explanation given of the connection between
moral knowledge and the will stands in need of revision. I have my own story to tell
about how moral reasons motivate an appropriately motivated agent, which I will
present in Part II of this book.⁶ More radically, however, it seems to me that the
account offered of moral knowledge, as it stands, suffers from a major defect, high-
lighted in the powerful criticism offered by Bernard Williams in his insightful discus-

sion of the view and its limitations in Ethics and the Limits of Philosophy.⁷ I interpret
Williams’s criticism as follows: the proposed form of cognitivism can give an excellent
account of particular forms of ethical reasoning and judgement as they arise within a
given historical community with its culturally specific concepts, practices, and forms
of reasoning. However, in its original form, the theory fails to allow for the possibil-
ity of a certain kind of radical, distinctively modern, form of reflection in which we
⁵ The term of art of ‘the logical space of reasons’ is prominent in John McDowell, Mind and
World (Cambridge, Mass.: Harvard University Press, 2001). Its contemporary use originated with
Wilfrid Sellars, Empiricism and the Philosophy of Mind (Cambridge, Mass.: Harvard University
Press, 1997), §36. Sellars in turn drew on the discussion of ‘logical space’ in Ludwig Wittgenstein,
Tractatus Logico-Philosophicus, trans. David Pears and Brian McGuiness (London: Routledge and
Kegan Paul, 1961), 3.42 ff.
⁶ Specifically I discuss this point in Ch. 4, below. Another philosopher who shares my view
that cognitivism must be revised so as to accept the internal reasons constraint is Adrian Moore
in his recent work Noble in Reason, Infinite in Faculty (London: Routledge, 2003). See e.g. sect. 3
of the ‘Introduction’, sect. 5 and 6 of the ‘First Set of Variations’, and sect. 9 of the ‘Second Set
of Variations’. However, the main difference between the argument of this book and Moore’s
is that Moore is more concerned to vindicate a broadly Kantian picture of practical reasons.
Such an approach places cognitivism within the wider framework of constructivism about ethical
judgements. For a constructivist it is only within a perspective of engagement with the world that
is broadly practical (or in Moore’s terms ‘sense making’) that issues about the truth of particular
claims arise. I discuss Moore’s views in ‘Maxims and Thick Concepts: Reply to Moore’, presented to
the Central Division of the American Philosophical Association, 2005 and available for download
at < />⁷ Bernard Williams, Ethics and the Limits of Philosophy (London: Fontana, 1985).
Introduction 3
take a critical stance towards the practices of our own historical community, or are
challenged by the practices and ideals of other communities.
The danger is that we will not be able to face up to such challenges rationally.
If our only purchase on moral properties is from within our ongoing form of life,
how can we respond if it seems that there is more than one such form of life, each

offering access to its own range of moral properties? This is the problem posed to
cognitivism by a certain kind of ethical pluralism. This is the kind of pluralism that
goes beyond the pluralism born of reasonable disagreement that one could expect to
find within morality as it has developed within any cultural setting.⁸ It is the radical
pluralism that seems inherent to any account of morality in a distinctively modern
society.
We are, collectively, in the grip of a high degree of ‘epochal self-consciousness’, to
use Bernard Yack’s helpful phrase.⁹ We think of ourselves as modern in a way that
invites scepticism, but in a way that seems, similarly, irresistible. Even if, at the end
of the day, we are not convinced by the claim that the ethical is faced with unpre-
cedented challenges in a modern world, the claim is certainly recognisable. It is the
claim, or threat, of a kind of radical pluralism that goes beyond a reasonable plural-
ism within morality, where that latter idea can be seen as a reassuring complement
to our ideal of autonomy. The radical pluralism that is, by contrast, a threat, over-
whelms us with options that are in some way incomparable, incompatible, and an
avenue not to freedom but nihilism. Our modern aspiration to radical freedom has,
on this view, spun out of control.¹⁰ If there is anything to this line of concern, and it
seems to me foolhardy to deny it, then I believe that addressing it allows one to isolate
and to criticize one of the key assumptions driving Williams’s critique of the cognit-
ivist position. Anyone sympathetic to cognitivism ought to address it. The project of
this book is to take Wiggins’s and McDowell’s original proposals, apparently com-
pletely derailed by Williams’s critique, and put them back on track by arguing that
they can be developed to meet this challenge. However, the way in which cognitivism
needs to be developed has to be handled carefully.
⁸ Reflection on the existence of this kind of objective value pluralism is associated particularly
with the work of Isaiah Berlin and Stuart Hampshire. Berlin, Four Essays on Liberty (Oxford: Oxford
University Press, 1969); Hampshire, ‘Morality and Conflict’, in Morality and Conflict (Cambridge,
Mass.: Harvard University Press, 1983), 140–70; ‘Morality and Convention’, in Amartya Sen
and Bernard Williams (eds.), Utilitarianism and Beyond (Cambridge: Cambridge University Press,
1982), 145–57.

⁹ Bernard Yack, The Fetishism of Modernities: Epochal Self-Consciousness in Contemporary Social
and Political Thought (Notre Dame, Ind.: University of Notre Dame Press, 1997), ‘Introduction’,
1–16.
¹⁰ This critique of the modern notion of freedom has generated a substantial literature, of which
the best is still Iris Murdoch, The Sovereignty of Good (London: Routledge and Kegan Paul, 1970).
For a more general context and a more optimistic analysis there is Robert B. Pippin, Modernism as a
Philosophical Problem, 2nd edn. (Oxford: Blackwell Publishers, 1999). I have also benefited from the
contrasting assessments of Hegel’s views on modernity in Michael Allen Gillespie, Hegel, Heidegger
and the Ground of History (Chicago: University of Chicago Press, 1984) and David Kolb, The
Critique of Pure Modernity: Hegel, Heidegger and After (Chicago: University of Chicago Press, 1986).
4 Introduction
I will dissent from one well-established strategy. One way of defusing this worry
about the corrosive effects of modern radical pluralism is to argue that all our modes
of reasoning and judgement face this problem. The problem comes with the terrain,
as it were, and given that ethical reasoning and thought is no worse off than any
other area of our knowledge, one can afford a relaxed attitude to the problem. Our
aspiration to objectivity is limited, urges the interlocutor, and since it is limited every-
where, it is possible to reimmerse oneself in ethical thought and life with good faith.
This move is generally developed from within the perspective towards the realism
debate developed by internal realism, associated particularly in the context of eth-
ics with Hilary Putnam and Sabina Lovibond. I explain in Chapter 6 why I do not
find this position satisfactory.¹¹ I agree with Bernard Williams and with others who
have further developed and refined his position, such as Adrian Moore, that while
our knowledge is indeed conditioned by our perspective on the world, absolute or
non-perspectival knowledge is possible in the way that the internal realist denies, in
spite of the ubiquity of such human, all too human, conditions on our knowledge.¹²
Both our scientific practice and, indeed, our ordinary concept of knowledge impli-
citly allow that while our ordinary grasp of the world is ineluctably conditioned by
our perspective, certain aspirations to knowledge contain the idea that we can detach
from any such local perspective. This detachment allows us to form a more inclus-

ive perspective that we hope will be maximally independent of our peculiarities as
knowers.
The implication of this view is that physics, for example, can transcend our his-
torical point of view in a way that seems wholly inappropriate for moral reasoning,
which seems more tied to a local perspective and its peculiarities. One response to
this position is to argue that it once again invites a relaxed diagnosis, perhaps a form
of sophisticated relativism. Our local view, conditioned by its peculiarities, is ours,
whereas other societies and historical periods have their own particular view. Given
that each view is, as it were, hermetically sealed from the others and, further, given
that we have abandoned the aspiration for a more than local perspective on ethical
practices, the resulting position may be practically uncomfortable, but it poses no dif-
ficulties at the level of theory. Such is the position of the postmodern ironists, such as
Richard Rorty and Jean-Franc¸ois Lyotard.¹³
¹¹ A position set out in the following works: Hilary Putnam, Realism with a Human Face
(Cambridge, Mass.: Harvard University Press, 1992); The Many Faces of Realism (La Salle, Ill.: Open
Court Publishing, 1987); Renewing Philosophy (Cambridge, Mass.: Harvard University Press, 1995);
The Collapse of the Fact/Value Dichotomy and Other Essays (Cambridge, Mass.: Harvard University
Press, 2002); Ethics Without Ontology (Cambridge, Mass.: Harvard University Press, 2004); Sabina
Lovibond, Realism and Imagination in Ethics (Oxford: Basil Blackwell, 1983). John McDowell also
seemed initially drawn to this line of response to Williams’s arguments: see esp. ‘Review of Ethics
and the Limits of Philosophy’, Mind, 95/379 (1986), 377–88.
¹² Bernard Williams, Descartes (London: Penguin Books, 1978); Adrian Moore, Points of View
(Oxford: Oxford University Press, 1997).
¹³ Richard Rorty, Philosophy and the Mirror of Nature (Oxford: Basil Blackwell, 1979); Jean-
Franc¸ois Lyotard, The Postmodern Condition: A Report on Knowledge, trans. Brian Massumi
(Manchester: Manchester University Press, 1984).
Introduction 5
I am not happy with this view, not because I regard it as shockingly amoral, but
rather because I believe a more attractive alternative is available. There is one central
claim of Rorty’s and Lyotard’s with which I agree: namely, that we can no longer

overcome this particular intellectual predicament, which has a long history, by appeal
to a ‘grand historical narrative’ of a Hegelian kind. In this book I aim to give this
claim a precise sense. I interpret it as the view that while we can form a conception of
what it would be for one framework of ethical judgement to be superior to another,
we cannot iterate this conception to yield the idea of that framework that cannot be
surpassed by any other. This would be an instance of the chain fallacy of arguing from
‘all chains have an end’ to ‘there is an end to all chains’.¹⁴ Expressed in this limited
way, the denial that we possess a grand narrative of the legitimation of ethical beliefs
steers clear of an obvious pitfall: self-refutation through paradox. From the standpoint
of which historical interpretation, the sceptic asks, does one announce the end of all
historical narratives, if not from that of the very kind of narrative whose existence is
being denied?¹⁵
This paradox is a variation of a paradox that may be felt to be problematic for
Rorty’s and Lyotard’s deflationary project in its original version: it certainly looks as if
the unavailability of a standpoint outside our various language games, the ‘God’s eye
point of view’ of a ‘final interpretation’ is being asserted from somewhere, and a some-
where remarkably similar to the traditional standpoint outside our various language
games.¹⁶ This book tries to avoid this paradox, too, by arguing that our conception of
objectivity and our conception of what, in a particular case, constitutes an appropri-
ately objective set of standards governing a given subject matter evolve reciprocally.¹⁷
I will discuss this issue throughout the book.
The challenge I have set myself, then, is to find a way of defending the rationality of
moral reasoning across entire frameworks of moral belief, across cultures and historical
periods, whileallowing forthe kind ofcritical insightinto moralpractices thatWilliams
urged was lackingin theoriginal versionof moralcognitivism. Yetthis positiveaccount,
while acknowledging the historical contextualization of moral knowledge, must also
dispense with the Hegelian resources of a grand historical narrative. It is a difficult
challenge, but I hope that the contextualist argument that I offer can meet it.
Contextualism is best introduced as a genuine alternative to both foundationalist
and coherence theories of knowledge.¹⁸ Both of these alternative approaches to know-

ledge make the error of imagining that we can treat the system of knowledge as a
¹⁴ Elizabeth Anscombe identified an instance of the chain fallacy (controversially) in the opening
chapter of Aristotle’s Nichomachean Ethics, trans. J. A. K. Thompson (London: Penguin Classics,
2004): Anscombe, Intention (Ithaca, NY: Cornell University Press, 1957), 34; Anscombe, An
Introduction to Wittgenstein’s Tractatus
(London: Hutchinson, 1967), 15–16.
¹⁵ The idea that the very idea of postmodernism is insuperably beset by paradox is discussed by
Yack, Fetishism of Modernities,ch.1.
¹⁶ Bernard Wiliams, ‘Auto da F
´
e’, in Alan Malachowski (ed.), Reading Rorty (Oxford: Basil
Blackwell, 1990), 26–37.
¹⁷ This preserves one (pretty minimal) sense to the philosophical term of art, ‘dialectic’.
¹⁸ For a general perspective on the epistemology of contextualism there is an excellent survey
in David B. Annis, ‘A Contextualist Theory of Epistemic Justification’, American Philosophical
Quarterly, 15 (July 1978), 213–19.
6 Introduction
whole. Both view justification as a matter of bringing the entire system of knowledge
to bear on particular knowledge claims within it. The standard metaphors used to
explain the structure of knowledge are the coherentist sphere and the foundationalist
pyramid. In contrast to these metaphors I suggest the contextualist image of know-
ledge as a ‘crazy’ or a ‘patchwork’ quilt. From a contextualist perspective, the system
of knowledge is made up of loosely related contexts. Within each one, some beliefs are
taken for granted, and provide a fixed framework for the evaluation of other beliefs.
However, these framework beliefs are not unchallengeable, as such, because they may
be the objects of enquiry in a different context. Indeed, we can be motivated to move
to a further context by problems generated within the original context.
I view the ethical understanding of a given society as forming such a loose collection
of contexts. From within any particular context we cannot intelligibly ask whether
our whole outlook should be scrapped and completely replaced. However, we are

in a position to entertain specific challenges either from within our culture or from
outside it. In the latter case I argue that we can draw a further distinction, deploying
for this purpose Williams’s account of the relativism of distance.¹⁹ Williams argued
that relativistic confrontations take two forms. In the first a person or group faces a
choice between two systems of belief that are genuine alternatives for them. In the
second a person or group faces a choice between two systems of belief that they could
not mutually inhabit without radical self-deception. Williams calls the former ‘real’
and the latter ‘notional’ confrontations. I use this distinction to argue that not every
challenge to our ethical understanding originating from beyond itself poses the same
kind of challenge to the outlook that we currently endorse. When we observe different
ways in which other people have felt, thought, and acted differently not every such
case is a challenge to us. Some of them are too far, metaphorically speaking, to be
real options for us, alternatives in practice to what we already do. However, when
such alternatives are close enough, we have the resources of reason, sympathy, and
imagination to assess the alternatives in the light of our own best practices.
The account of moral reasoning I offer has elements in common with theses
developed by Alasdair MacIntyre and Charles Taylor.²⁰ They have both been drawn
to a form of historicism as an account of practical and moral reasoning. They
each have a very similar picture of such reasoning to mine: it is argument pitched
at the level of historically conditioned frameworks of belief. However, alongside
this convergence between their position and my own I identify several points of
disagreement which I would trace to the residual influence of a form of Hegelian
grand historical narrative on their work.²¹ This is precisely the kind of narrative that I
agree with Rorty and Lyotard we do not possess. Specifically, I will argue that it is too
optimistic to view our evolving moral understanding as increasingly comprehensive,
¹⁹ Bernard Williams, ‘The Truth in Relativism’, in Moral Luck (Cambridge: Cambridge
University Press, 1981), 132–43.
²⁰ Alasdair MacIntyre, After Virtue (London: Duckworth 1981); Whose Justice? Which Rationality?
(London: Duckworth, 1988); Three Rival Versions of Moral Enquiry: Encyclopaedia, Genealogy and
Tradition (London: Duckworth, 1990); Charles Taylor, Sources of the Self (Cambridge: Cambridge

University Press, 1989).
²¹ I will argue for this claim, particularly with regard to Taylor, in Ch. 9, Sect. three.
Introduction 7
at the service of an ever-deepening insight, or as capable of overcoming all the
conflicting commitments of a modern ethical outlook. ²²
The view I defend may, unfairly, be accused of the besetting vices of any ‘internal-
ist’ view of ethical reasoning: such views are accused of being inherently relativist and
inherently conservative. Both criticisms depend on the belief that the contextualist
cannot accommodate the correct form of social criticism: one with sufficient detach-
ment from ideological beliefs to be able to identify and explain them, but without
such detachment that it marks a lapse back into a foundationalist perspective. I
devote Chapter 10 to explaining why I believe that the contextualist is particularly
well placed to explain the nature of social criticism.
I conclude my main argument with a coda that examines the consequences of my
view of moral knowledge for social and political philosophy. Contextualism has not
been as neglected an option in political philosophy as it has been in moral philosophy,
although I believe the argument that I have developed casts further light on the issues
involved. I argue that there is a close connection between the model of knowledge
that I have defended and Rawls’s political liberalism.²³ Rawls’s late views seem to me
to be a thorough statement of the consequences of adopting contextualism in polit-
ical philosophy. For his own purposes, Rawls does not quite see things this way, so I
explain in Chapter 11 why I think this apparent disagreement emerges. I then explain
how Rawls’s explicit adoption of contextualism allows one to defend the ‘politiciza-
tion of justice’ in his later work from both internal and external criticism. I go on,
in the final chapter, to explain how this form of political liberalism can accommod-
ate key emphases taken to be the exclusive preserve of the communitarian, centrally
the concepts of citizenship and of civil society. The result is a view that I call liberal
republicanism, a form of liberalism that seems to me correctly to exemplify our best,
that is, most contextually justified, political theory.
²² Taylor’s Sources of the Self ostensibly goes no further than a phenomenology of contemporary

pluralism, but there is, I will argue, enough evidence to suggest that Taylor envisions a final recon-
ciliation of this kind, at least when we have closely examined which of our contemporary outlooks
lack the kind of justification that Taylor demands of them.
²³ John Rawls, Political Liberalism (New York: Columbia University Press, 1993).
This page intentionally left blank
PART I
MORAL KNOWLEDGE AND
MORAL REASONS
This page intentionally left blank
1
The Problem of Moral Knowledge
§1: The non-cognitivist positions of Mackie and Harman are interpreted
and criticized. Five issues concerning moral knowledge are highlighted as of
future importance to the argument. §2: Minimalism about truth is adop-
ted as a framework for discussions of objectivity, which is conceived of more
broadly than the issues of either realism or cognitivism.
The aim of this book is to argue that we possess substantial amounts of moral and
political knowledge. A necessary preliminary is to bring out some of the peculiarities
of the problem of knowledge in these two cases. I will, at this point, focus on the idea
of moral knowledge. Bernard Williams, whose views will be acting as a foil to my own
throughout this book, pointed out that philosophical discussions of knowledge usually
treat the problem in an ‘all or nothing’ manner.¹ To take one well-worn example, you
either know that there is a red tomato in front of you, or it is a visual delusion and you
do not. The way in which our knowledge of the external world stands or falls with such
mundane examples has attracted some of the most interesting philosophy about the
problem of exactly what scepticism amounts to in this case. However, the case of moral
and political knowledge is not like this.² Williams pointed out that moral knowledge
raises special problems independent of problems about knowledge per se, not least
that its existence or otherwise is properly to be viewed as a matter of degree. This is a
consequence, as I will argue in Chapter 6, of his view that social conditions may be

¹ Williams, Ethics and the Limits of Philosophy, 24–5. Williams contrasted his treatment of this
point, in the special case of ethics, with that offered by Renford Bambrough in Moral Scepticism
and Moral Knowledge (London: Routledge, 1979). Bambrough attempted a Moorean refutation of
ethical scepticism that pointed to the existence of a single moral fact whose certainty was greater
than that of the sceptic’s reasoning to his or her sceptical claim that there were no moral facts.
Williams pointed out that there are intelligible alternatives to an ethical life whereas it is not clear
that scepticism about the external world offers an intelligible alternative to our ordinary conception
of an objective world not of our own making. More generally, Williams argued that Bambrough
does not establish what makes his representative example truly representative.
² This problem of representativeness in samples arises similarly for a view indebted more to
Wittgenstein’s work than to Moore’s, namely, that of J. L. Austin and Stanley Cavell. This argument
is well represented by the latter’s treatment of scepticism in The Claim of Reason: Wittgenstein,
Scepticism, Morality and Tragedy (Oxford: Oxford University Press, 1999). This view, too, focuses
on the peculiar examples used to sustain what Cavell calls ‘the sceptical recital’. Cavell’s ingenious
attempt to undercut the sceptic will be discussed critically in Ch. 7, Sect. 1 and more positively in
Ch. 10, Sect. 4. At this point I note Barry Stroud’s observation that this approach to scepticism, too,
founders on the failure to specify what counts as representativeness in epistemological examples.
Stroud argues for this conclusion in ‘Reasonable Claims: Cavell and the Tradition’, in Understanding
Human Knowledge (Oxford: Oxford University Press, 2000), 51–70.
12 Moral Knowledge and Moral Reasons
more, orless, hospitableto theexistence ofsuch knowledge. (Williams believed that the
kind of modern society that we live in is distinctively inhospitable to such knowledge.)
In this opening chapter I will consider the views of two philosophers who argue
that the existence of moral knowledge is of direct ethical relevance; all the more unfor-
tunate, then, that no such knowledge exists. These are the positions of the non-cogni-
tivist and of the error theorist. In this chapter and the next I will set out the challenges
posed by non-cognitivism and error theory to those, such as myself, who want to
defend the existence of moral knowledge and will begin to suggest criteria that any
satisfactory account of moral knowledge must meet. However, this question—that
of the existence of moral knowledge—must first be detached from two others with

which it is often confused. These two cross-cutting issues are those of authority and
of freedom. I believe that many people who are resistant to the idea that we possess
moral knowledge are often motivated by other moral commitments. Those commit-
ments seem to them, paradoxically, to demand the moral rejection of the idea that we
have moral knowledge (although, unsurprisingly, that tends not to be the way that
these objectors would phrase their objection).
The first question concerns the role of authority within moral thinking. One line
of resistance to the idea of moral knowledge claims that the existence of such know-
ledge would place some people in a position of authority over those epistemically less
well placed than themselves. Indeed, the non-existence of moral experts has been used
as an argument against the existence of moral knowledge.³ My response to these argu-
ments is to argue that these issues are quite separate from the issues of the existence
and extent of moral knowledge. The problems surrounding the epistemic authority
of morality can arise only when the case for and against moral knowledge has been
separately decided.
Whether or not there is moral knowledge does not make people authoritarian
or tolerant. There are plenty of instances of intolerant moral non-cognitivists and
equally of tolerant cognitivists. Suppose, for example, that you believe that there is
moral knowledge, but that it is an integral part of the good life that people should
realize this knowledge in their own lives in their own way. You might, plausibly,
believe this on the grounds that you believe an ethical ideal of autonomy is very
important. A view of this kind makes it obvious that it is at least possible to be a
cognitivist, but to believe that it is very misguided to force moral truth upon other
people (if such an idea even makes sense).
My response to the related ‘no moral experts’ argument is, once again, to separ-
ate it out for further consideration. One might, for example, meet it by arguing that
if there is moral knowledge, it is not very difficult to obtain and does not require a
major cognitive feat on the part of the mature moral agent. Perhaps it simply involves
what Kant called ‘mother wit’. I am not sure that there is very much one can say about
the qualifications for being a moral agent that is both general and informative: people

seem to draw on a range of cognitive capacities in moral judgement. A provisional list
would include sympathy, appropriate emotional response, careful and conscientious
³ See e.g. Bernard Williams, ‘Who Needs Ethical Knowledge?’, in Making Sense of Humanity
(Cambridge: Cambridge University Press, 1995), 203–12.