Post-liberalism

Post-liberalism
Studies in political thought

John Gray

London and New York

First published 1993
First published in paperback 1996
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
270 Madison Ave, New York NY 10016
Transferred to Digital Printing 2006
© 1993, 1996 J. Gray
Typeset in Times by LaserScript, Mitcham, Surrey
All rights reserved. No part of this book may be reprinted or
reproduced or utilized in any form or by any electronic,
mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.
British Library Cataloguing in Publication Data
A catalogue reference for this book is available from the British Library
Library of Congress Cataloguing in Publication Data
A catalogue record for this book has been requested

ISBN 0-415-08873-9 (hbk)

ISBN 0-415-13553-2 (Pbk)

Publisher's Note
The publisher has gone to great lengths to ensure the quality of this
reprint but points out that some imperfections in the original may
be apparent
Contents

Preface vii
Acknowledgements x

Part I Thinkers
1 Hobbes and the modern state 3
2 Santayana and the critique of liberalism 18
3 Hayek as a conservative 32
4 Oakeshott as a liberal 40
5 Buchanan on liberty 47
6 Berlin's agonistic liberalism 64

Part II Critiques
7 The system of ruins 73
8 The delusion of glasnost 85
9 The academic romance of Marxism 90
10 Philosophy, science and myth in Marxism 99
11 Against Cohen on proletarian un freedom 123
12 Totalitarianism, reform and civil society 156
13 Western Marxism: a fictionalist deconstruction 196
14 Post-totalitarianism, civil society and the limits of the Western
model 202
15 Political power, social theory and essential contestability 216
16 An epitaph for liberalism 238
17 The end of history - or of liberalism? 245

Part III Questions

18 The politics of cultural diversity 253
vi Contents
19 Conservatism, individualism and the political thought of the
New Right 272
20 What is dead and what is living in liberalism 283
Notes 329
Index 349
Preface

First published in 1993, Post-liberalism: Studies in Political Thought was

intended as a companion volume to my earlier volume, Liberalisms: Essays
in Political Philosophy (Routledge, 1989). In that earlier collection, I had
considered recent versions of fundamentalist or doctrinal liberalism,
according to which liberal institutions framed the only fully legitimate
regime for the whole of humankind, and found all of them flawed irrepar-
ably. The liberal form of the Enlightenment project - which was the project
of grounding liberal values, so as to confer on them a universally compell-
ing authority in the demands of reason - was, I concluded, bankrupt. That
earlier volume ended on a sceptical note, with little being said positively as
to what a successor to the liberal project might look like. In Post-liberalism,
I tried to give a more substantive statement of a successor-theory to
liberalism, in which liberal practice is defended not as the application of a
universal political morality but as an historical inheritance which meets
human needs in the context of late modernity. In this post-liberal per-
spective, the failures of foundationalist liberal political philosophy are
accepted as final and not to be lamented, since we have all we need in the
core institutions of liberal civil society, now spreading to many parts of the
world in the wake of the communist collapse. I concluded that it was to a
consideration of the details of the historical inheritance of liberal practice
that we should now repair, rather than to any of the theoretical vistas
offered by liberal theory, as we entered the early post-modern period.
The relation to liberalism of the post-liberal perspective I developed
when this book was first published was consciously dialectical and tacitly
ironic. A post-liberal position was advanced as being better able than any
variety of liberal political philosophy, in the historical context of late
modernity, to address the problems which liberal thought had sought to
resolve during the early modern period. These were problems arising from
the need to find a sustainable modus vivendi among people having
divergent world-views and value-perspectives that are rationally
viii Preface
incommensurable - that is, whose differences cannot be settled by any kind
of rational arbitration. The post-liberal position advanced in the first edition
of this book defended liberal institutions as being, almost universally, best
equipped to achieve such a modus vivendi in late modem conditions, while
rejecting the foundationalist claims made on behalf of liberal practice by
contemporary exponents of the Enlightenment project. The 'post-modem
liberal conservatism' argued for in this book resembles the position of
Richard Rorty, by which it was avowedly inspired, in attempting a radically
historicist reformulation of liberal theory, in which the near-universal
authority of liberal institutions is preserved.
In subsequent work I have considered that post-liberal view more care-
fully, and found it wanting. In Beyond the New Right: Markets, Government
and the Common Environment (Routledge, 1993), I examined the applica-
tions of a paleo-liberal theory to public policy during the brief episode of
New Right political hegemony in several Western countries, notably
Britain and the United States. In that study I argued that the intellectual
'foundations' of New Right policy were insubstantial in the highest degree,
dissolving at the first touch of serious criticism: but that New Right political
thought was significant, despite is ephemeral character, as a symptom of
errors which pervade liberal theory far more generally. From the standpoint
of the argument advanced in Beyond the New Right, the deeper lessons to
be learnt from the New Right episode do not concern the errors involved in
making of market institutions articles of faith rather than modest artifices
for the achievement of human purposes. They concern what the liberalism
of Nozick and Hayek has in common with the liberalism of Rawls and
Dworkin. It is this - the use of an abstract, ahistorical cipher to represent the
human subject, the neglect of the particular communal attachments that are
always constitutive of human identities, and the legalist project of deriving
a system of basic liberties or fundamental rights, which acts a limitation on
political practice everywhere - that is most deeply symptomatic of the
poverty of recent political thought. Beyond the New Right concludes with
the hope that a species of traditional conservatism, in which the local and
particular character of human identities and the embeddedness of the
human subject in diverse forms of common life are fully recognised, can be
reaffirmed.
In EnLightenment's Wake: PoLitics and CuLture at the CLose of the
Modern Age (Routledge, 1995). I take the argument of Post-liberaLism one
step further. I contend that there is no reason why the liberal project of
devising a modus vivendi for people who hold to incommensurably
different conceptions and values should everywhere be best achieved by the
adoption of liberal institutions. The true successor to the liberal project is
pLuralism, the position which affirms that the diversity of forms of ethical
 Post-liberalism ix
life is legitimately mirrored and reflected in a variety of institutions and
regimes, both liberal and nonliberal. Pluralism is a response to the diversity
of incommensurable values and perspectives that is a peculiarly prominent
feature of early post-modernity. It relinquishes the universalist ambitions of
the Enlightenment project and of liberal theory, and maintains that the
terms of peaceful coexistence must be worked out locally and in practice,
vary considerably along with cultural and historical contexts, and will only
sometimes encompass the construction of liberal institutions. Nearly all
contemporary political thought is a set of variations of Enlightenment
themes, including conservative thought, which has wholly surrendered its
scepticism about the Enlightenment conception of progress towards a
universal civilisation. The criticism of the Enlightenment project developed
in Enlightenment's Wake is, accordingly, a critique of conservatism, among
other things. It suggests that the historical space in which a coherent form
of conservative political practice could occur no longer exists in most
Western countries: it has been destroyed by New Right policies whose
effect has been to accelerate and deepen all the forces in late modernity
which weaken its ties with its past. For this reason, there can be no return
to tradition as a solution for our ills. A return to tradition is not a possibility
for another reason, which is the dependency of the Enlightenment on far
older Western traditions - Socratic and Christian, for example - which are
dissolving even as the Enlightenment project is dissolving. The Enlighten-
ment project was self-undermining, in that the weapon of critical reason
which it turned against its enemies has been, at length, applied to its own
foundational commitments; but it is also culturally irreversible, in that it has
permanently displaced earlier modes of thought, religious or secular, in
which enlightenment values are given a 'foundation' that is not local and
accidental. In the enlightenment cultures of Europe, at any rate, there is no
possibility of a re-enchantment of the world of the sort which romantic and
fundamentalist movements envisage. Our task, imposed upon us as an
historical fate, is that of improvising common institutions and the rudiments
of a common life, in an historical context in which what we share are the
shards and fragments of earlier worldviews and traditions, bequeathed to us
as fractured perspectives and provisional practices.
This post-Enlightenment, pluralist views is intimated at many points in
the essays collected in Post-liberalism, which for that reason can be read as
a prelude to later work, as well as an immanent criticism of the liberal
project in its dominant contemporary forms.
John Gray
Jesus College, Oxford
November, 1995
Acknowledgements

'Hobbes and the modem state' and 'Santayana and the critique of liberalism'
appeared in The World andI (Washington, D.C.) in March 1989 and February
1989. 'Hayek as a conservative' and 'Oakeshott as a liberal' appeared in The
Salisbury Review (London) in July 1983 and January 1992. 'Buchanan on
liberty' appeared in The Journal o/Constitutional Political EcolWmy, vol. 1,
no. 2 (Spring/Summer 1990), 149-68. The following appeared in the Times
Literary Supplement: 'Berlin's agonistic liberalism', 5 July 1991; 'The system
of ruins', 30 December 1983; 'The delusion of glasIWst', 21 July 1989; and
'The academic romance of Marxism', 24 February 1989. 'Philosophy, science
and myth in Marxism' appeared in Marx and Marxisms, ed. G. H. R.
Parkinson, Royal Institute of Philosophy Lecture Series, vol. 14 (Cambridge:
Cambridge University Press, 1982). 'Against Cohen on proletarian unfree-
dom' appeared in Social Philosophy & Policy, vol. 6, issue 1 (Autumn 1988),
77-112. 'Totalitarianism, reform and civil society' appeared in Totalitarianism
at the Crossroads, ed. E. F. Paul (New Brunswick: Transaction Books, 1990),
97-142. 'Western Marxism: a fictional deconstruction' appeared in Philo-
sophy, no. 64 (1989), 4OH. 'Post-totalitarianism, civil society and the limits
of the western model' appeared in TheReemergence o/Civil Society in Eastern
Europe and the Soviet Union, ed. Zhigniew Rau (Boulder, San Francisco and
Oxford: Westview Press, 1991), 145-60. 'Political power, social theory and
essential contestability' appeared in The Nature 0/ Political Theory, ed. D.
Miller and I. Siedentop (Oxford: Oarendon Press, 1983). 'An epitaph for
liberalism' appeared in the Times Literary Supplement, 12 January 1990. 'The
end of history - or of liberalism?' appeared in The National Review (New
York), 27 October 1989,33--5. 'The politics of cultural diversity' appeared in
Quadrant, November 1987, 29-38, and in The Salisbury Review, September
1988, 38-45. 'Conservatism, individualism and the political thought of the
New Right' appeared in Ideas and Politics in Modern Britain, ed. J. C. D. Clark
(London: Macmillan, 1990). 'What is dead and what is living in liberalism'
is published for the first time in this volume.
Part I
Thinkers
1 Hobbes and the modern state

Thomas Hobbes lived in, and wrote for, an age of civil and religious wars.
For this reason, it may seem that we have little to learn from him today that
we do not already know. To be sure, anyone who reflects upon the intract-
able religious conflicts of Northern Ireland or the Middle East will see that
wars of religion are as much an evil of our age as they were of Hobbes's,
and civil peace as precious a good. Beyond these commonplace reflections,
it would appear that Hobbes has indeed little to teach us. His entire system
of thought, conceived and developed at a time when the scientific revolu-
tion was barely under way, may have the aspect of an anachronism, of an
intellectual construct whose terms and postulates are so far removed from
our own that we are hard put to make sense of it, and cannot put it to worlc
to illuminate the dilemmas that confront us today. On this conventional
view, we may read Hobbes's writings as literature (for he is one of the
greatest prose stylists in English) or as history, but we will not tum to them
for instruction or enlightenment.
Much as might be said in support of this conventional opinion, it is
radically misguided. It is true that Hobbes's system of ideas encompasses
extravagances we find hard to credit, and that his entire mode of thought
has an archaic character, recalling medieval ways of reasoning more than
the methods of modem science by which he supposed his theorizing was
governed. Yet there is an arresting contemporaneity about many of
Hobbes's insights that we can well profit from. Nor is this relevance to our
age a matter of surprise, for Hobbes wrote at the start of the modem age (he
was born in 1588 and died in 1679), and few have seen further to the bottom
of the dilemma of modernity than he. By comparison or contrast with
Hobbes, John Locke is truly a remote thinker, Kant hollow and Burke not
much more than a nostalgist. Far from being an anachronistic irrelevance,
Hobbes's thought is supremely relevant to us, who live at the end of the
modem era whose ills he sought to diagnose. With all of its limitations and
its excesses, Hobbes's thought goes far to account for the maladies of the
4 Thinkers
modern state, and does so in ways that are as surprising and paradoxical as
they are instructive. The lesson of Hobbesian theory for us is that the
modern state is weak because it aims too high and has grown too large.
Worse, the modern state has failed in its task of delivering us from a
condition of universal predation or war of all against all into the peace of
civil society. Modern democratic states have themselves become weapons
in the war of all against all, as rival interest groups compete with each other
to capture government and use it to seize and redistribute resources among
themselves. In its weakness the modern state has recreated in a political
form that very state of nature from which it is the task of the state to deliver
us. In this political state of nature, modem democratic states are riven by a
legal and political war ofall against all, and the institutions of civil society
are progressively enfeebled. And the conflicts which rack the modern state
are not merely economic in origin. All modern democracies, but especially
the United States, have transformed the state into an arena of doctrinal
conflicts, wherein under the banners of fundamental rights or social justice
contending political movements vie for supremacy. Worst of all, the totali-
tarian states of the Communist world have created a monstrosity that even
Hobbes (for all his vaunted, and genuine, pessimism) could not have
envisaged - a political order in which the state of nature and a lawless
Leviathan are inextricably intertwined. Distracted from its true purposes by
the political religions of our time, Communism and liberalism, the modern
state has become a burden on civil society, sometimes (as in Communist
regimes) a burden that is too heavy to bear. Even in the Western demo-
cracies, it has become more of an enemy of civil society than its guardian.
We have been delivered from the lawless chaos of society without the state
into an anarchic servility of unlimited government.
That modern governments are weak because they do too much and claim
authority they do not have, and so are at the furthest remove from Hobbe-
sian states, is a paradox, but like every genuine paradox it is a truth
expressed in the form of an apparent contradiction. We can discern the
reasons for this truth, and the outlines of an Hobbesian critique of the
modem democratic state, if we look again, with fresh eyes, at Hobbes's
chief doctrines about man, society and government. Consider first
Hobbes's conception of human nature. What is most distinctive in it is its
departure from the classical conception of man, most notably theorized by
Aristotle and given a Christian rendition by Aquinas. In the classical
conception, man is like everything else in the world in having a natural end,
telos or perfection, which it is his vocation to realize. By comparison with
Hobbes's view, the content of this perfection - a life of contemplation in
Aristotle and salvation in Aquinas - is less important than its affirmation.
In Hobbes's view, mankind has no supreme good which it is called to
 Hobbes and the modem state 5
achieve, no summum bonum, but only a summum malum, a supreme evil,
which we aim to elude. A good life for a human being consists, not in
possessing any final or supreme good, but simply in satisfying our restless
desires as they spring up. Hobbes tells us that
Continual success in obtaining those things which a man from time to
time desireth, that is to say, continual prospering, is what men call
Felicity; I mean the Felicity of this life, for there is no such thing as
perpetual tranquility of mind, while we live here; because Life itself is
but Motion, and can never be without Desire, nor without Feare, no more
than without Sense. t
For Hobbes, as his most distinguished twentieth-century interpreter,
Michael Oakeshott, has put it, human fulfilment is to be found,
not in pleasure - those who see in Hobbes a hedonist are sadly wide of
the mark - but in Felicity, a transitory perfection, having no finality and
offering no repose. 2
As Hobbes puts this view emphatically:
the Felicity of this life, consisteth not in the repose of a mind satisfied.
For there is no such finis ultimus (utmost aim) nor summum bonum
(greatest good) as is spoken of in the Books of the old Moral Philo-
sophers. . . . Felicity is a continual progress of the desire, from one
object to another; the attaining of the former, being still but the way to
the later. The cause whereof is, that the object of man's desire is not to
enjoy once only, and for one instant of time; but to assure forever, the
way of his future desire.3
For Hobbes, then, the human good is life in movement, which is pursuit of
one's passing desires. For human beings, it follows, the greatest evil can
only be immobility, or death, since that is the cessation of all movement and
all desire. It is indeed Hobbes's view that men avoid death, and above all a
violent and painful death, before all else. Just as Epicurus had seen in the
absence of pain rather than in pleasure the good for men, so Hobbes gave
the avoidance of death and not life itself as the pre-eminent human end.
What do these claims imply for man in society? They suggest an incessant
competition for the means whereby death can be avoided, or delayed.
Human life, Hobbes avers, can be compared to a race - a race which has no
other 'goal' or 'garland' than 'being foremost'.
In it to endeavor is appetite; to be remiss is sensuality; to consider them
behind is glory; to consider them before is humility . . . to fall on the
sudden is the disposition to weep; to see one outgone when we would not
6 Thinkers
is pity; to see one outgo when we would not is indignation; to hold fast
by another is love; to carry him on that so holdeth is charity; to hurt
oneself for haste is shame . . . continually to be outgone is misery;
continually to outgo the next before is felicity; and to forsake the course
is to die. 4
The consequence of this competition, as well as its primary cause, is
a general inclination of all mankind, a perpetual and restless desire of
power after power, that ceaseth only in Death. And the cause of this, is
not always that a man hopes for a more intensive delight, man he has
already attained to; or that he cannot be content with a moderate power:
but because he cannot assure the power and means to live well, which he
hath present, without the acquisition of more. S
Hobbes's master thesis is that, in the absence of a sovereign power which
binds them to peace, men's natural condition is a condition of conflict or
war. As he famously sums up his chain of reasoning:
Whatsoever is consequent to a time of Warre, where every man is
Enemy to every man; the same is consequent to the time, wherein men
live without other security, than what their own strength, and their own
invention shall furnish them withall. In such conditions, there is no place
for Industry; because the fruit thereof is uncertain; and consequently no
culture of the earth, no navigation, nor use of the commodities that may
be imported by Sea; no commodious building; no instruments of mov-
ing, and removing such things as require much force; no knowledge of
the face of the Earth; no account of time; no Arts; no Letters; no Society;
and which is worst of all, continual fear, and danger of violent death.
And the life of man, solitary, poor, nasty, brutish and short. 6
No less famous than this passage is Hobbes's conclusion: human beings can
be delivered from their natural condition of war, only by the creation of 'an
artificial man', a sovereign authority which is empowered to do all that is
necessary to bring into being, and keep, a civil peace. This sovereign must
be unlimited in his powers of action, and must be a unitary authority, since
otherwise dispute will be engendered as to the limits of the sovereign's
authority, and peace will be put in jeopardy. The Hobbesian political
remedy for the natural misery of mankind is then a state whose authority is
unlimited, save by its task of keeping the peace. More, Hobbes believed that
his analysis would find support among men, who would be driven by
prudence and experience of the fearful consequences of weak or divided
government to enter into a covenant which instituted Leviathan, the
Hobbesian state.
 Hobbes and the modern state 7
What is to be said of Hobbes's argument? The overwhelming pre-
ponderance of critical opinion goes against his analysis at almost every
point. Hobbes has remained what he was for his contemporaries, a scandal
and a scapegoat for the false consciousness of his critics. As Oakeshott
delightfully observes,
Against Hobbes, Filmer defended servitude, Harrington liberty, Claren-
don the Church, Locke the Englishman, Rousseau mankind and Butler
the Deity.7
Much, if not most in the traditional criticisms of Hobbes's doctrine repre-
sents it crudely, and so goes astray. It is said, truly enough, that human
behaviour is not governed predominantly by the imperative of death-
avoidance. If it were, how could we explain the heroic resistance of the
Afghans against Soviet conquest in our own day, and many another similar
instance in history? It is suggested that Hobbes held to another falsehood
about human nature, the theory of psychological egoism, according to
which whatever any man does he does at the behest of his own interests.
And it has even been alleged that the model of a Hobbesian state is a
totalitarian government, from which civil society and autonomous institu-
tions have been extirpated. Despite occasional inadvertencies in Hobbes's
writings, all of these charges are wide of the mark. As a recent com-
mentator 8 has shown beyond reasonable doubt, Hobbes held not that all
human conduct was motivated by self-interest, but only that self-interest
was a powerful human passion, and sheer concern for others, especially
those who are not near to us in our affections, is rare and politically
insignificant. More precisely, it is the interest in self-preservation that
Hobbes supposes to be decisive in human conduct - and, most precisely,
the interest in avoiding a violent death at the hands of other men. This last
point is not clear in Hobbes's literary masterpiece, Leviathan, where it is
death itself that is specified as the supreme human evil, but it is explicit in
his De Homine, where Hobbes observes:
the greatest of goods for each is his own preservation. For nature is so
arranged that all desire good for themselves. . . . On the other hand,
though death is the greatest of all evils (especially when accompanied by
torture), the pains of life can be so great that, unless their quick end is
forseen, they may lead men to number death among the goods. 9
Hobbes's is not, then, the mechanistic theory in which an egoistic account
of human nature is conjoined with the claim that merely avoiding death is
the passion which most decisively governs men. Nevertheless, Hobbes is
certainly open to criticism for his neglect ofthe moral and political import-
ance of collective identification - the pervasive human phenomenon in
8 Thinkers
virtue of which personal identities are constiblted by membership in some
nation, religion, tribe or other collectivity. The fact is that the man who
conceives himself as a solitary individual, whose identity is unencumbered
by any collective identification, though he is real, is vanishingly rare. As
the novelist and aphorist, Elias Canetti, author of a fascinating sbldy in
Hobbesian political psychology, Crowds and Power, has said of Hobbes:
He explains everything through selfishness, and while knowing the
crowd (he often mentions it), he really has nothing to say about it. My
task, however, is to show how complex selfishness is: to show how what
it controls does not belong to it, it comes from other areas of human
nature, the ones to which Hobbes is blind. 10
Hume had rightly criticized Hobbes for his radical individualist picblre of
human beings as solitaries, who contract as much into society as into
government. Such a picture must be a distortion, he urged, since we are
born into families, which are social instibltions, and come to consciousness
speaking a language, which presupposes that we have a common life and a
shared history. Anticipating the twentieth-cenmry arguments of Ludwig
Wittgenstein for the impossibility of a private language, II Hume pointed
out that our ability to contract into government presupposes that we already
have the practice of promising. Home's point against Hobbes is that we are
social beings au fond. Canetti's is the subtler one that our identities as
persons are not natural but artifactual and are formed in a matrix of
collective identifications (of which the crowd is perhaps the crudest).
Because we derive our self-conception from membership in common forms
of life, self-interest cannot be a primordial motivation in our lives in even
the qualified way we have seen Hobbes make it so. The extravagant indivi-
dualism of Hobbesian psychological theory, which flies in the face of much
in experience and historical knowledge, comes in part from the method he
sought to adopt in political theorizing, which aimed to be rigidly deductive.
As he avows in De Homine,
politics and ethics (that is, the sciences of just and unjust, of equity and
inequity) can be demonstrated apriori; because we ourselves make the
principles - that is, the causes of justice (namely laws and covenants)
whereby it is known what justice and equity, and their opposites in-
justice and inequity, are. For before covenants and laws were drawn up,
neither justice nor injustice, neither public good nor public evil, was
natural among men any more than it was among beasts. 12
Hobbes, like many another rationalist after him, here neglects the origins in
natural social relationships of the moral and legal norms that sustain and
constitute the state. He neglects, in other words, the natural roots of the
 Hobbes and the modem state 9
artificial virtues, whose relations with each other are theorized so well by
Hume. Like subsequent rationalists, such as his near-contemporary
Spinoza, Hobbes's project was a sort of moral geometry, whereby moral
theory (if not moral life) could be purged of all that was merely contingent,
historical local and ephemeral. His adoption of the geometrical method
encouraged him to ignore, when he did not altogether deny, the historicity
of the various forms of moral and political life - a truth about them whose
implicat~ons for philosophy were not elaborated upon until Hegel
developed them (though they are prefigured in the work of Vico U ).
Even though traditional criticism of Hobbes is misplaced, it is true that
his thought is disabled by a rationalism which obscures the social and
historical character of the phenomena which are the subjects of his investi-
gations. It is far from being the case, however, that Hobbes's rationalism-
which he may have derived from Descartes, and which is in sharp contrast
with the empiricism of the nascent scientific revolution of the seventeenth
century - leads him to a conception of government that anticipates
twentieth-century totalitarianism. As Oakeshott has again put it,
It may be said ... that Hobbes is not an absolutist precisely because he
is an authoritarian. His scepticism about the power of reasoning, which
applied no less to the 'artificial reason' of the Sovereign than to the
reasoning of the natural man, together with the rest of his individualism,
separate him from the rationalist dictators of his or any age. Indeed ...
Hobbes, without being himself a liberal, had in him more of the philo-
sophy of liberalism than most of its professed defenders. 14
Oakeshott's observations are much to the point here. Hobbes was not a
proto-totalitarian just because the task of the state was for him the limited,
but crucial one of securing the peace of civil society. Though he denies that
subjects possess fundamental rights to liberty of conscience (or to anything
else, apart from self-preservation), Hobbes is insistent that the control of
belief by the promotion of an orthodoxy is no part of the duty of a
sovereign, and may well run counter to his only duty of keeping the civil
peace. The sovereign cannot in any case (and here Hobbes is at one with
Spinoza) control belief, since belief is not a matter of will. The sovereign
ought not (because it is beyond the purposes of government) to interfere
with opinion or expression, except in so far as it may threaten peace.
Nothing could be further from Hobbes's intention (or more remote from
anything he could have imagined) than the thought-police of twentieth-
century totalitarian states, with their vast propaganda machines and inter-
minable investigations into the ideological probity of their hapless subjects.
The paradox of the Hobbesian state is that, whereas its authority is
unlimited. its duty is minimal- the maintenance of civil peace. Civil peace
10 Thinkers
here comprehends more than merely the absence of civil war - it en-
compasses that framework of civil institutions whereby men may coexist in
peace with one another, notwithstanding the diversity of their beliefs and
enterprises and the scarcity of the means whereby these are promoted. Civil
peace, that is to say, embraces the institutions of private property and
contractual liberty, it presupposes a rule of law and it entails that the
subjects of a civil society are bound to no common purpose but instead
enjoy the freedom to pursue their own purposes. The liberties of the
subjects of a civil society are not, as Hobbes conceives of them, absolute or
inalienable rights, since they may be circumscribed by the requirements of
a civil peace in the absence of which they are altogether extinguished. The
liberties of the citizens of a civil society are not, in other words, pre-
ordained constraints on the authority of the state. They are intimated by the
spirit of civil society itself, which is held together only by recognition of the
authority of the sovereign and lacks any common purpose into whose
service citizens might be coercively pressed. Maximal and indeed un-
limited as the discretion of the Hobbesian state (or its ruler, which might be
a person or an assembly) certainly is, its authority derives entirely from its
contribution to the renewal in peace of civil society. It should be clear
enough that the Hobbesian conception of the state in no way prefigures or
models the dirigiste regimes which litter our century with their ramshackle
institutions and wayward projects. What does the Hobbesian conception
have to say, then, as to the character and condition of the modem state? Let
us see.
The first and cardinal truth about the state in Hobbes's conception of it
is that it possesses no resources of its own. It is sustained by a tax on
consumption, which is used to pay for a volunteer army, to defray the costs
of the legal system and to fund useful public works. The Hobbesian state is
not a minimum state of the doctrinaire sort theorized in the writings of
Herbert Spencer or Robert Nozick, and it does not exist to protect an
imaginary set of inalienable (and incorrigibly empty) natural rights of the
kind postulated in the philosophies of Locke and Kant. Hobbes's state has
in common with the minimum state of classical liberalism, nevertheless, the
crucial attribute that, since it has no assets of its own, it is not in the business
of distributing resources to its subjects. As with the classical liberal state,
the task of the Hobbesian state with regard to the institution of private
property is finished when it has specified the practices and conventions
which are to govern its acquisition and transmission and the procedures for
arbitrating disputes about it. The task of the Hobbesian state, accordingly,
is the demarcation or definition of property rights and the institution of a
rule of law for their adjudication. Once that task has been discharged,
nothing more remains to be done. Now consider the contrast with the
 Hobbes and the modem state 11
modern state. Every modern state owns or controls vast assets and most
modern states make a claim on approximately half of the wealth produced
by the civil societies which it is their task to protect. Above and beyond the
income required to fund national defence, the legal system and necessary
charitable and public works, all modern states operate vast redistributional
taxation and welfare systems, whereby income and wealth are transferred
coercively across a welter of interests and pressure groups. Further, all
modern states have entered into the business of wealth creation itself. By an
array of tariffs and subsidies, unsound banking practices, and a plethora of
regulations and regulatory authorities, modern states have invaded the
wealth-creating activities of civil society to a profound degree, moulding
and shaping the environment in which business enterprise functions, and in
effect becoming vast business enterprises themselves. Finally, every
modern state owns massive assets of its own, in nationalized or federally
owned business, and in land and plant of all descriptions. As a result of the
awesome economic power of the modern state, it pre-empts far more of the
income and wealth of its citizens than ever feudal rulers were allowed
(restricted as typically they were to one day in three of their serfs's labour)
and it exercises an invasive influence on every area of social life unheard
of since the absolutist monarchies of early modern Europe.
As to the causes of the growth of the modern state, we may speculate
reasonably that much may be accounted for by the advent of mass demo-
cracy. The contemporary neo-Hobbesian theorists of the Virginia Public
Choice School 15 have added much to our understanding of the expansionist
imperative of modern states by their analysis of the economics of political
life. The insight developed by the Virginia theorists is that. if we attribute
to political actors the same sort of profit-maximizing motivation ordinarily
ascribed to actors in the market-place, and if the environment in which
political actors operate is that of a mass democracy in which powerful
coalitions of interests form, then there will be an extremely powerful
tendency for the size of government to grow, and its control over civil
society to increase. For it will almost always be advantageous for politi-
cians to extend benefits to existing and new interest groups rather than to
curtail or withdraw them, since the losses to concentrated and collusive
groups of such defunding will always be politically more significant than
gains to widely dispersed groups. In modern mass democracies, accord-
ingly, states tend overwhelmingly to service private interests rather than
protect or promote the public interest. Contrary to the classical theory of the
state as the provider of public goods - goods, that is to say, which in virtue
of their indivisibility and non-excludability must be provided to all or none
- modem states are above all suppliers of private goods. Whereas in the
Hobbesian conception, the state exists to supply the pure public good of
12 Thinkers
civil peace, the modern state exists in practice to satisfy the private pre-
ferences of collusive interest groups. In so doing, it has become distracted
from, and has largely defaulted on its central functions of keeping the peace
and keeping in good repair the institutions of civil society.
The transformation of the modern state from a guardian of the public
interest and supplier of public goods - which it was, in Britain and the
United States, substantially until the First World War - has had profound
implications for civil society. Its chief impact has been to weaken the
vitality of the autonomous institutions which are the life-blood of civil
society. Charitable organizations, trade unions, educational institutions,
and large sectors of cultural life, which had hitherto enjoyed a significanl
measure of independence from government have been increasingly drawn
within the influence or control of the state. The sphere of free individual
activity, the sphere of contractual liberty, has waned, as the sphere of
hierarchical organizations, the sphere of status, has waxed. This is the
baleful process identified by such classic writers as Maine and Acton, and
illuminated in our own time (in different but mutually enriching idioms) by
Hilaire Belloc's The Servile State and F.A. Hayek's The Road to Serfdom.
It is the process - well advanced in all modern states, but reaching its
terrifying completion in the totalitarian states of the Communist blocs,
which I shall consider at the close of these reflections - whereby the free
subjects of civil society are transformed into dependent functionaries or
vassals of the state.
The consequences in modern democracies of the erosion of civil society
by an expansionist state has everywhere been the outbreak of a political
war ofredistribution. From being an umpire which enforces the rules of the
game of civil association, the state has become the most potent weapon in
an incessant political conflict for resources. Its power is sought, in part
because of the vast assets it already owns or controls, but also because no
private or corporate asset is safe from invasion or confiscation by the state.
From being a device whereby the peaceful coexistence of civil association
is assured, the state becomes itself an instrument of predation, the arena
within which a legal war of all against all is fought out. The rules of the
game of civil association - the laws specifying property rights, contractual
liberties and acceptable modes of voluntary association - are now them-
selves objects of capture. Corporate interests and pressure groups are
continuously active, by lobbying, colonization or cooption of regulatory
authorities, or just plain corruption, to mold these rules to suit their own
interests. Often enough, they are constrained to do so defensively, in the
sure knowledge that if they do not alter the legal and regulatory framework
to their advantage, their competitors will amend it against their interests.
Civil life soon comes to resemble the Hobbesian state of nature from which
 Hobbes and the modem state 13
it was meant to deliver us. As it has been restated by several recent
commentators, J6 Hobbes's state of nature has many of the characteristics of
the Prisoner's Dilemma explored in game theory, in that in it agents are
compelled to act against their own interests because of the uncertainty they
confront about the future conduct of others and the likelihood that others
will be similarly constrained to adopt self-destructive policies. Most typic-
ally, agents in a Prisoner's Dilemma are constrained to attack or cheat
others because of the rational suspicion they have of others' future conduct
towards them. The Hobbesian state is the classical solution of the Prisoner's
Dilemma in that the Hobbesian contract, by providing for agreed-upon
coercion to. obey known rules, releases its covenanters from destructive
conflict into the peace of civil life. In the modern state this order of things
is reversed. Individuals and enterprises are constrained to organize collu-
sively so as to capture the interventionist state, if only because they know
that if they do not, others will. As a result, their energies are distracted from
production, and displaced into the political struggle for redistribution. The
end result of this metamorphosis, in its extremity, can only be the im-
poverished and brutish life portrayed by Hobbes in the state of nature.
Matters have not yet gone so far in the major Western democracies. The
examples of Argentina under the Peronist dictatorship, or of Britain under
the last Labour Government, should however caution us against over-
complacency. Indeed, in subtler and less obvious ways, the stagnation
produced by the weakening of civil society and the overexpansion of the
state are increasingly evident as the United States follows Europe into a sort
of economic sclerosis, the result of which can only be further ill-judged
interventionism. The transformation of the modern state from a guardian of
civil association, whose conflict and best theorizing is found in Hobbes's
Leviathan, into a corporatist Behemoth, has implications which extend far
beyond the economic realm. The political war of redistribution is itself only
a facet, if the most insistent and easily perceptible one, of the debility and
weakness of modern government. The other arena of conflict in the modern
state is doctrinal rather than economic, and manifests itself in the pathology
of contemporary legalism. Once again a contrast with the Hobbesian con-
ception is instructive. In civil society as theorized by Hobbes - and, for that
matter, though less lucidly and profoundly, by Locke and Burke - all
subjects possess the same liberties under the rule of law. Only the sovereign
itself, as maker of law, may go beyond it, when the necessities of his
responsibility as keeper of the civil peace so dictate. The liberties
embedded in civil association are the old liberal freedoms - the freedoms
of occupation, conscience, contract, association and so on. In exercising
these freedoms, citizens create a myriad of intermediary associations and
spontaneous institutions in which the life of civil society is expressed.
14 Thinkers
Whether they be working men's friendly societies, gentlemen's clubs,
monastic retreats, missionary societies or eleemosynary organizations,
these spontaneous formations can each embody their own ethos while
keeping to the letter and spirit of civil association. They depend for their
existence and renewal solely on the commitment of their members, and
none enjoys any special privileges over the rest. At any rate, this was the
situation in England, the United States and most of Europe for most of the
time from the late seventeenth century until the First World War. In the
modern state, what intermediary institutions and voluntary associations
seek from government is not equal freedom under the rule of law, but
political recognition and legal priVilege. More, they seek a political
endorsement of the values and norms which their organizations embody
and express. What is most profoundly characteristic of social life in the
modern state relates to that aspect of human nature most conspicuously
neglected by Hobbes - its propensity to collective identification. In the
civil societies of nineteenth-cenrury Europe and America, it seemed that
this propensity could be fully satisfied, without recourse to political
coercion, through the spontaneous formations of voluntary association.
(Perhaps personal identities were more sturdy in those days, and less in
need of the dubious sustenance of political recognition.) In twentieth-
century democratic states, we see the need for collective identification
rmding overtly political expression in the form of contending projects for
the political protection of cultural identity.
This is the true significance of the manifold exercises in affirmative
action, anti-discrimination policy and minority rights which disfigure and
devalue much contemporary legislation. They are only incidentally, if at
all, remedies against injustice or past wrongs. They are projects of collec-
tive self-assertion, which seek to entrench and privilege a specific identity
by legal and political means. My expression 'legal and political means'
should be noted here, for it is intended to mark one of the most charac-
teristic vices of the modern state, which is the politicization of legal
process. In the United States, this has occurred through a ruinous inflation
of the rhetoric of rights, whereby every moral and political dispute and
debate is cast in the legalistic idiom of rights discourse. Accordingly, the
courts, and for that matter, the entire procedure of judicial review, have
become theatres of doctrinal conflict, in which rival political and cultural
movements contend through the medium of an intellectually bankrupt
tradition of rights discourse as to the proper content (never the appropriate
extent) of state intervention in civil life. Conflicts of interests and ideals
which were hitherto reconciled - as they still are reconciled, in the few
benighted polities (such as Great Britain and New Zealand) where the virus
of rights discourse has not yet become endemic - by bargaining, com-
 Hobbes and the modern state 15
promise and the political arts are in the United States intractably contested
through the processes of law. Distinct communities and ways of life which
had once been constrained to such a mutual accommodation by moderating
their claims on each other and on government have an incentive to engage
in all-out legal and political conflict Litigation substitutes for political
reasoning, and the matrix of conventions, subterfuges, and countervailing
powers on which any historic civil society must stand is weakened or
distorted. This, it is worth repeating, is the real significance of the incessant
contestation in universities, churches, municipalities and federal courts as
to the 'rights' of ethnic minorities, 'alternative lifestyles' and feminist
women. They concern, not the rightful extension to individuals denied them
of the liberties and immunities enjoyed by others in equal freedom under
the rule of law, but rather the political self-assertion of collective identities,
each of which seeks privileges and entitlements which cannot in their
nature be extended to all. So far gone is the United States in this degrada-
tion of law - which is the capital on which civil society must draw for its
daily support - that it is not hard to envisage the United States as heading
for an Argentine-style nemesis, in which economic weakness, over-
extended government and doctrinal excess compound with each other to lay
waste the inheritance of civility.
The worst nemesis of the modem state is not to be found in the Western
Democracies. It is not even that found in the so-called developing or Third
World, where arbitrarily constructed nation-states have completed the dis-
location of traditional ways of life under the aegis of a Eurocentric ideology
of modernization. It is in the Communist world that the true contrary to a
Hobbesian state has come about. There the overriding imperative is not
peace, but ideological uniformity and the totalitarian reconstruction of
social life according to Marxist-Leninist dogma. Under such circum-
stances, there cannot be the autonomous institutions which make up a civil
society, nor any sphere of individual liberty under the rule of law. The
coercive power of the state invades every nook and cranny of social life,
penetrating and weakening even the family, perhaps the ultimate unit on
which civil society stands. Two points are especially noteworthy when
contrasting the Hobbesian state with Communist states. First, and most
often neglected, is the lawlessness of life in Communist states. In the
absence of an independent judiciary and of proper legal procedures, and
given the ubiquitous presence of agencies prepared to wield extra-legal
coercion at their own discretion, no one can be sure of protection by the
law, or even know what the law may be. As an inexorable consequence, not
only are civil liberties impossible to exercise securely, but even the pre-
requisites of modem economic lifestyle - well-defined property rights and
reliably enforced contracts, for example - are ill-provided for. It is these
16 Thinkers
deprivations, as much as the extravagant Titanism of its central planning
apparatus, which account for the intractable waste, malinvestment and
sheer poverty which distinguish Soviet life.
Soviet totalitarianism is distinguished not only by the poverty it inflicts
on its subjects but by the war it compels them to wage upon each other. This
war of all against all among Soviet subjects is only the reverse side of the
civil war - the war on civil society - that the Soviet power has waged since
it struck the first blow in the Bolshevik coup d'etat.17 Since wealth, educa-
tion and health care are all in Soviet society positional goods - goods whose
radical scarcity means that their enjoyment or possession by one entails
their exclusion from others - each subject of the Soviet order must contend
with others in struggling to obtain them. In this respect Soviet society, like
Hobbes's state of nature, is a vast Prisoner's Dilemma, with each being
constrained to act against his own interest and, thereby, directly or in-
directly, to reproduce the order (or chaos) in which he is imprisoned. Thus
Soviet subjects are compelled to compete with each other in climbing the
rungs of the nomenklatura, pursuing the ordinary goods of life by party
activism or, in extremis, by informing or denouncing one another, and so
renewing daily the system that keeps them all captives. If cooperation is
found anywhere in Soviet society beyond the enclaves in which traditional
values of love, family piety and religious devotion are still alive, it is
perhaps in the enormous black, informal or parallel economy, where much
that permits the ordinary subject to stay afloat is transacted. Yet even this
nexus of voluntary exchanges has its dark side in that, by mitigating the
depredations of the Soviet planning system, it enables it (with the indis-
pensable nourishment of repeated inputs of Western capital and tech-
nology) to reproduce and perpetuate it. The promises offered by Gor-
bachev's project of perestroika are in all probability entirely delusive.
Every material incentive of the Soviet economy constrains its subjects to
collaborate in recreating it, with all of its unavoidable poverty and repression.
It is a salutary thought for the Western liberal mind that no Communist state
has yet definitely tolerated the re-emergence of civil society. In Poland the
countervailing authority of the Church may have for the time being succeeded
in forcing a retreat on the part of the totalitarian state, and in China we witness
a genuinely bold experiment whose outcome, however, remains unknowable.
So far, alas, we have no unequivocal reason for supposing that the indissoluble
linkage between a state of nature and a lawless government, once it has been
forged, is other than stable and self-perpetuating.
What of the prospects of the Western democracies? It cannot be said that
recent projects of streamlining overextended government and returning
many of its powers to the autonomous institutions of civil society, have
done more than slow or retard the twentieth-century tendency to ever-more
 Hobbes and the modem state 17
extensive government intervention. Neither the Thatcher government nor
the Reagan administration has succeeded in its project of withdrawing
government decisively from economic life and so restoring vitality to
independent institutions. Their project, like that of similar governments
elsewhere in the world, has been hampered by the inheritance of inter-
ventionism which they have faced in a burden of debt, overregUlation and
persistent inflation. The ultimate obstacle to the restoration in practice of a
Hobbesian forr..I of government may, however, not be economic at all. The
resistance of the modem state to confinement within the Hobbesian func-
tions of keeping the peace and maintaining civil society has its origin in the
humanistic hubris which animates the political religions of our time. Given
the waning of genuine transcendental faith, and its supplanting by secular
orthodoxies or religions from which the transcendent content has been all
but emptied, the modem liberal mind cannot accept that the state is a
desperately modest device, which does not offer salvation, but only a
barrier against the worst impediments to a tolerable human life. In the
Hobbesian conception, the state is but a makeshift shelter against the
primordial evils of the human lot. In its protective shadow, we may hope for
many forms of excellence to thrive. But the peace of civil association in a
limited state conceived on Hobbesian lines can only make such excellence
a possibility, it cannot guarantee it It is in the attribution to it of the
character of a political providence that the dilemma of the modem state
lies. Its resolution requires the relinquishment of the liberal illusion that the
state can be more than a necessary precondition of a good life whose
excellences will always be at risk.
Here, in civil associations, is neither fulfillment nor wisdom to discern
fulfillment, but peace. . . . And to a race condemned to seek its per-
fection in the flying moment and always in the one to come, whose
highest virtue is to cultivate a clear-sighted vision of the consequences
of its actions, and whose greatest need (not supplied by nature) is
freedom from the distractions of illusion, the Leviathan, that justitiae
mensura atqua ambitionis elenchus, will appear an invention neither to
be despised nor over-rated. II
The new Hobbesian dilemma of the modem state is that, in becoming the
prey of liberal dogma and the focus of a political war of redistribution, it
has become weak even as it has grown larger, and has neglected its true
tasks as it has become the servant of impossible dreams. It remains to be
seen whether a reversion to a Hobbesian humility, in which the state is at
once small and strong, is itself only a nostalgic vision, a dream that haunts
an age, like Hobbes's, that is deafened by the clamour of barbarous reli-
gions and possessed by a longing for peace.
2 Santayana and the critique of
liberalism

The thought of George Santayana has never had a wide influence. Both the
style and the substance of his philosophy go against the current of twentieth-
century sensibility and whatever echoes his philosophical writings had
among his contemporaries have long since faded into silence. Neglect of
Santayana's work by professional philosophers and by educated opinion is
unfortunate for many reasons. His prose style - condensed, aphoristic and
ornate at the same time - is beautiful and unique in twentieth-century
philosophical writing, bearing comparison only with that of such earlier,
and very different philosophical stylists as David Hume. Again, though his
work never formed part of any recognized tradition or school, Santayana's
contributions to a range of philosophical disciplines - the theory of knowl-
edge and scepticism, metaphysics and ethics - still contain something from
which we can learn, if only we are ready to read him with intellectual
sympathy. His contributions to these subjects have been ignored, partly as
a result of the vulgar academic prejudice according to which anyone who
can write exquisitely must be a belle-lettrist or prose poet rather than any
sort of serious thinker, and partly because the idiom of Santayana's writings
consorts badly with that of the analytic schools which have dominated
Anglo-American philosophy for most of our century. The chief reason for
neglect of his thought, however, probably lies in the circumstances of his
own life. It is in any case lamentable that his works are rarely seriously
studied nowadays, since we are thereby deprived of his thoughts on society
and government, which encompass one of the most profound and incisive
critiques of liberalism ever developed. It is a symptom of the intellectual
temper of our times that one of the very few systematic studies of
Santayana's philosophy to be published in recent years omits altogether any
consideration of his political philosophy. 1 It is hard to resist the suspicion
that Santayana's political thought is not only unknown to modem opinion,
but (where it has even been heard of) deeply unfashionable and indeed
 Santayana and the critique of liberalism 19
thoroughly uncongenial. For this reason alone, it may well have much to
teach us of the ironies and limitations of the ruling liberal world view.
The circumstances of Santayana's life were uncommon, and are more
than usually relevant to the understanding of his thought. Born in Spain in
1863 of Spanish parentage, Santayana retained his Spanish nationality
throughout his life and (though in fact he left Spain at the age of nine)
always regarded himself as a Spaniard. He was nevertheless educated in the
United States, wrote all his philosophical works and his poetry in English,
and had his only long-term association with a university, as a student and
later a professor, at Harvard. Santayana's intellectual temperament, aloof,
ironical and poetic, had little in common with that which, then as now,
dominated American culture, and he became increasingly disaffected with
American life. His distance from, and even disdain for, American culture
was evident to his colleagues at Harvard, and reciprocated by his polar
opposite there, William James, whose earnest, optimistic and sermonizing
outlook was outraged by Santayana's detachment, and who expressed his
condemnation of the style and substance of Santayana's thought famously
when he called it 'the perfection of rottenness'. Santayana's colleagues
were stupefied and outraged, when in 1911 he acted upon his estrangement
from American life and academic culture by resigning his professorship at
Harvard and departing forever for Europe, where he lived the life of an
independent scholar, residing in hotels and eventually a convent, and
remaining productive until the year of his death in 1952. There can be little
doubt that Santayana's repudiation of the academic milieu in which philo-
sophical inquiry had become professionalized and institutionalized,
together with his indifference to the moralistic and world-improving sen-
sibility which animated much in American thought and life during his
lifetime, reinforced the neglect of his work engendered by its peculiarities
of tone and content
Immensely erudite though he was in the history of philosophy,
Santayana appears to have been uninfluenced by any of the thinkers he
studied. In his later years, he tended to represent himself as a spokesman for
a human orthodoxy, as sceptical as it was naturalistic, which had hitherto
rarely found a voice, and then chiefly in poets, such as Lucretius. His
philosophical development is easily summarized. Aside from his doctoral
dissertation on Lotze's philosophy, his first significant publication was in
aesthetic theory. The Sense ofBeauty, published in 1896,2 contains many of
the themes which were to pervade his writings over the next half-century
and more - and, most particularly, an analysis of a spiritual value in
naturalistic, bodily and partly sexual terms which at no point descends to a
crude reductionism. The distinctive Santayanian conjunction of a sturdy
20 Thinhrs
Lucretian materialism about that which exists with a strong affinnation of
the spiritual reality of a realm of essences, is already manifest in this very
early work. and is worked out in systematic detail in Santayana's first major
philosophical statement, The Ufe ofReason,) published in five volumes in
190~. Santayana's metaphysical doctrine, a rather obscure and perhaps
ultimately incoherent combination of a materialist ontolOS)' with a pluralist
theory of essences, was further elaborated in the four volumes of his Realms
of &ing,4 published between 1927 and 1942. In addition to these major
treatises, Santayana published many collections of essays, of which Winds
of Doctrine (1913)5 and Soliloquies in England and Later Soliloquies
(1922), are among the most notable. Winds of Doctrine contains
Santayana's coruscating and devastating assessment of the early philo-
sophy of his friend Bertrand Russell, in which he comments on Russell's
belief - derived from G.B. Moore, and later abandoned by Russell under the
impact of Santayana's criticism - that there are objective moral qualities in
no way dependent upon the constitution of human beings for their content
- by observing that
For the human system whisky is truly more intoxicating than coffee, and
the contrary opinion would be an enor; but what a strange way of
vindicating this real, though relative, distinction, to insist that whisky is
more intoxicating in itself, without reference to any animal; that it is
pervaded, as it were, by an inherent intoxication, and stands dead drunk
in its bottle!'
Santayana's Soliloquies are noteworthy in that they contain some of his
most penetrating analyses of the failings of liberalism. and of the trans-
fonnations of opinion wrought by the catastrophe of the First World War. The
main themes of Santayana's political philosophy are scattered throughout these
and other essays, but it receives a complete and magisterial statement in his
Dominations and Powers, which appeared a year before he died.
What are the most essential elements in Santayana's political thought,
and how do they bear on his critique of liberalism? We find the very kernel
of Santayana's rejection ofliberalism in his observation that
this liberal ideal implies a certain view about the relations of man in the
universe. It implies that the ultimate environment, divine or natural, is
either chaotic in itself or undiscemible by human science, and that
human nature, too, is either radically various or only determinable in a
few essentials, round which individual variations play ad libitum. For
this reason no nonnal religion, science, art or way of happiness can be
prescribed. These remain always open, even in their foundations, for
each man to arrange for himself. The more things are essentially
 Santayana and the critique of liberalism 21
unsettled and optional, the more liberty of this sort there may safely be
in the world and the deeper it may run. 8
Santayana here illuminates the central dogma of modem liberalism, un-
reflectively and stubbornly held - the dogma that human nature is a fiction,
a chaos or an unknowable thing, so that it is not unreasonable for each
generation to start life afresh, to try every experiment in living again and
await what comes of it. The sceptical dogmatism of the modem liberal mind
is at the furthest remove from the outlook of the ancients, for whom human
nature was in most essential respects knowable and fixed, a stable matrix
within which variations among individuals and peoples might safely occur.
For the ancients, accordingly, liberty was not a fundamental right to in-
determinacy, founded on chaos, but a very definite thing:
When ancient peoples defended what they called their liberty, the word
stood for a plain and urgent interest of theirs: that their cities should not
be destroyed, their territory pillaged, and they themselves' sold into
slavery. For the Greeks in particular liberty meant even more than this.
Perhaps the deepest assumption of classic philosophy is that nature and
the gods on the one hand and man on the other, both have a fixed
character; that there is consequently a necessary piety, a true philosophy,
a standard happiness, a normal art.... When they (the Greeks) defended
their liberty what they defended was not merely freedom to live. It was
freedom to live well, to live as other nations did not, in the public
experimental study of the world and of human nature. This liberty to
discover and pursue a natural happiness, this liberty to grow wise and to
live in friendship with the gods and with one another, was the liberty
vindicated at Thermoplyae by martyrdom and at Salamis by victory.9
Classical liberty, for Santayana, was the freedom of self-rule, in which
civilized peoples pursued the arts of life in the assurance that knowledge of
the human good was (within limits, like everything else) achievable.
Liberal freedom, on the contrary, is the freedom of inordinacy, an hubristic
compound of antinomian individualism with a sentimental humanism
which the Greeks would have despised had they been decadent enough to
be able to imagine it. The contrast between classic freedom and the freedom
of liberal modernity is clear enough, but it should not lead us into mis-
interpreting Santayana's moral vision and political perspective. Santayana,
like the Greeks, believed human nature to be a stable and knowable thing,
with definite limits and a bounded range of variations. He differed radically
from the Greeks - at least, from the dogmatists among them, such as
Aristotle - in his conception of the status and content of the human good.
For Santayana, the human good was as various as the diverse kinds of
22 Thinkers
human beings who achieved it or failed to realize it. There is not one
supreme good for all men - contemplation, say, as Aristotle would have it,
or a life of work and prayer, as for Aquinas - nor is there a single form of
collective life, such as that of the polis, in which human flourishing may
occur. Instead there is a constrained but legitimate diversity of goods,
individual and collective, and it is a tyrannous impulse in political philo-
sophy (well exemplified in modem liberalism) to elevate any of them to the
status of summum bonum. Just as the good life for an individual may be one
of bourgeois productivity or aristocratic leisure, religious piety or the
pleasures of the senses, so monarchy and republicanism, free enterprise or
the feudalism of a traditional social order may be equally lawful facets of
the human good. What the human good encompasses in anyone time or
place is a matter for study and deliberation, not legislation, and there is
always a margin of contingency and caprice in it, which no theory can tame.
What is definitely excluded by Santayana's moral outlook is not the partial
relativity or indeed the limited subjectivity of the good, which he affirms,
but instead its infinite variability, its unknowable openness to unheard-of
novelty, its plasticity and malleability by human will and the utopian or
reformist imagination.
What is excluded by Santayana's moral conception is, in short, progress. It
is not that Santayana, in a spirit of misanthropic perversity, denies that human
arrangements are ever improvable. Nor does he subscribe to radical relativism:
he does not deny, but affirms, that there are better and worse forms of
government, virtues and vices that are common to all men. Santayana departs
from the parochial dogmatism of the classic Greek moralists when, in the
tradition of Democritus and Lucretius, he insists on a legitimate variety of
mores and ways of life, but he nowhere endorses the heresy of liberal toleration
- the heresy that there is nothing to choose between traditions and cultures. His
rejection of progress has other and more reasonable grounds. In the first place,
even when incontestable improvement takes place in human affairs, there is
nothing inevitable about it; it occurs as a matter of chance or human will, and
it is always reversible. As Santayana puts it:
Progress is often a fact: granted a definite end to be achieved, we may
sometimes observe a continuous approach towards achieving it, as for
instance towards cutting off a leg neatly when it has been smashed; and
such progress is to be desired in all human arts. But belief in progress,
like belief in fate or in the number three, is a sheer superstition, a mad
notion that because some idea - here the idea of continuous change for
the better - has been realized somewhere, that idea was a power which
realized itself there fatally, and which must be secretly realizing itself
everywhere else, even where the facts contradict it. IO
 Santayana and the critique of liberalism 23
The idea of a law of progress, or of an all but irresistible tendency to general
improvement, is then merely a superstition, one of the tenets of the
modernist pseudo-religion of humanism. Even if such a law or tendency
existed and were demonstrable, the liberal faith in progress would for
Santayana be pernicious. For it leads to a corrupt habit of mind in which
things are valued, not for their present excellence or perfection, but instru-
mentally, as leading to something better; and it insinuates into thought and
feeling a sort of historical theodicy, in which past evil is justified as a means
to present or future good. The idea of progress embodies a kind of time-
worship (to adopt an expression later used by Wyndham Lewis) in which
the particularities of our world are seen and valued, not in themselves, but
for what they might perhaps become - thereby leaving us destitute of the
sense of the present and, at the same time, of the perspective of eternity. The
idea of progress has yet another radical fault. It supposes that there is a
constant standard by which improvement may be measured, or at any rate
a consecutive series of such evolving standards. Now it is true that human
beings, like other animal species, have certain needs in common, and their
lot is improved the better these needs are satisfied. But in the human
species, if not in other species, the satisfaction of these basic needs evokes
others, and changes the standards whereby future improvement may be
measured. Nor is this self-transformation of mankind by the satisfaction of
its needs a linear or one-dimensional process, since different individuals
and peoples in different historical and cultural milieux suffer it differently.
And its effects are not always beneficent in terms of human well-being:
It is perhaps only in transmissible arts that human progress can be
maintained or recognised. But in developing themselves and developing
human nature these arts shift their ground; and in proportion as the
ground is shifted, and human nature itself is transformed, the criterion of
progress ceases to be moral to become only physical, a question of
increased complexity or bulk or power. We all feel at this time the moral
ambiguity of mechanical progress. It seems to multiply opportunity, but
it destroys the possibility of simple, rural or independent life. It lavishes
information, but it abolishes mastery except in trivial or mechanical
efficiency. We learn many languages, but degrade our own. Our philo-
sophy is highly critical and thinks itself enlightened, but it is a Babel of
mutually unintelligible artificial tongues. I I
Whereas it pervades modem culture, especially in America (where it ani-
mates much that passes there for conservatism), the idea of progress is
naturally most at home in liberalism. It is the central irony of modem
liberalism that a political creed devoted to liberty (as it was with the
classical liberals, with Tocqueville, Constant, Madison and the Scottish
24 Thinkers
Enlightenment) should by way of the idea of progress come to subordinate
liberty to the promotion of general welfare. 'The most earnest liberals', as
Santayana observes, 'are quickest to feel the need of a new tyranny: they
are the frrst to support vast schemes of world improvement, reckless of their
concomitant effects on the liberty which an earlier generation of wiser
liberals prized'. 'They [the most earnest liberals]' Santayana tells us, 'save
liberal principles by saying that they applaud it [the new tyranny] only
provisionally as a necessary means of freeing the people. But of freeing the
people from what? From the consequences of freedom' .12 Nothing in
Santayana's analysis suggests that this metamorphosis in liberalism from a
creed of liberty to a creed of progress through a new tyranny was avoidable,
or is somehow (as latter-day classical liberals such as Hayek imagine) now
reversible. Nor does he suppose the present regime of welfare-state or
corporate-capitalist liberalism (as we would now call it) to be stable or
enduring. 'They [the liberals] were no doubt right', he says, 'to be confident
that the world was moving toward the destruction of traditional institutions,
privileges and beliefs; but the first half of the twentieth century has already
made evident that their own wealth, taste and intellectual liberty will
dissolve in some strange barbarism that will think them a good riddance.'13
The liberal age is, then, for Santayana most defmitely a transitional one: its
ruling conceptions of progress and liberty degraded, confused or in-
coherent, its benefits in the growth of wealth and knowledge doubtful and
precarious, and the stability of the social order on which this expansion of
human powers depended fragile. The question arises: by what, in
Santayana's view, might liberalism be replaced?
Whatever he thought might supplant liberalism, Santayana was clear
that it would not be an extension of liberalism in some other form: for, as
he says, 'if anyone political tendency kindled my wrath, it was precisely
the tendency of industrial liberalism to level down all civilisations to a
single cheap and dreary pattern' .14 A post-liberal epoch, if one were to
come about, and mankind not simply relapse into barbarism, would not be
the anarchic Utopia that has always been a temptation of liberalism, and
which is naively theorized in the writings of those ultra-liberals, the anar-
chists, such as Fourier and Proudhon, and, in our own day, the epigones of
libertarianism. If anything, a post-liberal age would be the exact opposite
of such a play-filled Utopia. It would, to start with, shed itself of that spirit
of individuality, virtually unknown among the Greeks and Romans, which
is the gift of Judaism and Christianity to our civilization. Santayana per-
ceives, what is repressed in modem culture, that liberalism and the other
political religions of our age are only the illegitimate offspring of the
Judeo-Christian tradition, itself only the repression (in precisely the
Freudian sense) of our religious needs and sensibility, while political
 Santayana and the critique of liberalism 25
religions - liberalism just as much as Communism - are only the return of
the repressed needs for transcendence and the sacred. If, as the philosophes
of the French Enlightenment hoped and projected, European culture were to
shed its Christian inheritance, it would likely shed its individualism along
with it, and the result would be far from that dreamt of by liberal theory:
Health and freedom ... if recovered against the lingering domination of
Christianity, may reserve some surprises to the modem mind. The
modem mind is liberal and romantic; but a state of society and a
discipline of the will inspired by pure reason would be neither romantic
nor liberal. It would be sternly organic, strictly and traditionally moral,
military, and scientific. The literary enemies of Christianity might soon
find reason to pine for that broad margin of liberty and folly by which
Christianity, in merry Christian times, was always surrounded. They
could have played the fool and the wit to better advantage under the
shadow of the Church than in the social barracks of the future; and a
divided public allegiance, half religious and half worldly, might have
left more holes and cracks to peep through than would the serried
economy of reason. IS
Santayana himself seemed hardly to regret the passing of that spirit of
individuality which Christian faith had nurtured and sheltered. Without
sharing the primitive fantasies of Enlightenment rationalism, he inclined to
welcome a sort of ideal paganism in which religions would be candidly
accepted as local varieties of poetry, and the relativity of piety would
become a commonplace. The political form of this post-liberal epoch would
be a universal empire. Himself entirely a marginal man, through the prism
of whose thought were refracted many national cultures - Spanish, English
and American, among others - Santayana rightly abominated the tribalistic
passions associated with the modem nation-state. The form of economic
and social life he commended, one simpler, more local and more frugal than
that of liberal societies, was to be protected by the framework of a global
peace. His political ideal, and his prescription for a post-liberal order, is an
idealized version of the ancient polity, and is well characterized in an essay
of 1934, 'Alternatives to liberalism':
The ancients were reverent. They knew their frailty and that of all their
works. They feared not only the obvious powers bringing flood, pesti-
lence or war, but also the subtler furies that trouble the mind and utter
mysterious oracles. With scrupulous ceremony they set a watchtower
and granary and tiny temple on some gray rock above their ploughed
fields and riverside pastures. The closed circle of their national eco-
nomy, rustic and military, was always visible to the eye. From that little
26 Thinkers
stronghold they might some day govern the world; but it would be with
knowledge of themselves and of the world they governed, and they
might gladly accept more laws than they imposed. They would think on
the human scale.... In such a case, holding truth by the hand, authority
might become gentle and even holy. 16
Santayana's conception of a post-liberal regime, which has great strengths
as well as disabling weaknesses, has been criticized from a variety of
standpoints, some more vulgar than others. It has been attacked as proto-
Fascist, but this is surely a baseless criticism, with nothing to support it save
that Santayana was living in Fascist Italy at the time he published 'Alterna-
tives to liberalism', and may have thought the Mussolini regime preferable
to the realizable alternatives. The criticism of Santayana's post-liberal
vision as proto-Fascist is indefensible, if only because there are un-
equivocal evidences in his writings of his repudiation of Fascist doctrine
and practice. It is hard to interpret the following passage from Dominations
and Powers as other than an explicit critique of the vulgar Nietzchean
pretensions of the Fascist leaders:
Can it be that these Realpolitiker have forgotten the rudiments of morals,
or have never heard of them? Are these supermen nothing but ill-bred
little boys? ... The oracular Zarathustra, become prime minister, will sit
at his desk in goggles, ringing for one secretary after another.... Poor
superman! As things get rather thick about him, will he regret the happy
irresponsible days when, in the legend of a Borgia, he could publicly
invite all his rival supermen to a feast, in order to have them poisoned at
his table? Or will he remember how, distracted by the heat and nobly
fearless, he drank the poisoned wine by himself, and perished instead
most horribly? Ah, those bold romantic crimes were not really more
satisfactory than the entanglements of this official slavery. Both are
vanity. And in that case, what follows? It follows that these wild ambi-
tions (though some lovely thing may be summoned by them before the
mind) are themselves evil, at least in part; and that the misguided hero,
like a Damocles or like a poor ghost-seeing and witch-hunting Macbeth,
will lose his soul in gaining a sorry world, and his wishes, once attained,
will horrify him. These are trite maxims, and elementary: but I am
talking of children, to children: a pack of young simpletons led by some
young scoundrel. 17
The real criticisms of Santayana's vision lie elsewhere. There is in the first
place a decidedly arcadian quality about his conception of a post-liberal
order. It is not only post-liberal but also post-industrial. It is an obvious, but
also a decisive objection to such a conception, that it is incompatible with
 Santayana and the critique of liberalism 27
the maintenance of the world's existing population. A less crowded world
mjght well be a better one, our present population may indeed be un-
sustainable over the long run; but it is clear that it will be reduced only by
a catastrophe in whose wake not only the industrialism Santayana despised,
but also the possibility of a post-industrial order on the lines Santayana
envisaged, would be effectively destroyed. There are also deep difficulties
in Santayana's account of the imperial framework within which the local
and rustic economies he prized would be sheltered. His idea of a liberal
universal empire has many attractions. It acknowledges, what con-
temporary liberal thought cannot admit, that it is the empire, and not the
nation-state, that is the most appropriate political order for the realization
of the goals of individual liberty, the rule of law and peace among diverse
communities. As Santayana, outlining the idea of a universal liberal
empire, asks rhetorically: '
Why not divorce moral societies from territorial or tribal units, so that
membership in these moral societies, as in a free Church, should be
voluntary, adopted only by adults with a full sense of their vocation for
that special life, and relinquished, without any physical hindrance, as
soon as that vocation flagged, or gave place to some other honest
resolution?IB
Again, he observes:
Under such a Roman peace, as we call it, a further development is
possible. Not only may each nation, within its territory, preserve its
language and laws and religion under the imperial insurance, but where
different nations have intermingled, as often happens in great cities or in
provinces vaguely open to any immigrant, each may preserve all its
moral idiosyncrasy, its speech, dress and domestic life, side by side with
the most alien races. 19
Santayana's conception has the decisive merit of acknowledging the politi-
cal form of empire to be that best suited to maintaining the condition of
cultural pluralism - the variety of traditions and ways of life - advocated,
but everywhere discouraged, by modern liberalism. There is nevertheless a
vast unreality in his speculations on how this liberal universal empire might
come about. He rejects the idea that it might be a Pax Americana, on the
plausible ground that the very virtues of American culture - its evangel-
izing reformism and incorrigible optimism - ill fit the United States for an
imperial role. More disputably, he rejects the British Empire (as it existed
when he first wrote in 1934 what later became a chapter in Dominations
and Powers20 ) as a model for a Roman peace on the ground that the British
had a contemptuous indifference to the customs of their subjects that
28 Thinkers
rendered them odious as rulers. (The observation may be correct, but the
British attitude to their subjects had many points of affinity with that of the
Romans to theirs - and it at least protected the British from the folly of a
messianic liberalism.) Most absurdly, Santayana - not perhaps without a
motive of perversity - suggests that 'Perhaps the Soviets might be better
fitted than any other power to become the guardians of universal peace'.
With the utmost naivete (or disingenuousness), he goes on:
In regard to tenure of land, and to the management of industry and
communications, if the management were competent, a universal com-
munism, backed by irresistible armed force, would be a wonderful boon
to mankind .... The Soviets would ... have to renounce all control of
education, religion, manners and arts. We are proletarians and unwilling
Communists only in the absence of these things; in their presence, we ali
instantly become aristocrats. Everything except the mechanical skeleton
of society . . . must be left to free associations, to inspiration founding
traditions and traditions guiding inspiration. The local attachments of
such culture are important. and a just universal government would not
disturb them. Each nation or religion might occupy, as private property
under the common law, its special precincts or tracts of lands; or it might
live locally intermingled with other nations and religions; but each in its
own home would be protected from annoyance, and free to worship its
own gods with the homage of a complete life fashioned in their image. 21
Here Santayana's enmity to liberal culture has led him radically astray. It
was evident to any judicious observer by the mid-thirties, and certainly by
the time Dominations and Powers was published, that the Soviet system
was perhaps the most destructive of local traditions, and above all of
religious practices, in human history, and everywhere implemented a
revolutionary Communist messianism incomparably more tyrannous than
any identified by Santayana's strictures against the United States. For this
reason, if we were ever to fall under a Pax Sovietica, it would be the peace
of universal impoverishment and barbarism, and not the just liberal empire
of which Santayana dreamt.
In truth, Santayana was unable to give an account of any plausible
alternative to liberal society. He comes very close to admitting this, when
in •Alternatives to liberalism' he identifies liberal society as a transitional
state, but cannot specify its successor:
liberalism presupposes very special conditions. It presupposes a tradi-
tional order from which the world is to be emancipated. It presupposes
heroic reformers, defying that order, and armed with a complete innate
morality and science of their own by which a new order is to be
 Santayana and the critique of liberalism 29
established. But when once the new order has been thoroughly des-
troyed, that kind of heroic refonner may well become obsolete. His
children will have no grievances and perhaps no morality. Even the
abundance of their independent sciences, without an ultimate authority
to synthesize or interpret them, may become a source of bewildennent.
Add interracial war and a breakdown in industry, and there may seem to
be occasion for turning over a new leaf. As to what may be found, or may
come to be written, on the next page, no political programme can give us
any assurance. 22
If this diagnoses sapiently the ills of the liberal condition, Santayana is here
(confessedly) unable to offer us a panacea for them. His critique of
liberalism, however, retains considerable value. Santayana illuminated the
chief danger to liberal society in liberalism itself - in a political religion of
man-worship which had lost the humility, and indeed the scepticism, that
infonns the historic Western religions. With its incoherent doctrine of
progress, its inordinate and antinomian individualism, and its ultimate
subordination of the claims of liberty to those of an imaginary general
welfare, liberalism has at length become the enemy of the civil society it
once sought to theorize. As Santayana puts it in his apologia pro mente sua:
My naturalism and humanism seemed to them [Santayana's liberal
readers] to give carte blanche to revolution: and so they do, if the
revolution represents a deeper understanding of human nature and
human virtue than tradition does at that moment; but, if we make
allowance for the inevitable symbolism and convention in human ideas,
tradition must nonnally represent human nature and human virtue much
better than impatience with tradition can do; especially when this im-
patience is founded on love of luxury, childishness, and the absence of
any serious discipline of mind and heart. These are the perils that
threaten naturalism and humanism in America. 23
Santayana's political thought, despite his hopes of it, does not encapsulate
any continuing amount of a post-liberal order. What it most powerfully
suggests is the necessity (if the historical inheritance of liberal society
which is our patrimony is to be preserved and not squandered) of aban-
doning the romantic culture of limitless hubris for a classical ethos of
limitation and constraint. We are not, each of us, as our liberal culture
encourages us to imagine, limitless reservoirs of potentiality, for whom the
past is an incumbrance and the future a blank sheet of possibility. We are
finite, mortal selves, burdened by the evils of our history and the miseries
natural to the human condition, who achieve excellence and a measure of
well-being only in so far as we accept the disciplines of civilization.
30 Thinkers
Political wisdom is the sad business of prescribing for ordinary mortals, not
a recipe for adventures in infinite freedom. This humble view was not
unknown to the early theorists of liberalism, but it has been repressed and
obliterated by modern liberalism, with its hallucinatory perspectives of
open-ended world improvement and global betterment. We conceive
government aright, if we conceive it as Hobbes did, and as some of the early
liberals did, as a shelter against the worst of evils, civil war and the
reversion to nature. It is the limited government which we inherit as our
tradition - now far gone in desuetude - which protects us against such
misfortune, and which thereby allows the emergence of civil society as the
artificial remedy for our natural imperfections. Santayana reminds us, in his
remarkable essay on Freud, that
The human spirit, when it awakes, finds itself in trouble; it is burdened,
for no reason it can assign, with all sorts of anxieties about food,
pressures, pricks, noises and pains. It is born, as another wise myth has
it, in original sin. And the passions and ambitions of life, as they come
on, only complicate this burden and make it heavier, without rendering
it less incessant or gratuitous.24
Political wisdom lies in accepting this human lot, not in seeking to wish it
away by vast projects of reform. We alleviate the human lot as best we can
by repairing and renewing the traditional institutions we have inherited,
where these are themselves founded on a sane awareness of the human
condition. For us, this means returning to the minute particulars of the
liberal inheritance of civil society that we are at present frittering away in
crazed fads and fashionable experiments. Our danger is that, like an ageing
debauchee who has grown used to drawing on his patrimony, we may not
realize until it is too late that it has largely been consumed. It is such a
condition of which Santayana wrote when he observed that
we sometimes see the legislator posing as a Titan. Perhaps he has got
wind of a proud philosophy that makes will the absolute in a nation or in
mankind, recognising no divine hindrance in circumstances or in the
private recesses of the heart. Destiny is expected to march according to
plan. No science, virtue or religion is admitted beyond the prescriptions
of the state.... Here is certainly an intoxicating adventure; but I am
afraid a city so founded, if it could stand, would turn out to be the iron
City of Dis. These heroes would have entombed themselves in hell, in
scorn of their own nature; and they would have reason to pine for the
liberal chaos from which their satanic system had saved them.2S
The final insight of Santayana's critique of liberalism is not that liberal
society is inherently a transitional state of things. It is that the nemesis of
 SantayanD and the critique of liberalism 31
liberal society is its self-destruction by liberal ideology - by a frenzy of
theorizing which is willing to lay waste the inherited institutions of a liberal
order -limited government, private property, the rule of law - for the sake
of an imagined improvement of the human condition. Santayana's insight-
an insight he did not always himself grasp, captivated as he sometimes was
by delusive visions of radical alternatives to liberal society - is that the
preservation of the liberal inheritance has as its most necessary condition a
comprehensive disenchantment with liberal theory. Unless we can shake
off the hubristic illusions of liberalism, the spiralling decline in our civil-
ization is unlikely to be arrested, and may well end in the collapse of liberal
society itself.
3 Hayek as a conservative

Is Hayek a conservative? Many conservatives will quickly deny that he can

be anything of the kind. They will cite the famous Postscript to his Con~
stitution of Liberty,l 'Why I am not a Conservative', in which Hayek
disavows the characteristically conservative project of using the power and
authority of the state to protect endangered moral traditions and to shore up
threatened social hierarchies, and argues instead for a version of the classi-
cal liberal view, that the primary task is the curbing of all such political
power. Conservatives often invoke this and other evidence in support of a
picture of Hayek as a doctrinaire defender ofliberty, whose general outlook
is little different from that of Nozick, or a partisan of laisser-faire such as
Milton Friedman. It is a short step from radical libertarianism to an
ideology which, while centred on the defence of the market economy, is
neglectful of the moral tradition which makes a market economy possible.
Hayek's conservative critics take this step on his behalf, and condemn him
accordingly.
It is not hard to show that the standard conservative view of Hayek's
thought is ill-founded. Hayek's position is distinctive, to be sure. It
embodies the best elements of classical liberalism and also suggests a
criticism of many conventional conservative positions. At the same time it
derives from some of the most profound insights of conservative philo-
sophy, and puts them to work in an original and uncompromising fashion.
We must recall Hayek's birth and education in the last two decades of
the Hapsburg Empire, in whose defence he fought as an aircraftman, and
remember that his formative intellectual influences were not those of
English-speaking empiricist philosophy. Central among these influences
were the philosophies of Immanuel Kant and Ernst Mach, variants of which
dominated the intellectual life by which Hayek was surrounded in his
youth. Hayek's thought also bears the imprint of the Viennese critics of
language: of Karl Kraus, of the now almost forgotten Fritz Mauthner, and
of Hayek's half-cousin, Ludwig Wittgenstein. The reflections of these men
 Hayek as a conservative 33
on the decay of intelligence wroug'ht by the perversion of language have
always inspired Hayek, and have played a part in many of his later writings.
(Hayek's devastating analysis of the expression, 'social justice' ,2 in which
he illuminates its workings so as to make clear its lack of defmite sense,
may be best understood as continuing the Krausian tradition of resisting the
modern idolatry of general words.)
Most importantly, though, Hayek's writings reflect his lifelong aspira-
tion to come to terms with the debacle of the First World War, when the
high civilization and the rule of law established in Hapsburg Austria-
Hungary gave way to chaos and barbarism. As a result of this weakness,
Europe has been engulfed by vast movements dedicated to the repudiation
of the European inheritance. Hayek's attempt to synthesize the deepest
insights of conservatism with the best elements of classical liberalism
merits our closest scrutiny, if only because the experience which inspired it
- the experience of an apparently inexorable drift to dissolution and bar-
barism in all the central institutions of society - may not be so far from our
own. His researches into the ultimate sources of the current malaise of
civilized authority led him deeply into the theory of knowledge and into
philosophical psychology: for it is Hayek's view that the impossible ambi-
tions spawned by contemporary culture arise from a false understanding of
the human mind itself.
The well-spring of all Hayek's work in social philosophy and in eco-
nomic theory is, then, a conception of human knowledge. Hayek's theory
of knowledge can be understood, in the first place, as a sceptical variant of
Kantianism. For Hayek, as for Kant, our knowledge is not based in in-
corrigibly known sensations,] and the empiricist attempt so to reconstruct it
is forlorn. Our minds are no more passive receptacles for sensory data than
they are mirrors for reflecting the necessities of the world: they are creative
powers, imposing order on a primordial chaos by way of a built-in set of
categories. Philosophy cannot hope to step outside these categories so as to
attain a transcendental point of view, or to reach an Archimedean point of
leverage from which to assess or reform human thought. For Hayek, as for
Kant, philosophy is reflexive and critical rather than transcendental or
constructive: it plots the limits of the human understanding but cannot hope
to govern it.
Hayek's sceptical Kantianism has features, however, which take it far
from anything that Kant could have accepted and which give to it a wholly
distinctive turn. The organizing categories of the human mind are, for
Hayek, neither immutable nor universal; rather they express evolutionary
adaptations to a world that is in itself unknowable. Hayek's thought has
here a real point of affinity with that of his friend Sir Karl Popper, who has
long expounded a naturalistic and evolutionary theory, in which human
34 Thinkers
knowledge is regarded as continuous with animal belief. More decisively,
however, Hayek differs from Kant in denying that the governing principles
of our minds are fully knowable to us. We will always be governed by rules
of action and perception, which structure our experience and behaviour
down to their last details, and some of which will necessarily elude our
powers of critical inquiry. In recognizing these elements of our mental life
- these 'meta-conscious rules' of action and perception, as he calls them4 -
Hayek identifies a limit to the powers of reason more severe than any Kant
could have admitted. For if such rules exist, then (though we can at no point
learn their content), we can be sure that critical thought itself is governed
by them. Hence our own minds, no less than the external world, must in the
end remain a domain of mystery for us, being governed by rules whose
content we cannot discover.
Hayek is a Kantian, then, in denying that we can know things as they are
in themselves, or can ever step out from the categories which govern our
understanding. He goes further than Kant. in seeing the categories of our
understanding as mutable and variable, and in some major degree un-
knowable to us. Most distinctive in Hayek's sceptical and Kantian theory
of knowledge, however, is his insight that all our theoretical, propositional
or explicit knowledge presupposes a vast background of tacit, practical and
inarticulate knowledge. Hayek's insight here parallels those of Oakeshott.
Ryle, Heidegger, and Polanyi; like them he perceives that the kind of
knowledge that can be embodied in theories is not only distinct from, but
also at every point dependent upon, another sort of knowledge, embodied
in habits and dispositions to act. Some of this practical knowledge is found
in rules of action and perception imprinted in the nervous system and
transmitted by genetic inheritance. But much of the most significant part of
the practical knowledge expressed in our dealings with each other is passed
on mimetically, in the cultural transmission of traditions or practices, some
of which are bound to be inaccessible to critical enquiry. In all our relations
with the social world we are informed and sustained by these elements of
tacit knowledge, which we know to be pervasive in our thought and
conduct, but whose content we can scarcely guess at.
In his own view, and surely rightly, Hayek's conception of the human
mind as governed by rules, some of which must escape conscious scrutiny,
has the largest consequences for social philosophy. For it entails the bank-
ruptcy of the rationalist project, undertaken in different ways by Bacon,
Descartes, and Spinoza, of subjecting the mind to a systematic purge of
tradition and prejudice. We can never know our own minds sufficiently to
be able to govern them, since our explicit knowledge is only the visible
surface of a vast fund of tacit knowing. Hence the rationalist ideal of the
government of the mind by itself is delusive. How much more of a mirage,
 Hayek as a conservative 3S
then, is the ideal of a society of minds that governs itself by the light of
conscious reason. The myriad projects of modern rationalism - con-
structivist rationalism, as Hayek calls it - founder on the awkward fact that
conscious reason is not the mother of order in the life of the mind, but rather
its humble stepchild. AU of the modern radical movements - liberalism
after the younger Mill as much as Marxism - are, for Hayek, attempts to
achieve the impossible. For they seek to translate tacit knowledge into
explicit theory and to govern social life by doctrine. But only tacit knowl-
edge can engender government, and tacit knowledge may be lost by its
translation into overt, propositional form. Hayek is here developing, in its
political implications, a version of the thesis of the primacy of practice in
the constitution of human knowledge. The thesis has a distinguished pedi-
gree in the writings of a number of contemporary philosophers, of whoql
Oakeshott, Wittgenstein and Heidegger are perhaps the most notable.
The thesis of the primacy of practice leads Hayek to refine the argument
that rational resource-allocation under socialism is impossible - an argu-
ment which Hayek inherited from his colleague L. von Mises. In his
disputes with socialist economists, Mises had contended that, in the
absence of market-pricing of all factors of production, chaos in calculation
was bound to ensue, and could be avoided only be relying on world
capitalist markets and domestic black markets. (Saul Kripke has noted' an
interesting analogy between this argument of Mises's and Wittgenstein's
arguments against the possibility of a private language - an analogy I can
only remark upon, but not pursue here.) Hayek sees, as his colleague Mises
did not, that the knowledge which is yielded by market pricing cannot be
collected by a central authority or programmed into a mechanical device,
not just because it is too complex, nor yet because it is knowledge of a
fleeting reality (though this is closer to the nub of the matter), but rather
because it is knowledge given to us only in use. It is knowledge stored in
habits and in practice, displayed in entrepreneurial flair, and preserved in
the countless conventions of business life. Unhampered markets transmit
this knowledge, which is otherwise irretrievable, dispersed in millions of
people. One may almost say that, for Hayek, this practical knowledge
achieves a full social realization only when market-pricing is not interfered
with. For -like much traditional knowledge - it is holistic, a property of the
entire society, and not the private possession of any of its separate elements
or members.
Hayek's case against socialist planning, and in favour of the un-
hampered market, rests upon these considerations rather than upon any
Lockean theory of property rights, or upon a fanaticism for laisser-faire.
The impossibility of socialism, and of successful intervention in the
economy, is an epistemological impossibility (as well as a moral
36 Thinkers
impossibility). His differences with Keynes, for example, are imperfectly
understood if one does not grasp that the Keynesian macroeconomic
manager must claim knowledge which Hayek insists is available to no one.
The Keynesian planner may indeed achieve temporary successes, by
exploiting money illusion and manipulating business confidence, but this is
bound to be a short-lived victory. Such Keynesian policies work only in so
far as they are contrary to established expectations which they cannot help
eroding. Moreover, they take no account of the inevitable discoordination
of relative prices and incomes. Governments can do next to nothing to
remedy these consequences, since it is given to no one to know what is the
correct relative price-structure. Hayek nowhere suggests that market failure
is an apodictic impossibility; but he is surely on firm ground in arguing that
it is in the unhampered operation of the market process itself that we hav~
the best assurance of economic coordination .
. Socialism and interventionism, then, are but long shadows cast by a false
philosophy of mind. The order we fmd in society, no less than that which
prevails in our own minds and bodies, is an undesigned order, and not a
product of rational planning. The dominant superstition of the Age of
Reason is the belief that vital social institutions - the law, language, and
morality, as well as the market - must be or can become products of
conscious contrivance and control, if they are effectively to serve human
purposes. This modem superstition results from an anthropomorphic trans-
position of mentalistic categories to the life of society. Hayek's criticism
echoes a distinguished line of antirationalist thinkers, of whom Pascal is
perhaps the closest to him, in his celebrated distinction between ['esprit de
geometrie and I'esprit de finesse. It is because the rational principles of
social life are immanent in its practices that we cannot trust our reason in
its speculative projections for reform.
This is why, especially in his later writings, Hayek attaches so great an
importance to the spontaneous development of the law in the institution of
an independent jUdiciary. He goes so far as to see, in the contemporary
recourse to legislation, a major threat to liberty and to social stability. There
is, indeed, an important analogy between Hayek's arguments for the impos-
sibility of comprehensive economic planning and his criticism of a legal
system that is dominated - as all now are - by statute. Just as no economic
plan can approach the sensitivity and subtlety of the market process in
integrating men's plans and achieving coordination in the use of resources,
so statutory legislation cannot match the sensitivity of the common law in
responding to and adjudicating the concrete problems of man's social
existence; But the common law, which relies on the doctrine of precedent,
cannot survive without a strong, independent and decentralized judiciary.
 Hayek as a conservative 37
It is not that Hayek supposes that a modern state can altogether forswear
legislation,6 any more than it can wholly dispense with economic policy;
but in both cases the balance needs to be redressed, in favour of spon-
taneous order, whether that of the market or that of the judicial process. The
two issues of economic planning and the rule of law are therefore insepar-
ably connected for Hayek. He sees clearly that the rise of the administrative
state, together with the prevalence of grandiose projects for redistribution
and social welfare, pose a major threat to the rule of law, and therefore to
individual liberty. A government which seeks to regulate prices and
incomes is bound to transfer large powers to administrative authorities. In
the nature of things these authorities will exercise a terrifying discretion
over the lives and fortunes of the citizens. Such authorities may clothe their
arbitrariness in an ideology of social justice, or they may attempt to revive
the doctrine of the Just Wage. But their decisions cannot be contained
within the rule of law, for they crucially depend upon a claim to knowledge
which no one possesses - a claim which, in the nature of things, cannot be
adjudicated.
Aside, then, from the fact that policies of intervention in the market and
in the provision of social goods tend to expand as their failures are recog-
nized, such policies necessarily involve a transfer of authority over our
lives to administrators effectively unconstrained by law, and often uninhi-
bited by common moral sentiments. Hayek's criticism of the ambitions of
the administrative and welfare states should be less implausible in con-
servative circles than it was when his Road to Serfdom' appeared. We can
see now the accuracy of his prediction that the expansionist state will be
captured by movements and professions whose outlook and interests are
deeply at odds with the preservation of established ways of life.
What in turn may conservatives learn from Hayek's thought? His chief
importance, I think, is that he has freed classical liberalism from the burden
of an hubristic rationalism. He has thereby produced a defence of liberty
which aims to reconcile the modem sense of individuality with the claims
of tradition. Hayek shows that we are bound to rely primarily on inherited
traditions of thought and conduct in all our dealings with each other. The
inarticulate character of the great submerged part of our knowledge means
that we always know far more than we can ever say. It also means,
crucially, that the rational criticism of social life must come to a stop when
it reaches the tacit component of our practices.
There is an uncomfortable lesson here for conservatives, since Hayek's
diagnosis condemns the attempt to retard or reverse the flood of social
change, no less than it undermines the reformer's desire to remodel society,
according to some more 'rational' plan. Hayek would heartily endorse
38 Thinkers
Wittgenstein's remark, that trying to salvage damaged traditions by wilful
effort is like trying with one's bare hands to repair a broken spider's web.
The most we can do is to remove those artificial impediments to the vitality
of our traditions which have been imposed by the state. And with its
policies in education and housing, the state has surely been a far greater
destroyer of traditions and communities than has the market.
Hayek's chief lesson for conservatives, then, is that it is a delusion to
think that conservative values can be protected by a successful capture of
the expansionist state. The damage done to social life by an invasive state
is integral to its existence, and conservative governments are better occu-
pied during their tenure of office in whittling down the state and in restoring
initiative to the people, than in the futile enterprise of trying to convert the
state's bureaucracies to a conservative view of things.
This is not to say that Hayek's thought is not open to legitimate con-
servative criticism. At times he seems to subscribe to a doctrine of historical
progress which, though it was accepted by such conservatives as Burke and
Hegel, cannot be endorsed by any twentieth-century conservative. Here I
think we must tum to Michael Oakeshott's writingsB for the insight that
human history is not to be construed as a single evolutionary process, but
rather as a series of distinct adventures in civilization. Although Hayek is
perhaps right to see the conquest of the world by European individuality as
an historical fate, which it is idle to wish away, he may be too ready to see
this as a stage in a global progressive development. There is a faint echo in
his writings, which a conservative would wish could not be heard, of the
historical theodicy of the Enlightenment. This theodicy is an indefensible
and indeed pernicious part of the inheritance of classical liberalism, which
in most other respects we are wise to cherish.
Finally, Hayek's thought poses a dilemma for conservatives which few
of them have yet come to recognize. The dilemma is found in his percep-
tion, especially in his later writings, that the modem development of
age-old European moral and intellectual traditions has produced an outlook
that is deeply destructive of civilized institutions. The peculiarly modernist
outlook - a combination, I should say, of homeless moral passion with
rationalist fantasy - is now so pervasive as to have acquired deep roots in
popular sentiment and a secure place in virtually all the disciplines of
thought. It results in what Hayek calls 'unviable moralities'9 - systems of
moral thought and sentiment incapable of sustaining any stable social
order; in the bizarre intellectual constructs of contemporary sociology; and
even, as in architecture, in a corruption of the practical arts. Taken together,
these developments create a climate of culture which is profoundly hostile,
not only to its traditional inheritance, but even to its own continued exis-
tence. We confront the phenomenon of a culture permeated throughout by
 Hayek as a conservative 39
a hatred of its own identity, and by a sense of its purely provisional
character. This culture is not without sources in our most ancient religious
and moral traditions - for example, in Platonic rationalism and in Christian
moral [Link] In his writings on Mandeville, Hayek has made clear that the
defence of the market economy may demand a far from conservative
revision of ordinary morality. II His latest thoughts on the phenomenon of
intellectual and moral inversion 12 suggest that he has illuminated what is,
from a conservative viewpoint, an even greater problem: since much of
contemporary culture is possessed by a death-wish brought on by patho-
logical developments in some of our oldest traditions, a modem con-
servative must also be a moral and intellectual radical.
4 Oakeshott as a liberal

What may now be meant by the word 'liberal' is anyone's guess.

Michael Oakeshottl
The claim that the conception of political life that animates the thought of
Michael Oakeshott, clearly one of the most original and profound British
political philosophers since Hume and certainly this century's greatest
conservative writer, is a liberal conception may seem unacceptably para-
doxical or even wilfully perverse. Yet it is far from being novel. Com-
mentators such as W. H. Greenleaf and Samuel Brittan have seen in
Oakeshou's thought deep affinities with the intellectual tradition of classi-
cal liberalism; an entire paper (greatly admired by Oakeshott himself2) has
been devoted to interpreting Oakeshott as a liberal theorist; and both of the
two recent book-length studies of Oakeshott's work3 characterize his out-
look on politics as at least akin to that of a liberal. For all that, it still remains
unclear just what sort of liberal Oakeshott might have been, what it was that
he took from liberal thought and what he rejected, and how the liberal
element of his thought coheres with the rest.
It is not too hard to be clear about what Oakeshott rejected in liberal
thought. The liberal project of fIXing, by way of a doctrine or a theory, the
proper scope and limits of the authority of government, determinately once
and for all, as it was attempted by Locke or Kant, J. S. Mill or, in our own
time, by Rawls and Nozick, Oakeshott rejected as a prime example of
rationalism in politics. From this point of view, liberalism is (or was)
merely a species of ideology - of that rationalist abridgement of the
contingencies and vicissitudes of practice that aspires to be, but can never
succeed in becoming, an authoritative, prescriptive guide for practice. The
proper tasks and limits of government cannot be determined by reasoning
from first principles (supposing there could be such things); they can be
established, always provisionally and never indisputably, only by reason-
ings that are circumstantial and which invoke precedents, judgements and
 Oakeshott as a Uberal 41
practices that are already present in current political life. Oakeshott's
rejection of the rationalist (or fundamentalist) element in liberalism
expresses one of his most profound insights - that political deliberation and
political discourse are closer in their character to conversation than they are
to any sort of demonstrative reasoning. For Oakeshott, as for Aristotle,
political discourse is a form of practical reasoning, and in virtue of that
character can never issue in, or rest upon, the certainty found (according to
Aristotle, at any rate) in the theoretical disciplines. Political discourse, then,
though it may contain passages of argument, is not an argument, but a
conversation.
Oakeshott's repUdiation of this aspect of liberalism - perhaps its most
definitive or constitutive - has far-reaching implications for the current
understanding (or confusion) we have of the relations of political philo-
sophy with political practice, and it contains lessons which, if we could
learn them, might at least temper the barbarism into which political life has
long since fallen among us. The first lesson is that the hubristic project of
doctrinal or fundamentalist liberalism - the project attempted by all the
liberal thinkers mentioned earlier of fixing the boundaries of government
action by some principle or doctrine, be it laissez-faire, a specification of
allegedly natural (but patently conventional) rights, or a list of basic liber-
ties derived (as in Rawls and Dworkin4) from a conception of justice rooted
only in the fleeting local knowledge of the American academic nomen-
klatura - embodies a mistaken conception of philosophy itself. Whereas
philosophy can clarify the presuppositions or postulates of practice, and
thereby perhaps in some degree illuminate it, it can neither found practice
nor govern it. For it is practice that is always primordial. Accordingly, if
philosophical inquiry has any practical effects or benefits, they are oblique,
indirect and prophylactic. (And, of course, philosophy need have no other
goal than the pursuit of understanding.) Oakeshott's conception of philo-
sophy radically undermines the liberal foundationalist project, which seeks
to circumscribe the authority of government by a theory. It also serves to
chasten the ambitions of philosophers, who seem always to have been
astonished that political life could proceed without their constant
supervision.
Oakeshott's conception of political life as a conversation may have a
prophylactic role to play in respect of the barbarization of political dis-
course and practice that in our day is far gone. The tendency of much in
recent thought and practice has been to try to assimilate political discourse
to some other, supposedly superior mode, and thereby to deny to political
life its autonomy. This decline is, perhaps, farthest gone in the United
States, where in the context of a common culture whose resources are fast
depleting every political question is now couched in legalist terms. The idea
42 Thinkers
of politics as a conversation in which the collision of opinions is moderated
and accommodated, in which what is sought is not truth but peace, has been
almost entirely lost, and supplanted by a legalist paradigm in which all
political claims and conflicts are modelled in the jargon of rights. In such a
context, not only is civilized political discourse virtually extinguished, but
the legal institutions into which it is transplanted are corrupted. The courts
become arenas for political claims and interests, each of them inordinate
and resistant to compromise, and political life elsewhere becomes little
more than bargaining and log-rolling. The result of the American assimi-
lation of politics to law has, then, been the corruption of both. In the course
of this development, the profound insights of the authors of The Federalist
Papers into the limitation of government, not primarily by constitutional
devices, but by the civic virtues of a free people that has in common a
devotion to the culture and practice of liberty and which understands the
ground of that practice to be the imperfectibility of our species, have been
almost lost. In Europe, and in Britain, the virus of legalism has been less
pervasive, but the presence of other ideologies more powerful. If socialism
is everywhere on the wane, other ideologies are waxing - including ideo-
logical strands within conservative thought and practice, which ascribe to
governments the objectives of restoring a lost moral consensus, of reviving
a fragmented national integrity, or promoting maximal wealth-creation.
Here the autonomy of political life is compromised by its attempted assimi-
lation to the modes of discourse of history, economics or industrial manage-
ment. If we could recover the understanding of political discourse
Oakeshott has theorized for us, that understanding might - far more than the
institution of what Oakeshott, in the magnificent and invaluable new edi-
tion of Rationalism in Politics published by Liberty Press of Indianapoliss
and edited by Timothy Fuller, calls 'the absurd device of a Bill of Rights' -
protect us from the inordinacies and invasiveness of the modem state.
What we have seen so far is that the rationalist projects of doctrinal
liberalism, spawned by a false philosophy that pretends to govern rather
than merely to struggle to understand practice, have had the effect of
corroding our historical inheritance of civil society and of weakening
traditional constraints on the activities of government. We have yet to see
what Oakeshott retains of liberal thought. We find this, if I am not
mistaken, in the understanding of civil life he develops in his conception of
civil association, which may be taken to be the living kernel of liberalism,
and the chief element in Oakeshott's thought in virtue of which it deserves
the appellation 'liberal'. In his account of civil association, Oakeshott (as
Wendell John Coats, Jr, shows in his admirable paper 6) synthesizes a
variety of seemingly disparate understandings, Aristotelian and Ciceronian,
Roman and Norman, Lockean and Hegelian, to yield an account of that
 Oakeshott as a liberal 43
fonn of human association - emerging distinctively only in late medieval
Europe, and never without opposition in any modem European state - in
which men live together, not under the aegis of any common end or
hierarchy of ends, but instead by their subscription to a body of non-
instrumental rules, whereby they can (in all their variety and conflicting
purposes) coexist in peace. Civil association has, from the first, had a rival
in enterprise association - that mode of association constituted by shared
adherence to a common objective, the mode of association animating an
industrial corporation, say. What distinguishes a civil association from an
enterprise association, most fundamentally, is that the fonner is purposeless
in having no projects of its own, whereas the latter is constituted by its
projects. For Oakeshott, this is to say that civil association is a fonn or
mode of association that is non-instrumental and moral, whereas the latter's
authority derives solely from the projects that animate it.
Oakeshott is clear that modem European states have always partaken of
both modes of association. Rarely, if ever, has such a state been merely a
guardian of civil association. As he observes: 'Modem history is littered not
only with visions, in various idioms, of a state as a purposive association,
but also with projects to impose this character upon a state'.' Like the
minimum state of classical liberalism, civil association is an ideal type, or
a limiting case, not found in the real world of human history. In historical
fact, all modem European states have had elements - managerialist, mer-
cantilist or corporatist - in virtue of which they have acquired the character
of enterprise associations, and their role as custodians of civil association
has accordingly been compromised. Indeed, it may be said of twentieth-
century totalitarian states that they are nothing but enterprise associations,
committed for that reason to the destruction of civil association. And one of
the inheritances of liberal ideology, the idea of progress in history, has
made it hard for any modem state not to claim for itself the role of an agent
of world-bettennent. Indeed civil association itself has been defended as a
means to prosperity - a complete misunderstanding of it in Oakeshott's
tenns. One may say, in fact, that, though it has never been altogether
silenced, the voice of civil association has in our century never been very
loud: it has, most of the time, been shouted down by all those ideologies and
movements that perceive the authority of he state only in its success in
exploiting the earth's resources. It is this Baconian conception of the office
of government that, in Oakeshott's view, pervades all contemporary con-
ceptions of the state-as-enterprise-association, be they Fascist or Fabian,
Marxist or Manchesterist in inspiration.
We can now see what Oakeshott's thought has in common with classical
liberalism. For the classical liberal, the authority of the state did not depend
on its contribution, if any, to economic growth, any more than it depended
44 Thinkers
on the local cultural identity of its subjects. It depended on its being
recognized, not in tenns of desirable projects or particular cultural values,
but in tenns of law or justice - of lex. As Robert Grant has put it, in what is
by far the most perceptive and illuminating account of Oakeshott' s thought
we possess, 'A civitas (or civil association) need not be culturally homo-
geneous (thought doubtless a degree of homogeneity will help). The only
homogeneity which counts, and which is essentially "cultural", consists in
a common disposition to value lex above one's local cultural identity.'8 It is
in his account of the authority of the office of rule as being fonnal or
abstract in this fashion, then, not in his eloquent defence of individuality,
that the most fundamental affinity between Oakeshott and classical liberal-
ism is to be found. It shows him to have more in common with Kant, in the
end, than with Burke or Hegel.
It may well be that it is precisely in its kinship with classical liberalism
on the nature of the authority of the state that the Achilles heel of Oake-
shott's political thought lies. Both assert that the authority of a modem state
depends neither on its success in any substantive purpose nor on its
relationship with the cultural identity of its subjects. Both claims are
profoundly questionable. As to the fonner, less fundamentally important
claim, it is a matter of brute historical fact that no government can, in any
modem democratic state, long survive if it does not preside over sustained
economic growth. For this reason, every modem government is inexorably
drawn into the dreary business of aligning the economic with the electoral
cycle: all are in some significant measure mercantilist or corporatist in
practice. In short, all modem states are to a considerable degree enterprise
associations. The project of effecting a radical disseveration of government
from the conduct of economic life, which Oakeshott shared with the
classical liberals, has in common with classical liberalism a quality
otherwise entirely absent from Oakeshott's thought - namely, its Utopian
character. The project of denying to the modem state its character as, in
significant part, an enterprise association, is hardly a conservative one; if it
could be achieved, the transfonnation of political arrangements thereby
wrought would be little short of revolutionary. This is, in effect, tacitly
acknowledged by Oakeshott, when in his most overtly political essay, 'The
political economy of freedom' ,9 he endorses the radical proposals advanced
by the Chicago economist Henry Simons, in his Economic Policy for a Free
Society. to More typically, the debacles of Thatcherism and Reaganism sug-
gest that, for us, the enterprise-association state is an historical fate, which
we may indeed strive to temper, but which we cannot hope to overcome.
The second claim - that the authority of a modem state does not depend
on its subjects having a common cultural identity - is yet more question-
able, and more importantly so. It is the claim, found in Kant and in much
 Oakeshott as a liberal 45
recent American constitutional theory, that the recognition of the state as
embodying abstract or formal principles of law or justice is all that is
needed for its authority. Recent history does little to support this claim. On
the contrary, it may be observed that, in so far as a polity is held together
by little else than allegiance to abstract principles or procedural rules, it will
be fragmented and unstable, and its authority weak. The point may be made
in another way. The idea that political authority could ever be solely or
mainly formal or abstract arose in times when a common cultural identity
could be taken for granted. For Kant as for the framers of the Declaration
of Independence, that common cultural identity was that of European
Christendom. In so far as this cultural identity is depleted or fragmented,
political authority will be attenuated. We may see this ominous develop-
ment occurring in microcosm in Britain, where a minority of funda-
mentalist Muslims that is estranged from whatever remains of a common
culture, and which rejects the tacit norms of toleration that allow [Link]
society to reproduce itself peacefully, has effectively curbed freedom of
expression about Islam in Britain today. We may see the same somber
development occurring on a vast scale in the United States, which appears
to be sliding inexorably away from being a civil society whose institutions
express a common cultural inheritance to being an enfeebled polity whose
institutions are captured by a host of warring minorities, having in common
only the dwindling capital of an unquestioned legalism to sustain them.
Oakeshott himself observes that
the authority of an office of rule remains always a delicate matter of
current belief. It must be able to survive a rainy day; it must be proof
against disapproval and ridicule of the performances of the office and it
cannot be bought with good works; it hangs like a drop of dew upon a
blade of grass. II
For us, who live in an age of mass migrations and fundamentalist convul-
sions, and who are witness of the dependency of political allegiances on
resurgent local cultural identities, ethnic and religious, in the post-
Communist world, it is clear that the authority of a state that (however
inevitably compromised by its engagements as an enterprise association)
acts as a custodian of civil association is frailer even than Oakeshott per-
ceived, and cannot long do without the support given it by a common culture.
It may be that what Oakeshott has taken from the liberal tradition about
the authority of the state is not, finally, his most important borrowing from
it. It is his conception of civil association, in which the historical European
inheritance of civil society is cleansed of its theoretical accretions, such as
notions of natural rights, laissez-faire and the minimum state, that is likely
to be his most enduring gift to political philosophy, and it is one that can
46 Thinkers
plausibly be termed liberal. In Oakeshott's writings on education,12 again,
one finds a conception of education that is liberal in the best sense of that
term - a view of education, not as the inculcation of an orthodoxy, the
acquisition of information or the transmission of useful skills, but as initia-
tion into a cultural inheritance in the course of which we learn to become
civilized conversants. If, in very unOakeshottian vein, we were to ask
which element in his thought has most contemporary relevance, it would
perhaps be his idea of a liberal education as an unchartered intellectual
adventure, with no goal beyond its own inner telos - an idea which some of
us may recognize in our pasts, but which is threatened now on every side,
as much by managerial conservatism as by left-wing ideologues.
Of Oakeshott as a philosopher and as a man I have so far said nothing.
As to his philosophy, ably and comprehensively expounded by Paul
Franco 13 (himself a pupil of Timothy Fuller, America's best Oakeshott
scholar), it is his pluralist affirmation of the diversity of modes of discourse
and experience, of moralities as vernacular languages whose nature it is to
be many and divergent, and of the miscellaneity of practice, which no
theory can hope to capture, that embodies his most distinctive contribution
to philosophy. It is only in the later Wittgenstein (whose thought, though
influenced by Spengler, lacks the deep historical consciousness of Oake-
shott's) that we find a comparable critique of traditional rationalism and
contemporary positivism. It is a comment on the current state of academic
philosophy that Oakeshott's contribution to it has gone almost unnoticed.
If one had to express the spirit of Oakeshott' s thought in a single phrase,
one might say that it is a critique of purposefulness. The image of human
life that Oakeshott conveys to us is not that of a problem to be solved or a
situation to be mastered, it is the poetic (and religious) image of our being
lost in a world in which our vocation is to play earnestly and to be earnest
playfully, living without thought of any fmal destination. For those of us
lucky enough to have known the man, Oakeshott's writings will always
evoke the memory of his conversation, in which intellectual passion com-
mingled with a fathomless gaiety, and the dry reasonings of philosophy
became, as in the dialogues of Socrates, dialectical, and at last lyrical.
5 Buchanan on liberty

ThITRODUCTORYREMARKS
It may appear that I utter a truism when I say that the thought of James
Buchanan is a contribution to political economy. What is intended by that
remark, however, is far from truistic: it is that Buchanan's thought is
political economy in the classical sense - the sense in which it was prac-
tised by the Scottish School, and in which it encompasses social philosophy
as well as positive or explanatory economic theory. For, as with Smith,
Ricardo or indeed Marx, Buchanan's thought composes a system of ideas
and comprehends a distinctive perspective on man, government and
society. This is to say that, even without regard to its specific content,
Buchanan's thought is distinguished by its resistance to contemporary
disciplinary fragmentation and has thereby helped to reconstitute political
economy in its classical sense.
Buchanan's system of ideas is an original contribution to political eco-
nomy, synthesizing elements from a diversity of traditions - Wicksellian,
Austrian, and Chicagoan - into a powerful and coherent body of theory. It
is not my brief here to attempt to assess Buchanan's system as a whole - a
task that is in any case well beyond my powers. My brief is the narrower,
but nonetheless fundamental one of explicating and assessing the place of
liberty in Buchanan's work. For, whereas Buchanan's thought diverges
from classical liberalism in ways that are important and instructive, there
can be no doubt that. like classical liberal thought, Buchanan's is animated
by a profound moral concern for the fate of free men and free peoples.
Indeed, if one were to characterize Buchanan's thought as a research
programme in the Lakatosian sense, one would most appropriately charac-
terize it, at any rate in its normative dimension, as an inquiry into the
constitutional preconditions and constraints on individual liberty .
My examination of Buchanan on liberty will have six elements, or
phases. I shall, first, consider Buchanan's indeterminate contractarianism,
48 Thinkers
a form of contractarian political theorizing that has decisive advantages
over the more prescriptive theories of Gauthier and Rawls. Second, I shall
discuss Buchanan's critical constructivism - his version of critical rational-
ism and his critique oflibertarian and Hayekian conceptions of spontaneous
order and institutional evolution. Third, I will examine his subjectivist
conception of economic value, together with his associated criticism of
welfare economics and his general repudiation of notions of efficiency in
resource allocation as the principal functional value of market institutions.
Fourth, I shall investigate the foundational ethical values of Buchanan's
theory of liberty, arguing that the ethical foundation of his system is not
value subjectivism but a species of value individualism. Fifth, I shall
explicate Buchanan's perspectivist epistemology, in terms of which he is
able to defend the utility of the construct of Homo economicus without
attributing to it total explanatory or predictive power. Sixth, and last, I shall
address the most explicitly normative aspect of Buchanan on liberty, when
he tries to specify the conditions under which we may move 'from
Hobbesian despair to Humean hope'.
I shall not here attempt to summarize my conclusions, except in their
most general respects. The proceduralist character of Buchanan's indeter-
minate contractarianism has three sources, in his mitigated scepticism and
critical constructivism and in his value-individualism. This value-
individualism, in tum, is the ethical postulate on which his defence of
liberty stands. The postulate of value-individualism (the content of which I
later try to specify) is a distillation of the Western historic experience of
individuality. For this reason, Buchanan's contractarianism, like that of the
later Rawls, presupposes the cultural inheritance of Western individualism,
with its roots in Christianity and Stoicism. I shall argue that it is this cultural
context of Buchanan's individualism - a context which both shapes and
constrains the liberal individual - that enables him to make the hazardous
passage from Hobbesian despair to Humean hope. In the last context, I shall
argue that even the gossamer-thin veil of ignorance or uncertainty which
Buchanan's contractarian constitutionalism comprehends is unnecessary
when we come to theorize the real-world attempts of people in specific
historic contexts to contract new constitutional settlements and institutions.

Much recent contractarian theorizing suffers from a fundamental in-

coherence. On the one hand, contractarian theorists expouse evaluative
individualism, seeing in the autonomy of the individual the ultimate, foun-
dational postulate of contractarian political philosophy, and they conceive
of agreement under appropriately characterized circumstances as necessary
 Buchanan on liberty 49
and sufficient for legitimate authority. At the same time, most recent
contractarian theorists, notably Rawls and Gauthier, have made a require-
ment of strict determinacy in the upshot of contractarian deliberation a
major part of the agenda of contractarianism as a distinctive method in
political philosophy. In Rawls, indeed, the requirement is for unique deter-
minacy, as well as fixity, in the principles that emerge from the original
position. As I have argued elsewhere, 1 this agenda of determinacy, which
figures promint:ntly in all of Rawls's work, coheres badly with the recogni-
tion, explicit in his later work, that both the telos and the content of
contractarian reasoning arise against the background of a specific, historic-
ally concrete cultural and political tradition - roughly, the liberal tradition
of constitutional democracy, with its underlying matrix of individualist
moral culture. For, if the content of contractarian methods is given by a~
evolving and mutable historical tradition, it is hard to see why the upshot of
the method should be unalterably fixed, nor how it can reasonably be
expected to be uniquely determinate in its central results (such as the
specification of the basic liberties).
There is a basic incoherence in Raw lsian contractarianism, however,
that applies to it in all of its versions, and not only in its later, Deweyan-
historicist variety. This is an incoherence in its attempt to combine a
conception of justice as au fond purely procedural with the requirement that
contractarian method yield principles that are strictly, if not uniquely
determinate solutions to the dilemma of contractarian choice. The fusion of
these irreconcilable demands gives to Rawlsian contractarianism an in-
herent instability, marked by many observers, and commits Rawls to a
member of questionable, arbitrary and often plainly ad hoc manoeuvers
(such as the attribution of extreme risk-aversion to the hypothetical con-
tractors) in the effort to derive the principles (such as the Difference
Principle) on which he has independently settled. A methodological agenda
having these incompatible desiderata also opens his entire enterprise to
rival interpretations, such as the cognitivist and voluntarist interpretations
developed in Sandel's useful critique.2 This ambiguity in the interpretation
of Rawls's theory derives directly from its methodological contradictions,
and has subversive implications for Rawls's attempt to achieve a vantage
point of neutrality in respect of questions of realism in moral epistemology.
The ambiguities and difficulties in Rawls's work I have specified are in
no way intended to undermine Rawls's achievement in reviving con-
tractarian political philosophy as a systematic discipline. They are meant to
underscore the inappropriateness of the requirement of determinacy in
outcome for a methodology that is contractarian and individualist in politi-
cal philosophy. It is here, I believe, that Buchanan adheres more faithfully
than Rawls to the individualist value theory that is the ultimate postulate of
50 Thinkers
contractarian theorizing, when he explicitly disavows any agenda of deter-
minacy for the output of contractarian deliberation. As Buchanan has stated
his rejection of determinacy in canonical fonn:
Any distribution of tax shares generating revenues sufficient to frame the
relevant spending project passes Wicksell's test, provided only that it
meets with general agreement. Analogously, any set of arrangements for
implementing fiscal transfers, in-period, meets the constitutional stage
Wicksellian test provided only that it commands general agreement.
This basic indetenninacy is disturbing to political economists or
philosophers who seek to be able to offer substantive advice, over and
beyond the procedural limits suggested. The constructivist urge to
assume a role as social engineer, to suggest policy refonns that 'should'
or 'should not' be made, independently of any revelation of individuals'
preferences through the political process, has simply proved too strong
for many to resist. The scientific integrity dictated by reliance on indivi-
dualist values has not been a mark of modem political economy. 3
And, as Buchanan has stated of the divergence between his contractarian
perspective and other, more constructivist perspectives:
The whole contractarian exercise remains empty if the dependence of
politically generated results upon the rules that constrain political action
is denied. 4
The superior consistency of Buchanan's indeterminate contractarianism
gives it a decisive advantage over others, whether (as in Rawls) they seek
to specify an egalitarian upshot constrained by a classical liberal priorit-
izing of liberty, or whether, as in Narveson's recent work,5 they seek to
ground a minimal state, devoid of distributional functions, in contractarian
reasonings. Buchanan's approach is at once more internally rigorous and at
the same time less methodologically hubristic, in leaving the upshot of
contractarian deliberation to practice, as constrained by appropriate rules.

CRITICAL CONSTRUCTIVISM
Buchanan's approach resists the uncritical constructivism, with its implied
posture of godlike omniscience, that animates much in current political eco-
nomy and political philosophy, even in some of their contractarian varieties. It
differs from these uncritical constructivist approaches in leaving to practice the
upshot of contractarian choice at the post-constitutionallevel. It differs from
them, also, in working with a veil of ignorance that is gossamer-thin,6 screen-
ing out only knowledge of the contractors's final social state, and in taking the
status quo distribution of assets as its point of departure.
 Buchanan on liberty S 1
Buchanan rejects uncritical constructivism. even in the forms in which it
animates some contractarian theorizings, for the reasons I have given, and
to this extent he shares with Hayek a common conception of the species of
rationalism - ultimately Cartesian in inspiration, perhaps - that he wishes
to reject. Buchanan's evaluation of constructivism differs from Hayek's,
however, in several essential and fundamental respects. In the first place, as
an economist versed in public goods theory, the Prisoner's Dilemma and
free-rider problems, Buchanan perceives that in the absence of mutually
agreed and enforced constraints there is no guarantee, and indeed no
plausible prospect, of consistently cooperative behaviour. The spontaneous
coordination of human activities, manifest in the market-place, for
example, is not a primordial phenomenon; it is an artefact of legal and
political institutions that protect private property and facilitate contractual
exchange. Buchanan recognizes, in a way that Hayek only rarely does, that
there is little or nothing that is spontaneous in the social orders from which
we benefit; they are institutions, built up over the generations, often with
great difficulty, and they stand in need of recurrent repair and reform. As a
good scholar of Hobbes, Buchanan is more than sceptical about the spon-
taneity of social order: without denying that cooperative conventions may
evolve without design or intention, he is emphatic that social interaction in
the absence of clearly demarcated property rights formulated and enforced
by a common power is likely to result in a spontaneous disorder that is
wasteful and harmful to all. This point may be made in another way.
Buchanan recognizes that the undesigned coordination of the market-place
presupposes the public goods of peace and law - goods that, because of
their public character, can be provided only by the state. In this he is at one
with Hobbes, who rightly saw peace as the necessary condition, not only of
commodious living, but of liberty.
Buchanan's difference from Hayek may be stated in another, and per-
haps more radical way. Buchanan refuses to follow Hayek in generalizing
the observed phenonemon of undesigned coordination in market exchange
and extending and developing that observation into a theory of cultural
evolution or institutional Darwinism. For Buchanan, as for myself, such
generalization and extension are misguided, since they neglect the institu-
tional conditions under which alone an undesigned order that is beneficent
may be expected to emerge. The undesigned coordination that manifests
itself as order in the market does so in virtue of specific tendencies in
markets, and it is beneficent because voluntary exchange is not typically
zero-sum. Once the institutional framework to the market is absent, how-
ever, we have no reason whatever for supposing that competition - be it
competition of individuals for resources, or tradition or practice, or of states
- should be beneficent. To argue that it is so, is to assume or presuppose an
52 Thinkers
entire body of functionalist social theory that is at odds with the methodo-
logical individualism Hayek elsewhere expouses. It is to presuppose both a
unit of selection, and a mechanism for selection, akin to those in Darwinian
evolutionary theory, which have no real analogies in social theory. With
Hayek, Buchanan avows a mitigated scepticism, akin to Hume, and with
Hayek's friend Popper, he conceives of reason as in essence critical and
fallibilist rather than apodictic or axiomatic. Unlike Hayek (or Polanyi),
Buchanan is unwilling to allow that our historical inheritance of received
institutions should be exempt from critical appraisal. In fact, he maintains,
we have no option, we are bound, given the desuetude of many of our
institutions, and their colonization by collectivist interests, to subject them
to critical assessment. Only an historical theodicy could give man's random
walk in historical space the aspect of an orderly progression, when what is
disclosed to the critical eye are rare episodes of freedom surrounded by
long stretches of barbarism and tyranny. None of this is meant (by me or
Buchanan) to deny Hayek's massive achievement in illuminating the epis-
temic role of the market in generating and utilizing dispersed (and often
tacit) knowledge. It is simply to assert that we cannot rely uncritically on
the tacit wisdom of inherited institutions, but must instead try to reform
them with the aid of our best theorizing, when these institutions are in
disrepair, and perhaps embody the tacit error or ignorance of generations of
collecti vism.
These methodological objections to the Hayekian synthetic philosophy,
and to notions of spontaneous social order, have pertinent applications to
contemporary political developments. In the relations between market capi-
talist orders and totalitarian command economies, there are game-theoretic
predictions, and ample historical evidences, that competition may not pro-
mote market institutions, but may instead result in stable relations of
parasitism which serve to reproduce, and even to strengthen, the totalitarian
command economies. The limitations of institutional Darwinism are, how-
ever, perhaps most clearly evident in current developments within the
Communist blocs. For there - in Poland, Hungary, the Baltic States, China,
and perhaps in the Soviet Union - the inherited (or imposed) institutions of
central planning are collapsing without the evolutionary emergence of
successor institutions. True, in Poland and elsewhere, elements of civil
society that were formerly repressed by totalitarian bureaucracy have
successfully reasserted themselves in a spontaneous fashion, but the legal,
political and economic framework of civil society cannot there emerge as a
property of institutional evolution; it must be constructed, or else re-
constructed in the wake of revolution. The would-be covenantors of the
Soviet bloc countries are, in some respects, akin to the covenantors in
Hobbes's state of nature, save that (as Z. Rau has note(7) it is a state of
 Buchanan on liberty 53
enslavement that they seek to escape from. (I have myself argued elsewhere
that the dilemma of the subjects of a totalitarian order resembles the
Hobbesian dilemma in that one of the constitutive features of totali-
tarianism, which it shares with the Hobbesian state of nature, is lawless-
ness8). Hayekian institutional Darwinism has, then, no application to
historical contexts - like that of the Soviet bloc states - where extant
institutions are bankrupt and lack even latent successors. Admittedly, also,
public choice dilemmas may thwart the construction of new institutions, as
conflicts of interest emerge within the opposition and in the nascent civil
society; but it is still only the project of constructing a new social contract,
as embodied in new or reconstructed institutions, that offers hope of exit
from totalitarianism to a stable post-totalitarian order. In such circum-
stances, we cannot intrust ourselves to the providential designs of Burkean
or Hayekian institutional evolution. Rather, armed with the best theory we
can find, we must wager on a new social contract.

ECONOMICS, SUBJECTIVISM AND VALUE

The undesigned coordination of the market, together with its consequent
contribution to human well-being, are to be explained by the fact that
voluntary exchange - exchange, that is to say, conducted against a back-
ground of institutions defining and protecting private property rights and
enforcing the terms of contractual agreements - is typically positive-sum.
Without these background institutions, which are the matrix of the spon-
taneous order we observe in the market process, there is no reason to
suppose that competition, or exchange, should be other than zero-sum, or
even negative-sum. Indeed, it is precisely the character of the pre-legal state
of nature, in which most resources apart from those consumed for sub-
sistence are devoted to self-defence or to mutual predation, that such
exchange as occurs need not be, and often is not, a benefit to both parties.
In a market order that is constituted by the legal framework I have men-
tioned, on the other hand, there is a very strong presumption, grounded both
in theory and history, that the constantly iterated act of exchange will
benefit all parties to it.
To say this is to say that the market economy can be justified by its
contribution to the well-being of the individuals who participate in it. It is
of the utmost importance, both in my own judgement and in Buchanan's, as
I understand him, that this defence of the market in terms of its contribution
to human well-being not be conflated with a consequentialist or utilitarian
conception of the market as maximizer of collective or aggregate welfare.
We should resist this conflation, according to Buchanan, for several
reasons. In the first place, the idea of aggregate or collective well-being
54 Thinkers
involves a fallacy of conceptual realism - it confuses what is at best an
heuristic fiction with reality, in this case the reality of discrete human
individuals, each with their own distinctive desires, preferences and pro-
jects. This suggests a second reason for resisting the move from a well-
being-based justification of the market to an aggregative consequentialist
justification - which is the impossibility of aggregating individual pre-
ferences and satisfactions. Whereas comparative, on-balance judgements
of social welfare may be feasible in limiting cases, Buchanan rightly insists
that the pretensions of welfare economics founder on the incommensur-
ability of the various components of the well-being of persons. To say this
is not merely to reaffirm the classical (and in general insoluble) problems
of making well-founded interpersonal utility comparisons. It is also to note
the incommensurabilities that may arise among the components of even a
single person's well-being.
Aggregative, consequentalist justifications of the market have had, as
Buchanan has constantly reminded us, baleful results for theory and policy.
In theory, they have sponsored the conception of economic science as
having to do with wealth-maximization, or maximal allocative efficiency,
or productivity, reinforcing the Communistic fiction that views the catal-
laxy of the market as akin to an hierarchical organization or household. In
other words, aggregative conceptions have reproduced the disabling equa-
tion of market exchange with conceptions of maximization or economizing
that weakened the foundations of the classical economics of the Scottish
School. It is for this reason, among others, that Buchanan is an unequivocal
subjectivist as to economic value. As against the classical economists,
including Marx, Buchanan insists that economic value is not an inherent
property of any sort, it is subjective - an aspect of the choices and experi-
ences of individuals. And, since these choices encompass a multitude of
incommensurables, there is no way of summing up subjective economic
value into any judgement of collective or aggregate welfare.
In its uses in welfare economics, the aggregative conception has had a
damaging effect on policy. Like the fiction of perfect competition, the
aggregative models of welfare economics have encouraged policy-makers
to consider intervention and regulation wherever markets depart from some
theoretical norm of efficiency or maximization. They have thereby re-
inforced the vulgar academic perception of markets as impersonal pro-
cesses, sometimes chaotic or defective but having a logic of their own, and
of political processes as highly voluntaristic activities, barely constrained
by economic interests or sjtuational logic. Welfare economics has thus
contributed to the neglect of the economics of bureaucracy, government
and political life, and has tended to underestimate the importance of
government failure. In the most general terms, it has seen market failure as
 Buchanan on liberty 55
an occasion for government intervention rather than for the refonn of
market institutions (by, for example, the creation of property rights where
none have hitherto existed). The aggregative method of macroeconomics,
and, in particular, of welfare economics, has an inherent collectivist bias, in
that it seeks to enhance imaginary maxima rather than to refonn or extend
markets so as better to enable individuals to achieve their diverse and
incommensurable goals. ~uchanan has perfonned a signal service in setting
theory and policy on the very different individualist agenda of refonning
rules and institutions the better to facilitate the conditions of voluntary
exchange.

VALUE INDIVIDUALISM
The result of Buchanan's subjectivism about economic value is that, taking
market exchange as a system or process of reiterated transactions rather
than as any single act, the efficiency of market institutions is redefined in
procedural tenns. It is the voluntaristic character of market exchanges, and
not any assumptions about maximization or equilibrium, that assure their
beneficent contribution to the human good. It is here that I advance,
tentatively and provisionally, an interpretation of Buchanan's nonnative
theory that diverges somewhat from his often repeated affirmation of a
general value-subjectivism. Such subjectivism, with its distinguished
Hobbesian and Humean pedigree, is eminently apposite, not only in eco-
nomic theory, but also in political theory - at least, in the latter, as a
counterbalance to the conception of politics as an activity devoted to the
pursuit of truth on the model of truth-seeking and inquiry in the natural
sciences. It is more questionable whether a Hobbesian value-subjectivism
is presupposed or implied by Buchanan's political thought. Plausibly, the
foundational value is not that of preference-satisfaction (in which the good
is what is desired), but individual autonomy, conceived as an intrinsic good.
This, Kantian or Spinozistic, interpretation of Buchanan's system is
entirely consilient with subjectivism as to economic value. It is also com-
patible with a version of realism or objectivism as to the status of the value
of autonomy - although, as I shall later argue, this value is best interpreted
in Buchanan's thought as a distillate of our cultural tradition. My principal
point here, however, is that there is a position in normative theory, value-
individualism, that is distinct from value-subjectivism but consistent with a
subjective view of economic welfare. This individualist view, which
regards persons as themselves ultimate values (and not, as in utilitarianism,
passive receptacles of impersonal value) allows for a defence of the market
in tenns of its contribution to the autonomy and well-being of individuals.
And it does so without making any concession to the aggregative fiction of
56 Thinkers
consequentalism. For as in Hume, where this argument has its classical
formulation, the market benefits each, and thereby all - not, as in Bentham,
benefitting a spurious collectively and so (fallaciously) each.
In this account, the value of preference-satisfaction is not (as in some
varieties of value-subjectivism) ultimate, but derivative. It derives from the
value of autonomy, or, more precisely perhaps, of the autonomous person.
Buchanan favours market freedom under the rule of law, ultimately, as I
interpret him, not because preferences are most often satisfied under such
institutions, but because such institutions best protect and respect auto-
nomous individuals. That this distinction is not a trifling one is seen when
we consider those circumstances when populations appear not to value the
autonomy they may enjoy as individuals, and instead prefer the security
they expect (and regularly fail to receive) from arbitrary and unconstrained
government. It will be a later argument of mine that the normative basis of
Buchanan's political thought presupposes an inheritance of individualist
moral culture that may be on the wane. At this stage, I wish only to
comment that, whereas value-individualism may be affirmed in a way that
is neutral as to questions of moral epistemology, it is for that reason
compatible with a form of moral realism of the sort that Buchanan has
rightly rejected in other contexts.

EPISTEMOLOGICAL PERSPECTIVISM
Buchanan's resistance to macroeconomic modelling, and his extension of
microeconomic theory to political and other behaviours and institutions,
involves his invoking as an explanatory and predictive construct the con-
ception of Homo economicus. In terms of the methodology of the Virginia
School of Public Choice, of which he is the principal expositor, this entails
the adoption of a principle of parity of motivation in the economic and the
political realms, where 'economic' designates the sphere of explicit market
exchanges. Another way of expessing this methodological move is to say
that using the model of Homo economicus results in the development of an
economic theory of human behaviour in bureaucratic and political institu-
tions. It comprehends specifying and theorizing the incentive structures,
time preferences and goals of agents within these 'non-economic' contexts
and institutions, employing the postulate that their motives and deli-
berations differ in no essential respects from those of participants in regular
market exchanges - buyers, sellers, entrepreneurs, and so forth.
The result of this methodological move is to demystify the political
process and to theorize it in terms of the interests of the individuals and
groups (often acting collusively) who engage in it. Especially when applied
in conjunction with the theory of rent-seeking to such phenomena as
 Buchanan on liberty 57
industrial regulation and occupational licensure, this methodology has had
considerable explanatory and predictive power and has come closer than
any other to establishing a genuine positive political science. Having said
this, it is important to note that Buchanan has never subscribed to the
economic imperialism of those, such as Becker, who claim that the entirety
of human behaviour can be captured in the terms of economic theory.
Buchanan's differences from this approach are, as I understand his work,
severalfold. It is in the first place unclear from his writings whether he
accepts the specific content of Becker's methodology (such as the fixity or
stability of preferences) or whether he would endorse Becker's claim that
apparently non-deliberative behaviour, such as unreflective rule-following,
can always be explained as a strategic response to the information costs of
calculation. Buchanan's many sympathetic references to Austrian eco-
nomics suggest that the positivistic model of human action, and indeed of
economic theory, embodied in this Beckerite research programme would be
uncongenial to him.
The real difference between Buchanan's methodological commitment to
Homo economicus and Becker's is far more fundamental: it is episte-
mological. For, unlike Becker, Buchanan has explicitly disavowed the claim
that the content of Homo economicus captures the whole of human
behaviour, even in its political and catallactic dimensions. For Buchanan, the
public choice perspective is a window on the human world, a lens through
which much may be perceived and understood that which would otherwise
remain invisible or unintelligible, but it is not a mirror of social reality,
aiming to reflect it in its totality and within a single, holistic explanatory
scheme. The epistemological position or standpoint which best characterizes
Buchanan's view of public choice is perspectivism, a term often employed
in the interpretation of Nietzsche's theory of knowledge. Within the analy-
tical tradition in epistemology, the closest approximation to Nietzscheam
perspectivism is perhaps found in Nelson Goodman's irrealism,9 although
anticipations may be discerned in the pragmatist epidermologies of William
James, C. S. Pierce and indeed Quine. What are the central elements of
perspectivism in epistemology, and how do they affect the scope and limits
of the utility of Homo economicus in public choice theory?
Using the book The Reason of Rules,tO which Buchanan co-authored
with Geoffrey Brennan, as our principal authority, we may infer that the
per- spectivist epistemology of public choice has three important elements.
There is, frrst, the thesis that theory - clearly in the human sciences, but also
perhaps in the natural sciences - is underdetermined by evidence so that the
idea of science as a mirror of nature, a recorder of objective facts, them-
selves identified in a theory-independent fashion, is rejected. This under-
determination thesis is, in other words, a rejection of naive empiricism, or
58 Thinkers
at any rate, a statement of its severe limitations. There is, second, the thesis
that in theorizing the human world we may profitably have recourse to a
diversity of perspectives, none of which captures everything we wish to
understand. Thus, in social theory, the construct of Homo economicus may
be complemented by the Hayekian conception of man as at bottom a
rule-follower, and not a calculator, reckoner, or maximizer. This pluralist
aspect of perspectivist el>istemology may allow for fruitful research into
questions having to do with the pre-reflective or non-calculational condi-
tions of market exchange in Western societies, and it may assist us to
conceive of the possibility that in some societies (perhaps Japan) straight-
forward maximization may be an exceptional incident in economic life that
is otherwise comprehensively governed by cultural norms. The third ele-
ment of a perspectivist epistemology in public choice would be in the
proposition that our choice of a methodological perspective will be
informed, at least in part, by pragmatic concerns having to do with pre-
diction and control. In the specific context of public choice theorizing of
government failure, our programmatic concerns may also be ethical, nor-
mative concerns, in that we deploy the construct of Homo economicus to
attempt to understand, and so to predict and through constitutional reform
to control, the behaviours and incentive structures that recurrently give rise
to government failure. In its combination of the rejection of naive empiri-
cism, pluralism and methodological pragmatism, I see the perspectivist
epistemology of Buchanan as a tertium quid or via media between whole-
sale relativism or instrumentalism and the full-blown realism of Popper.
I do not aim here to attempt a philosophical assessment of perspectivist
epistemology, although I find it an attractive and plausible position, especi-
ally in the human sciences. My concern is rather to note that, because it is
advanced in this perspectivist fashion, and lacks the aprioristic character of
other research programmes for the extension of economic explanation into
'non-economic' areas of life, Buchanan's version of public choice in no
way commits him to denying the existence, or indeed the theorizability, of
phenonema such as Kantian principled behaviour, patriotism, moral vision
or altruism - even in political contexts. Indeed, it may be that the kind of
constitutional revolution needed in order to restore limited government
depends upon the existence, in leaders and segments of the population, of
motivations and commitments of sorts that Homo economicus does not
possess and which public choice does not aim to theorize. It is to the
prescriptive dimension, or implication, of Buchanan's thought that I now
tum, when I consider the necessary preconditions of a theory of individual
liberty in an historical context of inordinate government operating under
the imperatives of unconstrained majoritarian democracy.
 Buchanan Oil liberty 59
FROM HOBBESIAN DESPAIR TO BUMEAN HOPE
The nonnative telos of Buchanan's work: is the renovation of limited
democratic government, constrained in its taxing, transfer and spending
powers by a stiff regime of conventional rules. He has not engaged in
advocacy of specifIC measures, but instead regards this nonnative goal as at
once animating, and at the same time a natural implication, of his work in
developing constitutional political economy as a distinct discipline. Con-
stitutional political economy is distinctive, in Buchanan's conception of it,
not only in virtue of its methodology, but also because as a practitioner of
it Buchanan has consistently repudiated the role, so commonly assumed by
economists, of policy adviser to supposedly benevolent governments with
large directionary powers. Buchanan's role has been that of communi-
cating the importance of the rule-governed polity, of constitutionalism in its
Madisonian variety, and of showing its natural harmony with constitutional
political economy as a theoretical perspective.
What is Buchanan's account of the means whereby we move from the
Leviathan of unconstrained (or inadequately constrained) majoritarian
democracy to limited government? I do not mean to ask what are the
political steps to this end, but rather how that passage is to be theorized.
Several points are salient here. In the first place, Buchanan is clear that we
must take the status-quo distribution of assets as our starting point (Where
else, after all, could we start?) Working, as do Gauthier and Rawls, with
hypothetical baselines of equality or Lockean rights imparts into contrac-
tarian method non-procedural, or pre-procedural standards of justice which
are difficult, if not impossible, to defend in philosophical terms. This is to
say that, in adopting a status-quo baseline, Buchanan, like Hume, accepts
the historic distribution of titles without making fruitless inquiries into its
history or justice as appraised by controversial, questionable and often
conflicting standards. But second, Buchanan applies this rejection of non-
procedural conceptions of justice by arguing for a redistributional dis-
mantlement of the currently overextended state which satisfies the criterion
of Pareto-optimality: no one is left worse off. This is a most important
feature of his approach, since it presupposes the vital, but often neglected
point that redistribution is not always a zero-sum transaction. It need not be
so, when the redistribution effected involves dismantling most existing
transfer payments schemes, along with their bureaucracies, and disbursing
the resources so released to the public. This redistributional Parietianism is
interesting and important for many reasons, not least of which is that the
buying-off of rent-seekers which it incorporates is now an explicit part of
the policy of privatization and marketization now under way in Poland,
Czechoslovakia, Hungary and elsewhere in the post-Communist world.
60 Thinkers
On the theoretical side, Buchanan's account of the renegotiation of
entitlements incorporates the methodological device of a thin veil of ignor-
ance, screening out only knowledge of one's final social situation. I will
now suggest as a further development of Buchanan's thought, consistent
with his adoption of a status-quo baseline, that the device of the veil of
ignorance - even in the thin form he adopts along with other contractarian
theorists such as Kavb - is unnecessary in his system, and possibly
indefensible. For, if one is allowed knowledge of one's existing assets and
position, what is the rationale for being denied information as to one's
ultimate situation? A different way of expressing this thought is that the
actual risks of real-world renegotiation of property titles are what will
inform the contractors, and not any hypothetical uncertainty generated by
the thought experiment of the veil of ignorance. My suggestion is also
motivated by the fact that hypothetical circumstances cannot - despite all
that Rawls and Gauthier have to say - generate reasons for action for
persons in their actual circumstances. As Jeffrey Paul has persuasively
argued,l1 there is an is-ought gap in all hypothetical contractarianism for
just this reason. A natural development of Buchanan's thought, accord-
ingly, is to eliminate from it its last hypothetical or counterfactual com-
ponent, the veil of ignorance, and reconstitute his contractarianism as
entirely of the actual-contract variety .12 His public choice theorizing would
then fuse with his normative philosophy in providing a most powerful tool
for those in the real world who are seeking to achieve a new social contract
or constitutional settlement.
We come now to the crux of our entire inquiry: what is the place of liberty
in Buchanan's indeterminate contractarianism? Unlike Rawls's, Buchanan's
contractarian method does not privilege liberty from the start, and it does not
issue in a determinate list of basic liberties that are fixed and unalterable. Nor
is this altogether surprising, given the Hobbesian pedigree of his political
philosophy. Again, and for the same reason, Buchanan is clear that, in any
plausible real-world situation, contractarian choice would not (contrary to
Narveson) yield a Nozickian or Spencerian minimum state, and would not
necessarily issue in unencumbered Lockean rights at every point. Indeed, he
has argued forcefully that a limited government would surely have some
redistributional functions in providing for reasonable equality of opportunity
and inhibiting concentrations of wealth that might prove inimical to liberty.
In Buchanan's system, then, liberty is not given the apodictive priority it has
in Kantian-inspired contractarianism, nor is it the case that the role of the
state is defined by the protection of Lockean rights.
Nevertheless, it is plain that Buchanan's contractarianism is bound to
issue in most real-world contexts in the enhancement and protection of
individual liberty, and will in virtually all contexts protect central economic
 Buchanan on liberty 61
liberties. It does so in virtue of his opposition to aggregative theorizing and
his exploitation of the classical insight that voluntary exchange is mutually
beneficial. It does so, again, because his defence of the market economy is
in terms of its contribution to human autonomy rather than to any abstract
conception of general well-being or collective welfare. And it does so,
finally, in virtue of his insight that, provided there is a suitable framework
of law, the undesigned coordination of the market is superior to any that
could be generated by command or coercion.
It is true, however, that, so far as I can see, Buchanan's contractarianism
cannot give universal protection to the personal or civil liberties that are
central in Western individualist tradition. This apparent cultural limitation
of his method seems to me to be both inevitable and not undesirable. It is
far from self-evident, and sometimes plainly false, that the institutions and
civil liberties of even limited democratic government are always and every-
where appropriate and defensible. It may well be, for example, that in some
circumstances the transition from a totalitarian command economy can be
negotiated, and a stable post-totalitarian civil society achieved, only under
the aegis of an authoritarian state that protects market freedoms under a rule
of law, but in which the full panopoly of democratic and civil liberties is
absent. It is a strength, rather than a weakness, of Buchanan's system that it
can accommodate this possibility.
The dependency of Buchanan's social philosophy on the moral culture
of Western individualism seems to me, also, at once unavoidable and not
necessarily a weakness in it. Some aspects of Buchanan's thought, as of
Hobbes's, may have universal or near-universal scope and validity, in so far
as they concern the status of peace as a condition of any other political
good. Beyond the limits of Western individualist cultures, or of cultures
undergoing a process of modernization that involves assimilating Western
values, however, there is no guarantee that contractarian choice will issue
in anything more than the legal and economic framework of a market order.
We may go further. We may need to confront the disturbing prospect that,
in our own cultures, the morality of individuality has been so weakened by
generations of collectivism and governmental profligacy that contractarian
choice may not protect all of the liberties we associate with our tradition,
nor restore all of those currently curbed by interventionism and collec-
tivism. This disquieting possibility seems to me an inexorable result of the
virtue of Buchanan's contractarianism - its indeterminacy as to outcome.
Within Buchanan's system of ideas, there are resources which sustain
hope that a restriction of limited government is a possibility. Like Oake-
shott, \J Buchanan recognizes that, stated it its starkest form, wherein men
are theorized wholly in Homo economicus or egoistic terms, the Hobbesian
problem is insoluble. The problem is solved in Hobbes, only by the deus ex
62 Thinkers
machina of the proud or gallant man, who chooses to exit from the indig-
nities of the state of nature, and becomes flfSt performer in the social
contract from a sense of honour. Another solution is suggested by Hume,
who (while acknowledging all the realities of conflict theorized by Hobbes)
perceives that human beings are not the solitary predators hypothesized in
the state of nature, but social beings, born in families, bound together by
sexual and familial love, speakers of common languages in which common
forms of life are expressed. 14 Against this background conception of human
nature, Hume is able to illuminate the emergence of conventional norms
which allow men to act cooperatively and to mutual benefit. Such norms or
conventions are the submerged structure of any viable society, and they
cohere to form a common culture, of which legal and conventional rules are
the visible part. Where the common culture has been desolated by genera-
tions of collectivism, as perhaps in some of the countries of Latin America
(Brazil, Argentina), there can be little hope that a constitutional revolution
in the direction of limited government could be successful or stable. (Per-
haps the best that can be hoped for in those circumstances is that the
government will remain weak and ineffectual and allow the real business of
society to be transacted in the informal economy.) This is to say that, in
some historical contexts, the ruin of civil society is irreversible, and the
Hobbesian dilemma as insoluble as it is in the starkest account of the
Hobbesian state of nature.
In the Western constitutional democracies, with all their tendencies to
omnicompetent government, we are far from such a nemesis. Along with
the tacit error ingrained in the population by half a century or more of
overextended government, there is still a common stock of individualist
moral culture, which is the capital or patrimony, the historic inheritance, on
which a constitutional revolution in the direction of individual liberty under
the rule of law of a limited government can draw. (This common cultural
inheritance is possessed far more securely by the populace than by the
intellectual elites, whose work has not been to theorize but more often to
corrode or destroy it.) The forms in which we may hope that governments
withdraw from their current inordinacy will necessarily be various, as will
the new constitutional settlements that emerge. In states with written con-
stitutions, they will involve constitutional amendments (such as an amend-
ment mandating a balanced budget) via a constitutional convention. In
states, such as New Zealand and the United Kingdom, whose constitutions
are unfixed and unwritten, the return to limited government will involve a
reforging of the tacit economic constitution overturned by the disciples of
Keynes against the background of a politically irreversible shift of
resources from government to civil society. In the post-Communist blocs,
where the inordinacies of unlimited government have been most
 Buchanan on liberty 63
catastrophic, we may witness the inspiring spectacle of civil society
triumphing over totalitarian bureaucracy, and the reinstitution of a civil
society under the aegis of a social contract and a constitutional settlement
that guarantees market freedom, private property and civil liberty .

CONCLUDING REMARKS
None of these developments has the slightest degree of inevitability. In the
real world of insuperable unknowledge (to use a Shacklean neologism),15
the future is always dark and hazardous. At no point in his work has
Buchanan engaged in the hubristic exercise of forecasting the fate of
liberty. But if liberty has a future, it will have been fortified by Buchanan's
work. For the final message of Buchanan's thought, as I have interpreted it,
is that if we wish to preserve the precious heritage of Western individuality,
we are bound to engage in the project of theorizing the world as it is,
without illusion or groundless hope, even as Buchanan's teacher, Knight,
did in his writings. And then, armed with the theoretical understanding we
have achieved, we must wager on liberty by seeking, in all the vicissitudes
and contingencies of practice, a new constitutional settlement on limited
government in which free men and women may live in voluntary asso-
ciation - in liberty under the rule of law. 16
6 Berlin's agonistic liberalism

It is often said of Izaiah Berlin - not least, by Berlin himself - that he

abandoned philosophy for intellectual history, when (after a conversation
with the logician, H.M. Sheffer) he became convinced that he could not
make a truly significant contribution to the two areas of philosophy, logic
and psychology, which Sheffer has persuaded him would henceforth be at
the centre of the discipline. There is, no doubt, something to be said for this
common view. That Berlin has made an enduring, and thoroughly original
contribution to the history of ideas is not in any question. His essays in
intellectual history, brought together and published in five volumes (so far)
by his editor, Dr Henry Hardy of Wolfson College, Oxford, develop an
interpretation of modem European thought that is novel and profound. In
his work on the history of ideas, Berlin has exhibited a gift of imaginative
empathy with thinkers whose view of the world is deeply alien to his own
- thinkers such as de Maistre and Sorel, for example - in which he has no
peer among his contemporaries, and which is, perhaps, comparable only
with the imaginative clairvoyance whereby Keirkegaard was able to enter
into experiences and forms of life he had never himself known. More
particularly, our understanding of our intellectual inheritance has been
altered, and altered irreversibly, by Berlin's work on the thinkers of what
he calls the Counter-Enlightenment - those Romantics, fideists, and
solitaries, such as Vico and Herder, who broke into pieces that system of
ideas, rationalist and universalist, transmitted from the ancient to the
modem world, and embodied in the French Enlightenment, that formed the
bedrock of the Western tradition.
Endorsed as it has been by Berlin himself, this common interpretation of
his work seems to me nevertheless mistaken. For Berlin's essays in intellec-
tual history make a contribution to moral and political philosophy, begun in
the essays on liberty, that is of the first importance, subverting as it does the
ruling orthodoxies in those subjects. and raising an interrogation mark over
the very foundations of the Western tradition. The idea that animates all of
 Berlin's agonistic liberalism 65
Berlin's work, both in intellectual history and in political theory, is one
which, if true, as I take it to be, strikes a death-blow to the central, classical
Western tradition - and, it must be added, to the project of the Enlighten-
ment. Berlin's master idea is that ultimate values are objective and know-
able, but they are many, they often come into conflict with one another and
are uncombinable in a single human being or a single society, and that in
many of such conflicts there is no overarching standard whereby the
competing claims of such ultimate values are rationally arbitrable. Con-
flicts among such values are among incommensurables, and the choices we
make among them are radical and tragic choices. There is, then, no
summum bonum or logos, no Aristotelian mean or Platonic form of the
good, no perfect form of human life. which we may never achieve but
towards which we may struggle. no measuring rod on which different forms
of human life encompassing different and uncombinable goods can be
ranked. This assertion of the variety and incommensurability of the goods
of human life is not. it is worth noting. the Augustinian thesis that human
life is imperfect. and imperfectible: it is the thesis that the very idea of
perfection is incoherent. Nor are the conflicts among goods illuminated by
Berlin's value-pluralism occasioned. principally or merely. by such con-
tingencies as the brevity of human life. or the scarcity of worldly resources:
they are to be accounted for by the very natures of the goods themselves.
The incommensurabilities invoked by Berlin. and by which the notion of
perfection is destroyed, are what Joseph Raz - in his The Morality of
Freedom, 1 the only major work in political philosophy thus far in which
Berlin's pluralism is developed and applied - has termed constitutive
incommensurabilities. Consider an example from the arts. It makes no
sense to try to rank the excellence of the plays of Shakespeare, say. against
that of the French classicists, Corneille and Racine, or to put the beauty of
the Parthenon in the balance with that of Notre Dame. We lack the scales
on which these goods might be weighed. For this reason. though there may
be improvement or decline within specific artistic traditions, there can be
no progress in the arts that spans divergent and incommensurable traditions.
On Berlin's view. the same is true in ethics. The virtues of the medieval
Christian ruler cannon coexist, in one man, with the virtu of the Renais-
sance prince; nor can the virtues recognized in Homeric Greece flourish in
harmony with those of Whig England. In our own day, the virtues of a
devoted family man are incompatible with those of the man devoted to
pleasure. Within cultures, as within individual lives, we confront an
ineradicable moral scarcity, veiled or denied in most philosophies, which it
is our fate as humans to endure.
The distinctiveness and radicalism of Berlin's species of objective
pluralism are easily missed. Its distinctiveness was missed by Leo Strauss
66 Thinkers
when, with characteristic obtuseness and perversity, he condemned Berlin
as a relativist for whom all values were culture-specific and, in the end,
sUbjective. Throughout his writings, Berlin has constantly stressed that,
though their embodiments in specific forms of life will vary across cultures,
ultimate values are objective and universal - as are conflicts among them.
Berlin's variety of pluralism is a species of value realism, not of scepticism,
subjectivism or relativism. As Berlin puts it in his most recent collection of
essays, The Crooked Timber of Humanity: Chapters in the History ofIdeas:
Incompatible these (human) ends may be; but their variety cannot be
unlimited, for the nature of men, however various and subject to change,
must possess some generic character if it is to be called human at alP
Berlin's affirmation of generically human values that, despite their differ-
ent embodiments in different cultures, are universal, gives the lie to
Michael Ignatiefr s claim, made in his otherwise powerful contribution to
the current collection, that 'When Berlin speaks of universal values, he
means European values'. This interpretation, though it may indeed have
some slight textual support, runs flatly counter to the realist orientation of
Berlin's theory of value, and runs the risk of banalizing Berlin's standpoint
as in the end a variant of relativism. Everything suggests that Berlin's thesis
of value-incommensurability is meant to have universal force and that it is,
if anything, especially subversive of distinctively Western traditions of
rationalism and monism.
The implications of Berlin's value-pluralism for political philosophy
have gone curiously unnoticed by most, partly, no doubt, because they
undermine so much in recent liberal thought. It is a key feature of Berlin's
argument in his seminal 'Two concepts of Iiberty'3 that, just as there are
conflicts among ultimate values that are incommensurable - between
liberty and equality, equality and general welfare, say - so there are
conflicts no less radical within liberty itself - conflicts among distinct
liberties having incommensurable value. When the liberty of privacy com-
petes with freedom of information, a trade-off must be made and a balance
struck; but there is no comprehensive theory - no theory of the sort John
Stuart Mill thought he possessed when he sought to found the 'one very
simple principle' of On Liberty4 in the requirements of general utility - by
which such conflicts among liberties might be arbitrated. This is to say that
the legalist or constitutionalist project, which animates so much in recent
Anglo-American theorizing, of specifying a unique set of rights or basic
liberties that dovetail in an harmonious compossibility, the project - which
finds a clear expression in the work of Rawls, late and early - is grounded
in an illusion. (Is it subversion of this illusion that explains the fact,
otherwise extraordinary, that as yet no book-length study of Berlin's
 Berlin's agonistic liberalism 67
thought has been published?) Among liberties, as among equalities, there
may be, and are, radical choices to be made; gains made in one class of
liberties may be at the expense of losses in others; and there is no theory or
overarching principle whereby these hard choices can be made easy, or the
conflicts they express conjured away.
It is, perhaps, in injecting value-pluralism into the ideal of liberty itself,
rather than in its defence of the negative liberty of non-interference against
the claims of the positive liberty of self-mastery, that the true significance
of Berlin's 'Two concepts of liberty' may come to be seen. In insisting in
that essay that the liberty to live as one pleases and the freedom to have a
voice in collective deliberation are two things, not one, and that individual
freedom and political democracy need not, and do not, always complement
one another, Berlin uttered a truth, much against the current of the age, that
remains thoroughly unfashionable and fundamentally important. Again, his
distinction between negative and positive liberty remains a vital one, even
if one believes (as I do myself) that the chief value of freedom from
coercion by others is the contribution it makes to individual autonomy. The
permanent value of that essay may nonetheless lie elsewhere. It may be
found in the fact that. in founding the value of freedom in the opportunity
it gives us of navigating among incommensurable options and forms of
living, Berlin at the same time cuts the ground from under those doctrinal
or fundamentalist liberalisms - the liberalism of Nozick or Hayek no less
than of Rawls or Ackerman - which suppose that the incommensurabilities
of moral and political life, and of liberty itself, can be smoothed away by
the application of some theory. or tamed by some talismanic formula. It is,
perhaps, the chief virtue of a liberal society, as Berlin conceives of it, that
in it conflicts among values, including conflicts among liberties, are
perceived as ultimate truths about the human world, not to be suppressed or
explained away by spurious theorizing. It is in taking its stand on in-
commensurability and radical choice as constitutive features of the human
condition that Berlin's liberalism most differs from the panglossian liberal-
isms that have lately enjoyed an anachronistic revival. Unlike these,
Berlin's is an agonistic liberalism, a stoic liberalism or loss and tragedy. For
that reason alone, if there is any liberalism that is now defensible, it is
Berlin's.
Berlin is a philosopher of history as much as of liberty. His rejection of
historical inevitability is of a piece with his more general repUdiation of
human determinism. In the restatement of the ideas of Vice and Herder, he
has done much to restore plausibility to the idea that there is a mode of
understanding, peculiar to history and perhaps the other humane studies,
which is irreducible to that appropriate in the natural sciences. And he has
given contemporary plausibility to the view that, whereas human beings
68 Thinkers
share in a common fund of capacities and dispositions that constitutes their
nature, among these are the capacity to transform themselves - their needs,
their views of the world and their self-understandings - and the disposition
to constitute for themselves distinct and typically exclusive identities - as
Jews or Arabs, Frenchmen or Englishmen, and so on. It is here that a crux
emerges in Berlin's thought. It was a cardinal tenet of the Enlightenment
that, whatever cultural v'lriety the future of mankind would encompass, it
was reasonable to expect convergence on a universal civilization under-
girded by a shared, rational morality. If the upshot of Berlin's value-
pluralism is to undermine the idea of a rational morality, the effect of his
work on nationalism is to jettison the prospect that a time may come when
men's dominant allegiance is to the norms of a universal civilization. The
crux in Berlin's thought goes yet deeper than this. It is in a tension between
the idea of a common human nature and the idea of human self-creation and
self-transformation. If, as Vico, Herder and Berlin maintain, we are a highly
inventive species whose forms of life are radically underdetermined by our
common humanity, what reason is there to share the hopes of the Enlighten-
ment for an eventual convergence of liberal and humanistic values? Why
not expect, instead, endemic conflict among human beings, as they con-
stitute themselves into distinct and incommensurable cultures, each com-
mitted to its identity, and not all animated by anything resembling liberal
values? Do not recent developments in the Soviet Union, and in the
beleaguered state of Israel, suggest that this latter prospect is all too likely?
It is at this point that the kinship of Berlin with David Hume, remarked
upon in notable contributions to this collection by Richard Wollheim and
Stuart Hampshire, breaks down. For all his scepticism, Hume believed that
the moral and political judgements of men whose minds were unclouded by
enthusiasm or false philosophy would come to rest in a general con-
vergence. If only because of his conviction of the constancy of human
nature, he could think of human history only as an alternation of civilization
and barbarism - not as the exfoliation of incommensurably divergent
civilizations. Yet it is precisely that prospect, so threatening to the hopes
and expectations of the Enlightenment, that Berlin's radical value-
pluralism opens up. There is, then, a paradoxical kinship between the two
thinkers, after all: both are profoundly civilized men, defenders of the
values that animated the Enlightenment, whose philosophies render foun-
dationless the enlightened societies to whose defence they remain stead-
fastly committed.
One ofthe many merits of Edna and A vishai Margalit's rich and reward-
ing collection on Berlin and Berlinian themes 5 is the sense - very im-
perfectly conveyed here - it transmits of the tone of Berlin's writings and
conversation, of the multiplicity of his interests and the variety of his
 Berlin's agonistic liberalism 69
achievements. The fifteen sections of the book - encompassing essays by
Sidney Morgenbesser and Jonathan Lieberson, David Pears, Charles
Taylor, Richard Wollheim, Leon Wieseltier, Ronald Dworkin, G.A. Cohen,
Stuart Hampshire, Michael Ignatieff, Yael Tamir, Francis Haskell, Bernard
Williams, Alfred Brendel and Joseph Brodsky, together with a delightful
poem by Stephen Spender - touch on most aspects of the thought of this
many-sided thinker, only a few of which (though the most fundamental and
important ones) have been addressed here. The essays testify to the
character of Berlin's mind as a luminous prism, in which the cultural
traditions of Russia, England and Judaism are marvellously refracted. It is
their plurality of traditions in Berlin's own intellectual life that ultimately
prevents any assimilation of his voice to that of Hume. For, in the end, one
cannot help feeling that, with all that he shares with him in learning and
cultivation, Berlin's voice as it echoes through these pages is not Hume's.
If it shares with Hume's a profound intellectual gaiety and a relish fQr the
contingencies of history; it has in it resonances that come from other
aspects of Berlin's plural inheritance, altogether absent in the genial Hume.
The voice we hear is one that cleaves to mortal men and women in all their
unconsoled sorrow, and which refuses with a passion the mocking har-
monies of any theodicy. It is Job's.
Part II
Critiques
7 The system of ruins

As it has been disclosed to us in twentieth-century political history, the fate

of Marxism is to be the fmt world-view in human history that is genuinely
self-refuting. To be sure, all systems of general ideas about man and society
have unintended consequences when they are given practical effect, and it
is a commonplace that the distance between doctrine and practice is no-
where wider or harder to bridge than in political life. Further, it is a familiar
theme in political thought that social institutions may over the long run
have a self-destroying tendency in so far as they cannot help breeding
expectations they must fail to satisfy.
None of these traditional themes succeeds in capturing the thoroughly
paradoxical role of Marxian ideas in contemporary political life. The dis-
tinctive achievement of Marxism, peculiarly ironical in a system of ideas
committed au fond to the unity of theory with practice, is that its most
spectacular victories in the real world have afforded the most devastating
criticisms of its fundamental tenets. Accordingly, in installing in Russia and
in much of Asia new economic and political institutions to which nothing
in the old orders corresponded, the Communist regimes have exhibited
unequivocally that radical autonomy of general ideas in the political realm
which their official doctrine, no less than classical Marxism, tirelessly
denies. The stupendous successes of Communism in the real world have
given a practical self-refutation of the Marxian system, since in every case
the actual result of a revolutionary socialist victory has been to flout the
aspirations of the revolutionaries as it demonstrates once again the im-
possibility of Communism as Marx conceived it.
The self-refutation in practice of Marxism over the past half century was
not unanticipated in the theoretical writings of Marx's critics. In a rare
moment of realistic insight, the great Russian anarchist, Bakunin, predicted
that the outcome of a Marxian socialist revolution would be a form of
dictatorship more repressive and more exploitative than the bourgeois
political order it replaced. In a far more systematic fashion, Bohm-Bawerk
74 Critiques
in his Karl Marx and the Close ofhis System (1896)' dissected the errors of
Marx's economic theory and showed how they debilitated his account of
market capitalism, while B6hm-Bawerk's successors in the Austrian
School of Economics, L. von Mises and F.A. von Hayek, developed in the
19208 and 19308 powerful theoretical arguments explaining the failures in
resource-allocation of socialist systems. Apocalyptic though it has been,
the history of Marxism ;n practice over the past half-century has served
only to give concrete historical exemplification to criticisms of Marx's
ideas that were developed during his lifetime and in the first fifty years after
his death.
The ruin of Marx's system by the events of the past half-century has in
no way inhibited the production of Marxian theoretical literature in
Western societies. Throughout the past hundred years, Marxian ideas have
served in capitalist societies as weapons in the armoury of cultural criti-
cism, as tools in projects for [Link] history, and as postulates for
sociological research. In fulfilling this role of promoting self-criticism
within Western society, Marxian thinkers have been compelled to refine the
central notions of Marx's system beyond anything he could have recog-
nized or endorsed, and in so doing they have often obfuscated important
questions in the interpretation of his writings. It is one of the few hopeful
features of the flurry of activity surrounding the anniversary of his death
that a handful of books has appeared that give Marx's life and work the
benefit of a detached and scrupulous historical analysis. In this connection,
the Dictionary of Marxist ThoughP edited by Tom Bottomore is an invalu-
able aid in identifying the key terms in Marx's own work and distinguishing
their force in Marx from the uses made of them by later writers.
Bottomore's Dictionary is usefully complemented by GCrard Bekerman's
Marx and Engels: A conceptual concordance,] in which the crucial ideas of
the two writers are illustrated by quotations from their writings, carefully
chosen by Bekerman and skilfully translated by Terrell Carver. These
works of reference will prove indispensable to anyone who wishes to form
a reasoned judgement about the currently fashionable thesis that it was
Engels who made of Marx's subtle and eclectic thought a crude and
mechanical system.
A very different, but equally valuable service is performed by David
Felix's Marx as Politician.4 Felix's method is unique in Marxian scholar-
ship inasmuch as he develops his incisive criticism of Marx's theories
through the medium of a demystifying political biography. His strategy is
to deconstruct Marx's chief theoretical claims by illuminating their force as
acts in his struggles for political power over the emergent working-class
movements of nineteenth-century Europe and their rivalrous leaders. No-
where in Felix's elegantly and acidulously written book does he suggest
 The system o/ruins 75
that understanding Marx's theories in this way, as aspects of his political
practice, by itself devalues their claims to truth, but he shows convincingly
that we can best account for the manifest incoherences of Marx's system by
viewing it as a makeshift, constantly reworked according to the political
necessities of the moment. Again, without ever replicating the vulgarities
of psychohistory, Felix gives a psychological gloss to his political reading
of Marx's theoretical activity by displaying its roots in an ungovernably
assertive and domineering personality. Marx's virulent contempt for ethical
socialism, his rigid posture of opposition to all existing social orders and his
cynical dismissal of the claims of small nations and vanquished classes are
given a compelling interpretation by reference to his anomic and obses-
sional fascination with power. Felix's final assessment of Marx's political
vision grasps firmly a truth that has been stubbornly resisted by all of his
conventional biographers when he writes, "'Nazi" was the simplified acro-
nym for National Socialist German Workers Party. It was an accurate name
for the party Marx would like to have led in Germany in 1848-9, national-
istic, socialistic, and as anti-Semitic as tactically useful'.
The many affinities between Marx's political vision and the ideas and
movements of the radical Right which Felix identifies are profoundly
explored in Ernst Nolte's important collection of essays, Marxism.
Fascism. Cold War.' Since his seminal study, Three Faces 0/ Fascism,6
Nolte has been widely misread as a theorist of Fascism who conceives it in
Marxian terms as the radical anti-socialist response to capitalist crisis and
who seeks the elimination of the liberal category of totalitarianism in the
explanation of both Communism and Fascism. The discursive and wide-
ranging essays assembled in this volume should lay to rest any such
interpretation of Nolte's work, which is distinctive in representing con-
temporary Marxist practice as having authentic origins in Marxian doctrine
and instructive in perceiving the structural similarities of Marxian and
Fascist contestations of bourgeois society. Thus in identifying, in his brief
essay on 'The conservative features of Marxism',' the character of
Marxism (understood here to mean the doctrines held in common by Marx
and Engels) as a critique of modernity, Nolte helps us towards an explana-
tion of the encrusted cultural conservatism of all actual Communist regimes
that is more adequate than any to be found in the strained apologetics of
Western Marxian writers. The enmity of Communist governments to all the
most radical expressions of the modem spirit - in art and philosophy as well
as lifestyle and popular culture - is correctly perceived as emanating
directly from the anti-individualist animus which pervades the thought of
Marx and Engels alike. The repression in Communist states of all modernist
movements is not, then, an aberration or even an unintended consequence
of Marx's doctrine, but simply an expression of its original intent. In its
76 Critiques
application to the fascist phenomenon, Nolte's analysis is conclusive in
linking the Rousseauesque primitivism of Marx's fantasy of ending the
social division of labour with the Fascist rebellion against commercial
society. As Nolte drily observes:
Fascism can be directly compared with Marxism of the Soviet nature
only in its radical form, in respect of its inner solidarity and its appeal to
comrades of like mind in all countries; Italian fascism, in its phase as a
development dictatorship, and more than ever the Croatian Ustase and
the Romanian Iron Guard were in fact, on the contrary, more like many
of today's 'national liberation movements' than like late National
Socialism . . . ; there is nothing more grotesque than a 'theory of
Fascism' which denounces capitalism with much sincere indignation as
the root of Fascism, at the same time overlooking that the theory identi-.
fies itself with conditions which show all the formal characteristics of
Fascism. It is not astonishing that the liberal capitalist system produces
Fascism under certain circumstances, but it is astonishing that in the
great majority of cases Fascism has not succeeded in gaining power in
spite of certain circumstances. The explanation can only lie in the fact
that this social system with its peculiar lack of conception, its deep-
rooted divergencies, its inborn tendency to self-criticism, its separation
of economic, political and spiritual power obviously offers strong resist-
ance to a transformation to fascist solidarity, and is aware that the
deliverance which is promised would at the same time be loss of self.
Thus capitalism is indeed the soil of Fascism, but the plant only grows
to imposing strength if an exorbitant dose of Marxist fertilizer is added
to the soil. 8
The most important essay in Nolte's collection deals not with the question
of Fascism, however, but with errors in the historical interpretation of early
industrial capitalism which have been widely disseminated by Marxian
writers. Along with radical Tories such as Oastler, Sadler, Southey and
Disraeli, Marx and Engels associated the Industrial Revolution with the
pauperization of the masses and the devastation of their traditional ways of
life. By comparison with the factory system as it developed under laisser-
faire capitalism, pre-industrial life was pictured in almost Arcadian terms
of satisfying work, harmonious community and a reasonable sufficiency of
material goods. Nolte is assiduously specific in documenting how Marx and
Engels and the reactionary and Romantic critics of industrialism and the
factory system neglected the filth, squalor and waste of human life endemic
in pre-industrial society. In this Nolte's analysis parallels that of a number
of contemporary economic historians, among whom the most distinguished
is R.M. Hartwell, whose researches have gone far to establish that the
 The system of ruins 77
Marxian immiseration thesis is as false in respect of early industrialism as
it is of our own capitalist economies. An explosion of population involving
a massive decline in infant mortality rates, increasing consumption of
commodities hitherto regarded as luxuries, and many other empirical
factors point to the early industrial period in England as one of much-
enhanced popular living standards.
At the same time, Nolte is careful to specify the background of this
explosion in living standards in several centuries of European and, above
all, English political and cultural development, which preceded it. Noting
that 'European society is, from its beginnings in the early middle ages
onward, the society of a functioning or dynamic pluralism whose several
relatively autonomous powers, such as royalty and the aristocracy, the state
and the church, and also the individual states restrict each other, and yet
they remain even in sharpest struggle, related to each other and subject to
mutual influence', Nolte inverts the historical materialist thesis of the
primacy of technological and economic factors in accounting for social and
political changes and explains the technological development of early
industrialism as a variable dependent upon pluralist legal and political
institutions. In so doing he is concerned to stress particularly the import-
ance of the English example, wherein the Industrial Revolution was the
culmination of several centuries of agrarian development on a market
model. His account of the background and conditions of the Industrial
[Link] in England converges at several points with that given by Alan
MacFarlane in his fascinating Origins of English Individualism,' and it
would be encouraging to suppose that Nolte's book will do something to
subvert the legend, which the writings of Karl Polanyi and C.B.
Macpherson have made a central element in academic folklore, that the
seventeenth and eighteenth centuries in England encompassed a radical
transition from communitarian to individualist forms of social life.
The upshot of Nolte's analysis is that European capitalism is a historical
singularity, in no way the necessary or inevitable outcome of human social
development taken as a whole. It was as a lucky chance, the unlikely
outcome of a serendipitous conjunction of events, that market processes
were able to spread in the early Middle Ages and thus to lay down the
necessary conditions for the emergence of large-scale capitalist production.
This conclusion goes against one of the central tenets of Marx's thought,
and allows us to pinpoint one of its most disastrous errors. For all his
insistence on the particularities of specific cultures and on the unevenness
of economic development in different nations, Marx (and Engels after him,
albeit with fewer saving reservations) subscribed to a belief in something
like a law of the increasing development over human history of productive
forces. He asserted this not just as a brute historical fact nor yet as a mere
78 Critiques
trend, but as the unifying principle of human history. It is such a principle,
something midway between the statement of a trend and the enunciation of
a law that G.A. Cohen tenos the Development Thesis in his Karl Marx's
Theory of History: A Defence. tO It is one of the most noteworthy features of
Cohen's book, which sets standards of competence and rigour in argument
which have been matched by few twentieth-century Marxian thinkers and
which non-Marxian philosophers would do well to try to emulate, that his
defence of the Development Thesis is feeble and admittedly unsuccessful.
In the end Cohen is driven to invoke in its support a starkly Benthamite, and
for that reason wholly un-Marxian, conception of man as an economizer of
his efforts.
This move has to confront, however, the inconvenient fact that the
systematic and continuous expansion of productive forces over many cen-
turies appears to have occurred within capitalist Europe and its offshoots
and nowhere else. Explaining the singularity of capitalist development
generates a most fundamental criticism of the Marxian scheme of historical
interpretation. For, contrary to Cohen's attempted reconstruction of his-
torical materialism in Darwinian functionalist form, a mechanism for filter-
ing out inefficient productive arrangements exists only within the capitalist
mode o/production. Within a capitalist market economy, there is a power-
ful incentive for enterprises to innovate technologically, and to adopt
innovations pioneered by others, since firms which persist in using less
efficient technologies will lose markets, reap dwindling profits and eventu-
ally fail. Nothing akin to this selective mechanism of market competition
existed to filter out inefficient technologies in the Asiatic mode of pro-
duction, and it has no replica in existing socialist command economies.
Cohen's defence of the Development Thesis is bound to fail because it
attempts to account for the replacement of one productive mode by another
by invoking a mechanism which features internally in only a single mode
of production, market capitalism.
Cohen's argument has the virtue of confronting a central difficulty in
Marxian historical materialism which most Marxian writers prefer to pass
over. Thus the problem is mentioned by Alex Callinicos neither in his
propagandist tract, The Revolutionary Ideas 0/ Marx,l1 nor in his more
reflective and self-critical Marxism and Philosophy.t2 None of the writers
in David McLellan's Marx: The first hundred years\3 takes it up, even when
(as in the essays by Raymond Williams, Ernest Mandel and Roy Edgley)
their contributions focus more or less directly on problems and applications
of historical materialism. This omission is striking and lamentable, but
eminently understandable, since any recognition of the inadequacy of the
Marxian scheme of historical development is bound to undermine the
 The system of ruins 79
viability of Marxian socialism itself. If we acknowledge, as did Marx, the
essentially unconservative character of capitalist enterprise, we will fmd it
incongruous that he and his followers imagine that the prodigious virtuosity
of capitalism can be retained while its central mechanism, market com-
petition, is abolished. There is, in fact, no reason to think that the productive
achievements of capitalism will even be maintained, still less surpassed,
once market mechanisms for allocating resources are removed. It is this
insight which explains the vast chaos and colossal malinvestments which
are typical of all existing socialist command economies. In Marx's own
writings, in accordance with his refusal to engage in utopian speculations,
no proposal is ever advanced for the coordination of economic activity in
socialist or Communist societies: it is simply assumed, with the utmost
nalvet~, that an acceptable allocation of resources to particular uses will
emerge spontaneously, without the need for markets or pricing, from the
collaborative discussions of socialist citizens. It was indeed to this gigantic
evasion that Lenin referred obliquely, when he confessed that the principal
task of the Bolsheviks in the USSR was the construction of state capitalism.
Aside from the fact that it entails inexorably a concentration of power in
bureaucratic institutions which Marx always sought to avoid, but which
was realized fully in the Stalinist period, Lenin's project of a state capitalist
regime was bound to founder on the absence within it of the central
capitalist institution for resource-allocation.
In the event, the Soviet experience amply confirmed the predictions of
those economists of the Austrian School, particularly von Mises and von
Hayek, who argued for the impossibility of rational resource allocation
under socialist institutions. In the Soviet Union, working-class living stan-
dards after over sixty years of state capitalist construction are probably
lower than in Brazil, while elsewhere, in Hungary and in China, only the
expedient of reintroducing capitalist institutions is allowing wealth to grow
and incomes to rise. These developments exemplify in concrete historical
contexts the theoretical insights with which the Austrian economists pre-
vailed over their socialist opponents in the great debates of the inter-war
years. Yet, despite their intellectual victory, the Austrian arguments have
been ignored by generations of economists and their relevance to the Soviet
experience has been expounded in depth only by Paul Craig Roberts in his
vital and neglected book Alienation and the Soviet Economy.14 It is entirely
characteristic that in his contribution to the McLellan collection, Mandel,
after showing an awareness of the calculation debate that distinguishes him
from the bulk of his professional colleagues, should demonstrate his in··
ability to grasp the nature of the problem at issue when he remarks inno-
cently of von Mises's argument that it has 'in the meantime been taken care
80 Critiques
of by the computer' . As it has turned out, history has forced back on to the
intellectual agenda a debate which the intelligentsia for several generations
consigned to the memory hole.
The ruin of Marxism both as a scheme of historical interpretation and as
a theory of economic organization has evoked a variety of responses among
contemporary Marxian writers. The great majority has tried to prevent the
destruction of the doctrine by intractable facts through the elaboration of
protective ad hoc hypotheses. Accordingly, an effort has been made to
explain the catastrophic impact of Marxism in Russia by seeking out
continuities between the political culture and institutions of Tsarism and
those ofthe Soviet power, with the underlying insinuation that in Russia an
enlightened Western European creed of democracy and freedom was
corrupted by contact with tyrannous native traditions. Its culturally racist
features aside, this argument misrepresents Tsarism, which for the last sixty
years of its history was an open, progressive authoritarian system, far less
inhumane or repressive of individual liberty than the great majority of
member states in today's United Nations, and evolving in a context of
extraordinary economic growth and brilliant cultural achievement. The real
Russian tragedy was the reverse of that imputed by the conventional and
complacent view in that the blossoming civic traditions of Tsarism were in
1917 barbarized and destroyed by the incursion of a totalitarian ideology of
Western European origins.
On a more general level, this sort of protective manoeuvre within
Marxism must be criticized on the Popperian ground that it has the effect of
transforming what was in Marx's hands a living and corrigible body of
thought into an intellectual deadweight of reinforced dogmatism. Thus
every contribution to the Norman Fischer volume on Continuity and
Change in Marxism lS (with the partial exception of a cryptic and suggestive
piece by Kostas Axelos) reveals an abandonment of the empirical content
of Marx's thought in favour of a reassertion of its Hegelian essentialist
metaphysics. This metamorphosis of Marxism from a body of empirical
social theory and of historical interpretation into a self-enclosed meta-
physical system is most evident in the Frankfurt School, but despite all
protestations to the contrary it characterizes Althusser's Cartesian re-
construction of Marx's thought as well as Marcuse's Heideggerian varia-
tions on Marxian and Hegelian themes.
In fairness it must be said that the multiple ironies involved in this retreat
to metaphysical inquiry from a system of thought which at its height
promised an end to philosophy have not gone unnoticed by all Marxian
thinkers. The tension between the metaphysical tum in recent Marxism and
the anti-philosophical bent of Marx's own mature thought is at the heart of
Callinico's Marxism and Philosophy and it motivates Susan Easton's
 The system of ruins 81
search for affinities and convergences in Humanist Marxism and Wittgen-
steinian Social Phiiosophy.16 Easton's intriguing project of linking up a
form of Marxism in which human activity and not historical law is central,
with the Wittgensteinian conception of knowledge as embodied in social
practices, does not face its hardest difficulty in the biographical fact that
Wittgenstein's own political views were conservative, not to say reaction-
ary, and were never seen by him to conflict in any way with his developed
philosophical outlook. The most serious difficulty for this kind of Marxian
theorizing is its irresistible tendency to slip into an Idealist constructivism
about the social world of precisely the sort that Marx repudiated in his
attacks on Hegel and on Stirner. The metaphysical turn of humanist
Marxism is sure to be a dead end because it begins by shedding the realist
commitments which Marx himself rightly thought to be most distinctive of
his view of social life.
In their retreat from empirical theorizing to essentialist metaphysic!!, the
Hegelian Marxists forgo one of Marx's most ambitious projects: the
development of a comprehensive theory of ideology. Any theory of ideo-
logy, and above all a Marxian theory, incorporates a distinction between
appearance and reality in society which the Idealist implication of humanist
Marxism tends to occlude. Further, the abandonment of the claim to scien-
tific realism in Marx's thought suggests an obvious question about the
ideological character of humanist Marxism itself. This is a question that
haunts Jorge Larrain's meandering and inconclusive discussions in
Marxism and Ideology,t7 but which is posed decisively at several points in
Jean Cohen's Class and Civil Society: The limits of Marxian critical
theory.18 Cohen's is a luminously intelligent investigation of the limitations
of Marxian class theory which takes seriously the criticism of socialist and
Marxist thought as itself having the mystifying and repressive functions of
an ideology. She considers in this context not only the theory of Konrad and
Szelenyi, which echoes the predictions of the late nineteenth-century Polish
anarchist, Waclas Machajski, in representing Marxism in the Soviet bloc as
the instrument of a novel form of domination, but also Western theorists of
the new class such as Irving Kristol and Alvin Gouldner.
Cohen's own attitude to Szelenyi's class analysis of the Eastern bloc
societies - a most useful exposition of which Szelenyi gives himself in his
contribution to M. Burawoy and Theda Skocpol's Marxist Inquiries l9 - is
not free from ambiguity. She recognizes the truth in Szelenyi's and
Konrad's claims regarding the existence of an exploitative social stratum
which has arisen in the Communist regimes via its control of education and
of access of information, but she goes on to criticize their approach as
flawed because it adopts a strategy of analysis whose limitations are those
of Marx's class theory. The opposite situation seems to me to be the true
82 Critiques
one: the theory of the new class in its control of education and of access of
infonnation cannot be stated in Marxist tenns, but she goes on to criticize their
approach with Marxian class theory. That the new class is not a Marxian class
is a criticism of the theory of Szelenyi and Konrad only in so far as they see
themselves as completing Marxian social theory rather than abandoning it
whenever new forms of injustice and exploitation elude its grasp.
The general relevance of the theory of the new class is that it encourages
us to look at the ideological function of socialist thought itself. In so doing
we harness the critical intent which motivated Marx's analyses of the
classical economists to examine Marxian and other socialist system of ideas
as vehicles for the protection and promotion of the interests of specific
social groups. Essential to the theory of ideology, after all, is not only the
identification of a distance between reality and appearance in society, but
the demonstration that this distance is functional in enabling some social
interest to prevail over some other. Ideology, in short, facilitates domina-
tion and exploitation as ongoing social relationships. Not only in its mani-
festations in the Soviet bloc, but also in Western societies, socialist thought
invites ideological analysis as an instrument in the social struggle among
competing groups for access to state power and thereby to the resources the
modem interventionist state commands. Whereas a theory of the ideo-
logical functions of the socialist system promises much in the illumination
of the chronic legitimation crisis of the Communist regimes and of the
conflicts in our own societies, the project of developing fully such a theory
is one that even independent critical thinkers of the stature of Jean Cohen
seem to retreat from.
The undefended assumption that socialist goals stand in need of no
ideological demystification, even if socialist regimes sometimes do, is an
outstanding feature of Barry Smart's able exploration of the relations of
Foucault's thought with Marxism,20 and the inherent progressiveness of the
socialist ideal figures as a presupposition of analysis, inhibiting funda-
mental criticism, equally in George G. Brenkert's Marx's Ethics of Free-
dom. 21 It seems that the stance of radical criticism does not extend, so far as
these writers are concerned, to the socialist conventional wisdom of the
Western academic class.
A re-emergence of Marxism as a progressive research programme in
social theory may be predicated upon several rather exacting conditions. A
new Marxism worthy of serious critical attention would have to confront
the Austrian thesis that market competition and bureaucratic command
structures are together the mutually exhaustive means of resource-
allocation in complex industrial societies, with command economies hav-
ing ineradicable tendencies to vast waste and malinvestment. It would have
to consider the possibility that the economic chaos and political repression
 The system of ruins 83
characteristic of all socialist command economies are not mere aberrations,
but structurally inseparable results of such economies. It would, above all,
need to confront the repressed possibility that the Gulag represents an
unavoidable phase in socialist construction rather than a contingent inci-
dent in Soviet (and Chinese) experience. In order to face these hard ques-
tions, a new Marxism would demand a purer and more self-critical method
of thought than any varit:ty of Marxism has so far achieved. It would need
to engage directly with the moral theory of justice and exploitation and to
abandon the forlorn pretence that it can deploy some special, dialectical
logic to circumvent contradictions within its own theories. The central
concern of such a new Marxism - to link normative exploitation theory
with empirical class analysis - is in fact the subject matter of a powerful
new school of Analytical Marxism, led by such figures as G.A. Cohen, Jon
Elster and John Roemer, with whose works the future of Marxism, if it has
any, must henceforth be associated.
It is hard to imagine that the version of Marxian theory which looks like
being developed by these thinkers will do more than generate a few
scattered insights which are easily absorbed into normal social science.
Once the spurious claim to esoteric insight and omnicompetent method is
given up, Marxian thought confronts the same intractable difficulties in the
theory of justice and in the philosophy of social science which have
bedeviled non-Marxian thought, and it has little that is special of its own to
offer. The attraction of Marxism to the Western intelligentsia was, in any
case, never that of an analytically superior theoretical system in social
science. It was rather the appeal of a historical theodicy, in which Judeo-
Christian moral hopes were to be realized without the need for a trans-
cendental commitment which reason could not sanction. In the Communist
societies where Marxism has been institutionalized as the official ideology,
its mythopoeic elements have not indeed been especially prominent. For all
the paraphernalia of the Lenin cult, Marxist ideology has functioned there
in Hobbesian fashion, as an instrument of political discipline, and has had
no role in spiritual life. If anything, the inability of Communist Marxism to
function as a comprehensive view of the world has added a new twist to the
history of its practical self-refutation, as when the Soviet Buryat Mongols
appropriate the official legend of the Paris communards and pray to their
spirits, which have come to rest in the home of the Buryat's traditional
objects of worship under Lake Baikal. Yet the irony of Marxism's self-
effacement in the Soviet Union is unlikely to be altogether evaded in the
liberal intellectual cultures of the West, even if it does not take the beautiful
form of a Shamanistic metamorphosis of Marxist piety. Western analytical
Marxism will flourish and expand just in so far as it possesses those mythic
elements in Marx's thought that it is committed to shedding.
84 Critiques
At the same time. eliminating the mythic content of Marxism will rob it
of its distinctive power and speed its recuperation by bourgeois social
science. The final dilemma of Western Marxism is that. unless it represses
in the interests of criticism and objective knowledge the mythopoeic im-
pulse which explains its appeal over the past century. it can only present to
the rest of us the spectacle of an esoteric and barely intelligible cult, whose
devotees pass their time picking reverently among the shards and smit-
hereens of a broken altar.

December. 1983
8 The delusion of glasnost

Two world-historical tendencies are presently converging in Central

Europe - the rapid and inexorable dissolution of the post-war global settle-
ment and the continuing decomposition, as yet local, partial and reversible,
of socialism on the Soviet model. These are developments for which
Western policy and opinion are ill prepared. The revelations of glasnost
confound the certainties of generations of Western opinion-formers. In the
Soviet Union, in Poland and now in Hungary and the Baltic states, official
sources have revealed a spectacle of economic catastrophe, ecological
devastation, hideous social problems and massive popular estrangement -
phenomena long familiar to those who have ever lived in Eastern Europe,
but on which the Western mind has hitherto turned a senile and incurious
eye, clouded by fading images of a new socialist order. The inescapable
implication of such disclosures is that the prognostications of Western
scholars and journalists, who perceived in the States of Eastern Europe
blemished but progressive regimes bent on a task of constructive economic
and social planning, have proved to be (as Vladimir Bukovsky has said of
Western Sovietology) simply trash, less reliable and far more removed
from reality than the wildest tales of the most embittered ~migres. Nor has
Western conventional wisdom been noticeably upset by the harrowing
news now released daily. Instead, and entirely predictably, the profound
crisis to which glasnost and the reform movements attest is interpreted by
bien-pensont Western opinion as further evidence (if such were needed) of
the corrigibility of Soviet-style institutions, not as proof of their bank-
ruptcy. In the Soviet Union itself, glasnost has served only to reveal
intractable and perhaps insoluble economic problems, but this has not
prevented Western observers from discerning signs of large-scale re-
structuring. In our times, perestroika, like socialism, is in the eye of the
beholder.
The unravelling of the post-war settlement poses similar conundrums for
Western opinion. In a sort of political relativization of geography, the
86 Critiques
division of Europe at Yalta resulted in a loss of the sense of Europe's
historic boundaries. On the map, Prague is west of Vienna, and in cultural
and historical terms it is at the very heart of Europe; but Czechoslovakia did
not have the good fortune Austria had in being neutralized, so it is now part
of Eastern Europe, which is to say, hardly part of Europe at all. As far as
Western opinion is concerned, Central Europe disappeared down an
Orwellian memory hold at Yalta - it is no longer even a geographical
expression. Those who live in the forgotten lands of Central Europe may be
forgiven if they find the status quo less than entirely congenial. It is, in any
case, by now scarcely a status quo. With extraordinary subtlety and bold-
ness, Soviet policy has sought (and achieved) a return of the repressed
memory of the common cultural inheritance among the peoples of Central
Europe. It has done so, above all, in West Germany, which is the hinge on
which turns the future of Central Europe, and much else. Whatever the
immediate outcome of current negotiations, it is safe to assert that neither
the division of Germany in its present form, nor West Germany's current
relationship with NATO, can be sustained for long. As it stands, the
political and military posture of West Germany disregards both the realities
of history and legitimate German aspirations for reunification; and the
pressures for a separate settlement between West Germany and the Soviet
Union are probably irresistible. Such a Soviet-German understanding
would not be without precedent. It is, perhaps, not altogether fanciful to see
in today's accelerating rapprochement a repeat run of the Treaty of
Rapallo, under the secret terms of which Germany helped to train and arm
the Red Army, and which inaugurated a major programme of German
economic aid to the ruined Soviet economy. History was not kind to that
particular understanding, but everything suggests that the movement to
detach West Germany from its NATO allies cannot now be reversed.
The shape of a new German settlement is thoroughly obscure. We may
nevertheless ask, what would any such settlement portend for the nations of
Central Europe? Here differences in the conditions prevailing in the differ-
ent countries are of considerable importance. In Poland, the regime has
abandoned the vain attempt to maintain a classic Communist weltan-
schauung-state, thereby becoming the first authentic example of post-
totalitarianism, and is now embarking on an experiment in controlled
democracy. Developments in Poland have occurred principally in response
to crises within Polish society: they have been engendered by the ruination
of the economy and by the organized resistance of Solidarity and have been
facilitated by the alternative authority structures existing in the Catholic
Church. In Hungary, though it is also a response to economic catastrophe,
the reform movement has followed a different path, being initiated from
within the Party, itself doubtless responding to exogenous forces. Whereas
 The delusion of glasnost 87
in Poland the institutions of civil society have achieved a measure of
autonomy by asserting themselves successfully against the totalitarian
bureaucracy, in Hungary civil society is emerging in the wake of a project
of economic and political reform initiated from above. Again, the geo-
political environments of the two countries are manifestly different, with
neutrality on the Austrian model and EEC membership an outcome con-
ceivable for Hungary but unthinkable in the case of Poland, for which
Finlandization remains the utmost that can reasonably be hoped for.
Such differences between the two countries that are presently at the
cutting edge of reform in Central Europe should not lead us to overlook no
less significant features they have in common. Both are burdened by an
overwhelming foreign debt, which in all likelihood can never be repaid, and
which overshadows the faltering and inconsistent reform process in their
economies. In both countries, the ruling Communist parties are making
systematic efforts to contain and co-opt the oppositional movements, which
in each case are deeply divided and increasingly fragmented. Each country
has a powerful and gifted intelligentsia, which has a distinguished pedigree
of pre-Communist theorists of liberty and civil society, such as Istvan
Szechenyi and Jozsef Eotvos in Hungary, while in Poland the independent
journal Res Publica has given voice to a diversity of conservative and
liberal thinkers, Polish and foreign, contemporary and historical. Most
fundamentally, the two countries have in common what differentiates them
from their East European neighbours, where a bloodless totalitarianism still
rules: a project of reform which, in virtue of the intractable realities of
economic bankruptcy and popular estrangement, threatens constantly to
metamorphose into revolution. This is the truth concealed by slogans such
as market socialism - an inherently unstable equilibrium which can only
mutate into the politically explosive direction of the reinvention of the
central institutions of market capitalism, or else back into the repressive
machinery of the command economy.
The ambiguities, ironies and paradoxes of the reform movements in the
Soviet Union and parts of Eastern Europe are well explored in Jeffry C.
Goldfarb's timely and thoughtful Beyond Glasnost: The post-totalitarian
mind. I Discussing such diverse figures as Havel, Kundera, Baranczak,
Michnik, Haraszti and Konrad, Goldfarb finds in them all manifestations of
a post-totalitarian culture which, he claims, it is the strategy of glasnost to
emulate, denature and emasculate. It pursues this objective via the inven-
tion of yet another Newspeak, in which the rudimentary truthtelling that
distinguishes many of the thinkers and spokesmen of the oppositional
movements is distorted and corrupted. Elements of Goldfarb's analysis are
problematic and questionable. For one thing, the reality of a post-
totalitarian culture is at present clearly discernible only in Poland. It does
88 Critiques
not exist in Czechoslovakia (where, however, a pretotalitarian mind has
never disappeared, but has merely been repressed) and as for Hungary,
where the memory of a culture of fear remains recent and vivid, it is too
early to essay a judgement. What is important and profound in Goldfarb's
book is his depiction of the nations of Eastern and Central Europe as locked
in a chronic crisis in which, in various fashions and to different degrees, the
autonomous institutions of civil society constantly assert themselves
against totalitarian administration. This analysis marks a significant ad-
vance on the conventional Western perception of the reform movements as
expressions of self-criticism by the enlightened despots of the Soviet bloc.
The two movements which bring Central Europe back on to the agenda
of history - the unravelling of the post-war settlement and the dis-
integration of Soviet-style Communism - are interrelated at every point.
The historic injustice perpetrated by the Allies at Yalta was a necessary
condition of the imposition of Stalinism on Eastern Europe, just as earlier
intervention in the service of Wilsonian notions of national self-
determination and global democracy had given the coup de grace to the
Habsburg Empire. As a consequence of Europe's two civil wars, no institu-
tional framework is easily conceivable within which Central Europe could
become again a political reality. The empire of the Habsburgs is a nos-
talgist's dream, and many questions surround the prospect of a neutral,
unified Germany: what would be its borders? And its relations to Poland
and Czechoslovakia? The darker side of the dissolution of the post-war
settlement is in the prospect of most of Eastern Europe remaining in the
vice of an atavistic neo-Stalinism, with a permanent legitimation crisis in
Hungary and Poland, and West Germany prized loosed from NATO only to
inherit the rusting industries and indigent pensioners of the GDR. The
Soviet bloc would then face an indefinite period of decline, its reform
movements stifled by spasms of repression and economic collapse staved
off by recurrent injections of Western credit and technology.
It is a scenario on these lines - in which the Western powers throw their
weight behind the neo-totalitarian project of perestroika and the posture of
openness in the Soviet Union is abandoned - that seems the most likely, as
the economies of the Soviet bloc lurch into the abyss and nationalist
movements threaten to become uncontrollable. In this scenario, in which
the Soviet bloc is refinanced by the West without guarantees of real
economic reform in the direction of a full market economy, the prospect of
a rebirth of Central Europe will prove to have been a seductive mirage and
post-totalitarianism (except perhaps in Poland and Hungary) an ephemeral
phantom. The possibility of a genuine civil society, which the profound
crisis of the Communist regimes has briefly generated, will then have been
extinguished. This will have come to pass, in part, because of the intractable
 The delusion of glasnost 89
domestic problems of the East European economies, which cannot move to
market institutions without imposing costs on their populations that are
bound to test to destruction the fragile institutions of controlled democracy.
Even in long-established Western democracies, such as New Zealand and
the United States, the project of limiting government and reviving market
institutions has itself been limited by democratic pressures. (In the United
Kingdom. the upshot of Thatcherism is likely to be no different.) How can
we expect the long-suffering workers of Eastern Europe, condemned to
forty years of poverty by the socialist command economy, to consent to
decades of austerity as the condition of its dismantling? And is it reasonable
to expect the solidarity of workers and peasants to survive the conflicts of
interest that will emerge between them with the adoption of a market
economy?
Faced with such dilemmas, Western policy and opinion are likely to
react, not with constructive engagement, but with their characteristic mix-
ture of inept Machiavellianism and guileless credulity, and will opt for the
delusive stability of perestroika without glasnost. They will opt, in other
words, for economic reconstruction by authoritarian means. If they do, they
will be ~isappointed, since it is the lesson of China that economic re-
construction founders when it is not accompanied by a commitment to
political accountability and the rule of law, and recourse to repression is a
permanent possibility. In respect of Poland and the Baltic States, in any
case, it is far from clear that a return to repression is a realistic possibility.
Even if the present movement to civil society be reversed, it will not (as
many in the West continue to hope) inaugurate a period of stability in
Central and Eastern Europe, but only stave off change until a worse crisis
cannot be prevented. For, in truth, the situation in Eastern Europe is not one
in which orderly reform can be effected, it is more akin to a revolutionary
situation - but that is a truth which transgresses the limits of Western
glasnost.

July, 1989
9 The academic romance of
Marxism

The claim that the real history of the twentieth century can be illuminated
by the intellectual history of Marxism is, perhaps, one that could now only
be made in the United States. From the perspective of most Europeans,
including the vast majority of those living in the Soviet bloc and virtually
all who may once have been Marxists, any such claim will seem at best
heavily ironical, and at worst a piece of academic frivolity. For in none of
its many varieties did Marxist thought anticipate (nor has it yet even begun
to explain) the most decisive developments which occurred in Europe and
elsewhere in our century. It was paralysed by the trauma of the First World
War, in which the proletarian classes of Europe exhibited national and not
class allegiances, and it recovered for a decade only by retreat into mille-
narian fantasy about the new society (and, let us not forget, the new
humanity) which it supposed to be under construction in Russia in the
aftermath of the Bolshevik Revolution. The lessons of War Communism, in
which a systematic attempt to operate an economy without market
exchange may have cost as many lives as the later experiment in socialist
agriculture conducted by Stalin, were lost on the Marxian theorists of
revolutionary socialism - if, indeed, they troubled themselves to inquire
into these events. Instead, Marxist thought between the wars oscillated
uncertainly between witless Sovietism and self-deceiving piety towards
revolutionaries not compromised by collaboration in Stalinist terror, such
as the ineffectual and romantic Rosa Luxembourg and the vain and pitiless
opportunist Trotsky. Marxist thought was utterly unprepared for the succes-
sive catastrophes of the destruction of social democracy in Europe, the
Nazi-Soviet pact and the Holocaust. Equally, the role of slave labour in the
Gulag during the period of rapid industrialization under the Five Year Plans
in the Soviet Union was repressed for a generation or more in Western
Marxism, whose exponents found in the convulsions of the Cultural Revo-
lution in China (which probably cost more lives than the Stalinist terror, and
 The academic romance of Marxism 91
encompassed a cultural genocide in Tibet) an alternative solace for their
oppositional energies.
The coup de grace to Marxism in Western Europe has been given by
developments during the past generation, which have comprehended the
repression of a popular revolution in Hungary, an aborted refonn from above
in Czechoslovakia, the attempted destruction by way of a Pinochet-style
military dictatorship of the autonomous institutions of the Polish working class
in its project of asserting civil society (including the Roman Catholic OtUICh)
against the totalitarian bureaucracy and, last but not least, the strategy of
dismantling the institutions of central economic planning and engaging
Western finances in a project of capitalist economic development which, with
varying degrees of consistency, is currently under way in the Soviet bloc and
in People's China. Not one of these world-historical developments was pre-
dicted, nor has any of them been theoretically accommodated, by Western
Marxism, which has preferred to devote its attention to the question why no
socialist revolution has yet occurred in any capitalist society.
Like Marx's thought itself, the proposition that the intellectual history of
Marxism might be deployed to illuminate the crises of our time finds little
reception in contemporary Europe. If in Paris it would be greeted with
bored indifference, in Warsaw and Leningrad it will meet with incredulity
and downright contempt. It will command intellectual interest, if at all, only
on condition that it focuses on the causal relations between Marxian theory
and the catastrophes that have disfigured the societies in which Marxism
has acquired political hegemony - a topic virtually passed over in silence
by Martin Jay. How, then, does Jay attempt to justify the research pro-
gramme advanced in Fin-de-Siecle Socialism?l He tells us that it may be
'instructive to consider the possible parallels between the late nineteenth
century and the waning years of our own. For in so doing' , he asserts, 'we
may find some suggestions for the socialism that is likely to develop in the
21st century'. His statement is arresting, inasmuch as the suggestion that
the near future will offer a future to socialism in any of its radical variants
is likely to be other than risible only in a society which (like the United
States and unlike most of Europe) lacks any historical experience of revo-
lutionary socialism.
There is a deeper reason for the incongruity of Jay's project. For the
most part, American radical intellectuals in recent decades have, in default
of theorizing their own history, traditions and institutions, aped the spent
intellectual fashions of Europe. In general, American intellectual life in this
century has given evidence of prodigious vitality and originality, with
American philosophers developing the seminal work in pragmatism of such
thinkers as C.S. Peirce and William James, and American philosophy
92 Critiques
coming to dominance in much of the rest of the world. Again, American
thought has produced profound reflections on the American political
experience in the work of thinkers such as John Rawls, whose theory of
liberal justice embodies an unsurpassed distillation of American con-
stitutionalism, and the socialist historian Eugene Genovese, whose work on
the antebellum South has delineated the structure and dynamic of an
anticapitalist economy and culture in its relations with the commercial
civilization of the North.
By contrast, the American radical intelligentsia over the past decades
has typically been preoccupied with the ephemera of the European culture
market, whose most grotesque manifestations have enjoyed in American
academia a sort of stilted after-life - as evidenced in deconstructionist
literary theory and in the applications to law by the Critical Legal Studies
Movement of the banalities of the participatory democratic theory of the
1960s. Jay's project becomes intelligible, if not justifIable, when we note
its background conditions - an intellectual culture insulated from political
experience of radical socialism and an academic class which uses the
rhetoric and theorizing of the radical intelligentsia of Europe a decade or a
generation ago to legitimate its estrangement from its own culture. For
these reasons alone, we may confidently predict that the academic institu-
tions of capitalist America will be the last redoubt of Marxian theorizing
when it has long since been repudiated wholesale by the peoples and rulers
of the states where once it commanded power.
It must be acknowledged at once that Jay's intellectual history of
Marxism is always scholarly and is for the most part written felicitously and
in civilized tones. At times, indeed, he charmingly throws off intimations
of the absurdist aspects of his own project. Thus, as an aside, he notes that
the bizarre spectacle of autocratic rulers like Joseph Stalin, Josip Tito
and Mao Tse Tung holding on to power for life suggests, in fact, that
monarchy is a far more likely prototype for socialist politics in certain
20th century regimes than anything remotely derivable from the demo-
cratic tradition.
Inexplicably, Jay here omits any reference to Romania and North Korea,
where the institution of Communist autocracy promises (or promised) to
become hereditary - an even more delightful inversion of Marxian expec-
tations. Again, when Jay writes of Laclau and Mouffe's book, Hegemony
and Socialist Strategy 2, that they 'argue for a revised version of Gramsci's
notion of hegemony filtered through a post-structuralist critique of the
possibility of an identitarian subjectivity' , it is impossible to believe that he
is not satirizing the exotic argot spoken only by Western academic Marxists
at governmentally funded international conferences. Most centrally,
 The academic romance of Marxism 93
though, Jay's treatment in the four central chapters of the book of two key
figures in post-War Western Marxian theorizing - Jurgen Habermas and
Alvin Gouldner - betrays an intermittent awareness of the political mar-
ginality, social and economic context and inherent redundancy of Marxian
thought in the contemporary world.
Very properly, Jay observes that (in an authentic Marxian tradition)
Habermas has disavowed as utopian any effort at giving substantive con-
tent to a post-capitalist order and has instead concentrated on the procedural
conditions of human self-transcendence and emancipation - conditions
comprehending an 'ideal speech situation' of unmanipulated communi-
cative rationality, a reduction in the role of scientific-technological reason
and an enhancement of the aesthetic dimensions of human life. As
Habermas puts it, tersely: 'A reified everyday praxis can be cured only by
creating unconstrained interaction of the cognitive with the moral-practical
and the aesthetic-expressive elements'. Commenting on this and si~lar
statements, Jay remarks mildly that 'Habermas owes us a more explicit
explanation of the aesthetic-practical rationality he wants to defend in
modernism. It is difficult enough', he goes on, 'to grasp what a mediated
relationship among cognitive-instrumental, moral-practical and aesthetic-
expressive rationalities would look like, even if they all might be simul-
taneously reintegrated with the life-world. It is even harder if the rational
status of the third remains somewhat of a mystery.' Jay's comments on
Habermas may be put in somewhat less hermetic terms. The idea of full
communicative rationality in an ideal speech situation among people is an
opaque one in Habermas's thought. Whatever it means, it is associated with
the Enlightenment expectation that in an undistorted dialogue human
beings will come to convergence in their values, projects and perspectives.
At no point in his prolix and voluminous theorizings does Habermas give
this expectation any foundation in reason. It hovers in mid-air, founda-
tionless, like the Cheshire Cat's smile, its supportive body detached and
destroyed by the very modernism Habermas is so anxious to defend. Nor is
his expectation of convergence on values and perspectives as the outcome
of dialogic rationality sustained by the actual circumstances of modem
societies, which all in varying degrees harbour a diversity of cultural
traditions. An open dialogue among these traditions is far more likely to
result in the emergence of powerful rivals to Habermas's modernist
humanism than to reinforce it. In reposing his faith in open dialogue, he has
the character of a sort of bien-pensant Pascal, laying a wager on reason
which nothing in our (or his) experience warrants.
Jay comes closest to confronting the insuperable limitations and social
functions of Marxian thought in his two chapters on Gouldner, the 'outlaw
marxist' sociologist. In his Against Fragmentation 3 , Gouldner sought to
94 Critiques
develop earlier analyses in which he attempted to bring to consciousness
what Jay calls 'the suppressed secret of Marxism's embarrassing social
origins in the radicalized bourgeois intelligentsia'. It is Gouldner's embar-
rassing thesis that the radicalization of the intelligentsia is a response to a
blockage in their upward social mobility rather than the expression of any
commitment to critical reason. Jay puts it thus:
As blocked ascendants, radical intellectuals from Jacobins onward often
looked to state service as a possible way to salvage their positions, while
at the same time serving the general cause of social rationalisation.
Ironically, even the culture of critical discourse harboured its own form
of potential elitism. For although selfconsciously egalitarian in terms of
its adherents, it was hierarchically related to those outside its own boun-
daries. The absolute commitment to rationality, at least as a regulative
ideal, could easily conflict with the specific material interest of the
non-intellectual agents .... Indeed, the very notion of a material interest
suggested an undiscussable given outside the realm of rational discourse.
This is far from being a novel position, being anticipated by the Polish
syndicalist theorist, W. Machajski and, as Gouldner himself notes, by
Marx's arch enemy, Bakunin. But it is entertaining to find it advanced by a
Marxist.
Predictably, Jay is concerned to defend Marx against Bakunin's claim
that Marxism expresses the class interest of a nascent pedantocracy, but
Bakunin's claim is well borne out by the history of American academic
Marxism. In the United States, a radical intelligentsia which perceives its
social status to be lower than that of business, which is accordingly
estranged from the dominant values of its parent culture, and which is
substantially marginalized from the political process, has attempted to
legitimate its interests and objectives by the adoption of an anti-market
ideology. This stratagem may be better theorized in the terms of the
Virginia School of Public Choice as a classic instance of rent-seeking, or in
the Paretian categories of residues and derivations, but that it has many
affinities with the practices analysed in the Marxian theory of ideology
needs little elaboration. It is, perhaps, the chief point of interest in Jay's
book that, whereas he considers Gouldner's application to the radical
socialist intelligentsia of the Marxian theory of ideology, he is enough of a
spokesman for Western radical academics to refrain from pursuing
Gouldner's demystificatory critique.
John Roemer's Marxism4 is in a very different vein from that pre-
dominantly inspired by the Frankfurt School which Jay seeks to chronicle
and interpret. It is a species of analytical Marxism, which aims to dispose
of the Hegelian inheritance which encumbers Marx's own thought and to
 The academic romance of Marxism 95
reconstruct the central propositions of the Marxian critique of capitalism
using only the tools of mainstream ('bourgeois') economics and philosophy.
In the writings of G.A. Cohen and Jon Elster, this analytical school within
contemporary Marxism has produced work of considerable intellectual
quality, even if it is sometimes doubtful how far Cohen and Elster still
deserve (or wish) to be called Marxists. By contrast, Roemer in both
methodological and political terms is a far more orthodox Marxist, in that he
aims with the help of game theory to rehabilitate the notion of exploitation,
if not in the technical Marxist sense - a project which Cohen has apparently
abandoned. Further, unlike Elster, he endorses the classical Marxist critique
of capitalism as chaotic and inefficient and (apart from a few nervous
references to the pros and cons of market socialism) makes few concessions
to the allocative efficiency of markets and seems to conceive of Marxian
socialism in the traditional terms of central economic planning. Although he
is ready to abandon the technical Marxian idea of exploitation and with it the
labour theory of value, Roemer's project is to reconstruct the hard core of the
Marxian critique of capitalism with the resources of standard micro-
economics and neo-classical equilibrium theory.
The first point to be made about Roemer's reconstruction is that, just like
the neo-classical model it deploys, it stands at a very great distance from
any real-world economy. Indeed, Roemer's distance from reality is
signalled before he even begins to develop his model, when on the first
page of the book he tells us that Marxism has achieved 'intellectual hege-
mony' across a third of the world, in that 'the world-view of Marxism based
on class, exploitation and historical materialism is pervasive in these (com-
munist) societies'. This sordid piece of intellectual flatulence, which
betrays an inability to distinguish between cultural and political hegemony,
will provoke hollow laughter in Budapest and Prague (not to mention
Bucharest and Peking), where oppositional movements and indeed ruling
elites are publicly or privately contemptuous of Marxism even if they
sometimes remain socialist in outlook. Such bizarre statements are clues to
the basic weakness of Roemer's project, which arises from its borrowings
from the most abstract and least realistic portions of conventional economic
theory. Along with the majority of his fellow economists who are not
Marxists, Roemer works with an eqUilibrium model of economic life which
abstracts from its most fundamental features - the scarcity and dispersal of
knowledge and the role of market institutions in transmitting and utilizing
it via the price mechanism. Because it assumes full knowledge of others'
preferences and resources on the part of economic agents, general equili-
brium theory models no possible state of the world. It is not an ideal type,
but a barely coherent fiction whose centrality in economics has served only
to distract attention from the actual workings of market processes.
96 Critiques
Using this model, Roemer can demonstrate that under certain fonnally
statable conditions resources will be allocated more efficiently by planning
than in any real-world market system, and he is able to point to features of
markets in the real world which explain their failures. Nothing in these
reasonings tells us anything of value as to the comparative allocative
efficiency of markets and central planning as alternative institutional
arrangements. Roemer tells us that, for a variety of reasons, capitalist
economies exhibit 'a vast mizallocation of resources'. The invisible-hand
properties of market systems presuppose a tendency in them to reach
equilibrium, he asserts, and this is frustrated by the uncertainties about the
future that market participants confront - uncertainties that cause people to
behave irrationally. Now eqUilibrium in markets is, doubtless, an asymp-
tote, which they will not always have even a tendency to move towards, but
what justifies Roemer's implicit confidence that economic planning will be
any improvement on markets? (If 'people' behave irrationally in the face of
uncertainty, will not planners do likewise? Or will they abolish the un-
certainty arising from our ignorance of the future?) When we return from
the dream world of Roemer's theorems to history and common experience,
we discover that every market failure present in capitalist systems, such as
environmental pollution, discoordination, worker alienation and under-
employment, is replicated on a far greater scale (though often covertly)
under socialist central planning institutions. These facts, which are
commonplaces under the Soviet glasnost, are as yet too subversive to be
mentioned by Roemer.
The chaos, waste and malinvestment characteristic of central economic
planning, which is the common historical experience of those socialist
states which are now attempting to resurrect market institutions, was accur-
ately predicted in the writings of Mises and Hayek and others in the
Austrian School. The Austrian argument was always that, for the very same
reasons that a perfectly competitive market is a fiction, so central economic
planning is an impossibility. No government can possess the infonnation
about relative scarcities, time preferences and consumer tastes that is trans-
mitted (however imperfectly) through price signals, but without which
economic planning is an exercise in disaster. Nor do moves in the direction
of market socialism much mitigate these dilemmas, as the Yugoslav and
Hungarian examples show. In practice, socialist economies behave exactly
as Austrian theory predicts - they rely on historic and capitalist prices, they
depend crucially on recurrent injections of Western capital and technology
and on a parallel economy of which the official economy is only the visible
tip. It is probably unnecessary to note that Roemer's references to the
failings of Communist economies are scanty and evasive. In general, he
proceeds as if the moves towards reinvention of market institutions in
 The academic romance of Marxism 97
China and elsewhere are irrelevant to his argument. In this, and in the
fetishistic formalism of his entire method, he fails to take his own advice,
when he remarks apropos of invisible-hand theorems: 'Models are simplifi-
cations of reality, and judgement must be exercised before boldly claiming
that the simplifications are appropriate for the policy question at hand'.
It is in regard to exploitation that Roemer's theorizing is most resistant
to contact with history and social reality. As he defines it, 'a person is
exploited if the labour that he expends in production is greater than the
labour embodied in the goods he can purchase with the revenues from
production'. What is most significant here is not whether he is successful in
formulating a conception of exploitation that serves the Marxist critique of
capitalism without incorporating the defects of Marx's own conception, but
that Roemer's conception makes the actual economic well-being of the
worker a matter of theoretical and moral indifference. This is to say that,
even if we envisage a socialist economy that is not disfigured (as all
actually existing socialisms are) by networks of graft and corruption, by
Third World levels of proletarian housing and medical care and by an
exploitative nomenklatura, the worker who is not exploited will almost
certainly have a lower level of economic well-being than a comparative
worker under capitalist institutions - a result that flows inexorably from the
disastrous allocative influences of central economic planning. This result is
the more significant if we recognize that exploitation exists under socialist
institutions. Roemer goes so far as to admit this, but assures his readers that
much socialist exploitation is an inevitable and 'socially necessary' inci-
dent in socialist construction, while the 'questionable' forms of socialist
exploitation which arise from differential status are no worse than those
found in capitalist societies. It does not occur to him to ask whether such an
outcome is worth the millions of lives lost during the most active periods of
socially necessary socialist exploitation. The upshot of Roemer's neglect of
the real world is that, with all its gimmick-ridden formalism, his re-
construction of Marxian economic philosophy has the intellectual rele-
vance of a dated crossword puzzle.
Taken together, the books of Jay and Roemer exemplify the poverty of
American academic Marxism - a Marxism innocent of history, politically
irrelevant and marginal in its own culture. That these characteristics are not
confined to American Marxism is shown by the work of Roberto Unger.
with its hare-brained schemes for the transfer of productive assets to
'rotating capital funds' - schemes whose practical implementation would
rapidly transform the United States into a replica of Unger's native Brazil.
It is in regard to American Marxism, however, that the character of aca-
demic radicalism in the United States is most readily apparent - its function
as the ideology of an aspirant intellectual renlier class, an ideology
98 Critiques
despised in every society that has had the misfortune to be subject to its
political hegemony. and which in the United States compensates for its
manifest political nullity by seeking hegemony within academic
institutions.
10 Philosophy, science and myth in
Marxism

INTRODUCTION
Feuerbach resolves the religious essence into the human essence. But the
human essence is no abstraction inherent in each single individual. In its
reality it is the ensemble of social relations.)
It is a common belief, shared both by Marxists and by critics of Marxism,
that differences in the interpretation of this statement have important impli-
cations for the assessment of Marx's system of ideas. How we read it will
affect our view of the unity of Marx's thought and of the continuity of its
development over his lifetime, and it will bear crucially on our appraisal of
the epistemological status - metaphysical, scientific or mythopoeic - of the
various elements of the Marxian system. Among Marxists, members of the
Frankfurt School have emphasized the paternity of Marxian metaphysical
humanism in Hegel's conception of man as a self-creating being, while
Althusser and his disciples have seen in the extrusion from Marx's later
work of any such 'anthropomorphic' notion a guarantee of the scientific
character of his historical materialism. Among Marx's liberal critics, it is
widely agreed that he espoused an essentialist view of man and, often
enough, it is thought that this alone is sufficient to disqualify his system
from scientific status. No consensus exists, however, as to the cognitive
standing of the several components of Marx's thought That agreement
should be lacking as to the place in it of a conception of human nature is
hardly surprising. Different construals of the role of a view of man will
reflect divergent commitments, not only in the philosophy and methodo-
logy of social and historical inquiry, but in moral and political thought as
well.
I do not aim to canvass systematically all the salient philosophical
problems posed by Marx's assumptions about man. More modestly, I aim
to offer an assessment of the epistemic status and mutual relations of the
principal parts of Marx's system, but to do this indirectly and obliquely by
100 Critiques
way of a discussion of the contributions to Marxist thought of three
important writers. Each of these writers is in a genuine Marxian tradition in
that he develops a theme which is undeniably present in Marx himself, and
each of them presents the Marxian system as a totality belonging to a
different mode of discourse. First, I shall consider some aspects of the
writings of Herbert Marcuse. Against the attribution by Althusser of an
epistemological rupture in the evolution of Marx's ideas, I shall contend
that a definite metaphysical view of man runs consistently through all of
Marx's writings. Not only does evidence from the body of Marx's later
writings show that he never abandoned the philosophical anthropology he
endorsed in the Economic-Philosophic Manuscripts: it can easily be shown
that some such doctrine is presupposed by his materialist conception of
history. But whereas Marx's historical materialism depends upon his meta-
physic of human nature, the latter cannot adequately support the former.
Indeed, Marcuse's writings suggest that emphasizing Marx's metaphysical
humanism tends to undermine the claim of his system to be a theory of
world history. In Marcuse's case, it yields an elegiac Marxism, from which
class struggle has been all but eliminated, in which social revolution has the
character of an improbably Pascalian wager, and whose primary uses seem
to be those of a weapon in the armoury of cultural criticism. This result is
to be accounted for, not by reference to the metaphysical status of the
central postulate of Marcuse' s system, but in terms of specific incoherences
in its philosophical anthropology.
Second, I shall consider G.A. Cohen's recent defence of historical
materialism in functionalist terms. Important weaknesses in Cohen's argu-
ment suggest the conclusion that, if Marx's system contains a theory of
universal history and not just an explanatory model of capitalist develop-
ment, it must trade on teleological explanations of a sort which social
science cannot countenance. Thus Hegel's view that human history has an
overall intelligibility deriving from the fact that it has a telos or end-state, a
view which Cohen expounds sympathetically in the first chapter of his book
(see note 21 to this chapter) but whose bearing on Marx's historical
materialism he does not systematically examine, is actually indispensable
to it in so far as it contains a theory of the development of human productive
powers through successive economic systems. The collapse of historical
materialism as a theory of world history carries with it the ruin of its central
thesis about the primacy of productive forces in explaining social change.
The upshot of my criticism of Cohen's book is that, whatever incidental
contributions to social science it may contain, Marx's system does not in its
main elements belong to the scientific mode of discourse.
Finally, I turn briefly to examine the interpretation of Marxism
developed by Georges Sorel, a neglected Marxist thinker rightly described
 Philosophy. science and myth in Marxism 101
by Croce as the most original and important Marxist theorist after Marx
himself. In Sorel we find a construction of Marx's system, in which Marx's
own activist and Promethean conception of man and his doctrine of class
struggle are fully preserved, but in which his historicist and scientistic
pretensions have been decisively abandoned. In Sorel's writings, the myth-
opoeic character of Marxism as the ideology of proletarian class struggle is
explicitly acknowledged and its source in a definite moralmadition identi-
fied. Sorel's Marxism is not without the difficulties connected with any
form of relativism in social and political theory. It appears to be involved
in paradoxes of self-reference, and I shall touch on a couple of these whose
implications are serious for Sorel's idiosyncratic version of Marxism. I
shall point to a tension in Sorel's thought between its mythopoeic and its
scientific or diagnostic aspects and I shall suggest that developing a realist
science of society involves abandoning some of the most distinctive claims
of Sorel's Marxism.
The programmatic conclusion of the paper, whose cogency I do not aim
to demonstrate but merely to support indirectly by way of a survey of three
Marxian writers, is that Marx's is an explosively unstable system of ideas
each of whose components has a distinct epistemological status and stands
in need of a different kind of support. Elaborating each of these strands in
Marx's thought produces such radically diverse varieties of Marxism that
we are justified in regarding talk of Marx's 'system' as little more than a
/afon de parler.

MARXISM AS METAPHYSICS: MARX AND MARCUSE ON

MAN AS A SELF-CREATING BEING
The practical creation of an objective world, the fashioning of inorganic
nature, is proof that man is a conscious species-being.... It is true that
the animals also produce. They build nests or dwellings, like the bee, the
beaver, the ant, etc. But they produce only their own immediate needs or
those of their young; they produce one-sidedly, while man produces
universally; they produce only themselves, while man reproduces the
whole of nature; their products belong immediately to their physical
bodies, while man freely confronts his own product. Animals produce
only according to the standards and needs of the species to which they
belong, while man is capable of producing according to the standards of
every species....
It is therefore in the fashioning of the objective world that man really
proves himself to be a species-being. Such production is his active
species-life. Through it nature appears as his work and his reality. The
object of labour is therefore himself, not only intellectually, in his
102 Critiques
consciousness, but actively and actually, and he can therefore contem-
plate himself in a world he has himself created. 2
It is clear from this and from other, similar passages, scattered throughout
the corpus of Marx's writings, that when Marx asserts that man's nature is
that of a maker, he intends his reader to understand far more than that man
is a tool-using and a tool-fashioning animal. Certainly, that man's reliance
on tools in the reproduction of his life-activity is part of what distinguishes
him from other animals is never contested by Marx: it is a distinctive
feature of human life to which he often refers. It may be supposed that
when, in a well-known passage in The Gennan Ideology,3 Marx observes
that men may be distinguished from animals by their consciousness, their
religion or anything else, but '[they] distinguish themselves as soon as they
produce their means of subsistence', this is what he has in mind. The greater
part of Marx's meaning is more plainly evident, however, when he returns
to the question of man's differentia specifica in Capital,4 declaring that
'what distinguishes the worst of architects from the best of bees is this: that
the architect raises his structure in imagination before he erects it in reality'.
In such statements Marx appears to be insisting on a sharp contrast between
human life and that of other animals:
The animal is immediately one with its life-activity. It is not distinct
from that activity; it is that activity. Man makes his life-activity itself an
object of his will and consciousness. He has conscious life-activity. It is
not a determination with which he directly merges. Conscious life-
activity directly distinguishes man from animal life-activity. Only
because of that is he a species-being.s
What separates the life of men and women from the life of other animals, it
seems, is the element of Hegelian negativity, itself dependent on the capa-
city for abstract thought, which allows them to distance themselves from
the behaviour, instinctual or conventional, which they inherit directly from
nature or society. It is this negativity and critical reflexivity which forbids
any inference from man's fixed characteristics as a biological species to his
essential life-activity, and which licenses the claim that man's nature is that
of a self-creative being.
It is only this understanding of the claim that man is the producer of
himself and of his world which enables us to see how Marx could contrast
his own standpoint with that of the old, 'contemplative' materialism. For,
whereas he speaks of man as being a part of nature, Marx's standpoint is far
from the caricature of German positivism preserved in the writings of
Engels and Lenin. It is more akin to the Hegelian and radically anti-
naturalist standpoint, in which nature is given to man only as an artifact that
 Philosophy. science and myth in Marxism 103
he has himself produced. In this perspective, the relation between man and
nature is theorized, not as a process in which a domain of inert objects
confronts a passive subject in whose consciousness it is reflected, but as a
transaction in which it is their own practice which gives nature all those of
its features which are recognizable by men. Far from subordinating men's
lives and history to laws given independently of their purposes or self-
interpretations, Marx tends to treat the whole domain of nature as a preci-
pitate of }.uman activities. So strong does this tendency become, so
powerful the influence of the mystical idea (derived by Hegel from Bohme)
of nature as man's larger body, that there is some basis for Kolakowski's
imputation6 to Marx of a species-relativism. In general, of course, Marx is
far from an Idealist denial of the independent reality of the external world.
Rather, in standard pragmatist fashion, he tends to regard the concepts and
categories embodied in any view of nature as artifactual, useful fictions
whose truth-value derives wholly from their contribution to the success of
human struggles with the world. This is not to deny that man comes to
self-consciousness to find himself situated in a natural order. It is to say that
his picture of the world is not something given to him as a datum, but
merges in his practical struggles with it, which in their tum contribute to the
progressive humanization of the world in an objective and factual sense. I
shall call the thesis that man is in this radical respect his own maker and the
shaper of his world Marx's activist thesis.
So far Marx's view of the nature of man would seem to be largely a
development of some of the main preoccupations of German Idealism.
Marx also picked out as among man's differentiae specijicae another
feature, owing more to the Romantic movement, according to which man is
distinguished by a vital need for productive labour, going well beyond
anything required in the struggle with brute scarcity. For Marx, the
estrangement of labour in class and, above all, in capitalist society, consists,
not only or primarily in the expropriation of its product, but in the fact that
men's life-activity is governed by the autonomous power of the com-
modity. It is not just that men must work in order not to starve that
constitutes labour's estrangement. After all, a rentier who does not have
that necessity is judged by Marx to be no less disabled as a productive being
than is a proletarian. Marx's view, rather, is that labour ceases to be
estranged only when it is the direct expression of man's nature, only when
its character is poetic or artistic. This aspect of Marx's view of man is
productivist, not only in the sense that he acknowledges that human beings
must secure the means of their survival before they can do anything else,
but crucially in that self-expressive labour is conceived to be an endo-
genous imperative of man's nature. This feature of Marx's conception
distinguishes his thought sharply from the outlook of the ancient world, in
104 Critiques
which human fulfilment was believed to lie in contemplative absorption in
a natural or divine order. It separates him no less clearly from the Classical
Economists, by whom labour was conceived typically as a distasteful
incident en route to consumption, and whose social ideal is best epitomized
in J.S. Mill's 'gospel ofleisure'.
The expressive view oflabour (which I have termed Marx's productivist
thesis about man) is easily seen as a corollary or entailment of his activist
view of man as in essence self-determining. Thus far Marx might seem little
more than a radical individualist, a liberal humanist opposed to the fossil-
ization of social life in oppressive institutions. To construe the Marxian
system in this way, however, would be a grievous misunderstanding. For,
in a drastic extension of the Hegelian tradition, Marx supposed that self-
determination for man involves the suppression or dissolution of all
distinct, autonomous spheres of social activity and modes of intellectual
life. He states this view unambiguously in Capital: 7
The religious reflex of the real world can ... only then finally vanish when
the practical relations of everyday life offer to man none but perfectly
intelligible and rational relations to his fellowmen and to nature.
The life-process of society, which is based on the process of material
production, does not strip off its mystical veil until it is treated as pro-
duction by freely associated men, and is consciously regulated by them
in accordance with a settled plan.
Here Marx asserts that man's self-determination is inhibited just in so far as
the objects and relationships of the social world have a life and laws of their
own. It is not too much to say that, for Marx, social science withers away
with the ending of alienation, precisely because its subject-matter - a
densely textured ambience of autonomous institutions, conventions and
practices - has ceased to exist Similarly, it is only mankind's pre-history
that can on this view be law-governed. To acknowledge that for human
beings the social world must always remain a world that is not of their own
making, that men's identities will always be constituted by the roles and
activities in which they find themselves, would be for Marx to acquiesce in
an utter loss of human freedom. It is not easy, all the same, to envisage what
the end of estrangement would be like, given that so much that is con-
stitutive of our lives as we know them would vanish. There is no doubt that
the Marxian conception of a de-alienated community would have been
resisted by Hegel himself as asocial and abstract to the point of virtual
emptiness. However this may be, Marx's view that full human self-
determination is incompatible with the survival of any independent spheres
of social life yields an important clue to his central understanding of
capitalism and of its dialectical negation. It is not in virtue of the
 Philosophy. science and myth in Marxism lOS
distribution of income or the form of property that it involves that capitalism
is regarded by Marx as the supremely dehumanizing mode of production. It
is because capitalist production has the character of an impersonal process
by which human subjects are constrained, and in which their transactions are
frozen or crystallized into reified forms of the species-life, that Marx. sees it
as the radical loss or negation of the human essence. The alienation of labour
cannot survive the end of commodity-production just because estranged
labour is defined as that which occurs whenever productive exchanges fail
to be the direct expression of organized human will.
What defines socialism. accordingly, is the overcoming of commodity-
production and its replacement by planned production for use. It is planned
production for the direct satisfaction of human needs (including the vital
need for expressive labour), production no longer mediated by monetary
exchanges or by impersonal laws of supply and demand, which is definitive
of the socialist order.' It was taken for granted by Marx, no doubt, and it is
a natural implication of his account of self-determination, that economic
planning in a socialist order would be democratic. (I leave to one side here
difficult questions about the decision-procedures, majoritarian or other-
wise, appropriate to the democratic planning of a whole economy, and I
refrain from commenting on the Arrow-type logical dilemmas such pro-
cedures would certainly generate.) It is crucially important to grasp, how-
ever, that the self-regulation of a socialist economy is in every respect
diametrically opposed to that which obtains in a market order or catallaxy.9
For, whereas the order that emerges from a series of market exchanges,
though it has its source in human actions, owes its most important
properties to the fact that it is not the result of human design, the order of a
socialist economy in the Marxian conception of it is supposed to embody
only the intentions of its constituent subjects. It is a fundamental criticism
of Marx's idea of a socialist economy that, even supposing human pre-
ferences and purposes could in the absence of class stratification attain a
measure of compossibility sufficient to obviate serious conflicts over the
allocation of resources, the suppression of market processes would deprive
economic planners of much information useful to the rational imple-
mentation of agreed projects. Whatever its difficulties. it is in the idea of the
subordination to collective human will of spontaneous economic processes
that the thrust of Marxian socialism is contained, and it is this thesis about
the economic consequences of human self-determination that I propose to
call Marx's anti-autonomist thesis.
Each of these three facets of Marx.' s conception of man is fully preserved
in Herbert Marcuse' s version of the critical theory of the Frankfurt School.
The activist thesis is expressed in Marcuse's constant emphasis on the
creative role of man as the subject of history, the agent of the creation and
106 Critiques
continuous transformation of the fonns of his social life, and his rejection
of any image of man's self-development which pictures it in tenns of
submission to the necessity of historical laws. Like Marx, Marcuse argues
that it is the reduction of human relations to relations between things and,
in particular, to relations between commodities, that constitutes the essen-
tial inhumanity of capitalist society, and which, in reducing proletarian
life-activity to a commodity on the labour market, generates the 'deter-
minate negation' of the capitalist order. In Marcuse, as in Marx himself, the
conceptual foundation of historical materialism is a form of metaphysical
humanism, which conceives of history as the progressive disclosure of
man's defining species-powers, but which acknowledges that the activities
of empirical social groups and individuals may conceal as much as they
reveal of man's essential nature and possibilities. Marcuse is in Marx's own
authentic tradition in his strong emphasis on the functions of ideological
false consciousness in inhibiting men's reflection on their circumstances as
alienated labourers and thus in reproducing an irrational social order. In
elaborating an ambitious theory of ideology, Marcuse follows Marx in
presupposing a set of species-powers whose flourishing it is precisely the
role of false consciousness to prevent. Like Marx and Hegel, Marcuse takes
for granted that these powers of critical reflection remain latent in most
societies, and are manifestly present only in relatively highly complex
societies; but he is in no doubt that some such account of man's species-
being is presupposed by any theory of ideology. His work shows clearly
enough that both the logic of historical materialism and its development as
a theory of ideology require a philosophical anthropology of just the sort
Althusser's disciples seek to suppress from Marxism. In Marcuse, too,
fmally, we find Marx's productivist emphasis on the intrinsic value of
self-expressive labour, and the anti-autonomist conception of a rationally
reconstructed and holistically administered economic order.
In laying emphasis on the metaphysical humanist aspect of the Marxian
system, however, Marcuse retreats from its historicist claims and abandons
some of its most distinctive theses about the role of proletarian class
struggle in the transition to socialism. The dilemma in which Marcuse's
abandonment of Marxian historicism places him is well illustrated in his
characteristic combination of repeated emphatic statements of the invin-
cibly totalitarian character of modem capitalism with frequent affirmations
of the real possibility of total social revolution and reconstruction. On the
one hand, Marcuse has declared that 'the vital need for revolution no longer
prevails among those classes that as the "immediate producers" would be
capable of stopping the capitalist production'. 10 'Marx's conception of
revolution', he continues 'was based on the existence of a class which is not
only impoverished and dehumanized, but which is also free from any
 Philosophy. science and myth in Marxism 107
vested interest in the capitalistic system and therefore represents a new
historical force with qualitatively different needs and aspirations'.1\
According to Marcuse, the emergence of such an 'internal negative force'
is blocked in advanced industrial society - 'not by violent suppression or by
terroristic modes of government, but by a rather comfortable and scientific
coordination and administration'. 12 As a result of contemporary society's
'highly effective scientific management of needs, demands and satis-
faction' ,\3 'the internal historical link between capitalism and socialism ...
seems to be severed' ,14 so that 'the idea of the availahle alternatives eva-
porate into an utterly utopian dimension in which they are at home' .15 Here
Marcuse is asserting that, owing to the success of modem techniques of
demand management, in which even the unconscious needs and instincts of
the population become liable to manipulation, the traditional oppositional
class in capitalist society has become attached to the circumstances of its
servitude. Thus alienated labour may, with the perfection of industrial
relations techniques, come to be experienced as rewarding and fulfilling.
On the other hand, social groups exist which will seek to exploit the utopian
possibilities which Marcuse believes to be inherent in modem capitalism.
In One-Dimensional Man, Marcuse discovered the last vestiges of revolu-
tionary protest within 'the substratum of the outcasts and outsiders, the
exploited and persecuted of other races and other colours, the unemployed
and unemployable' .16 In An Essay on Liberation, Marcuse opines that, as a
result of the advance of automation, 'a general unstructured, unorganized
and diffused process of disintegration may occur'. 11 Marcuse proposes that
an intellectual vanguard exploit the new contradictions of late capitalist
society by implementing a policy of discriminating tolerance toward con-
servative and reactionary forces and forge an alliance with other opposi-
tional groups. The new society will be inaugurated by a government
consisting of an educational dictatorship of free men and women which,
guided by a radical transvaluation of values and inspired by a new sensi-
bility, can take advantage of the vast technological resources of modem
society and avoid the repression which has disfigured all previous revolu-
tions. In Marcuse's last writings, the social revolution powered by the entry
into radical politics of a minority of powerless outsiders, which in One-
Dimensional Man he had described as 'nothing but a chance', 18 is again
treated as a utopian possibility against which virtually all existing institu-
tions and social forces militate.
Marcuse's thought clearly illustrates the tensions inherent in any
revision of the Marxian system in which metaphysical postulates are given
epistemological priority over social theory and historical interpretation.
Socialism becomes a free-floating possibility, an abstract and avowedly
utopian prospect, grounded only in a speculative distillation of the human
108 Critiques
essence. Shorn of the Marxian theory of transition, in which the proletariat
is acknowledged to be the class in which mankind's universal interests are
embodied, Marcuse's Marxism resembles nothing so much as a Left
Hegelian radical humanism. In detaching the transition to socialism from
any sort of proletarian activity, again, Marcuse (like Stalin) revives a
pre-Marxist Jacobin tradition in which workers become passive objects of
their own liberation. The dilemma of Marcuse's thought results from his
recognition of the superannuation of the Marxian system conceived as a
theory of capitalist development and a scheme for the interpretation of
history coupled with his refusal to regard any empirical development as
constituting any sort of criticism of Marxism's metaphysical core. Against
this criticism, no doubt, it will be urged that Marcuse is at once more and
less authentically Marxian than I have pictured him as being. Thus he has
always maintained that working-class activity and initiative are indispens-
able to socialist revolution, just as Marx did. On the other hand, his theory
of human nature owes much to Freud, and his theory of a 'technological
eros' is oriented around the concept of play rather than that of labour. In
reply, I would point out that Marcuse's writings contain some of the same
tensions that haunt Marx's work. Like Marx, Marcuse seems to vacillate
between a conception of man as Homo laborans whose nature has been
thwarted by millennial class exploitation and a view of man in which his
possibilities are realized only by emancipation from labour into playful
freedom. In the one case, the ending of economic alienation is seen as
necessarily involving some sort of self-management and transformation of
daily labour into an inherently valuable experience, while in the other the
emphasis will be laid on the progressive reduction of labour time. Both
strands of thought about work are authentically Marxian, and it is un-
necessary to invoke Freud's influence on Marcuse to account for the latter's
emphasis on the intrinsic virtues of playful activity.
With regard to Marcuse's insistence on the necessity of working-class
initiative in the making of a socialist revolution, it must be said that this has
never carried much credibility. If the working class has any active role in
Marcuse's scenario, it is after the socialist revolution - in the period of
socialist construction. In denying to the working class any important role in
the earlier phases of socialist revolution, Marcuse is merely reasserting his
conviction of the integration of working-class people in advanced capitalist
society. Surely, however, even a sympathetic critic of Marcuse will feel that
a more perceptive response to evidences of this integration would have
been to undertake a revision of some of his most fundamental Marxian
commitments.
Now I do not mean here to invoke any naive falsificationist criterion of
the adequacy of Marcuse's (authentically Marxian) metaphysic of human
 Philosophy, science and myth in Marxism 109
nature. To call Marcuse's view of man a metaphysical view, after all, is to
say that no criticism of it which appeals solely to experience can ever be
decisive; it is to say that it is a view which structures experience and which
gives a framework to empirical evidence. This is not to deny that empirical
evidences can have salience to a criticism of Marx's view of man, which
may be evaluated in Lakatosian fashion as being part of a progressive or a
degenerating research programme. Such an evaluation will not be attemp-
ted here, but an outline of it is easily sketched. We will have reason to
regard the conception of man expressed in Marx's writings, and in such of
his followers as Marcuse, as being part of a degenerating research pro-
gramme, if there are philosophical reasons for revising it which bear on
available empirical evidence. To put the point dogmatically, Marx's view
of man as in essence an unconditioned and self-determining agent brings
with it all the confusion of Idealist logic and ontology. Kamenka' s criticism
of Marx on this point is worth quoting:
Man, as Marx in his metaphysical moments portrays him, is (potentially)
the unconditioned being of the scholastics (i.e. God), whose un-
conditionedness is one of his perfections, essential to his (true) nature,
and therefore to be deduced from it. It is from the scholastic view of God
that Marx unconsciously derives the conception of man as (properly)
always a subject and never a predicate. It is from scholastic logic that he
gets the otherwise unsupported notion that the self-sufficient, the self-
determined. the always active. is morally superior to the conditioned, the
determined. the also passive. '9
More plausibly, perhaps. Kolakowski has suggested20 that Marx's view of
man is bound up inextricably with a Platonist ontology in which the world
is conceived as a great chain of being. in which degrees of reality are
recognized and are conflated with degrees of goodness. and in which the
human species recapitulates in an historical theodicy the return to an
undifferentiated unity pictured as the destination of Spirit in neo-Platonic
and Hegelian writings. The objection to Marx's view of man, then. is not
the crass empiricist objection that its content has an essentialist or meta-
physical aspect, but rather a philosophical objection of a substantive sort. It
is not one that can be defended here, and I will not try to defend it. Perhaps
it will suffice to say that the claim that Marxism constitutes a degenerate
research programme rests on a conjunction of philosophical criticism of its
underlying logic and ontology with evidences and reasonings about politi-
cal life which suggest the incoherence of the Marxian idea of a return to a
lost unity in a conflict-free union of civil and political society. The crucial
objection here has been well put by Kolakowski in his critique of the myth
of human self-identity as foundational in the Marxian project of a
110 Critiques
post-political society. This is a criticism which need not draw directly on
philosophical reasonings, since it may be argued that the transcendence of
self-division, which is the project of Marx no less than of Rousseau, is
unachievable except at a price which Rousseau acknowledged but Marx did
not - namely, a retreat to much lower levels of critical self-awareness in a
far simpler social order. This suggests an ironical criticism of Marx's
historical interpretation in virtue of its failing to take seriously the irrever-
sibility of cumulative conceptual enrichment through the increasing moral
complexity and cross-cultural sensibility of human society. There is a
further strand of criticism salient here too, one which draws on the writings
of Durkheim and Weber, in which it would be argued that role-occupance
of the constraining sort Marx condemns is not merely functionally in-
dispensable to industrial societies, but actually largely constitutive of
human self-identity in any conceivable society. Despite their abstractness,
these are quasi-empirical rather than straightforwardly philosophical pro-
positions and ones we are in a better position to evaluate than Marx was.
This is to say merely that, if we are correct in judging Marx's system to
embody a degenerate research programme in our own context, nothing
follows directly as to its credentials in Marx's own time.

MARXISM AS SOCIAL SCIENCE AND THEORETICAL

mSTORY: COHEN ON HISTORICAL MATERIALISM AS A
SPECIES OF FUNCTIONALISM
In G.A. Cohen's recent restatement of Marx's historical materialism in
functionalist terms, the dependency of Marx's system on a view of human
nature is explicitly acknowledged, though its metaphysical content is
largely suppressed. Cohen notes that there is 'a Marxist tradition to deny
that there exists an historically invariant human nature', but insists that
there are 'enduring facts of human nature'21 - facts which Marx's system
depends upon, and which are not controverted by the insistence on the
partial self-transformation of man in history. As Cohen puts it, 'man is a
mammal, with a definite biological constitution, which evolves hardly at all
in some central respects throughout millennia of history' .22 How does
Cohen construe the place of a view of human nature in Marx? He states as
the central claim of historical materialism what he calls the primacy thesis,
which asserts that 'the nature of a set of productive relations is explained by
the level of development of the productive forces embraced by it (to a far
greater extent than vice versa)'.23 He tells us that the primacy thesis is
associated with a second thesis, which he calls the development thesis and
which he states as follows: 'The productive forces tend to develop through-
out history' .24 He specifies this as entailing more than that productive forces
 Philosophy, science and myth in Marxism III
have developed. For, whereas this might have happened 'for a miscellany
of unco-ordinated reasons',25 the development thesis asserts a universal
tendency to development of the productive forces. It does not entail that the
forces always develop, since, as Cohen notes, citing the decline of imperial
Rome as an example, there are 'exceptions to the generalization that the
productive forces, though indeed capable of stagnation, do not, barring
natural disaster, go into reverse'. Thus, as Cohen puts it succinctly, the
development thesis 'predicate[s] a perennial tendency to productive pro-
gress, arising out of rationality and intelligence in the context of the
inclemency of nature' .26 As to apparent counter-examples, a theory of
history is not answerable to abnormal occurrences, Cohen tells us, though
he admits that there is an as yet unsatisfied need for criteria of normalcy for
human societies.
Cohen tells us that the development thesis, for which criteria for abnor-
mality in society or history seem to be indispensable, is itself a necessary
support for the primacy thesis. What then supports the development thesis?
Cohen declares that 'we put it as a reason for affirming the development
thesis that its falsehood would offend human rationality'. 27 The develop-
ment thesis gains plausibility, he says,28 if we note three facts about men:
that they are somewhat rational; that their general historical situation is one
of scarcity; and that they possess intelligence of a kind and degree which
enables them to improve their lot. Cohen observes that the thesis that men
are irrational if they fail to exploit the opportunity to expand productive
forces whenever the growth of knowledge allows this has two difficulties.
We do not know 'the relative magnitude of man's material problem and
consequent interest in its solution, by comparison with other human prob-
lems and interests. . . . Whether the falsehood of the development thesis
would offend rationality demands a judgment of the comparative import-
ance of potentially competing interests' .29 Cohen identifies a second diffi-
culty when he observes that 'it is not evident that societies are disposed to
bring about what rationality would lead men to choose. There is some
shadow between what reasons suggests and what society does' .30 Cohen
says of his defence of the development thesis that it is 'not conclusive, but
it may have some substance' .31 This modest disclaimer fails to do justice to
the difficulties in which Cohen puts himself. They all arise from the fact
that the conception of rationality with which Cohen works, though it is
intended by him to have a universal application, has the content he needs
only if the values of some elements of our own culture are fed into it.
Otherwise, it is an almost completely empty, indeterminate and pro-
grammatic conception, and the thesis that productive relations tend to alter
to allow for the most efficient use of productive forces becomes vacuous
rather than contingently false. The notion of more or less efficient uses of
112 Critiques
productive forces makes sense only when, as in problems of engineering,
there is agreement on objectives. Even within our own culture, it is a salient
feature of social problems, including problems of production, that typically
they arise from conflicts of ways of life rather than from the kind of
technical disagreement that might occur between engineers charged with
the completion of a common project. 32 Consider in this regard current
controversy about factory farming and about the elimination and replace-
ment of small family farms by large agricultural corporations.33 To suppose
that discussion of the merits of rival forms of agricultural enterprise can
proceed by reference to some common standard of efficiency is to suppose
what is patently false, that farming has a single undisputed purpose. Under-
standing current controversy in this area involves recognizing that rival
conceptions of farming are at issue, each of which is bound up with the
defence of a broader way of life. An appeal to some over-arching standard
of efficiency in satisfying express preferences could settle such dispute
only if there were a common market to which all demands could be brought
and a reasonable bargain struck.34 No criterion of efficiency could be at the
same time neutral in a conflict of this sort and yet authoritative for all
parties to it. For, while contending social movements and ways oflife often
come to a worlcing compromise, such a settlement is wholly different in
character from a market-place bargain, since it often represents only an
abatement in the struggle between incommensurable claims.
These difficulties are even more formidable when an appeal to effi-
ciency is made, not to settle practical conflicts within a given culture, but to
help to explain transitions from one social order or historical epoch to
another. When capitalist enterprise emerged from feudal societies in
Europe, protagonists of the rival forms of economic life did not self-
evidently share objectives and problems for which capitalism proved to be
the most cost-effective solution. Those who think that human history as a
whole can be understood as a movement from less to more efficient uses of
expanding productive forces take for granted a common standard of effi-
ciency where one is not even conceivable. None of this is to suggest that
immanent criticism of a culture or a way of life cannot proceed by invoking
efficiency. It is not to endorse some quasi-Parsonian view in which the
rationality of any social arrangement is indistinguishable from its sheer
persistence. It is to say that a standard of efficiency is lacking which can be
applied cross-culturally or even (within a single society) between ways of
life expressive of incommensurable values.
These difficulties are not met by Cohen's observation that the develop-
ment thesis depends on a judgement about the relative importance in men's
lives of material and other interests. A sharp distinction between material
and other interests is apt only in a secular and post-traditional society of just
 Philosophy, science and myth in Marxism 113
the sort in which capitalist enterprise is typically dominant. Even within a
society of this sort, applying such a distinction presents problems, not of
imprecision or of open texture, but of sheer indeterminacy. If an affluent
but harassed businessman relinquishes a high income for a more relaxed
and leisurely life in the country, is he settling a conflict between his
material and his non-material interests in favour of the latter? If workers
object to the introduction ofTaylorian techniques of time-management into
the life of the factory, and their opposition is not dampened by a credible
offer of higher wages, is this a case where material interests are foregone
for the sake of other goods? And are we to understand that the path of
rationality lies in always favouring material interests, supposing a rough
and ready demarcation criterion between these interests and others is avail-
able? If not, what sense is there in the claim that the falsity of the develop-
ment thesis would be an impeachment of human rationality? Invoking
criteria of normalcy in history or in society (there is a disturbing slippage
here) is unhelpful, since we lack criteria which satisfy Cohen's requirement
of yielding the development thesis without question-beggingly pre-
supposing it. Finally, even if a workable classification of human interests
existed, and it were true that men always tended to favour their material
interests over others, still the development thesis would not follow. Cohen
alludes to Marx's apparent belief that it was population growth which, in
disrupting the balance between needs and natural resources distinctive of
primitive Communism,35 spurred the adoption of what he calls 'a more
aggressive technology'. An equally effective way of remedying the im-
balance caused by demographic growth, and a way which has been
favoured by the overwhelming majority of traditional societies, is the
institution of population controls. Unless a parochially European pre-
occlJpation with human mastery of the natural environment is written into
the conception of rationality with which Cohen works, I do not see how
such a response can be disqualified as irrational. The conception of Homo
economicus which Cohen's conception of rationality comprehends has, in
fact, no tendency to support Marxian productivism.
That Cohen works with an ahistorical, unMarxian and almost Bentha-
mite conception of rationality emerges from his cursory consideration of
the problematic aspects of any criterion for the assessment of productive
power. Productive power, he tells us, is 'the amount of surplus production
the forces (of production) enable', where surplus production is understood
in this context to signify 'production beyond what is necessary to satisfy the
indispensable physical needs of the immediate producers, to reproduce the
labouring class'. Thus, he declares, 'the development of the productive
forces may be identified with the growth of the surplus they make possible,
and this in tum may be identified with the amount of the day which remains
114 Critiques
after the labouring time required to maintain the producers has been sub-
tracted'.36 Now it is no news to Cohen, any more than it was to Marx, that
criteria for 'subsistence' inescapably contain cultural elements: for, apart
from anything else, differing forms of social life will make differing
demands on men's bodies, and so necessitate differences in diet and so on.
Further, it is not obvious on the face of it what time-span is to be used in
assessing a productivity increase. These are problems of which Cohen is
clearly aware, and which might be thought to pose no fatal threat to his
proposed criterion. The real difficulties lie elsewhere. First of all, the
efficiency or productivity criterion he suggests needs to be argued for: why
adopt this criterion rather than any other? It is not a criterion whose salience
is self-evident. Men have nowhere consistently regarded a shortening of the
working day, or an increase in their capability to support ever larger
numbers of human beings, as constitutive of their welfare. How could the
priority of such goals over other goals - over goals to do with national
grandeur, spiritual development and the preservation of ancient traditions,
for example - possibly be constitutive of rationality? Second, in pre-
supposing our ability to make on-balance aggregative judgements about
socially necessary labour time, Cohen's proposed productivity criterion
evidently involves just that weighing of incommensurables which it was its
purpose to circumvent.31 The key point, however, is that, even if Cohen's
productivity criterion can be coherently spelt out, he has said nothing to
show that increasing productivity is always (or paradigmatically) rational.
He needs to do this, given the distance between the productivity criterion
and ordinary (culturally and historically variant) conceptions of rationality,
if his proposal is to have any intuitive forcefulness.
It may be from an awareness of these difficulties that Cohen affirms that
'the fact that capitalism did not arise spontaneously outside of Europe is a
serious problem for historical materialism' .38 Unless we begin by ascribing
to human history an end-state for which capitalism is an indispensable
condition, there seems no reason why capitalism should not be regarded as
a unique episode. This view of the matter is given added plausibility when
we recall that Marx himself regarded the expansionist and productivist
imperatives which Cohen deploys in an attempt to develop a scheme of
universal historical interpretation as distinctive of and peculiar to the
capitalist system. 39 The dynamism which capitalism displays as a system
cannot be used to account for its emergence in the first place, given what
Marx describes40 as the essential conservatism of all pre-capitalist modes of
production. On one occasion, indeed, Marx apparently disavows any
generalization of an explanatory model which is in place when it is applied
to the workings of the capitalist system to the domain of human history as
a whole. If this statement of Marx's4 1 is taken as an authoritative evidence
 Philosophy, science and myth in Marxism 115
of his intentions, we have reason to regard historical materialism as a theory
of capitalism rather than a scheme of historical interpretation.
If the development thesis lacks rational suPPOrt. what of the primacy
thesis which Cohen rightly takes to be the corner-stone of Marx's historical
materialism? Cohen specifies the explanatory relation postulated in his-
torical materialism as holding between productive forces and productive
relations as being a functional rather than a causal relation. In part this
claim, along with the construal of the Marxian model of society as tripartite
rather than bipartite in which it is embedded, is intended as a rebuttal of
some criticisms of historical materialism. Cohen aims to answer, especi-
ally, writers such as Acton, Plamenatz and Nozick, who maintain that
historical materialism's explanatory pretensions are nullified by the con-
ceptual impossibility of making the sorts of distinctions it presupposes.
Thus he maintains against such critics that a characterization of production
relations is available which is recht/rei in that it does not contain those very
things (normative or superstructural aspects of social life) that it purports to
explain. It is still unclear, I think, that Cohen's description of the rights and
powers of proletarians in a capitalist order, for example, is recht/rei in the
sense he needs if capitalist production relations are to be describable
independently of the legal and ideological forms in which they are found.
Cohen is probably right, then, when he contrasts his own account of
production relations with that which Engels stigmatized as the 'force
theory' in Anti-DUhring,42 but it may be doubted if the force theory (whose
absurdity is manifest when it is assessed as an account of the production
relations characteristic of a whole epoch or mode of production) can be
avoided without embracing an account which remains open to the criti-
cisms Cohen is concerned to rebut. I do not want to pursue these questions
here, however, since nothing in my argument turns on how they are
answered, but merely to look more closely at the functionalist re-
formulation of the primacy thesis which Cohen advances.
To say that production relations and legal and political institutions have
a functional connection with productive forces is to say that they allow the
latter to develop and expand according to their potential powers. This is a
technological reformulation of Marx's materialism which captures at least
a part of Marx's meaning, and which allows for critical discussion. It does
not treat technological innovation as explanatory of all great social changes
but as itself beyond explanation, but asserts that, once the disposition of
human beings to conquer natural scarcity has brought about a new set of
productive forces, production relations and the rest of society will adjust so
as to permit their most efficient deployment. What is lacking in the account
so far is any elaboration of the mechanism whereby this is supposed to
occur. Cohen tries to supply this with an Engels-type argument that the
116 Critiques
explanatory form of Marx's materialism is akin to that of Darwin's evolu-
tionary theory,·3 but the argument has two crippling disabilities. One of
them has been noticed by Peter Singer," when he points to the mis-
conception of Darwin's theory contained in Cohen's statement of it.
Darwin did not (as Cohen suggests) explain the long necks of giraffes by
saying that long necks have the function of enabling giraffes to survive, for
to say that would be to say nothing. Darwin's explanation of the long necks
of giraffes was in terms of the action of natural selection on giraffes with
necks of varying lengths - an explanation which displaces any functional
explanation. Contrary to Cohen, a theory of evolution by the natural selec-
tion of the products of random genetic mutations renders any functional
explanation otiose in these domains.
More fundamentally, Cohen's reformulation identifies no mechanism,
akin to that of natural selection of genetic accidents in Darwinian theory,
whereby more efficient uses of productive forces replace less efficient
ones. To be sure, such a mechanism exists within capitalist economies in
which firms behave as Marx, following the Classical Economists, thought
they would behave. In a competitive capitalist environment, it may well be
true (as Cohen suggests) that adopting more efficient methods for the
exploitation of available productive forces may be a condition of a firm's
survival. The same competitive pressure may indeed account for techno-
logical innovation and the spread of new technologies throughout the
economy as well as for fInDS seeking to make the best use of existing
technology. (This latter is a point Cohen does not make.) Where the
mechanism of market competition is lacking, we have no reason to expect
that productive forces will gravitate to their most efficient uses. Cohen tells
us that this might take place in the absence of market competition, if, for
example, the central planners of a socialist economy were to decide to take
advantage of economies yielded by increases in the scale of production.
That this could happen cannot be disputed, but - as the complaints of
dissenting Soviet-bloc economists confirm - socialist economies contain
no mechanism in virtue of which it is bound to happen, or which generates
any persisting tendency in favour of such a development. When Cohen
speaks of this, the only case he mentions of a move to more efficient uses
of productive forces happening in the absence of competitive pressure, he
gives the game away by calling it a purposive elaboration of a functional
explanation. It clinches the suspicion that, except when its subject matter is
purposive behaviour and its cybernetic simulations, functional explanation
is out of place in social science. A crypto-Darwinian (but not a functional)
explanation may be given for technological innovation and for production
by another which is more efficient. The productive efficiency of a metro-
politan capitalist economy may enable it to overwhelm less efficient
 Philosophy. science and myth in Marxism 117
peripheral economies by a variety of means, including military force, but
that is another matter. Consideration of productive efficiency cannot
account for the emergence of the capitalist mode in the first place, and it
would be fanciful to suppose that they explain its extension into hitherto
traditional economies. More unblinkingly than Marx himself, Cohen mis-
takes an imperative of the capitalist system for a universal tendency. There
is nothing in Cohen's argument which adequately supports his claim that
Marxian functionalism lacks the conservative complications of func-
tionalism's non-Marxian variants.
I do not claim that the corpus of Marx's writings contains nothing of
interest or importance in the way of scientific analysis of the capitalist
system. Marxian trade-cycle theory, for example, though it remains
eminently controversial, is part of a theory of capitalist crisis and break-
down which may still be worth studying, but nothing in it warrants the
belief that capitalism's successor will be a more efficient mode of pro-
duction. Even the full validity of the Marxian theory of capitalist crisis
would not guarantee the development thesis. All this suggests that, quite
apart from any of the difficulties I have mentioned in specifying the exact
content of Cohen's functionalist statement of Marx's historical material-
ism, the scientistic and positivistic version of the Marxian system which
Cohen develops cannot support its role as a theory of history. It might be
replied that the weakness of Cohen's defence lies in the view of human
nature with which he works - a view which he admits45 is not to be found
elaborated in Marx's writings, though he claims they contain much that is
consonant with it It might be thought in other words, that Cohen's failure
is to write into his view of human nature a strong endogenous productivist
imperative. Even if this were done, however, it would not yield anything
like an historical law , or even a persisting tendency, unless other, competing
human dispositions were reliably defeated. This could be thought to be so, I
suggest, only if the expansion of human productive powers were an end-state
ascribed to human history as a whole. Unless human history has an imma-
nent goal or purpose, unless teleological explanation is appropriate in respect
of vast multigenerational changes, the Marxian theory of history fails. The
conclusion is irresistible that historical materialism has the character of an
historical theodicy and not of a contribution to a science of society.

MARXISM AS THE IDEOLOGY OF PROLETARIAN CLASS

STRUGGLE: SOREL ON MYTH AND MORAL REGENERATION
IN THE SOCIAL REVOLUTION
In Sorel's Marxism we find a species of revolutionary socialist ideology
which, as Isaiah Berlin has observed in a luminously perspicuous essay,46
118 Critiques
breaks with two of the most fundamental beliefs of the Western intellectual
tradition - the Greco-Roman belief that knowledge liberates and the
ludaeo-Christian belief in an historical theodicy - that inform and shape
Marx's own thought to its very foundations. Yet Sorel's claim to be a
follower of Marx was neither fanciful nor perverse. In at least two impor-
tant respects, Sorel develops distinctive and valuable features of Marx's
thought.
First, Sorel expounds in a sharp and clear form the metaphysical concep-
tion of man as a creative and self-creative being which I have claimed is a
logical presupposition of historical materialism. Against the thinkers of the
French Enlightenment, against Bentham and his followers among the
Philosophic Radicals, Sorel affinned a view of man as active rather than
passive, struggling rather than enjoying, doing and making rather than
contemplating or absorbing sense-data. Human action he seems to have
conceived as erupting from an inner necessity to imprint a unique mark on
the world. Nothing could be further removed from the inertial conception
of human action as a response to external stimulus or internal deprivation,
which we find in Hobbes, Bentham or Holbach. Sorel greatly valued natural
science, which he pictured in Marxian pragmatist fashion as a weapon in.
which human order is imposed on an inherently formless natural world, but
he had nothing but disdain for projects for a social science wherein human
conduct would be subsumed under the intelligibility of impersonal and
abstract laws. Sorel takes from Marx, then, his view of man as a creature
standing apart from the objects of the natural order, as distinct from other
animals in that his life is, first and foremost, an expression of a primordial
creative impulse.
Sorel takes from Marx, second, what I might call an agonistic view of
society. It is viewed, not as a perpetual-motion machine in which a rational
harmony of interests produces continuing progress in the satisfaction of
wants, but as an arena of conflicting classes, moralities and world-views.
One implication of the pluralistic and conflictual model of social develop-
ment Sorel derived from Marx is that, for Sorel as for Marx, socialist
revolution has a social rather than a political character. It is not a coup
d'etat or a putsch, the fruit of a conspiracy of deracine intellectuals, but the
culmination of a long period of class struggle. More radically than Marx,
Sorel attributed an anti-political character to the social revolution. It was
not just that revolutionary syndicalism, as Sorel imagined it, repudiated the
prudent and temporizing politics of the Third Republic, but that social
revolution was construed as an expressive and not a purposive act, a gesture
of independence and solidarity to the assessment of which its consequences
were virtually irrelevant. In his strong emphasis on the internal moral
qualities of class struggle, Sorel is doubtless influenced by Proudhon, but
 Philosophy. science and myth in Marxism 119
his anti-Utopian indifference to blueprints for a future socialist order is
authentically Marxian. More decisively than either of the two Marxist
writers I have discussed, Sorel grasps Marx's distinctive insight that social
development is the story of conflict between distinct social groupings, with
different places in society's productive apparatus and having contending
moral outlooks and views of the world. He develops Marx's accent on class
struggle - an aspect of the Marxian system which tends to fall out of
Marcuse's and Cohen's reformulation of it.
So far, I have maintained that Sorel adopts Marx's activist view of man
and the pragmatist view of knowledge that goes with it in Marx. He
develops Marx's doctrine of class struggle into an anti-Utopian critique of
ideas of social unity and rational harmony among men which goes well
beyond Marx's own intentions. He shares with Marx an anti-instrumental
and expressive interpretation of human action and, above all, of human
labour, whose centrality in men's lives he doubted as little as Marx did.
Where Sorel departs from Marx he is, in my view, on his strongest ground.
He relinquishes any historiosophical pretensions for his conception of class
struggle and goes so far as to condemn Marx for an excessively determinist
account of social development. Sorel's voluntarism is related to a deeper
difference from Marx, namely the primacy he gave to will over intellect in
human affairs. In insisting on the importance of myth as a non-rational
moving force in society and history, and in belittling the role of prudence
and of rational calculation of self-interest, Sorel broke decisively with the
intellectualist psychology Marx inherited from the Enlightenment and
which he attempted (somewhat incongruously) to marry with the expres-
sive conception of action he borrowed from Hegel. Sorel is influenced by
Schopenhauer and by Nietzsche in treating the conceptions men form of
their social circumstance as being moulded primarily, not by disinterested
reflection on an available body of evidence which is neutral as regards
conflicting purposes, but by their hopes and needs. Sorel's conception of
myth is akin to Marx's conception of ideology and to Freud's notion of
illusion in that the truth content of beliefs, images and symbols so charac-
terized is not denied but regarded as irrelevant to their function in the
overall economy of human activity. Sorel's departures from the intellec-
tualist psychology of the Enlightenment, however, lead him to view myth
in a fashion radically divergent from anything that can be teased out of
Marx's writings on ideology. In Marx, one of the criteria whereby a belief
or conception is judged to have an ideological function is that, apart from
its supportive role in rationalizing men's interests, it is typically distortive
of their real circumstances and conceals from them the true character of
their activities. A distinction between appearance and reality in society is
the hinge on which any authentically Marxian theory of ideology turns. In
120 Critiques
Sorel's thought, on the contrary, the idea that there is a true or essential state
of society to knowledge of which men can attain, if it is present at all, is
extremely subdued. Sorel comes close to the view that men's social rela-
tions are so constituted by their ideas and beliefs, and especially by myths
which make an appeal to the will, that any idea of a science of society must
be rejected as incoherent. Revolutionary socialism could not then be a part
of such a science: it must be viewed rather as a moral outlook and a
conception of the social world, evolved obscurely (like its rivals) from the
unknowable depths of the minds of human beings sharing a common lot.
Sorel's construal of socialist thought as the conceptual form of pro-
letarian class struggle was, of course, a direct influence on Lukacs's History
and Class-Consciousness,·7 where (as in Sorel) it gives rise to well-known
difficulties and paradoxes. All these problems turn on a paradox of self-
reference. For the claim that socialism, like all other ideologies, is a
perspective on society generated at a specific vantage-point, and serving
specific interests, seems itself to have the aspect of a cognitively absolute
claim formulated at a point of neutrality between competing interests and
their associated world-views, whose very possibility the general theory of
ideology denies. Another way of stating the difficulty is to say that judge-
ments that such-and-such a belief is ideological in character are always
theory-dependent. Further, the theory presupposed by any such judgement
must itself be given a privileged immunity from the relativization which it
confers on its subject matter if it is not to have a self-defeating effect. These
difficulties are, if anything, more prominent in Sorel's theory of myth than
in Marx's account of ideology.
I do not propose to comment directly here on these questions, but one
observation is worth making about the way in which Sorel's successors
developed his thought. It is in the writings of Michels and his school that
we find both a continuation of Sorel's work and a resolution of some of its
difficulties which at the same time embodies an incisive criticism of some
of Sorel's excesses. The substantive burden of Michels's criticism of Sorel
is, of course, contained in his denial that the revolutionary syndicates could
reasonably be supposed to be agents of the moral regeneration or ricorso,
the renewal or recursion from decadence, that Sorel hoped for. In so far as
the syndicates acted as organs of genuine class struggle, with intermediate
as well as long-term aims demanding a consideration of strategic and
tactical advantage, their anti-instrumental and anti-political integrity was
inevitably compromised. The pressure of a quasi-military struggle itself
generates cSlitism and an oligarchy with interests distinct from those of the
proletarian majority. In short, Michels saw that there was no reason to think
the revolutionary trade unions were exempt from the fate that had befallen
the social-democratic parties.
 Philosophy, science and myth in Marxism 121
Michels' criticism exposes an epistemological tension in Sorel's thought
as much as it reveals its practical weakness. The tension is between those
areas of his writings in which Sorel acts merely as the exponent of a
proletarian world-view and morality and those in which he is the theorist of
its emergence and conditions. The latter role is arguably presupposed by
much of what Sorel says in the former capacity. Further, its culmination is
precisely a realist science of power and of social movements of just the sort
Sorel despised. Since these aspects of Sorel's thought, the diagnostic and
the prophetic, were never clearly distinguished, the one frequently corrup-
ted the other. Thus we find nowhere in Sorel any theoretical recognition of
the historical limitation of the proletarian outlook whose governing myths
he had poetically explored. Nowhere does Sorel acknowledge that the
outlook he is expounding is that of certain social groupings at a certain
stage of their development, in a specific (and soon to be transformed)
economic and political environment. This is the price of Sorel's abstention
from grand theories of history - that his account of the working-class
movement and its morality is hopelessly unhistorical. If the delusive
character of Sorel's hopes for the proletariat is ever conceded by him, it is
only implicitly in his late flirtation with radical nationalist groups whose
anti-bourgeois stance seemed to him to offer some prospect of the ricorso
and cultural renaissance which to the end of his days he continued to
conceive in political (if highly idiosyncratic) terms.

CONCLUSION
Each of the three Marxian thinkers I have discussed and criticized recalls
elements of Marx's thought and develops them in a legitimate and in some
respects a fruitful direction. Like Marx's thought as a whole, each of these
thinkers elaborates upon an aspect of Marx's contribution, and it turns out
under pressure to disintegrate into several categorically distinct segments.
The upshot of my criticism of each of these writers is that the episte-
mological instability of the Marxian system of ideas is such as to ruin it as
a contribution to rational inquiry. If it liberates thought from some forms of
absolutist limitation, it at the same time creates new obstacles to unfettered
inquiry in their place.
It may be asked, at this point, what conclusion may be reached as to the
place in the Marxian scheme of a conception of human nature? Much that I
have said might seem to support a view akin to Alasdair MacIntyre's,48
when he denies that social and moral conflicts can ever be arbitrated by
appeal to a view of human nature. No view of human nature can have
practical authority, he suggests, because each moral tradition carries with it
its own view of human nature. We will, each of us, see ourselves in terms
122 Critiques
of the moral practices in which we stand, and our conception of human
nature will itself be constituted by the moral practices to which we belong.
This radical sceptical view, though it may seem to be supported by what I
have said in criticism of Cohen about the incommensurability of different
cultures and forms of life, is not entailed by it. It would be Idealism of just
the sort against which Marx (and Cohen"9 ) struggle to suppose that man can
be whatever he tries to make himself become, and this is a view neither
entailed by Marx's system nor ever expressed by him. Certainly, it is an
important task to establish what are the biological constraints on the pos-
sible variety of human nature, and the theorists of socia-biology are making
a contribution to this question even if all their positive conjectures are in
error. Even a well-founded theory of man's biological limitations, however,
is not social theory, or any part of it. For, if a conception of human nature
is indeed indispensable to social theory, it is as its prelude and not as its
foundation.
MacIntyre is on the right track, however, in denying that a conception of
human nature (or a social theory embodying or presupposing a view of
man) can have practical or moral authority. This is a conclusion profoundly
subversive of Marxism in whatever epistemological mode its propositions
are framed. For, whether metaphysical, scientific or mythopoeic in
character, the central notions and theses of Marxism have always been
supposed by its exponents to be capable not merely of illuminating conduct
but of guiding and inspiring it. 50 The running together of theory and practice
in this way has led to a disastrous confusion of categories in modem
thought from which we are only lately recovering. But that is another story,
and one that cannot be told here. 51
11 Against Cohen on proletarian
unfreedom1

In a series of important papers, G.A. Cohen has developed a forceful

argument for the claim that workers are rendered unfree by capitalist
institutions. 2 His argument poses a powerful challenge to those (such as
myself) who think that capitalist institutions best promote freedom. Yet,
formidable as it is, Cohen's argument can be shown to be flawed at several
crucial points. It is not one argument, but three at least, and one of the goals
of my criticism of Cohen on this question is to distinguish and assess the
various separate lines of reasoning that together make up his case for the
unfreedom under capitalism of workers as a class. Cohen argues of workers
that they are rendered unfree by the institution ofprivate property on which
the capitalist system depends, that they suffer a form of collective un-
freedom under capitalism, and that they are forced to sell their labour
power under capitalism. Against Cohen, I will maintain that every sort of
unfreedom which he shows to exist under capitalism has a direct counter-
part under socialist institutions. For that reason, Cohen's arguments estab-
lish nothing about the distinctive bearings of capitalist institutions on
freedom. Indeed, the upshot of Cohen's arguments is a set of conceptual
truisms about individual and collective freedom and unfreedom, force and
justice, which have application wherever these concepts themselves find a
foothold. Within the framework of Cohen's reasoning - in which an obsoles-
cent philosophical method of conceptual analysis is applied to the terms of
conventional liberal discourse - these truisms are incontestable. At the same
time, they tell us nothing of substance about the impact of capitalist and
socialist institutions on the freedom of workers. Most particularly, Cohen's
arguments tell us nothing about the comparative freedom of workers under
capitalist and socialist institutions. I shall myself advance a number of
arguments for supposing that workers as a class under socialism are likely to
be far less free than they are under capitalism. But even if my arguments for
the greater freedom of workers under capitalism are inconclusive, Cohen's
argument for proletarian unfreedom demonstrably fails.
124 Critiques
PRIVATE PROPERTY, CAPITALISM AND LIBERTY
Cohen argues first that capitalism cannot constitue or be equated with
liberty (even economic liberty) because capitalist institutions rest upon, or
comprehend, private property. Private property, Cohen maintains, neces-
sarily restricts liberty. (For the purposes of my argument, I shall follow
Cohen's example in treating 'liberty' and 'freedom' as synonyms. I do not
intend to endorse any general thesis of their synonymy.) He asserts: 'free
enterprise economies rest upon private property: you can sell and buy only
what you respectively own and come to own. It follows that such eco-
nomies pervasively restrict liberty. They are complex structures of freedom
and unfreedom'.3 Or, as Cohen put it in a later essay: 'private property
pretty well ;s a particular way of distributing freedom and unfreedom. It is
necessarily associated with the liberty of private owners to do as they wish
with that they own but it no less necessarily withdraws liberty from those
who do not own it. To think of capitalism as a realm of freedom is to
overlook half its nature' .4 In the latter paper Cohen spells out his point more
systematically: 'For consider', he urges, 'if the State prevents me from
doing something I want to do, it evidently places a constraint on my
freedom. Suppose, then, that I want to perfonn an action which involves a
legally prohibited use of your property. I want, let us say, to pitch a tent in
your large back garden. . .. If I now try to do this thing I want to do, the
chances are that the State will intervene on your behalf. If it does, I shall
suffer a constraint on my freedom. The same goes for all unpermitted uses of
a piece of private property by those who do not own it, and there are always
those who do not own it, since "private ownership by one person pre-
supposes non-ownership on the part of other persons".'~ Cohen in his earlier
piece uses a similar example to make the very same point: 'Let us suppose
that I wish to take Mr Morgan's yacht, and go for a spin. If I try to, then it is
probable that its owner, aided by law-enforcing others, will stop me. I cannot
do this thing that I wish to do, because others will interfere. But liberty, as
Narveson [has] reasonably said, is "doing what we wish without the inter-
ference of others". It follows that I lack a liberty here. '6
Cohen's argument, then, is that private property institutions restrict
liberty, because their existence or enforcement involves interferences with
some persons doing as they wish. He adopts here Jan Narveson's rough
account of liberty as 'doing what we wish without the interference of
others' , saying that 'when a man cannot do what he wishes, because others
will interfere, he is unfree'.' For the purpose of assessing this argument of
Cohen's, I shall follow him in adopting Narveson's definition.
If this argument of Cohen's aims to show that the liberty-limiting effects
of private property institutions are peculiar to or distinctive of private
 Against Cohen on proletarian unfreedom 125
property, it fails. In part it trades on the truism that one person's freedom,
like one person's private ownership, always entails another's restraint or
unfreedom: if anyone has a freedom to do something, this means at least
that others do not prevent him from doing it. But this thesis of the corre-
lativity of freedom with unfreedom - the theory that having or exercising
freedom always presupposes or entails restraint of freedom - has no special
connections with the notion of property. One person's freedom to do
something has as its shadow the unfreedom of other persons to prevent that
person from (or interfere with that person) doing it, just as one person's
owning something has its shadow in others' not owning it. These are
perfectly formal truths within a certain discourse of freedom. They tell us
nothing about the weight or importance of the unfreedoms generated by
private property institutions.
This last remark may be stated in another way. All property institutions
- capitalist, socialist, feudal, or whatever - impose constraints on the
liberties of those who live under them. It is true of every system of property
that there will be many things that persons wish to do that they will be
interfered with (or prevented from) doing. This is only an entailment of the
evident truth that all systems of property are embodied in legal and moral
rules which create opportunities and limit the options of those who live
under them, where these opportunities and options will have a significant
impact on the freedoms of the various social groups affected by them. Most
obviously, in so far as the moral and legal rules that go to make a system of
property are enforced, those who wish and attempt to act in ways prohibited
by these rules will be restrained from doing so, or at least be under threat of
such restraint. Their liberty will thereby be curtailed. This is manifestly true
of any institution of property. Only a Hobbesian state of nature, in which no
one owns anything because there are no enforceable rules about property,
might appear to form an exception to this truth. Even there, where there is
no system of property but only the fact of possession, persons will often be
prevented from doing as they wish by others' possession of things the
former need in order to implement their plans. There is then a restraint of
liberty, not only under any institution of property, private or collective, but
in any society in which anyone possesses anything - that is to say, in
virtually any imaginable condition of social life.
This exceedingly obvious truth may be illustrated with an example.
Consider a society which holds its means of production in common owner-
ship and whose citizens have only use-rights as individuals over them.
Think of a coastal tribe which lives by fishing and whose means of pro-
duction are canoes. The canoes are owned by the tribal community as a
whole, but each individual tribesman has use-rights in them. How does such
an institution of communal ownership bear on the freedom of the tribesmen
126 Critiques
who live under it? It is clear that there are many things that individual
tribesmen may wish to do which they will be interfered with in or prevented
from doing. They cannot use them any more frequently, or for longer time
periods, than the usufructuary rules allow. No tribesman can (without threat
of sanction) take a canoe for his exclusive and permanent use, or give a
canoe away as a gift, or sell or rent a canoe to a non-tribesman. If a
tribesman tries to do any of these things, the property rules of this system
of communal ownership will be enforced against him. He may be
ostracized, or punished by loss of his use-rights in the canoes. Again, if the
property system in the tribe is truly communal ownership in the means of
production, and not one of corporate ownership, the tribesman will have no
'share' in the means of production which he can take with him ifhe chooses
to leave the community. The tribe might allow use-rights in canoes to be
sold, but to the extent that it did, its system of communal ownership of the
canoes would be attenuated. For this reason, a system of true communal
ownership has clear implications of the liberty of migration from the
community. Tribesmen might not be prevented from or interfered with in
leaving the community, but it would be harder for them to do so, since they
would leave with little or nothing. Investing in a new means of production
would be similarly difficult, or impossible, for the tribesmen as individuals.
These last points aside, it is transparently clear that, even as such a system
of communal ownership creates and enlarges freedom in some respects, it
restrains and curtails it in others. Communal ownership, one might say,
pretty well is a distribution of freedom and unfreedom. Like any system of
property rules, communal ownership is a complex structure of freedom and
unfreedom. To think of communal ownership as a realm of freedom is to
miss half its nature.
This point may be further illustrated by a consideration of Cohen's own
example of communal ownership: 'Neighbors A and B own sets of house-
hold tools. Each has some tools which the other lacks. If A needs a tool of
a kind which only B has, then, private property being what it is, he is not
free to take B' s one for a while, even if B does not need it during that while.
Now imagine that the following rule is imposed, bringing the tools into
partly common ownership: each may take and use a tool belonging to the
other without permission provided that the other is not using it and that he
returns it when he no longer needs it. or when the other needs, it, whichever
comes first. Things being what they are (an important qualification: we are
talking, as often we should, about the real world, not about remote possi-
bilities), the communising rule would, I contend, increase tool-using free-
dom, on any reasonable view'.8
Cohen's confident judgement that communal ownership of tools
enhances tool-using freedom on any reasonable view is hard to accept. He
 Against Cohen on proletarian unfreedom ] 27
acknowledges that 'some freedoms are removed by the new rule. Neither
neighbor is as as!.ured of the same easy access as before to the tools that
were wholly his. Sometimes he has to go next door to retrieve one of them.
Nor can either now charge the other for use of a tool he himself does not
require. But these restrictions will likely be less important than the
increased range of tools available. "It seems plain that other, more impor-
tant freedoms will also be lost by the communizing rule. There is, first of
all, the freedom to engage in long-term planning about the use of tools.
Since the tools can be taken without permission at any time when they are
not needed for use, neither household can effectively engage in long-term
planning of their tool-use. (The inability under Cohen's scheme of house-
holds to form and implement plans for long-term tool-use may be an
implication of a simpler difficulty of his scheme. The difficulty is that it
says nothing about what will be done under it when households 'need', or
wish to use, tools at the same time.) It is precisely this loss of freedom
resulting from the inability to make long-term plans that has in many
societies led to usufructuary rules of property being supplanted by ones
based on individual ownership, which allow for contractual arrangements
for the renting of tools and other capital assets. There is, in addition, the
freedom, which each tool-user hitherto enjoyed, of determining the rate of
depreciation of the tools. For, in the real world - and it is the real world we
are talking about, not remote possibilities - tools are worn out by the jobs
they do. Different jobs wear out tools at different rates and in different
ways, as do different ways of using tools. Under the communizing property
rule, no one has the freedom to decide on a rate of depreciation or to
implement the policies needed to secure it. Nor - and this is a distinct point,
though one that follows from the preceding one - does anyone have the
freedom to decide when the tools are to be replaced, whether because they
are worn out or because new tools have become available which do the job
better. These are freedoms possessed and exercised under private property
institutions which wither or disappear under the Communist rule. They are
not unimportant freedoms for tool-users. Their loss under the Communist
rule must diminish tool-using freedom in important respects not mentioned
in Cohen's account of it.
It might be objected that these criticisms depend on interpreting Cohen
as hypothesizing a system of full communal ownership, whereas what he
sketches is one of only partial common ownership. I am unsure how partial
common ownership of tools is to be understood, but, most naturally, it
would seem to mean that, when the need of the owner to use the tool
competes with that of his neighbour, the owner's need trumps that of the
neighbour. The owner does not have an unencumbered right of liberal
ownership in the tool, then, but he can always retain or retrieve the tool
128 Critiques
from his neighbour when he needs to use it. Against this interpretation of
Cohen's scheme, I maintain that it does not describe any system of property
rights that is workable in the real world. Who is to decide when the owner
needs the tool, and who is to arbitrate disputes as to when it is needed by
him? If the owner refuses to hand over the tool to the neighbour, because he
believes the neighbour's use of it will depreciate the tool to the point that
his own future needs for it will go unmet, is he acting within his rights under
the scheme of partly communal ownership? Would he be doing so if he
retained the tool on the ground that he is uncertain how long he will need it
for a job in which he is currently engaged, or about to start? In the real
world, it is reasonable to suppose, these questions would be answered by
the scheme of partly communal ownership collapsing into full private
property or full Communism in tools. The scheme of partly communal
ownership is, for this reason, an unviable half-way house, and cannot, be
invoked to answer the criticisms I have developed earlier. I accordingly
disregard this interpretation, and proceed on the assumption that, if Cohen's
scheme is to be a genuine alternative to private ownership in tools, it must
encompass fully communal ownership of them.
It might further be objected that the sense of freedom has shifted here
from freedom as non-interference with what one wishes to do to a sense in
which it designates the options available. lo The objection is not an absurd
one, but it is baseless. If each tool-user wishes to plan long-term use of the
tools, determine their rate of depreciation, or implement a decision as to
their replacement, he will be prevented form doing so unless all the other
tool-users accede to his wishes. Just as under private property institutions
an owner has a veto over non-owners as to the uses to which his property is
put, so under communal ownership every participant in the communal
ownership scheme has under the rules constituting the scheme a veto over
the uses to which the common assets are put. In Cohen's example, any
member of the communal ownership scheme can invoke the Communist
rule against any of the uses I have mentioned. To be sure, there is a logical
possibility that all the co-owners will come to agreement on the uses to
which I have referred, so that no one will in fact be prevented from doing
as he wishes. In the real world, given that persons have different purposes,
values, and rates of time preference, this is a most remote possibility, but
even if - granting a socialist transformation of human nature and a corres-
pondingly greater conformism in values among the co-owners - it were a
reality, it would be irrelevant to the issue at hand. For, as Cohen himself
appears to think, II freedom as non-interference is curtailed not only when a
person cannot in fact act as he wishes (because of interferences by others),
but also, counterfactually, when it is true that he would be interfered with if
he acted on desires he does not in fact possess. Thus, in Cohen's example,
 Against Cohen on proletarian unfreedom 129
tool-users are denied freedom to detennine rates of depreciation of tools,
and so on, even if they never in fact wish to so determine them. They are
denied that freedom because, if they so wish, the Communist rule could be
invoked and enforced against them. They would then be prevented from
doing as they wish to do. It is incontrovertible that this is a real and
important set of freedoms that has thereby been lost. Moreover, these are
freedoms that would be attenuated or diminished, if (in response to the
problems I have adduced) the Communist rule were qualified by another
rule - a majoritarian rule, say - for the adjudication of issues of capital
depreciation and investment. Even if it were so qualified by majoritarian
decision procedures, the Communist rule would still extinguish tool-using
freedoms possessed under private property institutions. True, if all accepted
the outcome of a majority decision, even where it went against the pre-
ferences of some, it could not be said that any had been coerced. The
freedom of some would still have been reduced, however, since (as we have
noted in the case of unanimous agreement) they would have been coerced
in the counterfactual case in which they did not accept the majority deci-
sion. It is a nice question for Cohen's account, though not one I can address
here, whether tool-using freedom is greater under the unqualified com-
munist rule, or under a Communist rule qualified by majoritarian (or other)
procedures. What is unambiguously clear is that important tool-using free-
doms are lost in either case.
It is hard, then, to understand Cohen's confidence in asserting that, in
this example, freedom for tool-users has on any reasonable view been
expanded. For myself, I incline to the contrary view: on any reasonable
view, freedom has been diminished for tool-users in Cohen's example. I
will not press this last point, however, since I will return to the question of
judgements of on-balance freedom in a later section of this chapter. Cer-
tainly, I do not deny that there are cases where it is reasonable for reason-
able men and women to disagree in their judgements of on-balance
freedom. Perhaps Cohen's example is such a case. It remains indisputable
that private and communal property systems each generate and distribute
freedoms and unfreedoms and that Cohen has said nothing which has a
tendency to show that communal ownership has the advantage from the
standpoint of liberty.

COHEN'S CONCEPTION OF FREEDOM: A DIGRESSION

It is worth pausing at this point to remark that the conception of freedom
that Cohen deploys throughout his writings on proletarian unfreedom is not
Marx's, but instead one derived from standard liberal discourse. As Cohen
theorizes it, freedom is non-interference with individuals acting as they
130 Critiques
wish to act (or might wish to act). This is a conception of freedom defended
forcefully by Bentham and developed with great power in our own times by
Isaiah Berlin.12 In the context of Cohen's argument, this liberal conception
of freedom has several salient features. In it, freedom is sharply dis-
tinguished from other values. Freedom is freedom, not justice, welfare or
whatever: it is one value among many. (Each thing is what it is, and not
another thing.) In general, in fact, Cohen seems to want to work with a
conception of freedom that is value-free and morally neutral, so that judge-
ments about freedom will be empirical rather than normative claims. This
is a point to which I shall return later in this chapter. Again, freedom is
attributed primarily to individuals, and not to groups, collectives, or classes.
This is so (or appears to be so) even when the freedom Cohen is theorizing
is a form of collective freedom. Finally, freedom as Cohen conceives of it
is a matter of degree. This is so, inasmuch as some persons may typically
be interfered with less than others in doing what they want. In a capitalist
society, for example, according to Cohen, owners of property are typically
freer than non-owners. It is precisely Cohen's thesis that in capitalist
society owners of property have freedoms which proletarians lack.
It is clear enough that this is a distinctively liberal conception offreedom
that is being invoked by Cohen. It would not be endorsed by critics of
liberalism such as Arendt or Marcuse, J3 and, if it accords with ordinary
usage, it is with the ordinary usage of liberal societies. It is no less clear that
the conception of freedom that Cohen employs is not Marx's. It has been
argued persuasively elsewhere that Cohen's view of freedom is not only
distinct from but incompatible and incommensurable with Marx's.14 I will
not rehearse these arguments here, but a few points, unavoidably brief and
dogmatic in character, are worth making about the many points of contrast
between Cohen's liberal notion of freedom and the conception employed
by Marx. It is, to begin with, important to note that Marx's conception of
freedom is on any plausible view not value-free but value-dependent: it
embodies or expresses a distinctive view of the human good. Very roughly,
this is the view that the good for man is found in conscious, cooperative
productive activity (and not, for example, in the individual pursuit of
pleasure or in contemplation). Again, freedom is not in Marx's view to be
attributed primarily to individuals. It is predicated of the human species
itself and of individuals only as instances of it (or, perhaps, as members of
the various social classes which constitute the historic self-disclosure of the
species in its pre-Communist manifestations). For this reason, it is wholly
unclear that one person's freedom can in the Marxian account (as it plainly
does in Cohen's liberal account) conflict with another's. In class society,
the freedom of a proletarian to do something may indeed conflict with that
of a capitalist, but in Communist society, as Marx sketchily conceives of it,
 Against Cohen on proletarian unfreedom 131
it seems that there will be no important instances where freedoms conflict.
Finally, it was not Marx's view that, whereas proletarians were rendered
unfree by capitalist institutions, capitalists themselves were free (or even
freer) under them. Marx's view, surely, as that both capitalists and pro-
letarians are unfree under capitalism, even if (or precisely because) the
freedom of a proletarian to do something is rendered nugatory by a con-
flicting freedom of a capitalist For Marx, freedom is a collective good
rather than an attribute of individuals, and (at any rate in Communist
society) it is otiose to consider how conflicts among individual liberties are
to be arbitrated or resolved. It is because Marx conceives freedom in this
way that he can claim to be developing a genuine critique of bourgeois
notions about liberty.
Cohen's argument, by contrast, aims to be an immanent criticism of the
liberal understanding of freedom. He wishes to show that capitalist
societies contain important unfreedoms by the standard of liberal freedom
itself.15 The principal burden of my criticism of Cohen is that his argument
issues in claims of an entirely formal sort, which give no strength to a case
for the specific disadvantages of capitalism in terms of freedom. Accord-
ingly, save for the arguments of the section of this paper in which I attempt
a substantive assessment of capitalist and socialist institutions in terms of
workers' freedom, my own critique of Cohen is in tum an immanent
criticism. For the purposes of my argument I accept the liberal conception
of freedom. Whether it is an adequate conception is another question,
which I hope to address on another occasion.

INDIVIDUAL LIBERTY AND COLLECTIVE UNFREEDOM

Cohen's second argument for proletarian unfreedom turns on the claim that
any member of a group may be free to do something that every member of
the group is not free to do. Thus, whereas it may be the case that any worker
can become a capitalist, it does not follow that workers as a class are free
so to do. (I am here following Cohen's conception of capitalism, which in
tum follows Marx's. For them, a capitalist society is not simply one
exhibiting private property in the means of production - since many
societies, including societies of yeoman farmers, say, exhibit this feature -
but one in which a propertyless majority must sell its labour to a capital-
owing minority. At this stage, I will not dispute this understanding of
capitalism, since I will question its adequacy later in my argument.) For
Cohen, the unfreedom of workers under capitalism is in the fact that 'each
is free only on condition that the others do not exercise their similarly
conditional freedom', 16 so that 'though most proletarians are free to escape
the proletariat, indeed even if aU are, the proletariat is an imprisoned
132 Critiques
class'.17 In other words, if the freedom of any worker to leave his class
depends on the fact that most others do not also attempt to leave it, then the
class of workers is unfree even if every worker has the freedom to become
a capitalist.
These arguments, first stated in Cohen's paper of 1979, are developed in
his paper of 1983. There he cites Marx's remark that 'in this bourgeois
society every workman, if he is an exceedingly clever and shrewd fellow,
and gifted with bourgeois instincts and favored by an exceptional fortune,
can possibly convert himself into an expioiteur du travail d'autrui. But if
there were no travail to be exploite, there would be no capitalist nor
capitalist production'. 18 Cohen develops the thought contained in Marx's
remark by way of an example:
Ten people are placed in a room the only exit from which is a huge and
heavy locked door. At various distances from each lies a single heavy
key. Whoever picks this up - and each is physically able, with varying
degrees of effort, to do so - and takes it to the door will find, after
considerable self-application, a way to open the door and leave the
room. But, if he does so, he alone will be able to leave it. Photoelectric
devices installed by a jailer ensure that it will open only just enough to
permit one to exit. Then it will close, and no one inside the room will be
able to open it again.
It follows that, whatever happens, at least nine people will remain in
the room. 19
Cohen argues from this analogy that
Each is free to seize the key and leave. But note the conditional nature
of his freedom. He is free not only because none of the others tries to get
the key, but on condition that they do not. . . . Not more than one can
exercise the liberties they all have. If, moreover, anyone were to
exercise it, then, because of the structure of the situation, all the others
would lose it.
Since the freedom of each is contingent on the others not exercising
their similarly contingent freedom, we can say there is a great deal of
unfreedom in their situation. Though each is individually free to leave,
he suffers with the rest from what I shall call collective unfreedom. 20
Cohen's argument here is, perhaps, a more substantive version, or a
concrete application, of the thesis of the correlativity of freedom with
unfreedom: the freedom of the few workers who become capitalists pre-
supposes the collective unfreedom of the many workers who do not. Now
it is worth observing that there is so far nothing very determinate in this
 Against Cohen on proletarian unfreedom 133
relationship of individual liberty with collective unfreedom under
capitalism. We do not know how many workers becoming capitalists would
overturn Cohen's argument. Suppose most workers had a period as capital-
ists during their lives. Would this render Cohen's attribution of collective
unfreedom to the proletarian class invalid? Cohen might allow that collec-
tive proletarian unfreedom no longer exists in a society in which most
workers spend part of their lives as capitalists, but deny that the economic
system in which this can occur is any longer clearly a capitalist system.
Such a deilial seems unreasonable. In our supposition of a society in which
most workers are capitalists for part of their lives, the constitutive institu-
tions of capitalism - private ownership of the means of production and
market allocation of capital and income - are by hypothesis fully preserved.
(That is the point of hypothesizing that most workers spend part of their
lives as capitalists.) Admittedly, Cohen might object that capitalism neces-
sarily presupposes a propertyless proletarian majority, and, if he did, his
objection would be an authentically Marxian one. That is not a sufficient
reason for accepting Cohen's objection, however, since it amounts to
identifying capitalist institutions with that nineteenth-century variant of
them studied by Marx and, thereby, ruling out a priori the possibility of a
proletarianless capitalism. I shall return to this last point later. Against my
criticism, Cohen might further argue that whether or not workers are free in
a system in which they spend part of their lives as capitalists depends
crucially on how much of their lives workers spend as capitalists. This
seems to be part of the force of Cohen's assertion that 'the manifest intent
of the Marxist claim is that the proletariat is forced at (time) t to continue to
sell his labor power, throughout a period from t to t + n, for some con-
siderable n'.21 The point here is not unreasonable. H most workers had to
sell their labour power for only a small part of their lives, if (in other words)
they spent most of their lives as capitalists, they would no longer be
proletarians. In this circumstance, indeed, the proletariat - and with it
proletarian unfreedom - would have all but disappeared. In the hypo-
thetical case I have invoked, however, the near disappearance of pro-
letarian unfreedom comes about in virtue of the spread of capital among the
workers rather than by the abolition of capitalist institutions.
It is, of course, a standard position in Marxian political economy that the
kind of society I have envisaged cannot exist, or cannot exist for very long.
Processes of market concentration and proletarian immiseration will pre-
vent the diffusion of capital among workers which alone could raise them
within capitalism from the status of proletarians. This is an empirical claim
in economic theory which is open for us to dispute: developments such as
the growth of pension funds and of labour unions with significant capital
134 Critiques
assets might be seen as giving evidence of a trend towards the sort of
capitalism-without-a-proletariat to which my hypothetical example points.
Cohen's model of capitalist property institutions, in which the capitalist is
the sole proprietor of the means of production, is an anachronistic one, best
suited to economic life in England in the early nineteenth century. It has
little relevance to the late twentieth-century reality, encompassing
employee shareholding, management buy-outs, and profit-related pay. It
may be premature or specUlative to suggest that these developments attest
to a trend to a proletarianless capitalism. Nevertheless, even were my
example to remain entirely hypothetical, it would demonstrate that, as a
matter of its structure or logic, proletarian unfreedom as Cohen conceives
of it may be overcome, abolished, or reduced in a variety of ways, of which
socialism is (at best) only one.
The indeterminacies in the relation between individual liberty and
collective unfreedom under capitalism have implications of another sort for
Cohen's argument. Cohen allows that 'collective unfreedom comes in
varying amounts, and it is greater the smaller the ratio of the maximum that
could perform it to the total number in the groUp'.22 (It seems to be an
implication of Cohen's view that collective unfreedom is a variable magni-
tude that collective freedom likewise comes in varying amounts. Whether
it follows from this that a circumstance of complete collective freedom is
impossible, since one collective freedom entails another collective un-
freedom, is a question I cannot here pursue.) If collective un freedom is thus
a matter of degree - and not, for example, a condition in which individuals
simply are or are not - then it matters vitally how large, and how variable,
are the opportunities of individual emancipation by the acquisition of
capitalist status. Perhaps the more proletarians there are who seek to
become capitalists, the more opportunities there will be for them to do so.
The possibility certainly cannot be excluded a priori, and, as J have already
argued, the empirical theory which wou1d deny it is at the very least
controversial and disputable. Cohen's argument seems to depend on the
assumption that, no matter how large the number of proletarians seeking to
become capitalists, the number of opportunities to do so will remain fixed,
or at least will not expand significantly. He makes explicit this assumption
when he discusses another example of collective unfreedom:
Suppose, for instance, that a hotel, at which one hundred tourists are
staying, lays on a coach trip for the fIrSt forty who apply, since that is the
number of seats on the coach. And suppose that only thirty want to go.
Then, on my account, each of the hundred is free to go, but their situation
displays a collective unfreedom. 23
 Against Cohen on proletarian unfreedom 135
Cohen goes on:
The coach case is a rather special one. For we tend to suppose that the
management lay on only one coach because they correctly anticipate
that one will be enough to meet the demand. Accordingly, we also
suppose that if more had wanted to go, there would have been an
appropriately larger number of seats available. If all that is true, then the
availablr. amount of collective freedom non-accidentally accords with
the tourists' desires, and, though there is still a collective unfreedom, it
is, as it were, a purely technical one. But if we assume there is only one
coach in town, and some such assumption is required for parity with the
situation ofproletarians, then the tourists' collective unfreedom is more
than merely technical. 24
Cohen's assumption of fixity in the range of opportunities for workers'
escape from proletarian status, which he admits is essential to the argument
for proletarian unfreedom, looks entirely arbitrary. It is defensible, if at all,
only by reference to propositions in Marxian economic theory, which are
nowhere argued by Cohen. These are propositions having to do with the
systemic scarcity of capital under the capitalist system. It is this alleged
scarcity, presumably, which explains the claim that there are so few points
of access from the proletariat to the capitalist class. I do not think it can
fairly be said that there is any plausible contemporary statement of Marxian
economic theory which supports the claim that capital is subject to a sort of
endemic scarcity under capitalism. Much empirical evidence would in any
case count against such a claim. Indeed, there is evidence that suggests a
conjecture directly opposed to that of Marxian theory - the conjecture that,
the more workers succeed in becoming capitalists, the greater are the
opportunities of the remainder to do so. This is to say that the analogy of the
locked room from which only one person can escape may be wholly
misleading. A better analogy might be that of a group seeking to climb a
mountain, where the more members of the group who succeed in doing so,
the easier it is for those left behind to be raised up by the strength of those
who have gone on before. There is nothing to say that such an analogy does
not fit the case of modem capitalism better than Cohen's. In any event, in
the absence of a plausible theory which gives to the relation between
individual liberty and collective unfreedom under capitalism the deter-
minacy or fixity which Cohen admits to be necessary to his argument, we
have no reason to think proletarian unfreedom a necessary feature of
capitalism. Or, to put the matter in different terms, the indeterminacies of
the relationship that Cohen postulates between individual workers and their
class situation are abated only by invoking an empirical theory he gives us
no reason to accept
136 Critiques
The most fundamental objection to this argument of Cohen's is an
altogether different one. It is one that focuses on the condition of simul-
taneous or joint access to some opportunity, action, or status which Cohen
specifies as a necessary condition of freedom in respect of it. Cohen
specifies this condition, negatively and by implication, when he specifies
the necessary and sufficient condition of collective un freedom. He tells us:
'Collective unfreedom can be defined as follows: a group suffers collective
unfreedom with respect to a type of action A if and only if performance of
A by all members of the group is impossible' .25 He clarifies further: 'A
person shares in a collective unfreedom when, to put it roughly, he is among
those who are so situated that if enough others exercise the corresponding
individual freedom, then they lose their individual freedom'.26 Cohen's
definition of collective unfreedom seems radically at variance with much of
our standard thought. We do not usually suppose that, unless any subscriber
to a telephone system can use it at the same time as every other or most
other subscribers, then the entire class of telephone users is rendered unfree
by the system. Perhaps Cohen would maintain that this example, though
technically a case of collective unfreedom, is not an interesting one. Let us
concede the point. I do not think that that could be said of many social
institutions of which his definition would yield a description in terms of
collective unfreedom. Consider, in this regard, the institution (proposed by
C.B. Macpherson21 and others as a device for diminishing risks to indi-
vidual liberty under socialism) of the guaranteed minimum income. The
guaranteed minimum income, like many other socialist institutions,
depends for its existence on its being used at anyone time by only a few.
The freedom of any worker to take up his guaranteed income, and to live on
that alone, depends on the unfreedom of most others, whose labour sustains
the guaranteed income scheme.
On Cohen's definition, all workers in the socialist society are rendered
unfree by the guaranteed income scheme. Recall in this connection his
statement that 'a group suffers collective unfreedom with respect to a type
of action A if and only if performance of A by all members of the group is
impossible' .28 We may go further. Each socialist worker is free to live on
the guaranteed income not only because most others do not, but on condit-
ion that they do not. Then each socialist worker is free (to live on the
guaranteed income) only on condition that the others do not exercise their
similarly conditionalfreedom. Not more than a few workers can, at anyone
time, exercise the liberty they have. If, moreover, more than a few were at
any time to exercise it, then, because of the structure of the situation, all the
others would lose even this conditional freedom. Since the freedom of each
socialist worker is contingent on the others not exercising their similarly
contingent freedom, we can say that there is a great deal of unfreedom in
 Against Cohen on proletarian unfreedom 137
their situation. Though each is individually free to live on the guaranteed
income, he suffers with the rest from a form of collective unfreedom.
Like the collective unfreedom suffered by proletarians under capitalism,
the collective unfreedom of socialist workers in respect of the guaranteed
income is suffered by them as individuals. It is not a group unfreedom. 'A
person shares in a collective unfreedom when, to put it roughly, he is among
those who are so situated that if enough others exercise the corresponding
individual freedom, then they lose their individual freedoms'.29 By contrast
with genuine collective unfreedom, the unfreedom of the proletariat to
overthrow capitalism is a group unfreedom, since 'no individual proletarian
could ever be free to overthrow capitalism, even when the proletariat is free
to do SO'.30 By this criterion, the unfreedom of the socialist worker in
respect of the guaranteed income is a genuine one and not a group un-
freedom. Structurally, it is no different from proletarian unfreedom under
capitalism.
I am leaving aside here the important question of whether, when collec-
tive freedoms are lost by individuals, what is lost are individual freedoms.
For Cohen's argument to go through, it must be the case that collective
freedoms are not only freedoms possessed or lost by individuals, but also
instances of individual freedom as he conceives of it. This is to say that
collective unfreedom and freedom may be analysable in terms of inter-
ference and non-interference with individuals' opportunities to act as they
wish. Yet there is in Cohen no systematic argument for the reducibility of
collective freedom and unfreedom to individual freedom and its absence.
(perhaps Cohen thinks that collective and individual unfreedoms are exten-
sionally equivalent, but not collective and individual freedoms, but he does
not argue this, and it is hard to see how such an asymmetry could be
sustained.) Nor is this surprising, since the two sorts of freedom seem on the
face of it categorically distinct, and possibly even embody incommensur-
able notions of freedom. I mention this gap in Cohen's reasoning, not in
order to try to demonstrate that there are indeed two notions of freedom at
stake in his argument, but simply to remark that his argument founders
unless one kind of freedom (or unfreedom), the collective sort, is reducible
without remainder to the other, individual sort of freedom and unfreedom.
The analogy between proletarian unfreedom under capitalism and
worker unfreedom under socialism (in respect of the guaranteed income),
though close, is not exact. In the case of proletarian unfreedom, as Cohen
conceives of it, there are a few that are free in capitalist society - the
capitalists (including those proletarians who succeed in becoming such).
The reference group of those suffering a collective unfreedom to become
capitalists cannot encompass the capitalists themselves. In the socialist
society, all are unfree with regard to the guaranteed income. (There is an
138 Critiques
asymmetry here between the collective unfreedoms at issue if, as I have
earlier argued, a capitalist society might exist in which all are capitalists,
and if, as seems self-evident, there can be no socialist society in which all
live on the guaranteed income and nothing else.) This brings out again an
important point - that collective unfreedom, no less than individual un-
freedom, is for Cohen a matter of degree. Individuals, as members of
societies or groups, may suffer varying degrees of collective unfreedom. In
the last section of this paper, I will argue that socialist institutions may
generate more collective unfreedom for workers than do capitalist institu-
tions. This is so even if collective unfreedom is (as I have argued earlier) a
pervasive feature of social institutions generally. At this point, I want to
observe only that, whereas not all are unfree under capitalist institutions (to
become capitalists), there is at least one collective unfreedom which all
suffer under socialism - that relative to the guaranteed income. It is unclear
to me whether Cohen would consider this a lesser degree of collective
unfreedom than that which obtains for workers in capitalist societies.
In fact, Cohen appears to think that not all collective unfreedoms are
undesirable simply on account of the unfreedom they contain. He tells us
that 'some collective unfreedom, like some individual unfreedom, is not
lamentable'. And, crucially, he goes on: 'It is what this particular un-
freedom forces workers to do which makes it a proper object of regret and
protest. They are forced to subordinate themselves to others who thereby
gain control over their, the workers', productive existence' Y This is a
crucial passage inasmuch as it aims to specify what is bad or wrong in the
collective unfreedom of proletarians under capitalism. It is an extremely
unsatisfactory passage, at the same time, since subordination to others and
control by these others of workers' productive life is not peculiar to, or even
distinctive of, capitalist institutions. Workers may, as I shall later argue, be
forced to work for others in a socialist system, thereby losing control of
their productive lives. What is distinctive of, or peculiar to capitalism, is
that workers sell their labour - an option denied them under feudal and
socialist institutions. Cohen has said nothing, however, to show the special
evil of such selling of workers' labour. If, on the other hand, the lamentable
aspect of proletarian collective unfreedom is in the fact that the option of
not working at all is an undesirable one, so that workers are forced to work,
then it must be observed that this is a collective unfreedom found in all
societies, albeit in crucially variable degrees. Cohen here in fact comes
close to conceding my argument that collective unfreedom is a pervasive
property of social institutions32 and is to be found as clearly in the institu-
tions of a socialist society as it is in those of capitalism. Thus, nothing of
importance appears to follow from Cohen's arguments about the collective
unfreedom of proletarians. In the absence of further considerations, there is
 Against Cohen on proletarian unfreedom 139
nothing to say whether their unfreedom is lamentable or not, or, in general, to
show that it is a proper object of moral concern. He achieves the result that
proletarian collective unfreedom is an interesting and lamentable unfreedom,
rather than a technical or trivial one, only by invoking the altogether indepen-
dent claim that workers are forced to subordinate themselves to others who
thereby gain control over their productive lives. Without this further con-
sideration, the argument that proletarians suffer a collective unfreedom, though
perhaps valid. is uninteresting and indeed trivial.
This conclusion follows inexorably from the logic of Cohen's argument.
Like any other sort of unfreedom, collective unfreedom is not for Cohen
necessarily an evil. For, like the concept of freedom in general, the concept
of collective freedom is supposed to be value-free. Accordingly, the bad-
ness, wrongness, or moral importance of a specific collective unfreedom
must depend on other considerations. (Presumably, it is these other con-
siderations that are invoked when we make comparative judgements of
collective unfreedom.) In Cohen's account, the importance of collective
proletarian unfreedom is explained by reference to the claim that workers
are forced to work for capitalists. It is therefore to Cohen's arguments about
force and freedom in the workers' situation that I tum for illumination.

LmERTY, FORCE AND JUSTICE

Cohen believes, with Marx, that workers are forced to sell their labour
power to capitalists, and are thereby rendered unfree. He thinks that
workers under capitalism may be in some important respects freer than
Marx supposed.H But he defines workers (under capitalism) as 'those who
are forced to sell their labour-power'. He asks: 'Is the stated condition
necessary and sufficient? Certainly not all who sell their labour power are
proletarians, but the condition is that one be forced to sell it. Still, it must
be admitted that plenty of salaried non-proletarians are as much forced as
many workers. So the condition is not sufficient' . He goes on further to ask
'whether the condition is a necessary one: are proletarians forced to sell
their labour power?' He informs us that 'Robert Nozick answers nega-
tively', but that 'Nozick's objection to our condition rests on a false
because moralised account of what it is to be forced to do something' .34
Let it be noted that Cohen has not so far given us any reason infavour of
supposing that proletarians are forced to sell their labour power, and
thereby are rendered unfree. He has adduced instead an argument against
Nozick's claim that their having to sell their labour power fails to render
them unfree - the argument that Nozick's conception offreedom is 'moral-
ised' and so 'false'. Cohen spells out this argument against Nozick by
asserting that 'to prevent someone from doing something he wants to do is
140 Critiques
to make him, in that respect, unfree: I am unfree whenever someone
interferes, justifiably or otherwise, with my actions. But', Cohen goes on
'there is a definition which is implicit in much libertarian writing, and
which entails that interference is not a sufficient condition of unfreedom.
On that definition, which I shall call the moralised definition, I am unfree
only when someone does or would unjustifiably interfere with me.'3S He
further develops his argument against Nozick in a later piece, contending
that:
Robert Nozick might grant that many workers have no acceptable alter-
native to selling their labour-power.... But he denies that having no
acceptable alternative but to do A entails being forced to do A, no matter
how bad A is, and no matter how much worse the alternatives are, since
he thinks that to have no acceptable alternative means to be forced only
when unjust actions help to explain the absence of acceptable alter-
natives.... Nozick's objection to the thesis under examination rests
upon a moralised account of what it is to be forced to do something. It is
a false account, because it has the absurd upshot that if a criminal's
imprisonment is morally justified, then he is not forced to be in prison.
We may therefore set Nozick's objection aside. 36
Nozick's argument that workers are not forced to sell their labour power is
to be rejected then, according to Cohen, because the idea of unfreedom or
forcing which it deploys conflates interference with justifiable interference,
which in turn has the absurd result that the justifiably imprisoned man is not
forced to remain in jail.
Cohen's argument against Nozick is feeble and sloppy and fails for
several reasons. To begin with, Nozick's argument is not that justiflllble
interferences are not interferences with liberty. It is that the domain of
liberty is specified by principles of justice, so that ajusticizable interference
with liberty is an impossibility. What Nozick's view excludes as a possi-
bility is not, then, justified restraint of liberty, but justicizable restraint of
liberty - that is to say, restraint of liberty justified in terms of justice.
Accordingly, justified violation of the liberty demanded by justice remains
a violation of liberty. Nozick's account of violating side constraints which
protect liberty so as to avert a moral catastrophel7 tells us this: when we
violate side constraints and thereby commit an injustice, we do indeed
curtail liberty, but we justify doing so by invoking the larger morality
within which justice is usually (but not in this case of potential moral
catastrophe) paramount. So, whereas justice and liberty cannot compete
with one another, moral considerations may in extremity justify a violation
of the liberty that is demanded by justice. (I pass over the possibility that
the scope of justice is bounded or limited by circumstances of moral
 Against Cohen on proletarian unfreedom 141
catastrophe in such a way that rights violations are impossible in such circum-
stances, since this seems plainly to be a possibility which Nozick does not wish
to envisage.) For these reasons, it is mistaken to hold, as Cohen does, that
Nozick conflates restraint of liberty with justified restraint of liberty.
It remains true, nonetheless, that Nozick's is a normative conception of
liberty. In Nozick's account, the demands of liberty are given by a theory
of justice - by the theory of entitlements he sets out in his book. By contrast,
Cohen wishes to use only a 'neutral' or 'non-normative' conception of
freedom. Thus he tells us that 'whatever may be the correct analysis of 'X
is free to do A', it is clear that X is free to do A if X would do A if he tried
to do A, and that sufficient condition of freedom is all that we need here' .J8
It is this 'non-normative' account of freedom 39 that Cohen invokes against
Nozick. What are we to make of it? Cohen argues that Nozick's view is
'unacceptable' because it yields the 'absurd' conclusion that 'a properly
convicted murderer is not rendered unfree when he is justifiably
imprisoned'.4O The absurdity of this result is, presumably, in the fact that it
conflicts with ordinary-language uses of terms such as 'force' and 'free-
dom'. Cohen's argument itself rests on a confusion. Consider the case of
the would-be rapist who is forcibly prevented from raping his victim. On
Nozick's account, it is indeed true that forcibly preventing a rapist from
committing the act of rape deprives him of no freedom, since the act of rape
encompasses an unjust assault on another person's body and liberty, and is
therefore an act which no one is entitled to perform. I, for one, do not find
this result particularly counterintuitive. It seems to me to square better with
our ordinary linguistic and moral intuitions than the Benthamite view
which conceives of laws against rape as restricting the liberty of rapists (if
only a worthless or disvaluable liberty) for the sake of protecting the liberty
of their victims. I will not press this point, however, since Cohen's con-
fusion arises elsewhere. It is plain to all of us that justice demands the
restraint of rapists. Now, on Nozick's account, there is no liberty to commit
rape, and so no rapist liberty to be restrained. The rapist is forcibly pre-
vented from committing rape, but he is not thereby rendered unfree. At the
same time, the rapist's liberty may yet be restrained ifhe is imprisoned for
his crime. To think otherwise is to move illegitimately from the injustice of
the act of rape to the justice of the penalty of imprisonment. The mere fact
that rape is an unjustice, taken by itself, has nothing to tell us as to the just
punishment for rape. It does not even tell us that punishment is called for in
justice. If it is, then perhaps, as in Islamic law, the just punishment of rape
is not loss of liberty but, instead, loss of life. Only a full theory of retributive
and corrective justice can tell us what justice demands as the legal remedy
for rape. Nozick's entitlement theory of justice makes no claim to com-
prehend the demands of justice addressed in retributive and corrective
142 Critiques
theory: its subject matter is a different one. For this reason, it might very
well be that imprisonment for the crime of rape constituted a restraint of the
rapist's liberty, even if forcibly restraining him from the act of rape did noL
In Nozick's account, liberty has primacy among the demands of justice
(as it does in Rawls's theory of justice as fairness). Liberty and justice are
linked in Nozick's theory, inasmuch as forcibly restraining A from unjustly
restraining B is not to restrain A's liberty. A may be forced to do something,
or to refrain from doing something, then, and still not be rendered unfree by
that forcible restraint. One must have a liberty before it can be restrained.
One may be restrained, without it being true that it is one's liberty that has
been restrained. This Nozickian view that forcible prevention of an unjust
action is not a restraint of liberty is an entirely general thesis which in no
way presupposes Nozick's own account of justice. It is a feature, also, of
Rawls's theory, in which private ownership of the means of production is
not among the requirements of justice. Against this view of liberty or
freedom as a moral notion, Cohen invokes his 'neutral' or 'non-normative'
conception, stated earlier. Cohen's conception of freedom as a value-free
notion seems to me unacceptable for a number of reasons. I will not
consider these here, however, since I will address them later in this section.
Also, they arise from a position in philosophical method held by Cohen
which I shall expound and cliticize in the next part of this chapter.
What are Cohen's arguments for the claim that workers under capitalism
are forced to sell their labour power'? Cohen does not deny that they are
unfree to do so, only that they are not free not to do so. Indeed, he insists
that workers are free to sell their labour power inasmuch as one is in general
free to do what one is forced to do. Being forced to sell their labour power
entails that workers are free to sell it. He argues:
before you are forced to do A, you are, at least in standard cases, free to
do A and free not to do A. The force removes the second freedom, but
why suppose that it removes the first'? It puts no obstacle in the way of
your doing A, and you therefore remain free to do it. We may conclude,
not only that being free to do A is compatible with being forced to do A,
but that being forced to do A entails being free to do A. Resistance to this
odd-sounding but demonstrable result reflects failure to distinguish the
idea of being free to do something from other ideas, such as the idea of
doing something freely. I am free to do what I am forced to do even if,
as is usually true, I do not do it freely, and even though, as is always true,
I am not free with respect to whether or not I do it.41
Thus, workers in a capitalist society, unlike slaves in a slave society, are
free to sell their labour, even though they are forced to do so (and so are
unfree not to do so).
 Against Cohen on proletarian unfreedom 143
The unfreedom in workers having to sell their labour power is for Cohen
in the fact that they have no acceptable alternative. They may in a pure
capitalist society choose to beg, or starve, but those are not acceptable
alternatives: they are accordingly forced to sell their labour power. On the
other hand, Cohen allows that, objectively speaking, most proletarians
could (as some immigrant groups have done) acquire sufficient capital to
rise from proletarian status: 'Proletarians who have the option of class
ascent are not forced to sell their labour power, just because they do have
that option. Most proletarians have it as much as our counterexamples (the
immigrants) did. Therefore most proletarians are not forced to sell their
labour power' .42 Cohen here allows that most proletarians are not, in objec-
tive fact, forced to sell their labour power. As he admits: 'One would say,
speaking rather broadly, that we have found more freedom in the pro-
letariat's situation than classical Marxism asserts' .43 His argument that,
though most proletarians are not forced to sell their labour power, they are
nevertheless unfree, is one that appeals solely to the collective proletarian
unfreedom created by the conditionality ofeachproletarian'sfreedom to
leave the proletariat. This is demonstrated by the fact that, having dis-
cussed the case of the people in the locked room, Cohen concludes 'by
parity of reasoning, that although most proletarians are free to escape the
proletariat, and, indeed, even if every one is, the proletariat is collectively
unfree, an imprisoned class'.44 This is to make his argument circular. We
saw at the end of the last section that the proletariat suffered an interesting,
lamentable collective unfreedom, only if it could be shown that it is forced
to sell its labour power. We find now that, since most proletarians
admittedly are not forced to sell their labour power, the argument for their
being unfree not to do so depends on their suffering collective unfreedom
as an imprisoned class. We find, in short, that Cohen advances no argument
(apart from the weak argument from collective freedom) for the thesis that
workers are forced to sell their labour.
The point may be put in another way. The claim that workers are forced
to labour under capitalism, because they are not free to sell their labour
power, has a corresponding application to the situations of workers under
socialist institutions. Under a pure capitalist system, according to Cohen,
the worker has no acceptable alternative to working for the capitalist The
proletarian may choose to beg or starve, but, anciently, it is only an irony at
his expense to say that he is free to do so. It is ironical to suggest that the
proletarian is free to do these things, since we all know them to be un-
desirable (and, indeed, unacceptable) alternatives. However, note that, on
Cohen's own conception of freedom as being value-free, the proletarian in
truth is free to beg or starve, even though the statement that he has these
freedoms may have an ironical flavour. The proletarian is free to beg or
144 Critiques
starve because no one in a pure capitalist system will interfere with him or
prevent him from doing these things. In Cohen's account, therefore, the
proletarian is free to leave the proletariat by refusing to sell his labour to
the capitalist. He has this freedom so long as he may without interference
beg or starve.
Though he is free to refuse to sell his labour power to the capitalist, the
proletarian is nonetheless forced to do so, since he has no acceptable
alternative. He is free to do what he is forced to do, Cohen has already told
us, because being free to do something is a necessary condition of being
forced to do it. The logic of Cohen's argument requires him to go further
than this. It requires him to accept that the proletarian is forced to sell his
labour power even though he is free to refuse to do so. This is so, at least in
part, because 'force', unlike 'freedom', is for Cohen a normative concept.
It embodies standards of desirability and acceptability. It is only the con-
junction of the value-free notion of collective unfreedom with the value-
laden notion of force that gives Cohen his conclusion that workers are
non-trivially and lamentably unfree under capitalism.
His argument, then, turns on the claim that workers are forced to sell
their labour power under capitalism. (Some may well think that Cohen's
account of freedom is unacceptable because it has the absurd upshot that
proletarians are free to leave the proletariat by turning to beggary or
submitting to starvation. I let this pass.) But is it true that they have no
acceptable alternative? It would be true if their only alternatives were
beggary or starvation. As we have seen, however, Cohen admits that,
objectively speaking, most proletarians are in the same position as immi-
grant groups - they can become capitalists by accumulating capital. Like
immigrants, then, most proletarians have another exit from their proletarian
status. They need not become lumpenproletarians, but may become capital-
ists. Having made this admission, Cohen's only argument for proletarian
unfreedom is the claim that most workers cannot exercise this freedom at
the same time. This is the argument from collective unfreedom I have
already criticized, and it is the only argument Cohen has to offer. If workers
have other, acceptable alternatives to leaving the proletariat apart from
becoming beggars or starving, then the collective unfreedom which the
argument establishes is only a trivial one.
Let us, though, in a spirit of charity, set these arguments aside, and allow
that there is a sense that the worker is forced to sell his labour power to the
capitalist. Let us proceed to compare the proletarian's situation with that of
the worker in a socialist state. By hypothesis, the latter cannot acquire
private capital and live off that, and, since beggary will presumably be
illegal, he cannot live by begging. Unless he is disabled or ill, and so
(presumably) in receipt of government benefits, his only source of regular
 Against Cohen on proletarian unfreedom 145
livelihood is his income from his work. Perhaps family or friends can
support him for a while, but let us suppose that, like beggary under
capitalism, this is an unacceptable alternative - always supposing, contrary
to the historical experience of actually existing socialist states, that it is an
alternative that is not legally forbidden. Since the socialist worker has no
acceptable alternative to working for the socialist state, he is unfree not to
do so. Because he is not free not to work for the socialist state, he is forced
to do so. This js true whether the political form of the socialist state be
democratic or authoritarian, and whether the economic unit of the socialist
state is a public corporation or a worker cooperative. All that is required is
that state or public authorities be the sole permissible source of employ-
ment. The unfreedom of proletarians not to work for capitalists has as its
mirror image the unfreedom of socialist workers not to work for the state -
or, more precisely, for those who exercise power through the state. Like
proletarians, socialist workers fmd themselves in a situation in which their
productive lives are subject to control by others. It is ironical to suggest
otherwise.

THEORIES, CONCEPrS AND PHILOSOPHICAL METHOD

The upshot of my argument is that every charge of unfreedom that Cohen
can intelligibly make against capitalism may be made just as well against
socialist institutions. If workers are rendered unfree by the system of
private property on which capitalism depends, they are also rendered unfree
by a communal property system. If they suffer collective unfreedoms under
capitalism, they suffer other collective unfreedoms under socialism. If they
are not free not to sell their labour power under capitalism, if they are
forced to sell their labour power under capitalism, they are not free not to
work under socialism. This is to say that, in every one of the three respects
in which capitalism renders workers unfree, there is a clear parity with a
feature of socialist institutions in virtue of which they are rendered unfree.
Nor is the a reason for this parity hard to find. In so far as they establish
anything, Cohen's arguments give support to a series of conceptual truisms.
This is, indeed, the manifest character of Cohen's arguments, which all of
them consist in an application of a method of conceptual analysis.
For several reasons, Cohen's adoption of this method in political philo-
sophy is misconceived. In the first place, as I have already argued, it can
yield only conceptual truths, if it yields anything at all. Such truths can tell
us nothing substantial about the fate of freedom under rival institutional
arrangements. To find out about that, we must look to a theory of the real
world. As things stand, Cohen appears to suppose that, by eliciting con-
ceptual truths, he is in fact saying something about freedom in the real
146 Critiques
world. It is hard to explain otherwise his recourse to such portentous
expressions as 'complex structures of freedom and unfreedom'. It is a
delusion, however, to suppose that conceptual analysis can tell us the way
things are in the world.
The philosophical method which Cohen adopts is in any case super-
annuated. The 'analysis' of 'concepts' would be defensible as a strategy in
philosophical inquiry if it were the case, as Kant supposed, that the cate-
gories of the human understanding were invariant and universal. Otherwise,
the object of analysis can only be words, in all the miscellaneous diversity
of their usages. Among us, at any rate, there is no clear uniformity of usage
in respect of the key terms of moral and political discourse. Consider, for
example, Cohen's idea of liberty. The notion of freedom within which he
works is not an unequivocal deliverance of ordinary language, which
contains 'normative' as well as 'descriptive' uses of 'freedom' and 'liberty'
and their associated expressions. As its most sophisticated exponents have
long recognized,4s a neutral or value-free definition of liberty is a term of
art, a technical expression developed as part of a reconstruction by stipu-
lation of the terms of ordinary discourse. Further, even if ordinary usage
were consistent, it is wholly unclear that it would have authority for
political thought. As Joseph Raz has observed, 'linguistic distinctions ...
do not follow any consistent political or moral outlook.... What we need
is not a definition nor mere conceptual clarity. Useful as these always are
they will not solve our problems. What we require are moral principles and
arguments to support them. '46 And further: 'It is ... important to remember
that that concept [the concept of political freedom] is a product of a theory
or a doctrine consisting of moral principles for the guidance and evaluation
of political actions and institutions. One can derive a concept from a theory
but not the other way round.'47
Because ordinary language embodies no consistent outlook, the appeal
to ordinary language cannot even begin the task of developing a theory of
freedom. Even if ordinary usage did disclose a coherent conception of
freedom - say, a conception of the liberal sort which Cohen deploys
unreflectively and without criticism - the conceptual truths derivable from
an analysis of such usage would amount only to local knowledge of our
current linguistic practices. It is wholly unclear to me why Cohen supposes
that local knowledge of this sort should or could be authoritative in moral
and political philosophy. Many ordinary-language usages, or 'concepts',
embody and express repugnant moral judgements. (Consider the rich
terminology of popular racism.) Why should the deliverances of ordinary
thought and practice have any claim on reason, or any weight at all in
philosophical inquiry? How, in particular, are they supposed to be able
to support principles for the assessment and regulation of acts and
 Against Cohen on proletarian unfreedom 147
institutions? They might do so, if (once again with Kant) it were supposed
that only one set of practical maxims would emerge from an application of
the categories of our understanding. So far as I am aware this heroic
supposition is not one that Cohen has endorsed, but without it, the results of
the analysis of ordinary language are likely to be inconclusive or merely
conservative. Finally, even if such criticisms could be countered, the
archaic Rylean methodology which (at least in his work on proletarian
unfreedom) Cohen practices fails to address the powerful scepticism about
meaning voiced by many recent philosophers, such as Quine and Kripke.
What moral and political philosophy demands, instead of the pursuit of
illusory concepts, is the construction of theories - theories which at once
latch on to features of the real world and track our dominant moral con-
cerns. Granted, such theorizing should be conducted with as much clarity
as we can achieve. It should also, and crucially, help us to understand why
freedom is restrained in the real world. The philosophical method which
Cohen employs encompasses a disseveration of concepts from theories and
theories from values in political thought. Now it may well be that there are
some terms in our discourse, such as 'power' ,48 which are best theorized in
value-free terms, but I doubt that freedom is such a term. Everything
suggests we are on firmer ground if we think of freedom as a moral notion.49
The task of a theory of freedom is to give freedom a definite content by
reference to a larger moral and political theory. Most particularly, it is to
specify the liberty that is demanded by justice. The demands of justice are,
further, to be explained in terms of the requirements of the well-being of
individuals - which need to be spelled out in terms, not only of their basic
human needs, but also by reference to their cultural traditions and historical
[Link] The structure of liberties demanded by justice will vary
according to these aspects of individual well-being. The task of theorizing
freedom is, then, one which necessitates a close familiarity with the actual
needs of human beings and with the cultural and historical contexts in
which these needs are shaped. This task is not advanced by a bankrupt
philosophical method in which descriptions of local linguistic behaviours
masquerade as fundamental truths of moral and political life.
A theory of freedom, accordingly, will treat freedom as a moral notion.
It will embed that notion in a larger theory. That larger theory will have to
do, in significant part, with what human beings are like and with the way
things are in the world, but it will also express our evaluation of human
beings and the world. A theory of freedom should have explanatory power
as well as moral force. It should help us to understand the world as much as
it equips us to assess it. In particular, it should do what Cohen's account
signally fails to do - help us to assess the on-balance freedom achieved by
rival institutional frameworks.
148 Critiques
CAPITALISM, SOCIALISM AND ON-BALANCE FREEDOM
How, then, are the rival institutions to be assessed? As Cohen himself
acknowledges, making such an assessment is no easy matter. 'Each form of
society (capitalist and socialist) is by its nature congenial and hostile to
various sorts of liberty, for variously placed people .... Which form is
better for liberty may depend on the historical circumstances' .51 As Cohen
again acknowledges, there are two distinct questions about capitalism,
socialism and freedom. 'The first, or abstract question, is which form of
society is, just as such, better for freedom, not, and this is the second, and
concrete question, which form is better for freedom in the conditions of a
particular place and time' .52 In respect of the first, abstract question, Cohen
offers 'the following intractably rough prescription':
Consider, with respect to each form of society, the sum of liberty which
remains when the liberties it withholds by its very nature are subtracted
from the liberties it guarantees by its very nature. The society which is
freer in the abstract is the one where that sum is larger. 53
Cohen's prescription, as it stands, is not easy to apply. There is no
mechanical way of individuating liberties, and so no mechanical way of
computing which society has the greater sum of liberties. If by 'liberties'
Cohen (in consistency with his general conception of liberty as non-
interference) means acts which persons are not interfered with by others in
doing, such liberties are very different in the importance they have to those
who possess and exercise them. Even if a mechanical procedure were
available whereby we could individuate and count liberties, such a pro-
cedure would for this last reason (that it could not attach weights to the
liberties so specified) fail to tell us what we want to know - namely, how
much liberty each society contains. This is only to state once again a result
of recent liberal thought - that judgements of degrees of freedom on-
balance cannot as a rule be made without invoking standards of importance
in respect of the liberties being evaluated. 54 This is to say that, even though
there are central cases where we have no difficulty in making assessments
of on-balance or comparative liberty, a libertarian calculus is an
impossibility.
This is not to say we are left without resources in the attempt to weigh
capitalist and socialist institutions as to the comparative freedom they
contain, whether in general or for workers specifically. We have in fact
several methods of proceeding. One is to make the move of disaggregation
or decomposition of liberty into basic liberties which John Rawls does in
his A Theory of Justice and subsequent writings. 55 For the most part,
Rawls's basic liberties are the civil liberties to which Cohen refers when he
 Against Cohen on proletarian unfreedom 149
tells us that these freedoms of speech, assembly, worship, publication,
political participation, and so forth are not necessary concomitants of
capitalism. 56 If we accept Rawls's list of the basic liberties, we can assess
capitalist and socialist societies in respect of them. In the real world, though
it is true that capitalism is not always accompanied by the basic liberties, it
is no less true that the basic liberties have never been found in the absence
of capitalism. There is not a single historical example of a socialist or, in
general, a non-capitalist society in which these basic liberties are respected.
In the historical circumstances with which we are familiar, there is no doubt
which form of society is better for liberty. And, in so far as workers value
the basic liberties, there is no doubt which form of society is better for
workers and their liberty.
Against the Rawlsian move, which seems entirely unambiguous in its
results, Cohen may make a number of objections. He may object that,
however true it is in our current historical circumstances that socialist
societies fail to protect the basic liberties, the argument from the basic
liberties against socialism is nevertheless inconclusive. Perhaps we ought
to regard our current historical predicament as something to be rejected or
overcome, rather than simply accepted, and perhaps a socialist society is
achievable in which the basic liberties are protected. This is an objection I
will not address, though there are powerful reasons in positive political
theory57 for rejecting it, since my own argumentative strategy on this
question is satisfied if it is admitted that our present historical context
answers the question of on-balance freedom decisively in favour of capital-
ist institutions.
Cohen might make another, different objection to the Rawlsian move I
have made. The Rawlsian move depends on excluding entitlement to pro-
perty in the means of production from the set of basic liberties. Since
property rights in the means of production, individual or communal, do not
figure among the basic liberties as Rawls theorizes them, the choice between
capitalism and socialism cannot tum on how their respective property insti-
tutions create and sustain liberty. (The choice may be, and should be,
informed by a theoretical conjecture about the causal role of capitalist and
socialist property institutions in sustaining the basic liberties. That is another
question.) If we bring the property system back into the assessment of
on-balance liberty, it may be that the unequivocal result we earlier obtained
no longer holds. A society might curb some of the basic liberties and yet,
because its property institutions extended liberty in important ways, do
better from the standpoint of liberty than a society in which the basic liberties
are perfectly protected. This is the view of many Leninists about the Soviet
Union. I do not say that that is Cohen's view of the Soviet Union, but it does
give him a way of resisting the upshot of the Rawlsian move.
150 Critiques
Such a countennove does not save Cohen's case, however. Let us
forswear the idiom of the basic liberties and adopt the conception of
freedom as non-interference which Cohen deploys. Let us accept the result
which Cohen derives by applying that notion of freedom to capitalist
institutions - the result that capitalists are freer than proletarians. They are
freer in the sense that, because they have resources in the means of pro-
duction, they are less often (or less significantly) interfered with in doing as
they wish, and so (let us say) have more or better options at their disposal
than proletarians do. Accepting Cohen's conception of freedom and the
resultant inequality of liberty under capitalism, how does workers' freedom
fare under socialism? There seem to be clear respects in which workers'
options will be severely diminished under socialist institutions. It is an
important argument against socialism that, in transferring the control of
employment from a diversity of competing employers to a single public
authority, it will unavoidably curtail the options of workers. The point has
been well put by Hayek:
That the freedom of the employed depends upon the existence of a great
number and variety of employers is clear when we consider the situation
that would exist if there were only one employer - namely, the state-
and if taking employment were the only pennitted means of livelihood
. . . a consistent application of socialist principles, however much it
might be disguised by the delegation of the power of employment to
nominally independent public corporations and the like, would neces-
sarily lead to the presence of a single employer. Whether this employer
acted directly or indirectly he would clearly possess unlimited power to
coerce the individual. s8
Or, as Leon Trotsky, one of the chief architects of the Soviet system, put it
pithily: 'In a country where the sole employer is the state, opposition means
death by slow starvation. The old principle, who does not work shall not
eat, has been replaced by a new one: who does not obey shall not eat.'S9
Cohen has told us that, under capitalism, workers 'face a structure
generated by a history of market transactions in which, it is reasonable to
say, they are forced to work for some or other person or group. Their natural
rights are not matched by corresponding effective powers. '60 It is an impli-
cation of Hayek's argument that, in a socialist economy, workers will face
a structure generated by political power in which, it is reasonable to say,
they are forced to work for the state. Their legal rights are not matched by
corresponding effective powers.
Against Hayek's argument, it might be objected that it fails to take
account of the possibility of a fonn of market socialism, in which there is a
 Against Cohen on proletarian unfreedom lSI
diversity of worker cooperatives and so a variety of employment oppor-
tunities for workers. It is not to be doubted that such a decentralized socialist
system would likely be better for liberty than any centralist system could.
Still, there will be important respects in which workers' liberty will be
curtailed under market socialism. The problem is clearest when we consider
workers who belong to a minority group. Communal systems of ownership
of productive resources may be expected to find it hard to permit such
minorities to advance, perhaps most particularly when they are operated by
democratic procedures. Consider, by way of example, what would likely
have been the fate of immigrants of alien cultural traditions, in Britain and
the United States, if they had had to gain access to productive resources
solely through a democratic political process. Whenever there is a pre-
judiced majority, such minorities are likely to have narrower options even
under market socialism than they do under market capitalism, where a few
self-interested capitalists are sufficient for them to be able to borrow or rent
resources on the basis of which they can build up capital of their own.
Even when discriminating majorities are lacking, workers' options are
limited by the necessity each faces of having in his cooperative to secure
the agreement of his fellows for the introduction of any novel practice. As
Hayek has again put it, 'action by collective agreement is limited to
instances where previous efforts have already created a common view,
where opinion about what is desirable has become settled, and where the
problem is that of choosing between possibilities generally recognized, not
that of discovering new possibilities' .61 This point is reinforced when we
realize that, because of the fusion of capital ownership with job occupation
under market socialism, the dependency of the individual worker on his
cooperative will be considerably greater than his dependency on his
employer in a capitalist society.
As for the worker cooperative itself, it too will be in a circumstance of
dependency. If market socialism is to remain a form of socialism, pro-
ductive capital will have to be subject to political allocation rather than
private provision. Workers will have access to capital only through the
agencies of the socialist state. Because private capital is forbidden, no
worker can individually acquire enough capital to live on. Even as a
member of his collective, he is always dependent on the capital that is
allocated him by government. Perhaps worker-cooperators will have dis-
cretion over how much of their profits they set aside for investment in the
enterprise and, to that extent, a part of their capital might be self-generated.
But it seems clear that start-up capital and capital needed to stave off
bankruptcy, or permit large-scale expansion, will have to be politically
allocated. Workers under a market socialist system will confront a situation
152 Critiques
in which sources of capital are concentrated in one, or a few, state investment
banks. They will have far fewer sources of capital than existed under
capitalism. It is hard to see how this fact can avoid restricting workers' options.
The general point behind these arguments is that the freedoms generated
by capitalist institutions are not only those enjoyed or exercised by the
owners of capital.62 This is a point recognized by Cohen when he notes that
capitalism is a liberating system by contrast with its predecessor systems,
such as slavery and feudalism.63 He omits to note that many of these
freedoms, depending as they do on a multiplicity of buyers of labour and
sources of capital, would be extinguished by state socialism and diminished
under market socialist institutions. Capitalist institutions may, then, be
defended as institutions that promote workers' freedom better than any
realizable alternative. This may be so even if it be conceded that workers
are less free than capitalists. For it is still arguable that they are freer than
they would be under any other institutions. All forms of socialism, in
particular, appear to diminish that control over the disposition of his own
labour and person which the worker gained with the advent of capitalism.
It remains unclear on Cohen's account what are the freedoms conferred on
the worker by socialism which might compensate for this loss.
If I am right in arguing that workers, even as a class, are better off in
terms of liberty under capitalist than under alternative institutions, this
supports a defence of capitalism that is to be conducted strictly in terms of
its liberty-promoting effects. This argument is that the inequalities in
liberty which Cohen fmds in capitalist institutions are those which maxi-
mize the freedom of workers. Capitalism might then be defended as an
economic system which satisfies a variation on Rawls's difference
principle - a variation in which it applies only to liberty (and not to the
other primary goods). Unlike Cohen, I do not suppose that an argument for
capitalism from freedom can or should be severed from one that invokes
justice. Nor, indeed, do I imagine that the demands of justice can be
specified independently of an account of the well-being of those concerned
- though I have not tried to give any account of well-being myself. The
considerations I have advanced suggest that capitalism may nevertheless be
defended by reference to an argument of freedom alone. The result of these
considerations as to substantive liberty on-balance under capitalism and
socialism is that capitalism does best for liberty, even for those within
capitalism who have the least liberty.
The comparative assessment in terms of freedom of socialism and
capitalism may take a third form - by considering collective unfreedom
under each set of institutions. Let us, in this connection, go so far as to
concede that most proletarians are collectively unfree to become capitalists,
thereby putting aside the possibility of a capitalism-without-a-proletariat
 Against Cohen on proletarian unfreedom 153
which I mentioned earlier. If we do this, we may say (in another idiom than
Cohen's) that being a capitalist is in a capitalist society a positional good 64
- one that cannot be possessed or enjoyed by all. It is just in virtue of its
positionality that Cohen argues for the collective unfreedom of proletarians
in respect of it. Now perhaps, in the real world, being a capitalist is a
positional good, as Cohen and other Marxists suppose. A real collective
unfreedom is thereby generated for workers. This tells us nothing, however,
as to the relative or comparative collective unfreedom of workers under
capitalist and socialist institutions. It does not even adequately explain the
moral significance of the collective unfreedom of the proletariat under
capitalism. It shows only that under capitalism workers suffer a collective
unfreedom - the unfreedom to become a capitalist - which, it is supposed,
socialism would abolish. I have argued earlier that socialism would, in
effect, universalize and entrench this unfreedom. Let us accept Cohen's
argument to his conclusion, though, and see how workers fare as to their
collective freedom under socialism.
Under socialism, then, we shall allow, workers will be rid of the collec-
tive unfreedom to become capitalists. At the same time, it is likely that a
good many other collective un freedoms would be spawned under
socialism. It is a feature of all forms of socialism that resources which are
allocated by market processes under capitalism are subject to political
allocation instead. This is true even under market socialist institutions, in
which at least investment capital is politically allocated. In the real world,
however, political power is itself a positional good. It cannot be had or
exercised by all equally. The positionality of political power entails that,
when resources are subject to political control and allocation, they too
acquire the attributes of positionality. Thus in 'actually existing' socialist
societies, education, housing and health services exhibit a degree of posi-
tionality which likely far surpasses that possessed by analogous services in
Western capitalist nations. Workers in such socialist states suffer a degree
of collective unfreedom in respect of the Communist ~lite' s access to good
apartments, hard-currency stores, higher education for their children, ade-
quate medical care, and so forth, which plausibly exceeds any similar
positionality in capitalist societies.65
Attempts to explain this fact and which rely on ad hoc claims about
historical backwardness, illiberal traditions, or inauspicious circumstances
in the societies concerned neglect systematically the role of political allo-
cation of resources in shaping the incentives of individuals in such
societies. By contrast with societies in which resources are primarily allo-
cated by markets, individuals in socialist societies have an almost irresist-
ible incentive to seek positions of power in the party apparatus from which
they can assure benefits to themselves and their offspring. Hence the
154 Critiques
predictable transfer of command positions and their associated benefits
across generations of nomen1clatura in the socialist states. By contrast, in so
far as access to goods is in capitalist societies mediated via the market, it is
inherently likely to display less positionality than in socialist states. This is
so because, except where they are highly monopolistic, markets tend to
redistribute resources unpredictably across economic agents and, most
especially, across generations.
In general, we may expect that, with the politicization of resource
allocation that socialism brings, goods which hitherto had not been posi-
tional will become so. In so doing, they will generate collective unfreedoms
in respect of them which had not existed before. The pattern of collective
unfreedoms in society will mirror that of the possession of the supreme
positional good, political power. In this situation, socialist workers will be
imprisoned as a class, even if (dubiously enough) each of them is free to
join the ruling 6lite. The collective unfreedom which they will suffer in
respect of the positional good of political power will, in its tum, spawn
collective unfreedom in respect of all the goods that political power allo-
cates - that is to say, most of the goods of economic and social life. It is
difficult to resist the conclusion that workers under socialist institutions
will suffer a degree of collective unfreedom unknown in capitalist societies.

CONCLUDING REMARKS
The arguments Cohen adduces as to proletarian unfreedom are mostly
formal. They support truisms which teU us nothing of substance about the
advantages in terms of freedom of capitalist over socialist institutions.
When we turn from conceptual analysis to realistic considerations, we find
powerful reasons for supposing that socialist institutions will do worse than
capitalist institutions from the standpoint of workers' freedom. We find that
the parity of unfreedoms between the two systems that is supported by
Cohen's atavistic philosophical methodology is, in fact, entirely delusive.
In other words, if the goal of Cohen's argument is to demonstrate the
existence of unfreedom under capitalism, it succeeds - but only because the
unfreedom he discusses exists in every institution of property (and probably
in every imaginable society). It seems decidedly implausible that Cohen
should devote five papers, which develop a novel and implausible con-
ception of collective unfreedom and contrive to make a host of minute
distinctions in the discourse of freedom, to establish so banal and trifling a
result. If, on the other hand, Cohen's goal is the more interesting one of
showing that worker's freedom, or freedom in general, is greater under
socialist institutions than within capitalist institutions, his argument to this
result demonstrably fails. When we consider the substance of things, we
 Against Cohen on proletarian unfreedom 155
have every reason to think that workers' freedoms will flourish best under
capitalist institutions. Oddly enough, this supposition is amply confirmed
by historical experience. And that is another argument in favour of a form
of theorizing which, unlike Cohen's, seeks to explain and assess the free-
dom and unfreedom we find in the real world.
12 Totalitarianism, reform and
civil society

What must be true for post-totalitarianism to be a possibility? This question

- a question as to the conditions under which totalitarian political orders
may mutate, or successfully transform themselves, into orders, authori-
tarian, liberal or otherwise, in which the constitutive features of totali-
tarianism have been transcended or suppressed - presupposes a conception
of totalitarianism itself. The history of the idea of totalitarianism, however,
is a history of controversy. Presented initially by theorists and practitioners
of Italian fascism as a positive notion,l the idea of totalitarianism acquired
in post-War literature2 an unequivocally pejorative connotation, being used
to capture the common features of National Socialism and Soviet Com-
munism in their worst and most terroristic periods. It was later subject to
criticism and attack, both by Marxists and by others,3 who sought to deny
its theoretical coherence and utility and who tried to represent it as an
ideological construct of liberalism and conservatism which was deployed
for the purposes associated with (or attributed to) the Western powers
during the period of the Cold War. My inquiry is based on the rejection of
this latter view and on a commitment to the validity and usefulness of
totalitarianism as a central category in contemporary social and political
theory. This does not mean that I endorse the conventional conception of
totalitarianism, as developed in the post-War Anglo-American literature.
Legitimate as it may be, this conventional usage obscures a number of
interesting questions. Is totalitarianism necessarily or inherently terroristic,
or is terror a developmental phase which some totalitarian orders pass
through? This is a question with obvious relevance to the possibility of
post-totalitarianism, since it allows that a totalitarian regime may sustain
and reproduce itself by means other than the threat of violence. Can the
institutions of democratic capitalism be characterized as encompassing a
totalitarian order, as Marcuse and others4 have alleged? Is totalitarianism a
uniquely modem phenomenon, or are there instances of it in ancient times?
 Totalitarianism. refonn and civil society 157
Can post-totalitarian orders in which civil societies have clearly emerged,
be subject to successful retotalitarianization? Or is the move from totali-
tarianism typically irreversible?
What of the cultural roots of totalitarianism: is it alien to the Western
tradition, an aberration within it, or a development of elements of the most
central, ancient and fundamental Western traditions? I shall approach these
and related questions by way of a critique of the received notion of totali-
tarianism and of its standard applications to the Soviet case. The standard
conception of totalitarianism equates it with such phenomena as mass terror
and a charismatic leader with a cult of personality. It suggests that, in the
absence of clear evidence of these phenomena, we have a post-totalitarian
regime. By contrast, the conception of totalitarianism I shall advance
enables us to recognize as neo-totalitarian the forms of social and political
control that have emerged recently in the Soviet Union, and to distinguish
such a neo-totalitarian regime from the genuine post-totalitarianism that
exists in Poland, Hungary, and, more precariously, in the Baltic states.
The danger of the conventional conception is that it may blind us to the
varieties of totalitarianism. This is not to say that totalitarianism is a sort of
cluster-concept, with the variety of totalitarian regimes being unified only
by a pattern of family resemblances. For I shall maintain that the totalitarian
project is constituted by a single objective - that of merging state and
society in a new order from which the conflicts of interest, purpose and
value which are found in all historic societies have been extirpated. This
definitively totalitarian objective may be stated with greater precision. The
totalitarian project is the project of suppressing civil society - that sphere
of autonomous institutions, protected by a rule of law, within which indivi-
duals and communities possessing divergent values and beliefs may coexist
in peace. Implementing this project involves another, and even more stu-
pendous project: that of remaking the identities of those who have come
within the sphere of totalitarian power. A number of theorists, such as
Heller,S have seen in this project of making over human nature on a new
model the most essential, definitive feature of totalitarianism. I shall argue
that, whereas the totalitarian project has in many instances succeeded in
destroying civil society, it has nowhere forged a new humanity. I shall take
the destruction of civil society to be the chief historical result of totali-
tarianism. Indeed, on the view I shall defend, the suppression of civil
society is not so much a consequence, or a side-effect, of totalitarianism, as
its very essence.
It is for this reason, again, that I shall affirm the radical modernity of the
totalitarian phenomenon. Totalitarianism is uniquely modem, not only
because it presupposes a contemporary technology of repression, but
chiefly because it expresses a revulsion against the distinctively modem
158 Critiques
institutions of civil society. It did so even in China, where Communism
sought to legitimate itself as a project of modernization. I shall affirm the
radical modernity of totalitarianism as a political order, even though I will
maintain that its cultural origins are in elements of the most ancient and
central Western values and ideas (and not, for example, in the traditional
cultures of Russia and China.)
In particular, interpretations of the Soviet Union in terms of the antique
categories of tyranny or empire miss the mark by invoking concepts to
capture realities of which Montesquieu or Aristotle could have had no
conception. When Pipes argues that 'the techniques of police rule, intro-
duced piecemeal by the Russian imperial regime, were first utilized to their
fullest potential by their one-time victims, the revolutionaries',6 he neglects
the very radical discontinuities (which I shall later spell out in detail)
between the magnitude of repression between the Tsarist and Bolshevik
regimes, and, above all, he passes over the fact that the Tsarist secret police
was never employed to liquidate entire social groups.
Equally far from the mark is the claim that totalitarianism existed in
ancient times. When Barrington Moore cites the adoption in ancient China
of 'the famous pao-chia system of mutual surveillance, which resembles
and long antedates modem totalitarian procedures' he confuses totali-
tarianism with a technique which has been adopted by a variety of types of
government. When he goes on to refer to 'the hsiang-yueh system of
periodically lecturing the population on Confucian ethics' as another
'revealing precursor of modem totalitarian practices ... they demonstrate
conclusively that the key features of the totalitarian complex existed in the
premodern world',1 he neglects the revolutionary character of totalitarian
ideological indoctrination.
Totalitarian ideologies are instruments of social transformation and seek
to displace traditional systems of belief. That is why the academic canard,
that Communism is a modem secular or political religion, is true only in the
sense in which occultism and theosophy are religions - the sense in which
all are modernist projects of supplanting the mysteries and tragedies of
inherited faiths with gnostic techniques of liberation, represented as appli-
cations of scientific method. There is an inherent paradox in totalitarianism
in that it deploys modem ideology in the service of an anti-modernist project.
Because it embodies a revolt against modernity, totalitarianism can only be
its shadow. If totalitarian orders are to be reformed, accordingly, such reform
can be achieved only by way of a rediscovery of the most distinctive
institutions of the modem world - the institutions of civil society.
The question of the possibility of post-totalitarianism, accordingly, is the
question whether totalitarian regimes may so alter their nature as to allow
the re-emergence or re-invention of those institutions of civil life to whose
 Totalitarianism. reform and civil society 159
destruction they were committed. Indeed, since the emergence of civil
society encompasses a mutation in the very nature of totalitarianism, a
question inevitably arises whether totalitarian orders are capable of reform,
and, if they are not, how are current developments to be understood?
It is enmity to civil society, with its constitive institutions of private
property and the rule of law, rather than opposition to liberal democracy,
which is most nearly definitive of totalitarianism. 'Civil society' here refers
to the domain of voluntary associations, market exchanges and private
institutions within and through which individuals having urgent concep-
tions and diverse and often competitive purposes may coexist in peace. It is
this conception, I believe, that is held in common by such otherwise very
different theorists as Locke and Hegel. In both, civil society is distin-
guished sharply from the state. But each recognizes that the institutions of
civil society need definition and protection by law, and so cannot flourish
without the shelter of government. There are many types of government,
however, under which a civil society may exist - as many varieties of
authoritarianism, for example, as there are of democracy. Only the un-
limited government of totalitarianism is incompatible with civil society.
For this reason, non-totalitarian, and indeed post-totalitarian regimes may
come in a variety of forms, of which liberal democracy is only one. In all of
its varieties, however, post-totalitarianism represents the inversion of the
totalitarian order, not a terminal phase in its development. For a totalitarian
regime to cease to be such, it must be able to recall the civil life it has lost,
or else to fashion anew the institutions of civil society. My purpose in this
paper is to speculate about the conditions under which actually existing
totalitarian regimes might successfully negotiate this transformation and so
resurrect, or adopt, a civil society.
In fact. my project here will be the more limited one of conjecturing as
to the prospects of such a metamorphosis in Soviet-bloc regimes. I pass
over as not centrally germane to my purpose the example of Nazi Germany,
in which a totalitarian regime perished through military defeat, and all of
the Fascist states, none of which was genuinely totalitarian (even where, as
in Italy, it claimed to be). I pass over these other potential instances to
totalitarianism, primarily because the Soviet example constitutes the para-
digm case. In terms of chronology, the Soviet Union was the modem
world's first totalitarian state. Further, its example animated every subse-
quent totalitarian (or would-be totalitarian) state. As Schapiro has said:
The originator of the technique of mass manipulation, in modem times,
was Lenin; not only Stalin, but Mussolini certainly and Hitler (through
the German Communists) indirectly, learned it from him.... Lenin
never became (or ever wanted to become) a Leader such as Hitler or
160 Critiques
Stalin; but in the course of building up his technique of mass mani-
pulation as part of the process of ensuring victory for his party he
provided a model for these very different men.·
Lenin initiated not only totalitarian techniques of mass manipulation, but
also most of the other distinctive institutions of totalitarianism - such as the
concentration camp and the extra-legal powers of the secret police. As
Nolte has argued,!I there is also a case for ascribing to the Bolsheviks the
responsibility for pioneering the techniques of mass extermination which
the Nazis later emulated in their genocides. For these reasons, then, I feel
justified in focusing on Soviet-type instances as paradigm cases of totali-
tarianism. For a different reason, lack of space, I do not discuss in any
systematic fashion the potentially very important case of Communist China
other than by brief comments on the contrasts between traditional forms of
despotism in China and Chinese Communist totalitarianism and a short
comparison of the prospects of economic reform in China and the Soviet
Union. I restrict the scope of my inquiries to the topical, but also globally
decisively important question: Can the reform movements within the Soviet
bloc - such as perestroika in the Soviet Union, Solidarity in Poland, and the
various experiments in mUltiparty pluralism in Hungary and in the tolera-
tion of oppositional movements in the Baltic states - succeed to the extent
of re-establishing civil societies, or something akin to them? My project
here is not the hazardous one of prediction, though I will speculate as to the
likely failure of perestroika, but that of theorizing the conditions of, and
constraints upon, the emergence of post-totalitarian regimes within the
Soviet bloc. Pursuing this theoretical goal involves me in making signi-
ficant distinctio,}s between circumstances and prospects in different coun-
tries, particularly as to the causal origins of the various reform movements.
It also implicates me in the business of attempting to specify those features
of Soviet totalitarianism which give it its essential character and which are
most relevant to the possibilities of post-totalitarianism.
Here I shall make little use of the famous five-point syndrome of
totalitarian regimes proposed by Carl Freidrich in 1954,10 since it conflates
historically contingent aspects of Soviet-style totalitarianism with its con-
stitutive features. I shall focus instead on what I shall maintain are the three
key features of totalitarianism - its suppression of the institutions of civil
society, and especially of the rule of law, the central planning of the
economy and the character of the totalitarian regime as a weltan-
schauung-state. And, to anticipate my conclusions, I shall submit that
current movements for reform, such as perestroika in the Soviet Union, are
responses to the ruinous failure of the command economies - a failure that
has left even the servile nomenklatura of these regimes underrewarded. The
 Totalitarianism, reform and civil society 161
tactical objective of the Soviet refonn process is the refmancing of the
system by Western credit. The strategic objective is the stabilization of the
Soviet totalitarian order - which incorporates the goals of engineering a
shift in the distribution of military power in Europe and the forging of
subsidy relationships between the BEC and the Comecon countries. In all
important respects the Soviet perestroika embodies an authentically
Leninist strategy which includes a significant component of strategic dis-
infonnation. The crucial question - which I shall address towards the end
of this paper - is whether the new Leninist strategy of the Soviet leadership
risks unleashing uncontrollable forces in the Soviet bloc, and in the Soviet
Union itself, which threaten the system's very existence, or whether, des-
pite its colossal failures, the Soviet system can again renew itself through a
combination of selective repression and Western aid. There is also the
question whether, even granted massive doses of Western capital and
technology, the Soviet system can do anything more than stave off for a
while the prospect of a catastrophic economic collapse, with all the un-
predictable political consequences that would have. In any case, we need to
ask what are the implications for the prospects of post-totalitarianism in
Poland, Hungary, the Baltic states and elsewhere in the Soviet bloc. Before
we can even approach these questions, however, we need to consider what
are the constitutive features and decisive episodes in Soviet totalitarianism.

THEORIZING SOVIET TOTALITARIANISM: THE

CONVENTIONAL METHODOLOGY
We may begin by making some cautionary remarks on the methodology of
conventional Soviet studies. The conventional wisdom among Sovieto-
logists is that the Soviet state is a state like any other, whose distinctive
features are derived from Russian political and cultural traditions. Accord-
ing to this dominant view,1l the Soviet state is a tyranny, in whose fonna-
tion Muscovite traditions of authoritarian rule are at least as significant as
Marxist-Leninist ideology. If this is so, then the methods appropriate to the
study of the Soviet state are no different from those used with any other
state - the methods of standard political science, such as theories about
political culture, the study of interest groups, of the modernization of
institutions, and so on. It is no exaggeration to assert that this methodology
has been thoroughly unproductive and indeed counterproductive in its
impact on Western perceptions of the Soviet Union. It is radically defective
for a number of reasons. In the first place, this conventional methodology
is bound to exaggerate grossly the importance of cultural factors in explain-
ing Soviet totalitarianism. That this is so is demonstrated by the profound
similarities between Soviet-type regimes having very different cultural
162 Critiques
traditions. Despite their diverse political traditions, the East European
states exhibited during the Stalinist period structural affinities - such as the
campaign against religion and the attempt to impose a single weltan-
schauung, the attack on the peasantry, show trials, a terroristic secret police
and the regimentation of intellectual life - which are explicable only by
reference to their being modelled on the Soviet totalitarian state. Again, the
structural affinities between the periods of socialist construction in the
Soviet Union and in Communist China - the reliance on slave labour in a
vast Gulag, the catastrophic waste and malinvestment generated by a
command economy which promoted rapid industrialization at the expense
of the ruin of agriCUlture, and repression of minority peoples - overwhelm
any divergences between the two regimes which their different cultures
may account for. Yet again, Communist regimes remote from the Soviet
Union, and acting on radically different cultures, have emulated Stalinist
policies aimed at the destruction of peasant life - as in Ethopia, where the
Communist government goes so far as to call its policy for the forcible
resettlement of the peasantry by the term, 'systematization', used for the
same policy in today's Stalinist Romania. Other examples, from Cuba,
Mozambique and Angola, for instance, could easily be used.
finally, whatever their cultural contexts, all Communist regimes display at
present some decisive common features - the existence of an extensive parallel
economy, the allocation of such goods as education and housing by a conupt
and exploitative nomenklatura and a consequent pattern of extreme social
stratification - which dominate their respective cultural inheritances. This may
be affirmed, even if it be readily allowed that there are distinctive cultural
factors - in Russia, Poland and Bohemia, for example - which have modified
somewhat the development of the totalitarian regimes there.
My strategy of argument will be to subject the conventional metho-
dology of Soviet studies to an historical critique, in the first instance, by
confronting it with six aspects of Soviet totalitarianism that it syste-
matically neglects or underestimates:
• the economic, political and cultural achievements of late Tsarism;
• the dependency of the Soviet economy on Western capital and
technology;
• the central role of Marxist-Leninist ideology in Soviet development;
• the specifically Leninist origins of Soviet totalitarianism;
• the economic functions of the forced labour camps in the period of
socialist construction and the genocidal character of the policy of agri-
cultural collectivization; and
• the use by the Soviet leadership of techniques of strategic dis-
information.
 Totalitarianism, refonn and civil society 163
Having illustrated the poverty of the conventional methodology of Soviet
studies by these examples, I shall proceed to attempt to theorize Soviet
totalitarianism, more positively, by reference to the three characteristics
that distinguish it most radically from a civil society - its lack of the
institutions of market exchange and of their indispensable matrix, the rule
of law, and the continuing identification and legitimation of the Soviet state
by reference to the ideology of Marxism-Leninism. Soviet totalitarianism is
not a despotism or a tyranny, but an economic chaos contained in a political
state of nature. Only by so understanding it can we grasp its ruinous poverty
together with its awesome capacity for self-reproduction. We understand it
adequately, finally, when we see the necessity to the Soviet regime of its
third feature - its legitimation by a Marxist-Leninist ideology that at once
animates the present policy of perestroika and pro- hibits the only real
reform - the recreation of civil society.
The alternative methodology I shall adopt in this chapter is distinguished
by its taking seriously the role of Marxist-Leninist ideology in accounting
for Soviet policy. It postulates that the Soviet state has no legitimacy, and
(aside from the KGB) barely an existence, apart from that system of ideas.
It focuses on the features that distinguish the Soviet system from authori-
tarian states - in particular its control of information and its involvement of
its subjects in the processes and practices whereby it reproduces itself. In
seeking a better understanding of the Soviet system than that found in the
conventional methodology, I shall make use of two bodies of theory whose
salience to the Soviet phenomenon is persistently neglected in the con-
ventional wisdom - Austrian economic theory and Virginia Public Choice
Theory. Applying these bodies of theory to the Soviet system generates an
alternative perspective on it in which its most distinctive features are
captured and rendered intelligible. This alternative methodology does not
deny that there exist within the Soviet system the phenomena studied in the
standard approaches - it acknowledges that there are competing interest
groups, just as it emphasizes that there are markets, and it does not deny that
native Russian traditions are to be found within the Soviet system - but it
affirms that these phenomena tend to reinforce the stability of the Soviet
totalitarian order (often by being manipulated by it) rather than to weaken
it. By its virtual elimination of the private sphere within which such
phenomena arise and flourish in pluralist and in authoritarian states, Soviet
totalitarianism drastically modifies their typical effects. It creates an
altogether novel political order - a lawless Leviathan whose subjects are
victims of a recurrent Prisoner's Dilemma in which they are constrained to
reproduce the system that enslaves them.
164 Critiques
THE ACHIEVEMENTS OF LATE TSAIUSM
The dominant methodology involves a massive distortion of Russian his-
tory. It neglects the emergence, during the last half~ntury of the Tsarist
regime, of a strong and resilient (if never wholly dominant) tradition of
liberal legal and political philosophy, whose development has been defini-
tively analysed by Waliclci,I2 and which made an important contribution to
the constitutional reforms of the regime's last two decades. It grossly
exaggerates the character of late Tsarism as a police state (as Norman Stone
and others have shown)13 and underestimates the many characteristics it
had in common with other authoritarian states in Europe, such as Prussia.
And the conventional wisdom which represents the Soviet Union as the
continuation of Tsarist traditions disregards evidence of the extraordinary
economic, cultural and social progress which occurred during the last
decades of Tsarism. Let us be specific. In a book published six months
before the outbreak of the First World War, the French economist Edmond
Thery notes that in the five-year period from 1908-12, coal production
increased by 79.3 per cent over the preceding five years; iron by 24.8 per
cent; steel and metal products by 45.9 per cent. Even allowing for inflation,
the output of heavy industry increased by 74.1 per cent from 1900 to 1913.
Again, between 1890 and 1915, the rail network almost tripled. 14 As to the
magnitude of foreign investment in Russia - frequently, if often in-
consistently, cited as evidence of Russian backwardness - Norman Stone
has noted that on the eve of the First World War foreign investment in
Russia had declined as a proportion of total investment by from one-half in
1900 to a fifth in 1914. By 1909 Russia had already become the world's
fourth industrial power. 15
Nor was progress confined to industry. Thery reports that, in agriCUlture,
wheat production between 1908-12 rose by 37.5 per cent over the pre-
ceding five years; rye by 24 per cent; barley by 62.2 per cent; oats by 20.9
per cent; and com by 44.8 per cent. With good harvests, as in 1909 and
1910, Russian wheat exports made up 40 per cent of world wheat exports,
and even in bad years they amounted to 11.5 per cent. In social policy,
spending on education doubled in the decade between 1902-12, and, by
1915, over half of all children between eight and eleven years of age were
in school, and 68 per cent of all military conscripts were literate. Welfare
legislation in Russia followed a Bismarclcian pattern, although in Siberia
conditions approximated more closely to laissez-faire. As in Bismarck's
Prussia, various schemes of workers' insurance were introduced, and late
Tsarist Russia followed Prussia in pioneering the rudiments of the
twentieth-century welfare state. 16 As Dominic Lieven has noted, Russia
followed Prussia, also, in having a civil service that in the last decades of
 Totalitarianism, reform and civil society 165
the nineteenth century 'approached the Weberian ideal type'.17 Although
development was in many areas uneven, Tsarist Russia in its last half-
century was not the autocratic, Asiatic despotism that is the caricature in
conventional theturography, but instead a dynamic, progressive European
state.
During its last half-century, Russia experienced an explosion of cultural
energies, achieving a flowering in the arts - in literature, painting, dance
and religious thought, for example - which has rarely been matched in
European cultural history. The last decades of Tsarism saw much policy
that was clumsy, ill-conceived, or repressive - particularly in the policies
of Russification of the national minorities and discrimination and perse-
cution against Jews. The rapid progress achieved in the economy, in cul-
tural life and in some aspects of policy (but not, alas, in constitutional
affairs, where policy was lagging and faltering) is nevertheless especially
impressive, given the demographic explosion and consequent urban prob-
lems which government confronted. As Heller and Nekrich observe l ., the
statistics for these years are revealing - and when the October Revolution
occurred, it was a direct result of the First World War.
Perhaps the most striking discontinuity is in the area where the con-
ventional methodology finds most similarity between the late Tsarist and
the early Bolshevik regimes - that is, the discontinuity between the repres-
sive activities of the Okhrana and the Cheka - GPU - OGPU. The demon-
stratable discontinuities, detailed below, are most obviously in the
magnitude, scale and intensity of repression. A crucial discontinuity is also
in the quality of repression: the Okhrana, unlike any of the Bolshevik
security services, did not act extra-legally, but within a rule oflaw, and it did
not seek to liquidate whole social categories. It is against this background
that the quantitative comparisons made by Dziakl9 are to be assessed:

LATE-TSARIST PERIOD (1826-1917)

Executions
1826-1906: 894
1866-1917: 14,000 approx.
1866-1900: 48;94
1906 (six months ofthe Stolypin military field tribunals): 950
1907: 1,139
1908: 1,340
1908-12: 6,000
'Following the 1905-7 Revolution': 11,000
'Eighty years that preceded the Revolution in Russia': 17/year
(average)
166 Critiques
Deaths from executions, pogrom murders, and deaths in prison
1867-1917: 25,000

Convicts at hard labour

1913: 32,000 (year largest numbers were reached)

Political exile without confinement

1907: 17.000 (year largest numbers were reached)

Maximum number imprisoned (criminals and politicals)

1912: 183.949

EARLY SOVIET PERIOD (1917-24)

Executions by Cheka and tribunals
1917-23 200,000
1918 and Ist half of 1919: 8,389
1917-20: 12,733
'Civil War': 50,000
1918-19: 1,700,000
1918-23: 2,500,000 per annum

Deaths caused by Cheka

1917-22: 250,000-300,000

Deaths from the suppression of 'rebellions' and from prison and

camp treatment
1917-24: 300,000

Executions in the Crimea following General Wrangel's defeat and

evacuation
1920-21: 50,000-150,000

Hostages and prisoners in camps and prisons (1917-23)

1918: 42,254 hostages/prisoners in camps and prisons
1919 (to July): 44,639 hostages/prisoners in camps and prisons
1918: 47,348 hostages/prisoners in camps and prisons
1919: 80,662 hostages/prisoners in camps and prisons
1920 (late): 25,336 camp inmates plus 24,400 Civil War prisoners;
48,112 prisoners in RFSFR NKYU prisons; 60,000 NKYU
prisoners according to commissar of justice
1921 (Jan.): 51,158
 Totalitarianism. reform and civil society 167
1921 (Sep.): 60,457
1921 (Dec.): 40,913; 73,000 prisoners in NKYU prisons
1922 (Oct.): 60,000
1923 (Oct.): 68,297

It is unequivocally and undeniably clear from these figures that, in addition

to the arbitrary and extra-legal character of the Bolshevik terror, the quanti-
tative increases in the various kinds of repression in the Bolshevik regime,
as compared with those of late Tsarism, are so enormous as to amount
themselves to a qualitative transfonnation in the nature of the regime -
from authoritarian to tetalitarian.
It is not only that, as Dziak has put it, 'an unbroken patrimony between
Tsarist repression and Soviet terms cannot be claimed'. It is not even that,
as he also observes,
Even at the height of its repression against revolutionaries, Tsarist courts
offset Okhrana and Gendarme actions, thereby exercising a restraining
hand. In Lenin's system the courts were either ignored, or became
creatures of the Cheka.20
The crucial point is that the extra-judicial coercion and terror of the Cheka
were directed against whole social groups, and primarily against the
peasantry and workers. According to Soviet statistics used by Dziak, of the
forty thousand NKVD inmates for December 1921, almost 80 per cent were
illiterate or nearly so, and so from peasant or worker stock. As Dziak
concludes, decisively:
It simply was not Tsarist policy or practice to exterminate whole cate-
gories of people.... The Cheka's class war certainly was an 'aristicide'
of the leading sectors of tsarist society, but its more numerous victims
were the very classes it claimed to represent and serve. 21
There is, finally, the evidence of the size of the two security services, with
Richard Pipes noting that in 1895 the Okhrana had only 161 full-time
personnel, supported by a Corp of Gendarmes amounting to less than
to,OOO men, while by mid-1921, the Cheka accounted for approximately
262,400 men, not counting Red Army, NKVD and militiamen.
The external counter-intelligence activities of the Tsarist police were
restricted, as Piper notes, virtually to a single institution, the Russian
Embassy in Paris, which employed a handful of people to keep surveillance
on politically active Russian emigres. 22 A comparison of the Tsarist
external counterintelligence commitment with that of the Soviet KGB is
impossible, but also unnecessary, given the vast magnitudes of difference
168 Critiques
in scale that everything we know suggests. 1be Western conventional
wisdom, so deep-rooted in academic folklore as to be probably unshakable,
that Bolshevik terror was but a continuation of a Tsarist police-state, is
controverted by all available evidence. On the contrary, all the evidence
supports the view of late-Tsarist Russia as a civil society on the European
model, which it was the historic role of the Bolsheviks (animated, ironically,
by a European ideology) to destroy. Before it was destroyed by the Bol-
sheviks, the late-Tsarist regime was one that compares favourably with the
great majority of governments that exist today. It is highly probable, in fact,
that were the Soviet Union now what it was in 1913, it would come within
the twenty states that are most liberal and least oppressive in today's world.

SOVIET DEPENDENCY ON mE WEST

In truth the conventional image of late Tsarism better fits the early decades
of the Soviet Union in certain important respects. Throughout its history,
the Soviet Union has been crucially dependent on Western capital and
technology. Indeed, Western aid may well have secured the survival of the
regime during its periods of worst crisis. As Besancon has said
international society has on several occasions saved the Soviet regime in
moments of crisis. I need only mention Herbert Hoover's American
Relief Administration mission, which saved between five and six
million peasants starving to death during the famine of 1921, and
American aid in World War 2. Even during the most desperate War
Communism - the War Communism of the first Five Year Plans - the
West sent the Soviet Union considerable sums of money, technology
and engineers.23
Western aid was, however, no less important in the construction of Soviet
power than in preserving it from collapse. The Treaty of Rapallo, signed
between Germany and Soviet Russia in April 1922, was decisively impor-
tant in promoting both the economic and the military power of the Soviet
state. It provided for the establishment of joint Soviet-German industrial
and commercial firms, with more than two thousand German technicians
arriving in the Soviet Union in the wake of the treaty being signed. The
Treaty of Rapallo also initiated an important period of Soviet-German
military cooperation, enabling the Germans to circumvent the Treaty of
Versailles and rebuild the German army with state-of-the-art weapons
manufactured and tested in the Soviet Union in return for the provision of
training facilities for Red Army officers in Germany. The intimacy of this
connection may be gauged from the fact that German and Russian chemists
collaborated in the production of experimental poison gases.
 Totalitarianism, reform and civil society 169
This period of Soviet-Gennan cooperation came to an end in 1927 with
the Shakhty trial 24 of Gennan engineers, after which the Soviet Union
turned to the United States for assistance. Heller and Nekrich report that,
whereas in mid-1929 the Soviet Union had technical agreements with
twenty-seven German firms and fifteen American fmos, by the end of 1929
forty American fmos were cooperating with the Soviet Union. They go on:
The five-year plan could not have been implemented without foreign
assistance. In 1928 a group of Soviet engineers arrived in Detroit and
requested that Albert Kahn and Company, an eminent firm of industrial
architects in the United States, design plans for industrial buildings
worth 2 billion dollars .... According to an agreement with the Supreme
Economic Council of the USSR, the American firm agreed to design all
aspects of Soviet industry, heavy and light Foreign designers, techni-
cians, engineers, and skilled workers built the industrial units of the first
five-year plan. Primarily they were Americans, who pushed the
Gennans out of first place in 1928; after them came the Germans,
British, Italians and French. The dam on the Dnieper was built by the
firm of Colonel Hugh Cooper, a prominent American hydraulic
engineer; the majority of the largest Soviet power plants were equipped
by the British firm Metropolitan-Vickers; Western companies designed,
built, and equipped Magnitogorsk and Kumetsk, the Urals Machinery
Works, the Kaganovich Ball Bearing Plant in Yaroslav, among others
The full extent of Western economic and technological aid to the
Soviet Union will not be known until the Soviet archives are opened up.
The Western firms that collaborated with Moscow have concealed the
information almost as carefully as their Soviet partners. Nevertheless,
the American historian Anthony Sutton has come to the conclusion, on
the basis of German and English archives, that 95 percent of Soviet
industrial enterprises received Western aid in the form of machines,
technology, and direct technical aid.2S
Such direct Western aid to the Soviet power is a recurrent, almost a
continuous feature in its development. In recent times, it has included
economic and military assistance to Soviet satellite powers, as in the case
of British commercial and governmental aid to Marxist Mozambique. Side
by side with direct aid, there has been the explosive growth in recent
decades of Western bank lending to the Soviet bloc, with Western credits
amounting to 50 billion dollars by 1978. By the 1980s, the Soviet Union
was financing Western development of its resources with Western credit.
West European capital financed the construction of the gas pipeline from
the USSR to Western Europe, assuring an annual profit of 5-8 billion to the
Soviet Union, and in 1981 the Deutsche Bank and the Mannesman A. G.
170 Critiques
steel corporation entered a contract worth 16.5 billion with the Soviet
government to develop its Siberian energy resources.26 By the end of 1988,
Soviet foreign debt was estimated at about 43 billion dollars, much of it
loaned to the Soviet Union by Western banks at lower interest rates than
those offered to blue-chip multinational corporations such as IBM or
Shell.27 It is difficult to resist the conclusion that the Soviet Union, in its
demonstrated dependency on Western capital, technology and (let us not
forget) food imports, resembles the conventional caricature of Tsarist
Russia as an underdeveloped, backward power far more closely than did the
reality of late Tsarism.

THE CENTRALITY OF MARXIST-LENINIST IDEOLOGY IN

THE DEVELOPMENT OF THE SOVIET UNION
The dominant methodology of conventional Western Soviet studies
demonstrably exaggerates the continuities in political culture between late
Tsarism and the Soviet state and has passed over or suppressed the many
evidences of the dependency of the Soviet Union throughout its history on
Western aid. In its attempts to apply the categories and techniques of
Western political science to Soviet political life, the conventional discipline
of Sovietology has systematically neglected, or tried to interpret away, the
decisive role of Marxist-Leninist ideology in explaining the strategies and
tactics of Soviet leaders, and has consigned to an Orwellian memory hole
the formative impact of ideology in constituting and reproducing the most
distinctive institutions of the Soviet state.
Perhaps the most fundamental of the blind spots of the conventional
methodology concern the experiment in War Communism, the origins of
the Gulag and the functions of the Gulag in the period of socialist con-
struction, in each of which ideology played a decisive part. War Com-
munism was the project Lenin initiated in the Spring of 1918 of realizing
the utopian fantasies of his State and Revolution, written just before the
October Revolution. The conventional view, whose dominance was
secured by the work of Maurice Dobb and E. H. Carr,28 is that War
Communism was forced on the Bolsheviks by the exigencies of civil war.
More recent studies by Paul Craig Roberts, Thomas Remington and Silvana
Malle 29 have demonstrated that War Communism was not an emergency
measure, but a policy having the conscious goal of abolishing market
institutions. This policy, like many other aspects of Leninist totalitarianism
(including the institution of forced labour), has a direct ancestry in the
thought of Marx himself, for whom socialism and Communism were, first
and last, the abolition of commodity-production and market exchange. In
accordance with this Marxist project, a great leap forward was attempted,
 Totalitarianism. reform and civil society 171
in which money was suppressed, private trade was banned, compulsory
labour service introduced and grain requisitioned from the peasants at fixed
prices. The results of this experiment were catastrophic. Famine broke out
in the cities, industrial production collapsed and in 1921-2 famine followed
the collapse of food production in the countryside. But the most significant
collapse during the experiment in War Communism between 1918-22 was
that in population itself. According to Soviet sources, in excess of five
million lives w<!re lost as a result of the famine of 1921-2. In addition, at
least ten million lives were lost in the course of the Civil War. Whereas it
may be questionable to attribute the resultant total entirely to the experi-
ment in War Communism, the human cost in terms of lives of this period of
social revolution and revolutionary war is fully comparable to that of the
later period of Stalinism - and perhaps serves as a better illustration of the
practical consequences of Communism in the full rigour of its doctrinal
purity.
War Communism exemplifies a decisive characteristic of the Soviet
state from its inception in the Bolshevik dictatorship - its character as a
weltanschauung-state. Tsarist Russia was a monarchy in which the head of
state was also head of the Church - as in the case of the United Kingdom to
this day. It was never a state dedicated to the revolutionary transformation
of its subjects's beliefs and the inculcation of a new orthodoxy. Nor were
its policies, foreign or domestic, governed, primarily or consistently, by an
ideology. The Bolsheviks, by contrast, were guided by ideology, and that
primarily, from the start. As Dziak has observed in regard to the terror of
these years of War Communism: 'The Cheka operated under an all-
embracing plan, simple though it was: the bourgeoisie were to be exter-
minated. That this mass extermination was premeditated, and not merely,
as Soviets claim, a response to White reaction and foreign intervention, is
seen by its continuance well after the defeat of the Whites and the with-
drawal offoreign forces, that is, well into the 1920s: 30
The War Communism period exemplifies the totalitarian origins of
Soviet Communism in all three respects - in the institution of a weltan-
schauung state, in the destruction of the rule of law by an extra-judicial
secret police, and in the attempt to eradicate the institution of market
exchange. These are features of the Soviet state that it has exhibited
throughout its history, and which persist to the present day.

THE LENINIST ORIGINS OF SOVIET TOTALITARIANISM

Another distinctive feature of the Soviet state, ignored by the dominant
methodology, is the origin of the terroristic secret police, and of the concen-
tration-camp system, in Leninist theory and practice. This was a
172 Critiques
development fraught with significance for the future, since it heralded the
emergence of one of the constitutive features of Soviet totalitarianism - the
omnipresence of extra-legal coercion and the creation at the political level
of a lawless state of nature. Credit for first use of the tenn concentration
camp goes to Trotsky, but the idea of establishing forced labour camps as
part of a project of socialist re-education originates with Marx and the
institution was established by Lenin. Shortly before the October 1917 coup,
Lenin declared that he would adopt the institution of compulsory labour,
which would be more potent that the guillotine, which merely terrorized
and broke active resistance, while forced labour would break passive resist-
ance, particularly among proletarians. 31
Mass terror was characteristic of the Soviet state from its inception, with
Dzerzhinsky giving unlimited 'extra-judicial' powers - including powers
of execution - to all regional Chekas in September 1918. In January of
1918, Lenin had agreed that the Cheka might appropriately be renamed
'The Commissariat for Soviet Extennination', but judged it impolitic to do
SO.32 In 1921 the Cheka assumed responsibility for the problem of the
millions of homeless children in Russia, taking direct control of the camps
and homes where many of them were sent. In 1922 the GPU, the Cheka's
successor, acquired further extra-judicial prerogatives, which were once
more expanded in 1926. The GPU inherited from the Cheka, also, the
notorious Solovetsky concentration camps, whose existence and character
were revealed to an indifferent public opinion in the West by survivors in
the 1920s. The institutional framework of Soviet totalitarianism was con-
structed by Lenin - a fact which has been studiously ignored under
glasnost. The suppression of the patrimony of Stalinist totalitarianism in
that of Lenin under the Soviet glasnost to the present time is, in all
probability, inevitable, since acknowledging this unbroken historical con-
tinuity would be a death-blow to the legitimacy of the Soviet system. One
may state the same point in other words, by affinning that the acid test of
Soviet glasnost will come when it proceeds to demystify Lenin and
Leninism and recognizes in these the origins of the current pathology of the
Soviet state.

THE FUNCTIONS OF FORCED LABOUR CAMPS IN THE

SOVIET ECONOMY
A further blind spot of the dominant methodology is the economic role of
the Gulag, which is evaded or distorted in conventional Soviet studies. It is
represented as peripheral in importance, or else as an aberration of socialist
planning, rather than as an indispensable instrument of Soviet power during
the period of the Five-Year Plans. A classic example of this interpretation
 Totalitarianism. reform and civil society 173
is found in Nove's Economic History o/the USSR. In this book of over three
hundred pages there are only a few scattered remarks on the subject of the
camps. We are told, for example, that
in this period (the period of the Great Leap Forward) prisoners and
deportees, especially the latter, emerged as a significant factor in the life
of the country. For example, only a small portion of the inhabitants of
the new town of Karaganda went there of their own volition. 33
Later, Nove refers to the human cost of the period of socialist construction
'in the diversion to camps of unknown millions, of whom a high proportion
were above average in intelligence, energy and technical knowledge'. And
he admits that 'Especially after 1936, officials of the NKYD ... exercised
important supervisory functions through the economy, and they also ran a
big economic empire using forced labor, until the break-up of this empire
after Stalin's death'.34 Nove comments on the human cost of collectiv-
ization - in terms of deaths caused by famine and deportation - that
The Soviet population in 1926 was 142 millions, and for 1932 it was
officially estimated at 165.7, since it had been increasing at the rate of
about 3 millions a year. In 1939, seven years later, it was only about 170
million. Somewhere along the way well over 10 million people had
demographically disappeared. [Some, of course, were never bom.]3'
Nove concludes, judiciously:
Perhaps it is Russia's tragedy ... and a measure of her achievement that,
despite all that happened, so much has been built, and not a few cultural
values preserved and handed on to a vastly more literate population. 36
An incomparably more accurate account of the economic significance of
the Gulag, and of the human costs of collectivization, is given by Nekrich
and Heller. As they put it: 37
As early as 1929 all the camps had been placed under the direction of the
OGPU, which for years had directed the archetypal camp at Solovki.
The OGPU became the country's largest construction company. With a
virtually limitless supply of unskilled labor at its disposal, the OGPU
conducted massive arrests of engineers and technicians to manage the
unskilled laborers. Anew, purely Soviet institution arose, the sharashka:
a prison in which engineers, scholars, and researchers worked in their
fields of specialisation for the interests of the state. At the large-scale
building sites, in the super-factories, the specialists were monitored by
armed guards. The largest construction site of the First Five-Year Plan,
the Baltic-White Sea Canal, was built by prisoners under the leadership
174 Critiques
of 'engineer-wreckers'. Trotsky's dream of 'militarised labor' became a
reality under Stalin in the fonn of the 'penalisation of labor'. The gates
of the camps were adorned with Stalin's words; 'In the Soviet Union
labor is a matter of honor, power and heroism'.
The reliance on slave labour has continued to be a stable feature of the
Soviet system, as is demonstrated by the use of North Vietnamese con-
scripted labour in the construction of the Siberian gas pipeline over the past
decade or so.
On the genocidal character of agricultural collectivism. Nekrich and
Heller write:
The full story of this first socialist genocide has yet to be written.
Chronologically, the first genocide of the twentieth century was that of
the Annenians by the Turks. The massacre of Don Cossacks by the
Bolsheviks during the civil war likewise approached genocidal pro-
portions. The Turks destroyed a population of a different faith and
nationality; the Cossacks suffered during a fratricidal civil war.
The genocide against the peasants in the Soviet Union was unique not
only for its monstrous scale; it was directed against an indigenous
population by a government of the same nationality, and in time of
peace.
In 1945, after the defeat of Nazi Gennany and the public disclosure of
all its crimes, jurists, sociologists, psychologists, historians, and journal-
ists began the inevitable controversy over whether the Gennan people
had known about the Nazi crimes or not. There is no question that the
Soviet city people knew about the massacre in the countryside. In fact,
no one tried to conceal it. Stalin spoke openly about the 'liquidation of
the kulaks as a class', and all his lieutenants echoed him. At the railroad
stations, city dwellers could see the thousands of women and children
who had fled from the villages and were dying of hunger. Kulaks,
'dekulakised persons', and 'kulak henchmen' died alike. They were not
considered human. Society spat them out, just as the 'disenfranchised
persons' and 'has-beens' were after October 1917,just as the Jews were
in Nazi Gennany.38
Again, it must be observed that the political creation of an artificial terror-
famine with genocidal results is not a phenomenon restricted to the his-
torical context of Russia and the Ukraine in the 1930s, but is a feature of
Communist policy to this day, as evidenced in the 1960s in Tibet, and now
in Ethiopia. The socialist genocide of small, 'primitive' peoples, such as the
Kalmucks39 and many others, has been a recurrent element in policies at
several stages in the development of Soviet and Chinese totalitarianism.
 Totalitarianism, refonn and civil society 175
Once again, Communist policy in this respect faithfully reproduces
classical Marxism, which had an explicit and pronounced contempt for
small, backward and reactionary peoples - no less than for the peasantry as
a class and a form of social life. And the reliance on forced labour camps,
like agricultural collectivization, was an inexorable consequence of the
Marxist-Leninist project of eradicating one of the constitutive institutions
of civil society, market exchange, together with its legal matrix in the
institution of private property.

SOVIET TOTALITARIANISM AND STRATEGIC

DISINFORMATION
The conventional methodology neglects or suppresses the economic role of
the Gulag in the period of socialist construction and it is silent as to the
genocidal character of the policy of collectivization. This is so, in part at
least, because of the high degree of control over information achieved
under Soviet totalitarianism, and also because Western observers have
often been reluctant to publicize information in their possession about the
worst failings of the Soviet system. Thus the famine in the USSR in the
early 1930s was little reported in the Western media, and Eugene Lyons
writes in Chapter 15 of his autobiography, Assignment in Utopia,40 'The
Press Corp Conceals a Famine' , of his own participation in the suppression
of what was then known of the millions of deaths the famine was causing.
The collaboration of the Western media with the Soviet authorities in
concealing their worst atrocities is further exhibited in the silence which
reigned in Western media for about a quarter of a century as to the forcible
repatriation and subsequent fate of over two million people claimed by the
Soviet Union after the Second World War.41 It is likely that Soviet res-
ponsibility for the Polish Katyn massacre will be admitted, under glasnost,
by the Soviets themselves, before it is admitted by the British Fore~gn
Office, which continues to adhere to the fiction of Nazi responsibility for
that atrocity. The history of Viasov's Russian Liberation Army (ROA), its
role in the liberation of Prague and its statement of enmity to Nazism and
Stalinism in the Prague Manifesto, have only recently been given an accur-
ate historical statement.42 Western authorities have colluded with Soviet
authorities in concealing nuclear disaster and the existence of the nuclear
Gulag.43 It is only as a result of the Soviet glasnost that the almost apo-
calyptic ecological catastrophes in such areas as that surrounding Lake Aral
have been publicized in the West. One of the many ironies of the present Soviet
policy of glasnost is that, in focusing on aspects of Soviet history which in the
West have been discussed only in rare, obscure and neglected emigre journals,
it has served to reveal the scope aild limits of Western glasnost.
176 Critiques
The dominant Western methodology goes most seriously astray in its
neglect of the Soviet use of strategic disinfonnation. This blind spot in
Western perception and theorizing is particularly disquieting given the
similarities evident between the current glasnost strategy and the highly
successful disinfonnation exercise conducted during the period of the
New-Economic Policy (NEP). Under the NEP, market relationships were
partially restored in the economy, with peasants in particular being freed to
charge market prices for their products, and so to enrich themselves. (When
the NEP was ended by Stalin, the peasants found their new wealth liable to
confiscation - a fact with direct relevance to present Soviet experiments
with cooperatives.)
This disinfonnation exercise was inaugurated along with the NEP in
1921 by the fonnation by the OGPU inside the USSR of a false anti-
Bolshevik organization, the Monarchist Alliance of Central Russia, other-
wise known as the Trust, and operating under the cover title of the Moscow
Municipal Credit Association, which replaced an earlier and genuine group
destroyed by the OGPU in 1920. Unimportant in themselves, the historical
details of the Trust are illuminating in revealing the mechanism of what
Dziak has called 'the prototypical deception and protective operation in the
Soviet repertoire'.44 The Trust operation had been proceeded by an earlier
deceptive operation, usually known as Sindikat I, in which the OGPU
attempted to penetrate the organization of Boris Savinkov, fonner War
Minister in the Kerensky government and friend of Polish leader Pilsudski
and of Winston Churchill. A second operation, Sindikat 2, ran concurrently
with the Trust, and successfully lured Savinkov back into the Soviet Union.
Perhaps the first of the Soviet deception operations was that against the
British agent, R.H. Bruce Lockhart, in the Summer of 1918. It is possible
that, at the same time, the British agent Sydney Reilly was 'turned'. In any
case, Reilly claimed later to be convinced of the authenticity of the Trust,
and, like Savinkov, was persuaded to return to the Soviet Union, where he
probably perished. 45
The primary objective of the Trust was to weaken, divide and neutralize
the powerful anti-Soviet emigre movement that had arisen in Europe in the
wake of the great Russian diaspora after the Civil War. Numbering over a
million, of whom a quarter were White officers or men, the anti-Bolshevik
emigres constituted a threat the Soviet government could not afford to
ignore. A secondary goal of the Trust was to persuade the governments and
intelligence services of the Western powers that the revolutionary socialist
government of the Bolsheviks was on the brink of a metamorphosis, of
which the NEP was only an intimation, into a traditional Russian regime
which could safely and easily be integrated into the international
community.
 Totalitarianism, re/onn and civil society 177
The fonnation of the Trust was accompanied by the capture by the
OGPU of two important 6migre movements, the 'Changing Landmarks'
movement and the 'Eurasian' movement Apparently originating as early as
1918 among the old intelligentsia in Russia, the movement found expres-
sion in several voices in 1920. In the Summer of that year a fonner Cadet
leader and lawyer, Professor Gredeskul, undertook with the approval of the
authorities a speaking tour of the Soviet Union. Slavic Dawn, a Prague
6migre newspaper, in a statement characteristic of the movement, affmned
that the Bolsheviks were now defenders of the Russian national interest. In
the autumn of 1920 there appeared in Harbin, Manchuria, a seminal state-
ment of 'changing landmarks' ideology, a collection of articles entitled The
Struggle/or Russia by Nikolai V. Ustryalov, a former Kadet and supporter
of Admiral Kolchak. Written by six members of the 6migre community, of
which Ustrayalov was most prominent, the collection appeared subse-
quently in Prague, under the title Smena Vekh, in 1921. The theme of the
collection, which was taken up by sympathizers within the conservative
intelligentsia such as Shulgin, Efimovsky, and KJyuchnikov, was that the
Bolshevik government was in process of mutating into a nationalist
dictatorship on lines theorized in the writings of reactionary thinkers such
as Konstantin Leontiev and Joseph de Maistre.
The movement received substantial unofficial support from the Soviet
government, which facilitated the publication in Prague and Paris of a
weekly magazine, The Change 0/ Signposts, and in Berlin a journal called
On the Eve (Nakanune). In 1922 the Soviet government allowed the move-
ment to publish journals New Russia and Russia in Leningrad and Moscow.
Smena Vekh journals also appeared in Riga, Helsinki, Sofia and Harbin.
Nakanune survived with Soviet subsidies until it was closed in June 1924.
A year and a half later, the 'changing landmarks' movement was
suppressed in the USSR. As Dziak comments:
Nakanune faithfully reflected the Soviet party line and was of immense
value to Moscow as an 6migr6 instrument of conversion in the Soviet
cause ... like the Trust, it (the Changing Landmarks movement), when
it had served its purpose, was eliminated. 46
The Eurasian movement followed a similar pattern of evolution. In 1921 an
anthology was published in Sofia entitled Iskhod k Vostoku (Exodus to the
East) advocating Russian nationalism, the idea that the political culture of
Russia was as Asiatic as it was European and speculating that the Bolshevik
regime might be a proto-version of an appropriate authoritarian system for
Russia. Exodus to the East was followed by a further seven volumes, the
last appearing in 1931. A Eurasian Chronicle appeared in twelve volumes
between 1925 and 1937. The Eurasian movement flourished until 1928,
178 Critiques
when it split and began to decline, but its importance continued well into
the 1930s, when it was influential in gaining the collaboration of sym-
pathetic 6migr6s with the Soviet government.
Between 1921 and 1927 the Trust organization achieved major
successes in its campaign of disinformation. Leading 6migre leaders such
as Boris Savinkov and Generals Wrangel and Kutepov were contacted by
Trust agents and convinced of its authenticity. The 6migre leaders in tum
were decisive in convincing Western intelligence services of the genuine-
ness of the Trust. By the mid-1920s, it seems that no less than eleven
Western intelligence services were heavily dependent on the Trust for
information about developments in the Soviet Union. 47 For Western
governments, the policy implication of the Trust was clear - the dissident
forces within the Soviet state were to be aided by diplomatic recognition,
the expansion of trade and a curb on destabilizing anti-Soviet activism.
Everything should be done that might ensure the success of the NEP. As
Dziak has summarized this strategy:
the disinfonnation fostered through the Trust reinforced the initiatives of
the NEP, which was also overseen by Dzerzhinsky in his dual capacity
as chief of state security and chief of the Supreme Council of National
Economy. From this perspective, the NEP itself served a deception
purpose in that it helped to refinance Soviet industry at Western political
and economic expense. 48
From the standpoint of the Soviet leadership, the NEP and its associated
disinfonnation exercises were indeed successes. Some ten per cent or more
of the Russian 6migre community returned to the Soviet Union, including
several of its leaders such as Savinkov and Kutepov. The 6migr6 movement
was never again a significant threat to the Soviet regime. Western powers
assisted in the reconstruction of the Soviet economy, above all Gennany,
which also built up the Soviet war machine. The NEP period was also a
period of political consolidation for the Soviet regime. Vestiges of the old
political parties in the Soviet Union were eliminated. Religious activity was
brought finnly under political control in the pro-regime 'living churches'.
Powerful nationalist movements in Georgia, the Ukraine, Annenia and the
Asian republics were suppressed and these nations subjected to full annexa-
tion. Mongolia became the first Soviet satellite state. Twelve new com-
munist parties joined the Comintern. By the time Stalin closed down the
NEP and wound up the Trust, it had substantially achieved its goals 'to
prevent internal revolt, expand foreign trade, attract foreign capital and
expertise, gain diplomatic recognition from non-Communist countries, pre-
vent major conflict with the Western powers, help to exploit the contra-
dictions in and between the capitalist countries, neutralise the 6migre
 Totalitarianism, reform and civil society 179
movement, and help to promote world revolution through the Communist
movement' .49
There are ominous parallels between the disinfortnation exercises of the
NEP period and current policies of glasnost. The objective of financing the
reconstruction of the ruined Soviet economy with Western capital and
technology is the same. Now, as then, Gennany is likely to playa crucial
role in supplying capital. An activist diplomacy is gaining major con-
cessions from the Western powers, as a result of which the USSR can
proceed with its programme of modernizing its military forces while
appearing to reduce its offensive capability. Side by side with a diplomacy
which divides NATO, diminishing almost to vanishing point the Western
perception of Soviet enmity and so promoting the psychological dis-
armament of the West, the USSR has increased the activities of its intelli-
gence services, particularly in Germany and Britain, with the objectives of
expediting technology transfer and manipUlating opinion. (The reorgan-
ization of 1987 of key departments within the KGB and GRU is likely to
prove one of the more enduring achievements of perestroika.) Such are the
clear parallels between current developments and the NEP period. They
suggest that perestroika is best interpreted as a reversion to a Leninist
strategy of disinformation and activist diplomacy (the latter encompassing
exercises in tactical retreat, as in Afghanistan) and the abandonment of the
clumsy, costly neo-Stalinist strategies of the Brezhnev period. They
suggest that we are now witnessing the 'sixth glasnost', the boldest yet,
designed to destroy the Western perception of Soviet enmity and permit the
Western refinancing of the Soviet system, As Epstein has put it:
By 1989, at virtually no cost to Soviet power, the sixth glasnost had
provided the Soviet leadership with not only the tens of billions of
dollars in credits it required to further expand its industrial (and military)
capacity, but ... was perceived as less of a threat to Western Europe
than the United [Link]
Granted these parallels, how are present developments to be understood as
responses to the contemporary situation in the Soviet Union? I have speci-
fied as the chief defects of the conventional methodology of Soviet studies
its reliance on standard techniques of analysis in political science and, in
particular, its deployment of inchoate notions of political culture. This
methodology neglects the structural affinities displayed by all Communist
regimes. In the present sections, I have argued that the conventional
methodology neglects the potential for strategic disinformation possessed
and exercised by the Soviet regime (not only in the Trust episode, but in
later episodes such as the WIN operation in 1947-51.51 ) The alternative line
of analysis and theorizing developed here is that of the Soviet Union as a
180 Critiques
system constituted, from its inception, by Leninist ideology and a terroristic
security service. We may even go so far as to concur with Dziak, and refer
to the Soviet Union as 'a counter-intelligence state'.52 On this view, the
Soviet Union is a totalitarian order whose character as such is guaranteed
by the KGB, whose role in representing the Soviet Union to the external
world oUght never to be underestimated. Current policies of perestroika and
glasnost, on this alternative analysis, are authentically Leninist exercises in
strategic deception.
The parallels between the present period and that of NEP, suggested by
the alternative methodology I have advanced, are real enough. Never-
theless, they do not of themselves account for the crucial differences in the
historical contexts of the two episodes. By comparison with the NEP
period, the USSR now confronts far graver problems of ethnic conflict,
nationalist and separatist tendencies, religious and fundamentalist move-
ments and environmental degradation. For reasons I shall try to specify in
a later section of this paper, it is far from clear that the economic renais-
sance of the NEP period, brief and partial as it was, can be repeated. Nor is
it obvious, or even plausible, that infusions of Western credit can facilitate
the resurrection of the USSR's senile industries. The financial burdens of
the Soviet state, including an estimated 14 billion dollar per annum com-
mitment to non-military support of the East European Communist regimes
and analogous commitments to Cuba, are incomparably greater than those
of the fledgling Soviet state. Even if it was conceived as a grand deception,
the policy of glasnost is extraordinarily risky for the Soviet leadership.
Indeed, however it was conceived - and a Leninist interpretation of its
inception seems by far the most plausible - the current Soviet policy
confronts difficulties and hazards far greater than any that the Soviet
regime has ever faced in peace time. So as to theorize these phenomena, we
need to abandon the blinkered perspective of Kremlinology, which has
dominated Soviet studies for lack of any disciplined theoretical alternative.
We need to ask: what are the real-world constraints that govern and con-
strain the Leninist strategy and tactics of the Soviet rulers? In order to
answer this question, we must attempt, more directly and systematically, to
theorize the constitutive features of Soviet totalitarianism.

SOVIET TOTALITARIANISM: CALCULATIONAL CHAOS IN A

POLmCAL STATE OF NATURE
Central in the Soviet totalitarian order are the failure of central planning and
the political consequences of that failure. The War Communism episode
shows that the Bolsheviks took seriously the Marxian commitment to the
abolition of the institutions of market exchange. As Roberts and
 Totalitarianism, refonn and civil society 181
Stephenson have said: 'Public ownership of property is not the defining
characteristic of Marxian socialism; central planning is. In Marxian social-
ism, there is no exchange; therefore, there is no private property (rights to
exchange)' .53 The suppression of private property follows from the aboli-
tion of commodity production, not vice versa. The project of replacing the
institutions of market pricing with institutions of central planning confronts
massive and insuperable difficulties, well theorized in the work of the
Austrian economists Mises and Hayek.54 The argument of Mises is that
economic calculation presupposes market pricing: the information required
for rational resource-allocation - information about relative scarcities and
consumer preferences, for example - is so complex that no central planning
authority could possibly collect it. The number of transactions in a modern
economy, again, is so enormous that the planner will face an insuperable
problem of calculation if he attempts to simulate. market processes in a
mathematical model. Hayek's argument focuses on the epistemological
impossibilities of central planning rather than its practical difficulties: it is
a problem of knowledge, not merely a problem of calculation, which is fatal
to the socialist project of supplanting market institutions. The knowledge
which the planner requires is dispersed knowledge, scattered about society,
and it is often local knowledge of circumstances that are in their nature
ephemeral. Most importantly, the knowledge possessed by economic
agents is often practical knowledge, embodied in skills or dispositions, and
only slightly theorized by the agent. Indeed, the tacit knowledge on which
we all rely in our economic dealing may be only partly theorizable: it may
contain elements, such as entrepreneurial perceptions or traditional prac-
tices, which cannot be given any full theoretical statement. Because much
of it has this tacit or practical character, economic knowledge cannot be
retrieved and collected by the planner. This last argument was developed
independently by Michael Polanyi,5S who saw that it imposed an insuper-
able limit on the planning both of science and the economy. In economic
life, as in science, the planner is defeated by the fact that we always are
making use of knowledge of parts of which we are bound to remain
ignorant, and we always know more than we can ever say.
In this Hayekian and Polanyian account, the role of market institutions
is that of discovery procedures which allow for the disclosure and utiliza-
tion of dispersed knowledge. This is a very different model from that of
classical economic theory, in which the market is a device for relating
scarce resources with competing ends. In the Hayekian-Polanyian model,
no one has know ledge of the available resources: if the market exists to
economize on anything, it is on knowledge, the most irremediably scarce
resource of all.
182 Critiques
The implications for economic policy and institutions of the Austrian
calculation argument are radical. There is, first, the implication that in the
absence of market pricing of most factors of production there will be
calculational chaos. This is a result which puts paid, not only to the
discredited Lange-Lerner model56 of competitive socialism through the
medium of market simulating devices, but also to the anachronistically
fashionable idea of a market socialism of competing worker-cooperatives. 57
It demolishes the latter since, if the calculation argument applies at all, it
has its clearest application to capital: it suggests that only market institu-
tions can deploy the dispersed knowledge required if capital is to be subject
to an efficient allocation. This carries with it a second result for policy -
that most factors of production, including most capital, be held in decentral-
ized private ownership, or in other words, that there is no viable middle way
between the institutions of market capitalism and those of central planning.
The radical implication for current policy is that no programme of
perestroika, or economic reform, can hope to succeed which does not
encompass dismantling the socialist system of ownership and dispersing its
productive assets as private property. Such a policy of privatization is
feasible, however, only on condition that the dispersed assets may be
securely held and exchanged. This is in tum possible only under a rule of
law in which private property is protected from arbitrary seizure by govern-
ment. (It is the absence of the rule of law, and all that entails, that accounts
for the lack of economic progress in non-totalitarian states such as the
dictatorships of Latin America and Africa.) It is obvious that the institution
of a rule of law would amount to the suppression of one of the key structural
features of Soviet-style totalitarianism, since it would impose important
constraints on the hitherto unlimited discretionary power of the state. We
must ask ourselves, why might the ruling elites of a totalitarian state accept
such a constraint on government? Answering this question requires
examining the political environment of the Soviet economic chaos, and the
constraints imposed on feasible reform that it imposes.

mE INCENTIVE STRUCTURES OF SOVIET CENTRAL

PLANNING: A PUBLIC CHOICE APPROACH
The argument so far has been entirely epistemic. It has been the argument
that, whatever its goals or the motives of the ruling elites that control it,
central planning cannot achieve a rational allocation of resources. This is
not to say that it cannot succeed in achieving specific and limited goals: the
examples of the Soviet space program, and even more of the build-up of the
Soviet war machine, show that it can so succeed. Plausibly, however, it has
done so only with the aid of Western technology, and only at the cost of
 Totalitarianism. refonn and civil society 183
displacing scarce resources from civilian uses and thereby further im-
poverishing Soviet society, including elements of its privileged nomen-
klatura. The epistemic argument - the argument that central economic
planning is an epistemological impossibility - is important precisely
because it is insensitive to the incentive structures that govern the planners.
Its result is that the planners will fail in most of their objectives, whatever
they are - whether or not they recognize consumer preferences, for
example, aim primarily to reward the nomenklatura, or simply to modern-
ize the war machine. As James Sherr has put it in his authoritative study,
Soviet Power: The Continuing Challenge,
Whatever its ostensible purpose, the elimination of the market and the
creation of a command economy has one clear consequence: the who,
what, and how of economic relations is determined by planners, not by
those who produce and consume. It is the structure of the system that
demands this, not the unimaginativeness or selfishness of those
involved. An enterprise director cannot do what he considers best for
society since, without the market's signals, he cannot know what this is.
He may know the difference between a tractor that works and one which
does not, but he cannot know how many tractors are required, what sort
of tractors to build, and where they are most needed. 58
One large part of the rationale of perestroika is indeed to achieve the three
objectives specified above more effectively than is possible under central
planning. The result of the calculation argument, however, is that economic
reform is bound to fail, and these objectives will be unachievable, without
the reconstruction of a regime of private property under the rule of law.
Failing that, the objectives of perestroika can be achieved, if at all, only by
a massive Western refinancing of the Soviet system.
The predictive content of the calculation argument is that a centrally
planned economy cannot exist. What exists under central planning institu-
tions is a complex structure of parallel markets, relying on historic, capital-
ist or black market prices, subject to recurrent episodes of authoritarian
intervention. This theoretical result, worked out in Polanyian terms in Paul
Craig Robert's Alienation and the Soviet Economy59 is amply corroborated
by such excellent empirical studies as Peter Rutland's The Myth of the Plan.
Rutland concludes:
it turns out that the structure of the command economy is not what one
might naively expect to find - it is messy, overlapping and subject to
endless and obscure organisational mutation.60
At the present time, what exists in the Soviet Union is exactly that which is
predicted in the calculation argument - a chaos which reproduces itself by
184 Critiques
covert reliance on parallel markets and by continuous mutations in the
planning structures themselves.
We must consider the environment of incentives within which the
planners operate, independently of the problems of knowledge which they
confront, and the larger structure of incentives which surrounds those who
inhabit the Soviet state. For the planners, it is clear, all the incentives of the
soviet system point to conservatism and risk-avoidance rather than initia-
tive or constructive investment. As Sherr has said,
Virtually extinct in the Soviet Union after sixty years of socialism is the
linchpin of capitalism, the entrepreneur. (To the extent that he survives,
he is apt to make the illegal second economy his habitat.)61
Planners are bound to be risk-averse in their investment policies, and, since
no error-elimination mechanism for inefficient investments exists under
central planning institutions, they have an incentive to conceal mal-
investments by diverting further resources into them. Even aside from these
insuperable epistemic problems, the command economy contains an incen-
tive structure thr.t is biassed against efficiency. The very existence of the
massive planning bureaucracies creates a powerful interest against the
liberalization of economic life. As Sherr again puts it,
Replacement of plan by market is in itself a surrender of power, since it
means the elimination of planners. At present, Soviet agriCUlture, and
therefore the day-ta-day lives of Soviet peasants, are regulated by
hundreds of thousands of officials and thirty-three ministries. If pro-
duction and exchange were henceforth to be determined by those who
produce and consume, what functions would their former controllers
then perform? What powers would the state still posses over those thus
emancipated?
The economic interests of the planning bureaucracies only reinforce one of the
defining features of Soviet totalitarianism, which is the obliteration of the
economy as an element of an autonomous civil society. As Sherr concludes:
It is not the KGB, 'indomitable' as it is, which makes the Soviet Union
a totalitarian society, nor even the CPSU's monopoly of political power,
but the fusion of political and economic power. Monopoly of power -
Stalin's first justification of the planning system - will probably be its
last defence.62
The same incentive structure that guarantees the economic inefficiency of
the command economy also confers on it a powerful political stability.
The analysis we have developed so far has followed the Public Choice
School in attributing to planners the same motives we attribute to actors in
 Totalitarianism, reform and civil society 185
the market-place - it theorizes the bureaucrat as Homo economicus.
Another tool of the Public Choice School, game theory, may help to
illuminate the structure of Soviet totalitarianism as a generalized Prisoner's
Dilemma and so to explain its phenomenal stability.

SOVIET TOTALITARIANISM AS A POLmCAL STATE OF

NATURE AND THE NEW HOBBESIAN DILEMMA OF ITS
SUBJECTS
Soviet totalitarianism, like that in Communist China, differs from any kind
of traditional despotism in virtue of its near obliteration of civil life. This
has a momentous consequence - that for all of its subjects daily life must
be conducted within the institutions of the Soviet state. It is the state, or one
of its arms, that determines the job a person holds, the apartment he lives
in, the education available to his children and all of the other crucial
dimensions of his life-chances. In the Soviet Union, to a far greater extent
than in any Western country, housing, education and employment are
positional goods in that access to them is achieved primarily through
position or influence in the Party or its subordinate hierarchies. It is of vital
importance to understand that this is true, not only of the exploitative
nomenlclatura, but of virtually everybody in the Soviet system, and gives to
everyday life there an aspect of mutual predation lacking in societies where
most goods are allocated by markets. It means that, in the Soviet Union,
exchange is often a zero-sum transaction - what one gains another loses. It
means that, however estranged they may be from it, subjects of the Soviet
state must daily renew its institutions, and thereby perpetuate it in exis-
tence. This is so, even when they turn to the ubiquitous parallel economy
for sustenance, since the illegal economic networks which enable it to
reproduce itself exist only on its sufferance. For this reason, no subject of
the Soviet system can escape contamination by its practices.
The characteristics of the Soviet totalitarian order as a political state of
nature encompass a paradox, since in Hobbes and other contractarian
thinkers the state of nature is (by definition) pre-political. Like all true
paradoxes, however, this one contains an important truth. Soviet totali-
tarianism has in common with the Hobbesian state of nature, first and last,
lawlessness. Without legal order, there are no protected domains of inde-
pendence within which individuals may frame and enact their plans. Worse,
since there is no law, anything may be judged to be illegal: every Soviet
subject must live in a permanent legal twilight in which any act of his may
be arbitrarily criminalized. Given that the necessities of daily life demand
constant breaches of rules and regulations, every Soviet subject is per-
manently vulnerable for their infractions - but he would be so, even if by a
186 Critiques
miracle he had not breached any known rule, since the security organs still
retain extra-judicial powers. Further, some laws - such as that against
anti-Soviet activity - are susceptible to any interpretation. This condition of
lawlessness brings into being an environment of uncertainty which also
resembles that of the Hobbesian state of nature. For in it no one can be
bound to keep agreements with others, save by extra-legal means, and there
is an incentive for everyone to use all available resources of power to
protect his interests against possible attack by others. Power is vested in the
Party and its organs. Thereby arises that process of constraint mutual
predation, so characteristic of Soviet life, in which Soviet subjects use the
power of the Party to prey upon one another. Finally, even though it is far
from being pre-political (indeed everything in it is politicized), the Soviet
totalitarian order is akin to the state of nature in that lacks the institutions of
civil society. To say that it is a political state of nature, then, is to say that
it is a lawless polity, in which the war of all against all is conducted through
the medium of the Party and its auxiliary institutions.
It is the genius of the system that it constrains its victims to renew it. The
great majority of its subjects, including most of the nomenklatura, would
undoubtedly be better off without it, but they are compelled to recreate it
daily. Their exchanges are often not zero-sum so much as negative-sum -
they leave each worse off. In this the captives of the Soviet state resemble
the denizens of Hobbes's state of nature, condemned by their insecurity to
a war of all against all that none may escape. This is the Prisoner's Dilemma
in a classic form: rational prudence as engaged in by each produces in-
security and impoverishment for all. The Soviet state is a phenomenon
which even Hobbes's dark vision could not have foreseen - a Leviathan in
which lawless power and a predatory state of nature are inextricably
intertwined.
The darkest side of Communist totalitarianism - its ability to implicate
its victims in its worst atrocities - has been well expressed by Simon Leys:
If totalitarianism were merely the persecution of an innocent nation by a
small group of tyrants, overthrowing it would still be a relatively easy
matter. Actually, the extraordinary resilience of the system resides pre-
cisely in its ability to associate the victims themselves with the all-
pervasive organisation and management of terror; to turn them into
active collaborators and accomplices. In this way the victims acquire a
personal stake in the defense and preservation of the very regime that is
torturing and crushing them.63
As he acknowledges, Leys's observation parallels many made by
Alexander Zinoviev. We need not take literally Zinoviev's assertion64 that
Stalin's power was the ultimate expression of the power of the people to
 Totalitarianism, reform and civil society 187
accept the insight it contains - that one of the chief sources of the stability
of totalitarianism is its ability to implicate its victims in the terror and
repression to which they and their fellows are subject.
One explanation of the stability of totalitarian orders, then, is in the
morally compromised condition of their subjects. The endemic scarcity of
the necessities of life, and their control by the Party and its subordinate
organizations, condemns the Soviet subject to compete with his fellows in
collaborating with the system that oppresses him. The impact of this daily
necessity to lie, cheat, and distrust one's fellows, and thereby to renew the
system by which all are held captive, is not illuminated by standard
Arendtian notions'S of the atomization of social life in totalitarian orders.
Such atomization is real, but (as such observers as Havel66 have noted) the
impact of totalitarianism on personality is yet more destructive. It tends to
the very fragmentation of personality, to a pervasive demoralization,
anomie and disintegration of the person that perhaps only Orwell foresaw.
This is not to accept Alexander Zinoviev's extreme thesis that totali-
tarianism has indeed created a new man, Homo sovieticus,67 whose per-
sonality has been entirely collectivized. The evidence is that, though he
exists, Homo sovieticus is rare - most Soviet subjects retain their pre-Soviet
ethnic, religious and moral identities. Though they have thus survived
intact, for the most part, the identities of the subjects of totalitarian orders
have been injured in ways that further compound the difficulties of moving
toward a stable post-totalitarianism - a post-totalitarianism that is not
anarcby or chaos.

THE PROSPECTS FOR PERESTROIKA

How might the Soviet political state of nature be transcended? Only, it
seems, by the emergence of an Hobbesian sovereign. On the most benign
interpretation to which it is subject, perestroika may be seen as the project
of just such a sovereign - on the model, perhaps, of Peter the Great - aiming
to construct in the Soviet Union a framework of law within which enter-
prise and civil society might shelter. In support of this interpretation might
be cited Gorbachev's anti-corruption campaign and his frequent in-
vocations of socialist legality.
In fact, the available evidence supports the opposite interpretation - that
perestroika is in substance the project of suppressing the nascent forms of
civil society that had begun to emerge during the Brezhnevite era of
stagnation. During those two decades, there h~d occurred throughout the
Soviet Union, but especially in parts of Soviet Asia, the growth on a large
scale of illegal businesses, controlled and indeed owned by Party bosses
enjoying considerable autonomy. The forms of civil life represented and
188 Critiques
expressed in this 'mafia' were thoroughly distorted by the totalitarian
environment which they inhabited and on the basis of which their initial
power rested. Nevertheless, the extensive network of parallel institutions
which this mafia operated sustained the manifest prosperity of the
provinces in which it operated, most particularly Uzbekistan and
Kazakhstan, and it enabled a section of Soviet society to create a space
within which quasi-autonomous institutions could exist and in which
resources could be diverted from the control of the state to private ends.
In many parts of the Soviet Union, the so-called 'era of stagnation' was
in fact a period of illegal boom. It is precisely this prosperity which is being
threatened or destroyed by perestroika, with its campaigns against corrup-
tion. Here we reach an insoluble contradiction in the policy: its objective of
reasserting central Party control over local and regional satrapies is in
irreconcilable conflict with the objective of economic renewal. Free
markets and contractual exchange barely existed during the era of stag-
nation, but simulacra of them flourished on a grand scale in the parallel
institutions of the illegal economy. Like black markets everywhere, those
of the Soviet era of stagnation were inefficient, inequitable and unworkable
in enterprises requiring large injections of long-term capital; but they
mitigated the catastrophes of the central planning institutions, putting to
productive work resources that would otherwise have been wasted and
thereby generating a standard of living higher than any achievable by
planning.
From the standpoint of the totalitarian party, however, a flourishing
parallel economy means a diversion of resources into a sphere that is not
controlled by the state: a key objective of perestroika is therefore to reclaim
these 'privatized' resources for the state sector. In economic terms, the
results of this policy have been, and can only be disastrous. It entails
disrupting and often destroying the principal efficient institutions of pro-
duction and exchange in the Soviet economy. In political terms, perestroika
expresses the project of renewing totalitarianism, of reincorporating within
the totalitarian order the quasi-autonomous social forms that characterized
the Brezhnev period. As Francois Thom has succinctly observed:
The Law against Unearned Income has signalled the start of an offensive
against civil society in the purest Communist tradition. 68
The economic aspect of perestroika will almost certainly fail. It must fail,
not just because it tends to destroy the parallel institutions that have hitherto
partly supplanted the planning institutions, but also because it deprives the
planning institutions of signals from the parallel economy by which they
have long been guided. A real shift to a market economy - the only way in
which economic renewal can be achieved69 - is excluded because it would
 Totalitarianism, reform and civil society 189
entail the reconstitution of a civil society and the end of the totalitarian
fusion of the economy with the polity. It is in any case doubtful whether the
motives and dispositions needed to run a Western-style economy persist in
the Soviet Union - particularly when everyone knows that the profits of
enterprises established under perestroika are liable to NEP-style con-
fIScation whenever the policy is reversed. After two generations of civil war
against civil society, a market economy, along with other elements of civil
life, can probably only exist in the interstices of the totalitarian state. The
economic objectives of perestroika can be achieved, if at all, only in
enclaves of the Soviet economy where Western capital is under Western
management, and does not disappear into the abyss of the central planning
institutions.

THE FUTURE OF THE SOVIET UNION: OTTOMANIZATION?

If the Soviet case is treated as a paradigm, it has a clear implication:
post-totalitarianism, the dissolution of totalitarian institutions and the re-
constitution of civil society, is achievable by a process of internal evolution
only in states where some civil institutions have remained intact. This result
is corroborated by the Polish example, where the Church was never sub-
stantially incorporated into totalitarian administration, and by the Baltic
states, where a flourishing parallel economy, together with a strong sense
of nationhood, combined to limit the effectiveness of the post-War re-
construction on the Soviet totalitarian model. In these states, a reversion to
totalitarianism could probably not be accomplished even by recourse to
military repression, which (as in Poland in the early 1980s) would instead
achieve merely an authoritarian despotism. In the Soviet Union, a reversal
of the policy of glasnost in the wake of economic collapse is, by contrast,
bound to shatter the present precarious neo-totalitarian eqUilibrium and
precipitate a return to totalitarianism. It is profitless to speculate on the
forms that Soviet totalitarianism will assume in the period after the likely
termination of the present policies. A recurrence to Brezhnevite 'stag-
nation', and the 'Ottomanization'70 of the Soviet Union, with the provinces
and the republics regaining a measure of de facto independence, seems the
most benign outcome at present imaginable, but even that does not look
very likely. The economic situation in the Soviet Union probably prohibits
any such reoccurrence, with a catastrophic collapse, and consequent wides-
pread famine, being predicted by Soviet economists in mid-1989 as real
possibilities in the USSR two or three years hence.71 A policy of military
repression on the model of that applied in China in mid-1989, inaugurated
in response to nationalist and religious disturbances themselves mani-
pulated by local Party bosses, is well within the bounds of possibility. In the
190 Critiques
medium to longer term, a policy of exploiting Russian nativist sentiment -
already evident in the leeway permitted to the Pamyat organization - may
develop as a means of buttressing the totalitarian regime with populist
support. Whatever the specific form it might take, a reversion to classical
totalitarianism in the aftermath of economic collapse need not mean that the
present policy has altogether failed. If it yields diplomatic victories and an
influx of capital - with the current German-Soviet negotiations producing
something akin to a second Treaty of Rapallo - the policy of perestroika
might be assessed as having succeeded despite its economic failure, since it
will have secured for Soviet totalitarianism yet another lease on life.
There is, however, another side to this scenario. Even if totalitarianism
were to be re-imposed in the Baltic states, say, and the neo-totalitarianism
project of perestroika reversed (along with glasnost) in the Soviet Union in
the wake of economic collapse, the prospects for classical totalitarianism in
the USSR over the longer term of a decade or more are poor. For, even if
the Soviet Union were to emulate the Chinese model, it would not return to
wholesale Stalinist terror, which the majority within the Party would resist
It would instead rely solely on military force, together with selective police
repression. In such circumstances, the maintenance of ideological con-
formity - the reproduction of the weltanschauung-state - is likely to be an
increasingly low priority. In other words, if there is a reversion to classical
totalitarianism in the USSR, it is likely to be relatively brief, being followed
by worsening economic crisis and a slow shift to an 'Ottomanized' authori-
tarianism, held together (as in the era of stagnation) by military force, but
against a background of worsening economic decline and increasing
popular discontent. Retotalitarianization, if it occurs in the Soviet Union,
will not last long - almost certainly, less than a generation. If, as Walicki
has argued72 , the days of totalitarianism are numbered, once the nomen-
klatura comes to be concerned solely with self-reproduction, and is no
longer animated by a project of retotalitarianization, it bodes ill for the
repressive policies of the current neo-Stalinist leadership in China.
It may even be the case that the Soviet Union is itself on the verge of
launching into a transitional period of post-totalitarianism. The dramatic
collapse of the Communist monopoly of power in the GOR, following on
the dismantling of the Berlin Wall and the no-less-sudden blossoming of
popular opposition in Czechoslovakia, may be interpreted (if the analysis
developed earlier has any plausibility) as tactical moves in a Soviet strategy
of 'reverse Findlandization', modelled on the authentically Leninist para-
digm of the Treaty of Brest-Litovsk (which in 1918 ceded large parts of
Russia, including the whole of the Ukraine, to Germany). The aim of this
strategy is to secure the rapid decoupling of the United States and Western
Europe (which American economic weakness in any case renders
 Totalitarianism, reform and civil society 191
inevitable in the medium term) and the neutralization of most of Western
Europe at the cost of Soviet withdrawal from Eastern Europe. Even if, as is
likely, this strategy is successful, it is an extraordinarily dangerous one for
the Soviet Union to adopt. It risks strengthening secessionist movements
within the Soviet Union and triggering new movements for independence
in areas (such as Soviet Central Asia) which have, until now, been quies-
cent. If such a scenario of internal disintegration were to begin to unfold in
the USSR, the most likely outcome would not be a project of retotali-
tarianization, but instead recourse to classical authoritarian measures of
martial law and police repression. In that eventuality, however, the Soviet
Union would have become irreversibly post-totalitarian.
The prospects of the experiments in controlled democracy in Hungary
and Poland will depend mainly on the degree of success each achieves in
its policy of economic reform. Even in these comparatively favolH'able
cases, the shift to a market economy will involve dislocations and costs that
may destabilize the fragile political settlements achieved there. In Poland,
the economic programme of Solidarity is not wholly coherent, but even a
consistent policy of full marketization carries with it the risk of dividing
Solidarity in the country from Solidarity in the city (and for that reason
finds little favour outside the Party). In Hungary, which has passed from a
non-terroristic totalitarianism sustained chiefly (as was long the case in
Czechoslovakia) by economic sanctions to a controlled democracy without
any significant intervening phase of post-totalitarianism, the political
settlement may appear even more precarious. Recourse to repression in
Hungary may, however, be rendered less likely by the role that Hungary
may play in Gorbachev's European diplomacy. Here Hungary and Poland
are very different: though they have in common an economic crisis, includ-
ing an unsustainable foreign debt, they are to be distinguished in that
developments in Poland have arisen endogenously, whereas in Hungary
they occur at least in part in response to exogenous pressures from within
the Party and, perhaps, the Soviet Union. For this reason, repression could
be achieved more easily in Hungary than in Poland should policy undergo
a major reversal in the Soviet Union and its European diplomacy be
dislocated, or the domestic Hungarian political process be destabilized by
economic collapse.
In both Poland and Hungary, the danger exists of a recourse to a
post-Communist dictatorship as the necessary political mechanism for the
transition to a market economy. Such a 'Pinochet solution' to the transition
problems of dismantling the centralized economy is unlikely to occur in
Czechoslovakia or Eastern Germany. with the former returning to the
strong social-democratic traditions of the post-War years and the latter
eventually being integrated into West German democratic institutions as
192 Critiques
part and parcel of the inexorable process of reunification. In the rest of
Eastern europe, classical Communist totalitarianism is already showing
signs of weakness (as in Bulgaria), and even Romania may not prove
immune to change or collapse. In these parts of Eastern Europe, however,
the project is of re-Balkanization rather than of transition to civil society.
The model for such a prospect, in these parts of Eastern Europe (and,
indeed, in the Soviet Union), may be contemporary Yugoslavia, with its
intractable ethnic conflicts, profound economic problems, weak populist
governments, and chronic tendencies toward disintegration. In all these
areas, post-totalitarianism may mean political authoritarianism super-
imposed on social and economic chaos.
The answer to the question with which we began - what must be true for
post-totalitarianism to be a possibility? - has proved to be that important
elements of civil society must remain intact if the transition to a full civil
society is to be achieved. Where, as in the Soviet Union, these are lacking,
totalitarianism appears (at any rate for the medium term) to be a one-way
street. Even in Poland, where civil society was never fully suppressed, the
conflicts of interest generated by a policy of marketization could produce
considerable political instability, and endanger the otherwise well-
conceived policy of buying off the apparatchik class. The possibility sug-
gests itself, in the light of these considerations, that the task of reforming a
Communist economy confronts problems that are insoluble. Uke the Soviet
Union itself, most of the states of the Soviet bloc (even where a return to
classical totalitarianism is not a realistic prospect) cannot be expected to
fulfil the fIrSt condition of a stable post-totalitarianism, which is a stable
market economy. They must expect, not the reconstitution of civil society,
but Ottomanization - the process of decline, corruption and the waxing of
the institutions of the parallel economy that characterized the era of stag-
nation in the Soviet Union, but in a context of worsening economic con-
ditions for the entire bloc.
Except perhaps in Poland and Hungary, Czechoslavakia and East
Germany, the waning of totalitarianism, as it seems likely to occur over the
medium term, primarily for economic reasons but accelerated by ethnic and
nationalist conflicts, is unlikely to be accompanied by the waxing of a civil
society. For the foreseeable future, chaos and instability, contained only by
the recurrent threat and exercise of military force, seems the most likely
outcome for the USSR. The reform policy of perestroika founders, in the
end, on the overwhelming likelihood that the transition from totalitarianism
cannot, save in exceptional circumstances, be conducted in an orderly
fashion. Instead, it is to be seen as the prelude to a process of repression,
decline and instability that resembles the beginnings of revolution more
than it does any kind of reform.
 Totalitarianism, reform and civil society 193
THE CULTURAL ORIGINS OF SOVIET TOTALITARIANISM
On the account I have tried to develop, the Soviet totalitarian regime owes
its stability primarily to internal factors - to the public choice problems
generated by any move to a civil society and to the new Hobbesian dilemma
i..'lat constrains its sUbjects. I have also argued that Western aid has been
decisive in enabling the regime to recover from crisis and to expand. The
upshot of my account has been to confirm the soundness of Lenin's insight
into the nature of Western capitalist states - which, Lenin prophetically
observed, were bound to compete with each other in forming the Soviet
state. This competition, like the Prisoner's Dilemma that sustains the Soviet
totalitarian order internally, is doubtless statable in game-theoretic terms. It
goes far to explain the lack of effective Western strategy in regard to the
Soviets, and the strategic advantage the Soviets have in formulating long-
range policy in regard to the West.
The question arises as to why Western opinion has systematically mis-
understood the nature of the Soviet regime, and in particular why Western
opinion lacks any perception of the enmity of the Soviet Union to Western
civil societies. I submit that a major part of the explanation for this blind-
ness is in the fact that Soviet Communism has its roots in elements of the
most ancient, central and fundamental Western traditions - and not, as the
conventional wisdom supposes, in Russian political or religious culture.
What are the cultural origins of Soviet Communism? In order to answer
this question we must ask another: What are the intellectual antecedents of
Marxism? Kolakowski73 has located one source of Marxism in the Greek,
and especially the neo-Platonic preoccupation with the contingency of
human existence, which was transmitted to Marx via Hegel. In this account,
Marxism is a secularized version of the mystical soteriology of Greek
Platonism in which a return to the unalienated human essence replaces
reabsorption in the Absolute as the form of salvation and release from
contingency. Kolakowski's analysis neglects or underestimates the contri-
bution of Judeo-Christian traditions to the intellectual and moral formation
of Marxism. For it was Christianity, with its conception of human history
as a moral drama, which allowed the Platonistic soteriology to be trans-
formed into an historical theodicy. This is the insight captured by Voegelin
in his interpretation of modem political religions as gnostic immanent-
izations of Christian eschatology. As Voegelin puts the point:
The characterisation of modem political mass movements are neopagan,
which has a certain vogue, is misleading because it sacrifices the his-
torically unique nature of modem movements to a superficial resemb-
lance. Modem redivinisation has its origins rather in Christianity itself,
194 Critiques
deriving from components that were suppressed as heretical by the
universal church.74
The same point is made by Michael Polanyi:
Had the whole of Europe been at the time of the same mind as Italy,
Renaissance Humanism might have established freedom of thought
everywhere, simply by default of opposition. Europe might have
returned to - or if you like relapsed into - a liberalism resembling that of
pre-Christian antiquity. Whatever may have followed after that, our
present disasters would not have occurred.75
On this interpretation, Marxism is a Christian-historicist gloss on a Greek-
rationalist doctrine of salvation. It should be noted here, however, that
contrary to the conventional academic wisdom, Western Christianity is far
more implicated in th, generation of Marxism than Eastern Christianity.
For Western Christianity imbibed the elements of Aristotelian rationalism
as transmitted via Acquinas into the medieval world, together with the
humanist values of the Renaissance and the secularizing impact of the
Reformation, that had little impact on Russian Orthodoxy. For Orthodoxy,
with the exception of a few iconoclastic thinkers (such as the early
Berdyaev76), Marxism represented the incursion of a Western ideology into
Russian Christianity - and one which. furthermore, had emerged in the
West partly because of the decadence of Western Christianity. This is not
to deny that, as Besancon has shown,77 Soviet Communism has some
sources in an alienated Russian intelligentsia, - a point emphasized by
Solzhenitzyn78 and prophetically made by Dostoyevsky.79 It is to question
the common Westem belief that Soviet Communism was ever sustained by
elements in Russian religious life - against which it has waged a petpetual war.
It is from its expression of the central tenets of the European Enlighten-
ment, however, that Marxism, and its embodiment in Soviet Communism,
derives its essential appeal to Western intellectual opinion. It is here that the
Jacobin lineage of Leninism, and the character of the French Revolution as
the first precursor of twentieth-century experiments in social engineering
via the mass liquidation of entire social groups need to be noted. The recent
revisionist historiography80 of the French Revolution has noted its terrorist
nature, in particular the fact that, whereas at the time the Bastille was
stormed, it contained fewer than ten inmates, by the time the Terror had run
its course around half a million Frenchmen were incarcerated for political
reasons, many of whom would perish in jail. We know that Lenin was
himself much influenced by the Jacobism precedent as an early experiment
in what Talmon has well called 'totalitarian democracy' .81 There seems to
be good reason, then, to see a clear historical linkage between the two
 Totalitarianism, reform and civil society 195
revolutions, both as to their goals, their strategies and the types of institu-
tions they produced.
It is in the common origins in the secular faith of the Enlightenment that
the affinity of the two revolutions is most plainly seen. The ideas of a
self-consciously planned society and of a universal civilization grounded in
scientific knowledge are central elements of that religion of humanity that
is expressed in both Marxism and liberalism. They express, in a dis-
tinctively modernist fashion, values and beliefs - rationalistic and opti-
mistic - derived both from the Greco-Roman classical tradition and the
Judeo-Christian traditions that are coeval with Western civilization. It is in
this truth that the central paradoxes of totalitarianism are to be found: the
paradoxes of its enmity to the civilization that gave it birth and the paradox
that, though Marxism-Leninism is a modernist ideology, Soviet totali-
tarianism is at war with the most fundamental institutions of the modem
world as it has thus far developed. We may justly judge that von Laue
exaggerates greatly, when he avows that 'it was the West which, by the
model of its superior power, has shaped the Soviet dictatorship. Soviet
totalitarianism has basically no more than the caricature echo of Western
state and society, the best copy feasible under Russian conditions.'82 Von
Laue's extreme overstatement, like the claim that Western societies have
totalitarian aspects, nevertheless expresses a grain of truth - the truth in the
claim that totalitarian ideology has its roots in Western tradition, and
totalitarian regimes are episodes in the global process of Westernization,
aberrant and distorted not by the traditional societies which they destroy but
by elements within the Western tradition itself.
Western opinion's blind spot in regard to the nature of Soviet com-
munism is congenital and incorrigible. It expresses an integral part of the
Western world-view. One may even say, without too much exaggeration,
that Gust as totalitarianism is only the shadow of modem civil society) so
Soviet Communism is only the shadow cast by the European Enlighten-
ment. A realistic perception of Soviet enmity to Western civil societies
presupposes an insight into the defects or limitations of Western traditions
of which the animating ideology of the Soviet regime is an authentic
development. Nothing supports the hope that Western opinion is capable of
the self-criticism such an insight requires. If it comes to pass, the fall of
Soviet totalitarianism is most likely to occur as an incident in the decline of
the occidental cultures that gave it birth, as they are shaken by the
Malthusian, ethnic and fundamentalist conflicts which - far more than any
European ideology - seem set to dominate the coming century.83

October, 1989
13 Western Marxism: a tictionalist
deconstruction

The visits that Wittgenstein made to the Soviet Union in the late 1930s must
be among the least researched episodes in his life. Most of his biographers
mention the visit he made in 1935, and a few refer to a later visit in 1939.
None tells us anything of substance about what he did there, and, in
particular, none of them gives any clue as to how his experiences in the
USSR might have influenced Wittgenstein's philosophic'al development.
We learn that during his first visit Wittgenstein was offered a Chair in
Philosophy at the University of Kazan (where Tolstoy had studied) and that
for a while he considered seriously the possibility of settling in the Soviet
Union. We learn nothing, or little, ofWittgenstein's intellectual contacts in
the Soviet Union. It is only very recently, in fact, that we have come to
know of the most formative of Wittgenstein' s intellectual encounters in the
Soviet Union, which occurred in his conversations in 1935 and 1939 with
the neglected Hungarian Marxist thinker, L. Revai. 1
Since Revai' s life and work remain little known in the West, it is worth
sketching their main outlines. Born in Budapest in 1881, the first son of a
well-established banking family, Revai made a minor mark on Central
European intellectual life in the first decade of this century as a literary
theorist. His work at that time (now virtually unobtainable and nearly
forgotten) was derivative and unoriginal, being an eclectic weaving
together of a variety of currently fashionable themes. It reveals nothing of
the intellectual radicalism which distinguishes his mature theorizing and
amounts to a highly conventional application of Kantian and Schopen-
hauerian conceptions to familiar questions in the theory of culture. In the
1920s Revai published hardly at all. He had joined the Communist Party
shortly after its foundation in 1918, abandoning the romantic syndicalism
of his youth for a Leninism he was never to renounce, and seems to have
occupied himself for a decade or more in political and organizational work.
Little is known, even now, of Revai's thought during this period. We know
 Western Marxism: afictionalist deconstruction 197
that in 1933 Revai left Hungary for the USSR, and stayed there until 1945.
From the present volume we learn, for the first time, that in 1936 Revai was
incarcerated in a labour camp, from which (as one of the very few to have
returned from internment at that time) he was released in 1938. It is only
with the present volume, above all, that we learn that on several occasions
in 1935 and 1939 Revai met Wittgenstein, and engaged with him in con-
versations which left a lasting mark on the philosopher's later thinking.
Revai's life after the War was uneventful. He pursued his oblique and
elusive career in Hungary as a translator and occasional anonymous contri-
butor to Budapest cultural reviews. Scarcely known in his own country or
elsewhere, seeming to have acquired few, if any, disciples or followers,
Revai published nothing under his own name (except for a brief statement
of neutrality during the disturbances of 1956). He died in 1973. Throughout
nearly thirty years of obscurity in his native land, ignored by the authorities
and without even the dubious privileges of a dissident intellectual, Revai
worked patiently and indefatigably on his master-project - the development
of a Marxist theory of language.
Revai's life-work was never completed. The six essays collected in The
Word as Deed: Studies in the lAbour Theory of Meaning are only frag-
ments, embodying all that remains of a massive project, and constituting the
literary remains of a thinker whose input on the development of Western
thought has thus far been entirely esoteric. It is, indeed, only owing to the
resourcefulness of the collection's two editors that Revai's work has been
rescued from oblivion and its bearing on Wittgenstein's philosophy
revealed. In an extended biographical foreword to the collection, Olsen and
Kahn piece together an account of Revai's life and thought from the
evidences of the papers recovered (through the intervention of his sole
disciple) from his modest apartment in Budapest They are able to tell for
the first time how Revai, having by them emerged from a stay in the Gulag,
spoke with Wittgenstein about the conception oflanguage as an incident in
human labour which he had begun developing in the 1920s in Hungary, and
which his experiences in the camps had crystallized into a more definite
doctrine. In the camps, Revai told Wittgenstein, the complex grammar of
civil society was dissipated, and speech returned to its more primordial
function as an integrating mechanism in the human transformation of
physical energy. It was in the camps, where the forces of production were
developed in transparent social relations without the mediation of mystify-
ing ideological structures, that the adverbial labyrinth of language was
deconstructed into the aboriginal rudiments of imperatival speech. Revai's
discovery in the camps was of an Ursprache, made up of speech-acts whose
sense was exhausted in their uses in the ongoing praxis of labour. It was this
discovery which Wittgenstein exploited, when in the Investigations2 he
198 Critiques
experimented with the possibilities of a language consisting only of words
connected with a single activity.
The philosophical concept of meaning has its place in a primitive idea of
the way language functions. But one can also say that it is the idea of a
language more primitive than ours. Let us imagine a language.... The
language is meant to serve for communication between a builder A and
assistant B. A is building with building-stones: there are blocks pillars,
slabs and beams. B has to pass the stones, and in the order in which A
needs them. For this purpose they use a language consisting of the words
'block', 'pillar', 'slab', 'beam'. A calls them out; B brings the stone
which be has learnt to bring at such-and-such a call. Conceive this as a
complete primitive language.
In Wittgenstein's work, the manifest political content of Revai's discovery
is lost, sublated and recuperated in a reactionary and unhistorical reification
of ordinary usage. For Revai, by contrast, the Ursprache of the camps was
the problematic from which he was to develop his first formulations of a
materialist theory of meaning. And the discourse of the camps had in
Revai's thought another significance, elaborated fully by him only much
later, as a dialectical pre-figurement of the speech community of com-
munist society, in which words are only shadows cast reflexively by the
self-consciousness of deeds.
In part, no doubt, because the exercise of dating Revai's writings
remains speculative and conjectural, the collection is organized in concep-
tual rather than chronological fashion. It is in the first three essays that we
find Revai's chief theses on meaning, use and labour set out in pr0-
grammatic terms. His most radical thesis concerns the place of the most
primitive unit of meaning, the ergoneme, in the constitution of the act of
labour. Revai's insight here - an insight that he probably communicated to
his colleagues at the Marx-Engels Institute in Moscow, and which may
have been a factor in his subsequent incarceration - was that the decisive
step to the human species from its animal forbears cannot consist, as Engels
supposed, in the development of tools. If man is a tool-using animal, he is
defined (and defines himself) by the exercise of the most distinctively
human tool of all, that of language. What is only work in animals - the
expenditure of energy through the manipulation of matter - becomes
labour in humans, since it is from the first saturated with a semantic
superstructure. This is the dialectical counterpoint of Goethe's bourgeois-
humanist dictum, 1m Anfang war die Tat. Revai insisted that, whereas
speech is a moment in the constitution of the act of labour (and language
itself the shadow of speech) labour itself is the most primitive of all
 Western Marxism: aJictionalist deconstruction 199
speech-acts. Without the ergoneme, which is the reflexive self-recognition
of deeds, we have, not labour, but only work. But the most primitive unit of
meaning is not theorized, for Revai, after the fashion of bourgeois-semantic
atomism as an element from which complex structures of meaning are
constituted piecemeal. Rather, the ergoneme is found only in the holistic
semantic structures which arise along with the social relations of labour. It
is, for this and other reasons, a radical error to model language-use on the
reified metaphor of the speaker. To do so is to consecrate the fetish, central
to the bourgeois-Robinsonnade of Western linguistic theory, of the auto-
nomous language-user, and to suppress recognition of speakers as social
ensembles of ergonemes. It may be said that Revai's entire project was a
project of transcending the subjectivist-idealist problematic of the speaker.
As he puts it himself, magisterially: 'It is not speakers who labour, but
labour that speaks'.
For Revai, accordingly, the primitive unit of meaning, the ergoneme, is
a necessary constituent of the act of labour itself. Further - and it is in the
collection's second essay that Revai develops his critique of the ideological
construction of the speaker - units of meanings are always distributed over
complex semantic structures, which themselves mirror constellations of
labour power. It is within these structures that we must situate the activity
of speech. If we do this, we find, once again, not the speaker, but the
speech-community - the historically specific construction of labour
powers. It is here that Revai makes one of his boldest moves, and identifies
speech-communities with classes. The radical intent of such a move in the
USSR in the late 1930s should be obvious, and stands in need of no
elaboration. The third and fourth essays treat of the political economy of
meaning. Among the topics addressed are the unity of speech and act in
pre-class societies, the primitive accumulation of meaning, surplus mean-
ing and the expropriation of meaning, the sequestration and enclosure of
meaning in early industrial capitalism, the estrangement of word from
meaning in advanced capitalist orders, and related topics.
In the last two essays, Revai brings his theorizing to bear on theoretical
controversies being conducted in the USSR from the late 1930s onwards.
We learn from the editors that Revai's thoughts were stirred in the late
1940s and early 1950s by the debates surrounding the contributions made
by N. Y. Marr to linguistic theory. Stalin, we recall, had criticized Marr not
only because of his attempts to develop a theory of the class-specificity of
languages, but also because, in separating language from thought, Marr had
lapsed into Idealism. Revai' s paper bears the marks of his earlier thought,
but it appears to have been occasioned by Stalin's response to the contri-
butions of D. Belkin and S. Furer to the debate on Marr's linguistics. Stalin
200 Critiques
criticizes Bellcin and Furer for failing to appreciate the distinctively human
character of language-use, and so for neglecting to distinguish language-
use from the signs and gestures of animals. Observing that
linguistics treats of normal people possessing a language, and not of
anomalous deaf mutes who lack a language, [Stalin asks:] How do
matters stand with the deaf mutes? Does their thought function, do ideas
originate? Yes, their thought functions, ideas do originate. It is clear that
since the deaf mutes lack a language, their ideas cannot originate on the
basis of language material.... The ideas of deaf mutes originate and can
exist only on the basis of the images, perceptions and conceptions
formed in practice about the objects of the exterior world and their
relations among themselves, thanks to the senses of sight, touch, taste
and smell. Outside of these images, perceptions and conceptions,
thought is empty, devoid of any content whatever, i.e. it does not exist.'
Revai's commentary on this debate, never published in his lifetime, has as
its problematic the thesis of the semantic constitution of the labour-act and
(as its dialectical counterpart) the thesis of the construction of meaning by
labour power. From this problematic Revai derives his further thesis that
the question of the place of language in the 'superstructure' or 'base' of
social relations is, and cannot avoid being, wrongly posed - a position he
shared with Stalin. Much of the fifth essay is addressed to this question. It
is in the last of the essays, however, that Revai reveals his political intent,
when he comments, cryptically and suggestively, on the epoch of silence.
The society of deaf mutes, he argues, is the society of Western capitalism
in its fmal stages, in which language has been replaced by gestures, and
labour evacuated of its meaning-content. The 'empty chatter' of capitalist
verbosity, in so far as it expresses at no point the semantic content of labour,
embodies the silence of the proletariat as its dialectical condition. The
silence of the period of socialist construction, by contrast, is fecund with
labour content. Revai does not try to suppress reflection on the organization
by administrative measures of speech and language in the Stalin period, nor
does he aim to pass judgement on it, after the fashion of bourgeois
moralists. He sees the Stalin period, instead, as one in which the necessary
publicity of meaning was, for the first time, given concrete historical
realization. The fiction of the speaker, together with the Idealist shibboleths
of subjectivity and private language-use, were during this period subjected
to a decisive critique on the terrain of praxis. The socialist administration
of language-use, then, is an historically necessary phase in the elimination
of the problematic of the subject In theorizing it in this fashion, Revai
gestures in his last essay to the practice of speech in Communist society.
There, he maintains, the interiority of thought is transcended, and speech
 Western Marxism: ajictionalist deconstruction 201
utters itself in a dialogic context in which it is realized as the semantic fonn
of the community of labourers. It is superfluous to comment on the remark-
able affinities between Revai's conception of Communist discourse and
that developed later by Habermas.
In addition to an extended biographical foreword, the collection is
distinguished by a long analytical postscript, in which the editors consider
how Revai' s theorizing is to be assessed and developed. It must be said at
once that their own perspective is very different from Revai's, whose
classical labour theory of meaning they repudiate explicitly. Nor should
this be surprising, since one of them, G. Olsen, had already in his important
Sense and Reference in Marxian Semantics 4 abandoned many of the cen-
tral tenets of an orthodox Marxian account of meaning. By contrast with
Revai, Kahn and Olsen are methodological nominalists, whose project is
that of generating the complex semantic structures of Marxian linguistic
theory from individual speech-acts. They invoke here work by another
linguistic theorist of their school, P. Reimer,S who aims to reconstruct
classical Marxian meaning theory, without postulating the ergoneme,
solely on the basis on constellations of speech-acts. Most innovatively,
drawing heavily on work of Kahn's, they advance a refonnulation of the
orthodox Marxian theory of surplus meaning, in which the exploitative
extraction of meaning from workers is analysed and explained in tenns of
the thesis that the meaning-content of each labour-act is retained by the
labourer only on condition that that of every other be subject to expro-
priation. 'Each may speak, but all are silent' - so Kahn summarizes a long
chain of subtle reasonings whereby the silence of the proletariat is re-
affinned in the new Marxist theorizing. In the final section of the analytical
postscript, Kahn and Olsen present competing accounts of the relations
between labour power and semantic structure, with Olsen arguing against
Kahn that the attribution to semantic structure of an inherent development
tendency is an illicit global generalization from the historically specific
semantic structures of the capitalist mode of production. In so developing
rival theoretical paradigms within a shared problematic, Olsen and Kahn
give further evidence, if such were needed, of the progressive character of
the research programme on which they are engaged.
In Kahn and Olsen's postscript to this invaluable collection, the central
insights of Marxist thought are preserved by the appropriation of the most
powerful techniques of bourgeois thought. Until now, Revai's work had only
an occult impact on Western thought by way of its influence on the greatest of
twentieth-century philosophers. Given the creative development to which
Revai's thought is subject in Olsen and Kahns' postscript to this volume, it will
be extraordinary if Revai' s work does not come to exercise a commanding
authority over the most advanced sections of the Western academic class.
14 Post-totalitarianism, civil society
and limits of the Western model

INTRODUCTION
The sudden collapse of Communist power in Eastern Europe, and its
weakening in the Soviet Union, have been interpreted by many in the West
as evidences of the global triumph of democratic capitalism. At its most
extreme, this interpretation has been extended by Francis Fulcuyama into
the thesis that history, conceived as the history of conflict between rival
ideologies and systems of institutions, is over, and liberal democracy has
been revealed as the final form of human government. I The manifest
bankruptcy of Communist institutions has been interpreted as conclusive
evidence in support of the universal appeal of Western institutions and
values and the expectation has been engendered that it can only be a matter
of time before civil societies on the Western model come to exist in all the
post-Communist countries.
My aim in this paper is to contest this received view, or, more precisely,
to try to sift in it what is true from what is false. My conclusion will be
threefold. First, I shall hold that it is true that Communist totalitarianism is
in the throes of a terminal crisis, such that attempts to reproduce it (as in
China) are bound in the medium term to fail. Second, I shall argue that,
whereas the viable post-totalitarian regimes which emerge in the wake of
Communist totalitarianism must have the character of civil societies, they
need not (and often will not) resemble Western liberal democracies in other
important respects. Finally, I shall maintain that, although all talk of a 'third
way' between capitalism and Communism is an exercise in unreality, con-
temporary Western democracies are in several respects defective models
for the emergent post-Communist states.
It is important to begin by specifying the central, constitutive features of
totalitarian political orders. As I have argued elsewhere,2 the single most
important feature of totalitarian orders is their suppression (partial or com-
plete) of the institutions of civil society - the autonomous institutions of
 Post-totalitarianism 203
private property and contractual freedom under the rule of law, which allow
people of divergent values and world-views to live in peaceful co-
existence. Because they politicize economic life and repress voluntary
associations, and because they are weltanschauung-states seeking to
impose a single world-view on all, totalitarian regimes are constituted by
the project of destroying the key institutions of civil society. Their success
in this project varies from case to case, with civil society being much more
comprehensively desolated in Romania and the Soviet Union, for example,
than in Poland or Czechoslovakia. Whatever its degree of success or
completeness, totalitarianism is to be defined by its opposition to civil
society, not by contrast with liberal democracy. This point may be put in
another way. Civil societies may exist and flourish under a variety of
political regimes, of which liberal democracy is only one. The authoritarian
civil societies of East Asia - Taiwan, South Korea, Singapore and Hong
Kong - are instances in which economic and personal liberties are sub-
stantially protected, but political or democratic freedoms curtailed. Indeed,
historically, civil societies have been more commonly associated with
authoritarian regimes or limited democracies than with liberal democratic
institutions: consider the examples of Bismarckian Prussia, or Whig
England (where democratic participatory freedoms were extremely res-
tricted). It is a central thesis of my argument that civil society and liberal
democracy need not and often do not go together, and the most decisive
phenomenon in the collapse of Communism is not the adoption of demo-
cratic governance but rather the emergence of civil life.
It is important, also, not to conflate the re-emergence of civil society
with 'the triumph of the Western idea' (in Fukuyama's banal terminology).3
A flourishing, dynamic and progressive civil society existed in Russia in
the late-Tsarist period, which it was the mission of the Bolsheviks (acting
in the service of a European ideology) to destroy.4 Again, the fundamental
elements of civil society were built up in Japan, not in the wake of defeat in
the Second World War, but in the Meiji period, two generations earlier.
Further, in Japan as elsewhere in East Asia, the adoption of the institutions
of civil society was not accompanied by acceptance of Western indivi-
dualist values or the abandonment of Eastern cultural traditions. Accord-
ingly, although the first examples of civil society appeared in Western
Europe in the aftermath of feudalism, it is a fundamental mistake to equate
the emergence of civil society with Westernization, or with the spread of
democratic capitalism.
Given this understanding of totalitarianism, how do things stand in the
post-Communist societies? We need to distinguish different regimes at this
point, so as to be clear which of them are genuinely post-totalitarian.
Romania and Bulgaria, though in the medium term they are likely to
204 Critiques
become so, are not yet genuinely post-totalitarian since they are ruled by
neo-Bolshevik cliques which have not permitted the re-emergence or re-
construction of the institutions necessary for civil life. Czechoslovakia,
Hungary and Poland all appear authentically post-totalitarian, and are prob-
ably irreversibly so. The most intriguing case is that of the Soviet Union
itself. It cannot be affirmed that a civil society has emerged, full-blown,
anywhere in the Soviet Union. Nowhere have economic reforms re-
introduced the essential preconditions of a market economy - a law of
property and contract, private investment capital or an autonomous banking
system. The most essential elements of civil society, private property and
the rule of law, have yet to be instituted. Nevertheless, the Soviet Union no
longer resembles anything akin to classical totalitarianism, and it is very
doubtful whether a reversion to totalitarianism in the Soviet Union is
realistically feasible. This is so for two main reasons. First, there are now in
the Soviet Union powerful oppositional movements, having a large
measure of autonomy from the state apparatus, which could be brought
back within Communist control only by recourse to large-scale military and
political repression. Most important among these autonomous movements
are the independence movements in the Baltic States, Georgia, the Ukraine,
and elsewhere in the Soviet Union. So deep are the popular and nationalist
roots of these movements, that only repression on a massive scale could
defuse them. In the case of Georgia, paramilitary repression has done little
more than slow down the movement for independence. If the independence
movements were to be crushed, or driven underground, a policy of repres-
sion would be required that spanned the Soviet Union as a whole and
radically reversed the impetus of glasnost.
Nothing in Soviet history, or in the present circumstances of the Soviet
leadership, allows us to rule out such a recourse to repression as the only
effective measure for preventing the disintegration of the Soviet state. Even
if such a policy of large-scale repression were adopted, however, it could
not hope to restore classical totalitarianism. Glasnost itself has done irre-
parable damage to whatever ideological legitimacy Soviet power may
otherwise have retained. It has revealed (both to Western observers and to
the Soviet public) the catastrophic inheritance of over seventy years of
socialist central planning - an inheritance encompassing almost apo-
calyptic environmental degradation, Third World infant mortality rates,
woefully inadequate housing and medical care, and industrial plant and
techniques that are twenty or even fifty years out of date. The result of the
revelations of glasnost has been to exhaust whatever ideological capital the
Soviet regime may still have possessed. The Soviet Union is not now, and
is most unlikely ever to become again, a weltanschauung-state on the
classical totalitarian model. If the policy of glasnost is reversed, the upshot
 Post-totalitarianism 205
will not be retotalitarianization, but rather an authoritarian regime, shorn of
the trappings of ideology in all but name, which sustains itself by a com-
bination of selective repression, tactical concessions and Western aid.
If this is the likely scenario for the Soviet Union, how are we to interpret
developments in China? There can be little doubt that in China a project of
retotalitarianization is being attempted - in the aftermath of an experiment
in market reform in which Marxist-Leninist ideology was even more
comprehensively abandoned than has been the case in the USSR. We do not
at present know enough to do more than speculate as to current policy and
its likely upshot in China. In the short run, the policy of repression, and of
reinstituting a totalitarian regime, is likely to appear to be more successful
in China than any analogous policy could hope to be in the USSR. China's
difficulties with her national minorities, though real, are as nothing com-
pared with those in the USSR. Equally as important is the fact that popular
disaffection does not seem to have progressed as deeply in the peasant
masses in China as it has in the USSR's industrial workers. As far as we can
tell, China's peasant majority remained quiescent during the events leading
up to the Tiananmen Square massacre, and (so far as we know) there has
been nothing in China that resembles the strike of the Siberian miners.
Finally, there is little evidence that there exists in China the extensive
network of underground institutions which may exist in the Soviet Union
now, and which certainly existed in Poland after the declaration of martial
law in 1980. For the immediately foreseeable future then, the forces of
emergent civil society have been effectively paralysed in China.
The prospects for totalitarianism in China over the medium term are
nevertheless poor. All available evidence suggests that genuine Marxist-
Leninist ideological conviction is dead in China, and that it is being used
currently primarily as in instrument of repression and of political discipline.
In the absence of ideological conviction, China may resume many of the
practices of a totalitarian regime, but the reality will be that of an aged
nomenklatura clinging on to the perquisites and privileges of power.
Further, the economic consequences of Tiananmen Square are not to be
underestimated. Doubtless the Chinese leadership was correct in believing
that foreign economic aid would not long be withheld. At the same time, it
will face massive difficulties in motivating the Chinese people to resume
productive work. A plausible scenario is that, over five or ten years or so,
economic reform will be resumed, against the background of yet worse
economic conditions. It is also plausible to suppose that, in different pro-
vinces, differing degrees and mixtures of repression and liberalization in
political life will develop. Over a generation, it is difficult to envisage the
successive Chinese leaderships having the capacity to resist the devolution
of power within China, and the economic imperatives which will impel it
206 Critiques
in the direction of civil society. If the Soviet Union is irreversibly post-
totalitarian at present, China is likely to have reached the same stage of
development in a generation at most In both cases, however, the waning of
totalitarianism will not have been accompanied by the waxing of a stable
civil society, but instead by a far more uneven, complex and unstable
diversity of circumstances and political conflicts and settlements.
In most of the emergent post-Communist countries, it is inherently
unlikely that the reconstruction of civil society can be conducted under the
auspices of a democratic regime on the Western model. This is so, in part,
because few of these countries have democratic traditions that are long and
deep, even when they had centuries of civil life before Communist institu-
tions were imposed upon them. Of the post-Communist states, only
Czechoslovakia can claim a real democratic inheritance: all of the rest
existed for all of their recent history under one or another variety of
authoritarianism. Contrary to received opinion in the West, however, this is
not the fundamental reason why the transition to civil society is unlikely to
occur within democratic institutions. As I have argued elsewhere,' Western
opinion has gone astray in systematically overestimating the importance of
political culture in Communist regimes, and underestimating the role of
Communist ideology and institutions. For all their diverse cultural inherit-
ances, the Communist regimes of Cuba, China, Bulgaria and Eastern
Germany, for example, exhibit structural affinities that can only be
explained by reference to the Communist institutions and ideologies they
have in common. Equally, the likely incompatibility between the neces-
sities of the transition from totalitarianism to civil society and the pre-
conditions of liberal democracy arises, not principally from the cultural
inheritances of the various post-Communist states, but instead from the
massive human and economic costs of liquidating the bankrupt economies
which the post-Communist regimes inherit.
These can scarcely be exaggerated. In Eastern Germany, for example,
which is in many ways in the most favourable circumstances, it is estimated
that between a third and a half of the workforce will unavoidably be
dislocated in the transition to a market economy, with the cost of market-
ization running into tens of billions of dollars over the next five to ten years.
If this is the case for Eastern Germany, which is the uniquely fortunate
beneficiary of a peaceful Anschluss with one of the world's greatest eco-
nomic powers, what will be the lot of Poland, where the costs will be far
greater and available capital incomparably less? The danger is that the
dislocations and costs of the transition period will destabilize the fragile
political settlement achieved by the Solidarity-led government. Thus far,
the immense reserves of legitimacy of that government has enabled it to
implement extremely painful and unpopular measures, which lowered
 Post-totalitarianism '}J)7

inflation from its peak of over 1,000 per cent a year to a fraction of that. The
painful adjustments inseparable from rnarketization have, however, only
just begun, with the prospect of massive lay-offs looming ahead - without
the insurance of a welfare safety-net for workers who are displaced. In such
circumstances, one may not unreasonably fear syndicalist resistance to
market reform, perhaps organized by ex-Communist trade unions. In that
event, however, a second period of dictatorship could well ensue, with
market reform being implemented via a Polish version of 'the Pinochet
solution' - economic liberalization pursued under the auspices of political
authoritarianism.
Such an outcome is far from being inevitable in Poland. It is still less so
in Hungary and Czechoslovakia, where the economic inheritance of central
planning is less ruinous than in Poland. Even in the Soviet Union, it is
plausible to envisage the Baltic states surmounting the economic diffi-
culties of the transition period through the medium of democratic insti-
tutions, given the social and political solidarity generated by the strong
sense of nationhood in Lithuania and the political maturity of the Baltic
peoples as a whole. Even there, however, we can discern factors which may
bring about a divergence between the necessities of the transition period
and democratic institutions. Ethnic conflicts that are relatively mild in
Latvia foreshadow ones that are already murderous in the Caucasus.
Ancient nationalist rivalries, compounded by the Stalinist legacy of mass
deportations and relocation, make the prospect of a stable civil society
being introduced via the process of democratization delusive in many parts
of the Soviet Union. In Soviet Central Asia, democratic mass mobilization
risks awakening Islamic peoples from their long quiescence and exposing
them to the appeal of fundamentalism. It need hardly be stressed that
Islamic fundamentalism, with its denial of any legitimate realm of purely
secular activity, is incompatible not only with liberal democracy, but also
with civil society in any of its many varieties.
In order to understand the colossal difficulties that are bound to be
confronted in the period of transition, we need to look not only at history,
but also to bodies of theory which explain totalitarianism and its current
crisis. The economic failure of Soviet-style central planning is best theor-
ized by the Austrian and Polanyian analyses of Hayek and Roberts, 6 who
focus on the epistemic role of market institutions and processes as genera-
tors and transmitters of dispersed and often tacit knowledge. The Austrian
and Polanyian perspectives require supplementation by the insights of the
Virginia School of Public Choice,7 which theorizes the incentive structures
created by institutions of central planning, and explains how the insuper-
able epistemic limitations of central planning are compounded by the
perverse interests planners have in concealing malinvestments. The
208 Critiques
fundamental problem of the centrally planned economy, by contrast with
any market economy, is that it lacks an error-elimination mechanism for
misconceived projects. For anything resembling an efficient coordination
of economic activity to be possible, it is essential that enterprises be
allowed to fail, and for the costs of these failures to be borne by those
responsible for the failure. The policy implication of this Public Choice
insight is the radical one that economic efficiency presupposes not only
market pricing but also private property in (most of, at least) the assets and
resources that are produced and exchanged.
The perspective of Public Choice illuminates not only the theoretical
explanation for the failures of central planning but also the dilemmas of the
transition period. The chief contribution to theorizing the phenomenon of
[Link] made by the Public Choice perspective in its recog-
nition that many interests will be injured in the transition period, which (if
the theory is sound) may be expected to organize collusively to resist and
thwart market reform. In a democratic environment (or in a non-democratic
one, such as contemporary Romania) we may expect elements of the
exploitative nomenklatura to forge tactical alliances with groups of
workers who expect to lose out in the early phases of market reform. The
dilemma of the transition period is the classic dilemma theorized in Public
Choice - that collusive interests may successfully prevent changes that
would benefit nearly all, including themselves.
No model exists for the successful conduct of the transition period. The
model of Erhardt's sudden deregulation of the post-War German economy
is largely inappropriate, for a variety of reasons. Post-Nazi Germany, unlike
most of the post-Communist states, possessed largely intact much of the
institutional infrastructure of a civil society - a law of property and
contract, the remnants of a banking system, and so forth. Insulted and
injured as these institutions had been under the Nazi regime, they had not
been comprehensively destroyed, as had been the case in much of the
Soviet bloc. Further, defeat in war had the unforeseen and highly desirable
consequence that collusive interest groups were destroyed or dissolved in
post-War Germany. This is to say that post-Nazi Germany did not have a
massive nomenklatura, inherited from the earlier totalitarian regime. Para-
doxical as it may sound, then, the situation of post-War Germany was
immeasurably more favourable (from the standpoint of the re-emergence of
civil society) than that of most of the post-Communist societies.
Post-War Germany had another decisive advantage in the legitimacy of
its government and in the fact that the Germans could be confident that,
under the rule of law, any assets acquired by them through their productive
labour would not later be subject to arbitrary confiscation. By contrast, in
the Soviet Union the memory of the New Economic Policy remains vivid,·
 Post-totalitarianism 209
and even where genuinely free markets are opened up the fear of sub-
sequent sequestration of assets and profits destroys or weakens the incen-
tive to productive work. Only the institution of a real rule of law - which
means a genuinely independent judiciary and an end to extra-judicial
coercion of all sorts - could so alter the incentive structure of the Soviet
peoples so as to release their protective energies. We may confidently
predict the imminent ruin of the economic reforms of perestroika, if only
because no such rule of law has yet been instituted. If anything, the opposite
policy has prevailed, with cooperatives being subject to arbitrary regulation
and confiscatory taxation. The lesson of the Public Choice perspective,
then, is that market reform will fail, when the legal infrastructure of the
market economy is weak or absent. This is a lesson that applies not only to
post-totalitarian societies, but also to authoritarian dictatorships, in Africa
or Latin America, for example, where local kleptocracies are unrestrained
by law. In both cases, economic reform has no chance of success unless it
is preceded by a reform of the law. This is to say that. if (as all historical
evidence and theorizing suggests) a modem society cannot reproduce itself
without the institutions of a civil society, including a market economy,
those institutions themselves presuppose, as their ultimate guarantor, the
rule of law.
The implications of this result for the Soviet Union are radical and
pessimistic. A civil society will come to exist in the USSR, only if the
Communist party is denied control over the jUdiciary and the law enforce-
ment agencies. It will prevail, only if the organs of state security, above all
the KGB, are brought within the compass oflaw, at least to the degree that
their Western counterparts are subject to such control. Nothing at present
supports the hope that this is about to occur. Indeed, the Soviet authorities
can with some truth argue that in parts of the Soviet Union, such as the
Caucasus, only the unfettered discretionary authority of the military and
security forces keeps at bay the Hobbesian spectre of anarchy. In such
circumstances, the implementation of a rule of law is an almost impossibly
daunting task.
Within the Soviet Union, the prospects for civil society vary enor-
mously. In the Caucasus, they are slight, with the best prospect being that
of a Hobbesian peace maintained by military force. (I do not mean to deny
that ethnic conflicts may have been exploited by the Communist authorities
for their own purposes of 'divide and rule', but only to affirm that such
conflicts are real enough and in some instances amount to intractable
obstacles to a stable civil society.) In the Baltic states, the prospects are
much brighter, provided only that political and economic independence of
Moscow can be made a reality. It is as to Russia, and analogously Georgia
and the Ukraine, that it is hardest to wager any conjectures. We do not know
210 Critiques
if the Soviet state in its present fonn will withstand the pressures for
disintegration. The official Leninist policy of establishing a genuine con-
federation among the existing component states of the USSR seems bound
to flounder, even if it is seriously meant, given the popularity and radical-
ism of the various independence movements. There may well be a measure
of disinformation, moreover, in the more alannist reports that the Soviet
Union is on the brink of disintegration and even of nuclear civil war. What
is clear enough is that the present situation in the Soviet Union is des-
perately unstable. Further, as the anonymous Z has well written,9 Western
bien-pensant opinion goes badly astray in supposing that Western policy
can (via further massive transfusion of credit, for example) stabilize the
situation. In all probability, the Soviet Union a decade or so hence will
resemble Yugoslavia now - a state on the edge of disintegration, with
pockets of relative affluence and peace and vast areas of desolation and
ethnic conflict. The chief difference from Yugoslavia, and so the principal
limitation of the analogy, is that on present trends (which exhibit no real
reductions in defence expenditures) the Russian component of the USSR
will remain a m..ssively armed military superpower. This is the scenario
that, on current evidence, seems the most likely.
The result of the inquiry so far is that, except in a few limiting cases, the
transition to civil society cannot be negotiated via the medium of demo-
cratic institutions on the Western model. What then is to be done? We
cannot be content with a model that offers no hope of release from tyranny
for the peoples of the post-Communist world. We can find reason for a
measure of hope, I shall argue, if we abandon the model of contemporary
Western mass democracy, and return to earlier conceptions of civil society,
the rule of law and limited government. This requires of us an exercise in
self-criticism which the Western academic nomenklatura shows few signs
of engaging in. For the most part, Western liberal academia has responded
to the collapse of Communist totalitarianism by proposing market refonns
that stop well short of the necessary precondition of a viable civil society,
namely, private property in most productive assets. The commonest pre-
scription is market socialism of one sort or another. The unrealism of this
conception, its instability and its utopian character. have been so often
demonstrated lO that it would be tedious to rehearse here in detail the
arguments which support this result. 11 In brief, however, everything sug-
gests that. if market socialism could be instituted at all. it would exhibit (as
it has in Yugoslavia) massive malinvestment, sluggish growth and techno-
logical backwardness, structural unemployment and probably hyper-
inflation. Even this is to give market socialism the benefit of a doubt it does
not deserve, since its adoption on a large scale would entail sacrificing
massive economies of scale. and resisting or reversing the most powerful
 Post-totalitarianism 211
economic trend of our age - that is, it would entail the destruction in the
West of the transnational corporation. In the post-Communist world, a
move towards market socialism could only entrench syndicalist power and
delay the integration of the post-Communist economies into the global
economy. For these reasons, market socialism is a non-starter, whose only
real func- tion is to serve as a fig-leaf for the naked reality of the bankruptcy
of socialism in our age.
Market socialism is not among the viable options for the post-
Communist societies. For them, as for us, the real options span a range from
libertarian capitalism to egalitarian social democracy, all of which pre-
suppose acceptance of the core institutions of market capitalism. For the
post-Communist societies, at least for a generation, the libertarian capitalist
model is likely to be most salient and the Swedish egalitarian model
completely irrelevant. For Sweden has maintained its egalitarian welfare
state, only because it has the patrimony of over a century of capitalist
wealth accumulation (assisted by neutrality in two world wars). None of the
impoverished economies of the post-Communist world (with the exception
of Eastern Germany, and the possible and partial exception of Bohemia)
has analogous wealth at its disposal, or wiJ] have such wealth for decades
to come. Egalitarian social democracy simply is not a viable option for most
of the post-Communist societies.
This is not to say that the best prospect for the post-Communist societies
is in merely replicating Western capitalist institutions in their present
forms. Although acceptance of the core institutions of market capitalism is
an inescapable necessity for all of the post-Communist societies, the forms
these presently assume in the West are an historic accident, contingencies
in the development of these institutions rather than essential, constitutive
features of them. The different post-Communist societies may reconstruct
the core institution of private property in a diversity of ways, each appro-
priate to their environmental, cultural and economic circumstances.
Further, the degree to which they operate a mixed economy will vary
significantly - though they are likely, if they are prudent, to opt for a state
or public sector smaller than that in most Western countries, including the
United States.
For the post-Communist societies, then, their future does not lie in
slavishly replicating the mass democracies of the West. These are flawed
models in several respects. In the first place, despite recent exercises in
privatization, all of the Western countries are burdened with an over-
extended and omnicompetent state, with political life being dominated by
competition for the resources that the state commands. This has created in
Western States what I have elsewhere called the new Hobbesian dilemma l2
- which is that, if only as a means of preserving their own assets, citizens
212 Critiques
are constrained to organize collusively so as to capture governmental
power, to extract economic privileges from government and thereby to
pre-empt resources. In its worst, this new dilemma generates a political war
ofall against all, with the power of the state being the chief weapon in that
war. One consequence of this dilemma is that, except where policy has been
made by institutions partially insulated from democratic political com-
petition - institutions such as the Bundesbank in Germany, the Federal
Reserve Bank in America, and the Mm in Japan - economic life has been
dislocated by governments aiming to achieve an artificial alignment of the
economic and the electoral cycles. Even in the Western states, then, civil
society has been weakened and the economy substantially politicized,
yielding a mild version of totalitarian democracy.
Civil society has been weakened in another way in the Western demo-
cracies. Policies of affirmative action and positive discrimination have
generated a new class of group rights, politically created and allocated,
which is dubiously compatible with the underlying institutions of civil
society. Especially in the United States, where policies of affirmative action
have been implemented to the point of absurdity, voluntary association has
been restricted and the rule of law compromised by policies which confer
unequal privileges on favoured minorities within the population. The infla-
tion of the discourse of fundamental rights has, in effect, politicized judicial
interpretation of the law and substantially undermined the independence of
the jUdiciary. As a side-effect, it has rendered intractable issues (such as
abortion) which in political cultures less imbued with legalism have been
resolved by legislative compromise. In short, the revisionist liberalism
which dominates the mass democracies of the West - the new liberalism,13
emanating from some of the works of J.S. Mill and theorized by con-
temporary writers such as Ackerman and Dworkin, which has equality and
general welfare rather than individual liberty as its central constitutive
values - has produced weak and inordinate governments, whose constant
and arbitrary interventions have seriously weakened the autonomous insti-
tutions of civil society.
The post-Communist states will rapidly reach a dead end if they seek
merely to copy the institutions and practices of the Western democracies,
since their economic circumstances make unlimited democracy and its
excesses far more costly to them. They will be well-advised if, on the
whole, they look back to earlier traditions and conceptions of civil society
and limited government rather than at the practice of contemporary
Western states. Here I refer to the conceptions adumbrated in the writings
of the classical economists of the Scottish School, for example, in the
French classical liberal school of Torqueville and Constant, and in con-
temporary thinkers such as F.A. Hayek and James Buchanan. 14 The insight
 Post-totalitarianism 213
contained in all of these writers is that democracy must be limited if civil
society is to be preserved, and liberty and prosperity ensured.
What does this fundamental insight imply for the institutions and
policies of the post-Communist states? In part, no doubt, it simply endorses
the policy of privatizing state assets now under way in many of the East
European states, since such privatization, if pursued consistently, will bring
about a transfer of initiative from government to society that is virtually
irreversible. For this to be so, however, there needs to be constructed a body
of property and contract law in which ownership and the terms of voluntary
exchange are clearly defined. Contrary to Hayekian theorists of spon-
taneous order and institutional Darwinism, IS such law will not evolve of
itself in the post-Communist societies; it must be deliberately constructed.
Further, it must be interpreted and enforced by a judiciary that is genuinely
and thoroughly independent of any government or party. It is the indepen-
dence of the judiciary which guarantees the rule of law, far more than the
niceties of any constitution. Achieving such independence, and guaran-
teeing it for the future, should be the first priority of the post-Communist
states, with the framing of democratic-institutions being regarded as a task
which waits upon the successful achievement of a real rule of law. Nor
should the post-Communist states follow Western models in according
democracy unlimited competence. They will be wise if they follow the
example of those Western democracies that remove certain areas of policy
- above all, monetary policy - from democratic politics. Setting up an
independent central bank, preferably (as in New Zealand) constrained by
rules, is a measure that would bring immense benefits to all the post-
Communist states, since (as the Polish experience has shown) monetary
reform is a necessary prelude to any successful market reform of the
inheritance of central planning. In general, the post-Communist states
should aim to build institutions that constrain democracy rather than to
exalt it.
The institutions whereby the post-Communist states constrain demo-
cracy and shelter civil society will inevitably vary. In at least one, Romania,
the restoration of monarchy may be the only way in which legitimacy can
be regained by government. In many others, only authoritarian govern-
ments are likely to be able to reconstruct the institutions of civil life. Within
the Soviet Union, in particular, there will surely be instances in which,
paradoxically, the task of building up the rule of law will be one that history
gives to a post-Communist dictatorship. In all of the post-Communist
states, priority should be given to re-inventing civil society and limiting
govern- ment by the rule of law, and the temptation resisted to ape the
declining mass democracies of the West.
214 Critiques
CONCLUSION
The upshot of our inquiry has been that current Western practice and theory
is, in general, a defective model for the post-Communist states. An older
tradition of civil association l6 and limited government is better suited to
their needs and circumstances, which stresses the rule of law as the most
vital precondition of civil society. Western opinion has missed the mark in
focusing on democracy as the litmus-test of genuine post-totalitarianism.
Democratic political life can (as in Romania, Bulgaria and probably the
Soviet Union) be manipulated or faked in the interests of the ruling nomen-
klatura caste. Perhaps democracy could have prevented some of the most
catastrophic episodes in recent Soviet policy, such as the destruction of
nature (and the consequent massive ruination of human health) in the area
around Lake Aral. Even there, it is the devolution of power, rather than its
democratic governance, that is most essential. Once again, democracy on
the Western model is likely to prove at once impracticable and even
defective as an ideal for many of the emergent post-Communist regimes.
All this suggests the irrelevance to the post-Communist countries of that
now-dominant Western tradition of republican-democratic thinking
typified in our time by the thoughts of Hannah Arendt.17 This is an insight,
grasped rarely in Western circles, that even some Communist theorists
(such as Lagowski in PolandI.) have absorbed. On the positive side, our
inquiry has suggested that it is an older tradition of conservative and
classical liberal thought that has most to offer the citizens of the
post-Communist regimes. It is a central feature of this older liberal tradition
that, by contrast with the revisionist liberalism of Rawls,19 for example, it
is personal and economic liberties that are prioritized, not political liberties.
This is only to put in other terms the classical liberal insight that individual
liberty and democracy need not go together, and often do not. This is a
lesson that most of the emergent post-Communist states have yet to learn.
Among contemporary thinkers, it may wen be that the leaders of the
post-Communist states have most to learn from James Buchanan's thought.
To Hayek must go the credit for predicting, and for explaining, the collapse
of the socialist economies. Nothing in Hayek's thought gives guidance,
however, to those seeking to walk the long road back from socialist serf-
dom. It is in the metaphor of a new constitutional contract, powerfully
explored in Buchanan's writings,20 that the reality of the post-Communist
regimes is best captured. For the task of those emerging from the darkness
of totalitarianism is to forge a new social contract among themselves, in
virtue of which both totalitarian enslavement and the servile state of the
contemporary Western democracies are transcended. In forging this new
conflict, the leaders and peoples of the post-Communist states must avoid
 Post-totalitarianism 215
both the lawless legality of the Communist past and the politicized legalism
of the Western present. With all of its varieties, post-totalitarianism will be
a reality that is stable and irreversible, only when the autonomy of its
contrary, civil society, is defined and protected by the rule oflaw.
August, 1991
15 Political power, social theory
and essential contestability

The starting-point of my enquiry is an observation. In recent years social

theorists have been much occupied in controversy about power. Theorists
have argued as to whether power is to be attributed to social actors or to
social structures, about how the possession of power may be distinguished
from a disposition to exercise power on specified occasions, and so on. I
begin my enquiry with the commonplace observation that these con-
troversies in the theory of power express divergencies in the philosophy of
the social studies. Often, these controversies are interpreted as expressing
divergent views about the merits of positivism as a stance in the philosophy
of social science. Differences between positivists and the critics of posi-
tivism are posed differently in different traditions, but here 'positivism'
seems to stand for the claim that in its methods and goals the study of
society ought to replicate a received model of the study of nature. The
practice of natural science is understood to be insulated from the value-
commitments and from the philosophical bias of its practitioners and (at
least in its normal periods) to display agreement on rules of method and
convergence on a common body of theory. For positivists, there can be no
reason why the dispassionate study of society should not in due course yield
a like body of common theory, or why such theory should not serve as a
common resource for partisans of opposed policies and competing
interests. According to positivism's critics, the subject-matter of the social
studies disqualifies any natural-scientific model for their investigation. The
observational basis of the social studies (they insist) is not neutrally access-
ible to rival theoretical frameworks, it is captured differently by each of
them. Social theories in tum are not dissociable from the moral and political
commitments of their exponents. Often, critics of positivism maintain that
the theory-dependency of social facts along with the value-dependency of
social theories introduces a dimension of relativity or contestability into
social thought. It is in virtue of this character of relativity that social theory
 Essential contestability 217
can never replicate natural science. Social theory remains an area of intract-
able and inherent controversy for these critics of positivism, primarily
because it is unavoidably infonned by value-commibnents between which
reason cannot arbitrate.
This common story about controversy in the philosophy of social
science may be illustrated by reference to controversy in the theory of
power. Here the writings of Steven Lukes and W. E Connolly! are parti-
cularly illuminating for, despite several differences of emphasis in their
writings, they share three beliefs which go far to support the conventional
view of controversy in this area as expressive of debate for and against
positivism. First, they hold that the concept of power, along with many
other concepts in social theory, is essentially contestable. This is to say that
its proper range of uses is inherently a matter of irresolvable dispute. For
Lukes and Connolly, the essential contestability of the concept of power
derives from the fact that rival applications of it embody conflicting value-
commitments. So, inasmuch as it cannot help invoking some notion of
power, social theory (in Lukes's and Connolly's account of it) cannot avoid
being a normative exercise. Second, Lukes and Connolly agree that the
point of locating power in society is to fix responsibility. They insist that
responsibility, like power, is an essentially contested concept, and they
acknowledge that fixing responsibility involves making a moral appraisal.
Further, they wish to contrast exercises of power with the structural deter-
mination of social happenings: they assert that, though opportunities for
action are structurally constrained, the subject-matter of the theory of
power is human agency rather than the workings of social or historical
necessity. Power is an interstitial phenomenon, arising in gaps in the
structural determination of social events. Understanding the exercise of
power inescapably involves the application of moral notions connected
with responsibility and with the nature of persons. Third, Lukes and
Connolly argue for the dependency of rival views of power on competing
conceptions of human interests. These are normative notions inasmuch as
they have ratifying and validating as well as descriptive and explanatory
uses, but Lukes and Connolly insist that judgements about interests remain
none the less partly empirical in content. They are to be spelled out in terms
of the preferences that are displayed by autonomous agents who have a
clear apprehension of the relevant alternatives open to them. For Lukes and
Connolly, then, power is an essentially contestable concept depending on
opposed conceptions of human interests, where it is the preferences of an
autonomous man or woman that finally determine his or her real interests.
At the meta-theoretical level, the central thesis advanced by Lukes and
Connolly is that, in virtue of the essential contestability of its constitutive
concepts, any kind of social theory is a form of moral and political practice.
218 Critiques
I shall have much to say in criticism of the nuts and bolts of their accounts,
but at this point I want to observe only that there seem to be two entirely
distinct theses in their writings about power. One is the thesis that the
theory of power, like the rest of social theory, is beset by an ineradicable
conceptual relativism such that no account of power can compel rational
assent. The other is the thesis that social theorizing cannot avoid being a
normative activity, that theory-making about the social world is informed
by the theorist's values, so that making such theory is (whether self-
consciously or not) an aspect of its practitioner's moral and political life.
My comment here is that, whereas Lukes and Connolly cite the latter thesis
in support of the former, the supportive connection between the two
remains unspecified. Nothing that Lukes and Connolly argue about power
goes any distance towards linking the theory-dependence of social facts
with the value-dependence of social theories. The upshot of my argument
at this meta-theoretical level will be that relativism or scepticism in social
theory may have sources other than the claim that social theories are
value-laden. One reason that arguments about the value-neutrality of social
theories have been run together promiscuously with arguments about the
theory-neutrality of social facts is to be found in the recent vogue for a
jargon of essential contestability. I shall argue that, whereas claims about
essential contestability may once have had utility in a campaign against
certain obsolete positivisms in social science, the idiom of essential con-
testability constitutes an impediment to further advance in social theory.
Talk of essential contestability conflates a range of distinct problems and
insights in social theory whose careful disaggregation is a condition of
progress in understanding its prospects and limitations.
My programmatic conclusion will be that there are many reasons for
scepticism and humility in social theory, but none of them imply or pre-
suppose that theorizing about society is necessarily informed (or corrupted)
by the theorist's values. Before I can reach this conclusion, I must consider
in detail the account given by Lukes and Connolly of the relations between
power, human interests, and autonomy and their claim that judgements
about the exercise of power embrace judgements about responsibility. I
shall argue that there are not the logical links between the concepts of
power, interests, and autonomy postulated by these writers and that they are
mistaken in their belief that discourse about the distribution of power in
society need be connected with the moral assessment of responsibility. A
further result of my examination will be the claim that in both Luke's
account and that of Connolly, liberal and socialist positions in political
theory and sceptical and realist positions in moral theory coexist and
compete with each other. The account of power offered by these writers is
 Essential contestability 219
a theory divided against itself at more than one point, and I shall in
conclusion suggest how this self-division may best be resolved.

POWER, AUTONOMY AND INTERESTS

According to Lukes, the common core of all conceptions of power is found
in the notion of one agent affecting another. Further, when the notion of
power is invoked to capture the import of a social exchange, it always
comprehends the idea of significant, non-trivial affecting. Using John
Rawls's distinction 2 between a generic concept and the many specific
conceptions a concept may spawn, Lukes goes on to distinguish three views
of power as
alternative interpretations and applications of one and the same under-
lying concept of power, according to which A exercises power over B
when A affects B in a manner contrary to B' s interests. 3
Each of these three views of power emerges from and operates within a
definite moral and political perspective and carries with it a definite con-
ception of human interests. The first, 'one-dimensional' or 'pluralist' view,
defended by such writers as Dahl, Polsby, Wolfinger, and Merelman, for
whom power is exercised in observable conflict over policy issues, pre-
supposes a 'liberal' understanding of interests in terms of policy prefer-
ences expressed in political participation. The second, 'two-dimensional'
view, expounded by Bachrach and Baratz, holds that power is exercised,
not only when there is victory in a conflict over policies, but also when
potential issues are organized out of the political arena, and this second
view embodies a 'reformist' conception of interests which allows that they
may be submerged or concealed, distorted or deflected from full expression
by a series of non-decisions. According to the 'three-dimensional' view
which (like Connolly) Lukes himself defends, each of the conceptions of
interest presupposed by the first two views is defective. Power may be
exercised over men and women even when their wants have been fully
satisfied, and, in Lukes's 'radical' conception of power and interests,
people's real interests are not in what they happen to want, but in whatever
it is that they would want if they were to become fully autonomous agents.
As far as Lukes and Connolly are concerned, a power relation exists
whenever one agent affects another significantly and does so detrimentally
to the latter's interests. Before I try to probe the links they postulate
between power, interests, and autonomy, it may be worth making a few
preliminary observations about the conceptual analysis of power contained
in Lukes's and Connolly'S writings. To start with, it is not clear to me why
220 Critiques
significant affecting should be placed at the core of the concept of power.
Significant affecting is a feature of many sorts of social interaction - it
characterizes love and trade, among many other involvements - and it is not
plausibly represented as especially or distinctively characteristic of power
relations. Again, it is wholly unclear why the exercise of power should
always involve a significant impact on human interests, or, most crucially,
why power relations are defined as social relations involving conflicts of
interest. It seems clear - and this is a point to which I will return - that
power relations may involve a conflict of goals or of preferences without
thereby turning on a conflict of interests. As a corollary of this last point, I
cannot see why the attribute of power should be restricted to social inter-
actions where one agent affects detrimentally the interests of another. No
such restriction is found in ordinary thought and language, and imposing
one by stipUlation has counter-intuitive results.
The conceptual analysis of power advanced by Lukes and Connolly is
unduly restrictive in focusing attention on the significant affecting of
interests and in insisting that power has to do, only or centrally, with
conflicts of interest. It introduces into the concept of power the requirement
that the exercise of power occurs always in a circumstance of (manifest of
latent) social conflict. This is to sanction a conflictual conception of power
of the sort repudiated by those theorists (Arendt and Parsons are examples4)
who conceive of it as a collective resource, possessed in greater or lesser
magnitudes by different societies. Further, it writes into the notion of power
a presumption of its maleficence, a presumption commonly made in liberal
and socialist intellectual traditions, but one that is hard to sustain for all
that. The concentration on 'power over' to the exclusion of 'power to'
cannot be justified by the claim that such power is especially salient to
political contexts, since the claim already carries with it the indefensibly
restrictive (and distinctively liberal) conception of political life as some-
how inherently a domain of conflicts.
It is not hard to show that the requirement of any exercise of power that
injures the interests of the agent over whom it is exercised yields counter-
intuitive results. At a microsocial level, I may apply force or coercion to
another individual, not in order to protect or promote any interest of mine,
but with the aim of benefitting him. If I succeed in attaining my aim, the
exchange between us may be neutral, adverse, or favourable to my interests
while it benefits those of the other. At a macrosociallevel, Stalinist policies
in which working-class liberties and current living standards are repressed
have been defended, coherently if not at all plausibly, by invoking the
long-term and on-balance interests of the working class itself. What reason
could there by for denying the character of power to these relationships?
 Essential contestability 221
In citing these examples, I do not seek to rebut the contention that a man
always loses something when power is exercised over him, but only to
remark on its triviality. Presumably, everyone has an interest in acting on
their current wants, even where so acting prejudices their interests taken on
balance. Presumably, also, the interest that is damaged when anyone is
prevented by another from acting on his current wants is the interest in
autonomy. Still, it remains a commonplace that we all have interests other
than the interests we have in our autonomy: no one can plausibly deny that
these other interests might be promoted at the expense of our interest in
autonomy. This is only to utter the truism that the damage done to an
agent's interest in autonomy may well be compensated for by gains to the
agent's other interests and by an improvement in these interests as a whole.
All this begs a natural question about how someone's interest in autonomy
is to be weighted against his other interests when their claims compete, but
nothing important hinges here on this omission. A competition between the
interest in autonomy and a person's other interests can be circumvented
altogether only by the fraudulent expedient of attaching an infinite weight
to the human interest in autonomy (relative to other human interests). I take
it that, whatever the interest in autonomy comprehends, no one could
sensibly allow it to overwhelm or engross all other human interests. This
point is not met by the response that, when someone's interest in their
autonomy conflicts with other interests, it is up to that person to trade them
off against each other. Even making the heroic assumption that the person
decides always to promote his or her interest in autonomy against all other
interests, the possibility remains that weakness of will or poor calculation
on his part will thwart this policy. This possibility opens up the prospect of
an autonomy-maximizing paternalism in which an agent's autonomy in one
area is restricted so as to promote the agent's chances of autonomy on
balance. In the case I am considering, the hypothesis is that this paternalist
intervention promotes the agent's interests. Whether or not it does so, I
cannot see why an intervention of this sort should be refused the character
of an exercise of power.
So far in my discussion, the term 'autonomy' has been used almost as a
cipher. Its content has gone unspecified. We may take it for granted that its
provenance is in Kantian ideas about moral personality, though as my
argument proceeds it will become clear that, in Lukes's and Connolly's use
of the notion, matters are not so simple. At this point I want to observe that,
however the ideal of autonomy is in the end filled out, it [Link] any
account of it be a complex achievement, having a variety of elements and
capable of being realized in differing degrees. If this is so, and (as both
Lukes and Connolly are at pains to insist) autonomy should be regarded as
222 Critiques
a variable magnitude rather than a threshold phenomenon, then it should be
obvious that autonomy may be promoted (and even maximized) by an
intervention that restricts it. A policy of restricting an agent's autonomy
might, then, be licensed solely by reference to the value of autonomy. It
might be so justified if, as in the case of autonomy-regarding paternalism to
which I alluded earlier, the autonomy promoted is that of the agent whose
autonomy is restricted. Alternatively, it might be justified in this way if an
intervention reducing someone's autonomy on balance at the same time
increases the autonomy of all the others affected by it and maximizes the
value of autonomy in the circumstances under consideration. There are,
admittedly, questions that go unanswered in the sketch I have offered of
such a policy and its justification. We do not know how weighty is the
interest each man has in his own autonomy: is it weighty enough to support
a side-constraint against policies which promote autonomy (now or in the
long term) at the expense of his (current of lifetime) prospects of auto-
nomy? Nor do we know how, when once it is acknowledged that autonomy
is a complex achievement having various elements, we are to weight these
elements when a policy having regard only to the value of autonomy
promotes its constitutive elements in differing degrees: we have no guid-
ance as to how we are to make on-balance judgements even of one agent's
autonomy. But neither of these questions is answered in Lukes's or
Connolly's accounts, and I do not see that they impeach the drift of my
argument at all.
The observations I have made so far have all been about the conceptual
analysis of power held in common by Lukes and Connolly. I have ques-
tioned its special connection with significant affecting and argued that the
general link with interests is undefended. More controversially, I have
maintained that the requirement that exercises of power be always in
circumstances of social conflict. latent or manifest. disqualifies conceptions
of power as a collective resource. One general point illustrated by these
observations is that Lukes's and Connolly's account ofthe generic concept
of power is itself non-neutral in its implications for debate about the
appropriate conception of power. It establishes boundaries for any bona
fide conception of power, and such boundaries are certain to be highly
controversial. It is hard to see how matters could be otherwise, if power is
indeed an essentially contestable concept. There is no more reason to
expect agreement on the formal features or conceptual analysis of power
than there is to expect it on the merits of its various conceptions. The fact
that neither Lukes nor Connolly has anywhere satisfactorily defended their
conceptual analysis has led one commentator to observe that Lukes's
argument makes sense only if it is seen as reposing on a real definition of
 Essential contestability 223
power.s Given the conventionalist and relativist aspects of essential con-
testability theses, this is an ironical and paradoxical result.
A crucial claim made in both the accounts I am considering is that
judgements about an agent's real interests, and so about the agent's power
situation, retain an empirical dimension inasmuch as they are spelled out in
terms of the preferences of an autonomous agent. This is a crucial claim, for
on it depends the belief that the view of power it supports is a part of
empirical social theory and not merely an ideological assertion. If a power
relation always involves a conflict of interests, and if these interests may be
latent rather than manifest. then whether a judgement about a power rela-
tion is true will depend wholly on claims about the (hypothetical and
counterfactual) preferences of an autonomous agent. Everything then turns
on whether the conception of autonomy at work here can be satisfactorily
elucidated, defended, and given operational value.
An initial uncertainty concerns the nature of the link postulated in Lukes
and Connolly between the preferences or choices of an autonomous man
and his real interests. It is obvious that an agent's preferences when the
agent is (relatively highly) autonomous cannot be identified with his or her
interests for there can be no presumption that an autonomous agent will
have prudence as his or her dominant motive: such an agent may, after all,
choose autonomously to sacrifice his or her interests to some ideal he or she
cherishes. The link must be between an agent's real interests and the agent's
autonomous judgement of these interests, not between his or her real
interests and whatever it is that he or she prefers when they are autonomous.
This has the important implication that there may be occasions for conflict
among highly autonomous men and women, even if (implausibly) the
interests of such agents cannot conflict. There seems no reason why the
social conflicts engendered among autonomous men and women whose
interests harmonize but whose goals or ideals do not, should not express
themselves in power relations.
A deeper uncertainty concerns the nature of the link postulated in Lukes's
and Connolly's accounts between interests and autonomous choice. What an
agent chooses autonomously as to his or her own welfare might merely be
useful evidence as to where the agent's interests lie, or else it might serve as
the criterion of these interests: in the latter case, an agent's interests are
necessarily constituted by his or her autonomous preferences about his or her
own welfare. In general, though in both writers there are occasional equi-
vocations, Lukes and Connolly go for the view that autonomous choice is the
criterion of interests, not just good evidence about them.
But, if the necessary link is not between autonomous choice or pre-
ference and real interests, but between what an autonomous agent takes his
224 Critiques
or her interests to be and what they are, it begins to look again as if human
interests constitute a subject-matter independent of anyone's choices or
preferences. On the one hand, if the link between autonomy and interests is
criterial, then anyone's judgement of their interests will be ideally in-
corrigible, while on the other hand there must be room for the notion of a
mistake being possible in this kind of judgement if indeed it is a judgement
with an independent subject-matter. How is this difficulty to be resolved?
It might be suggested that this is not a deep problem. The distinction needed
might simply be that between a person's preferences for their own welfare
and those they have for the welfare of others: their interests are constituted
by the former. Whereas some such commonsensical distinction is in place
and will be presupposed by much of what I myself later argue, it does not
seem to me to meet the difficulty I have in mind. This is that there seem to
be two views of the rellltions between autonomy and interests at issue here.
One view is that an agent's preferences about his or her own prospects just
are his or her interests: an agent's only reason for action is that he or she has
goals, and the agent's interests are those of the goals or preferences that
concern the agent. The other view is that an agent's interests are a matter of
fact for the agent: they give the agent reasons for action whether or not he
or she knows about or assents to them.
At this point in their theory of interests Lukes and Connolly confront an
unavoidable parting of the ways between subjectivist and objectivist
accounts. In one perspective, argued for by Brian Barry,6 the concept of
having an interest may be analysed wholly in terms of opportunities for the
satisfaction of wants. In this view, the only relevant authority to which
appeal may be made to settle questions about an agent's interests is the
judgement of the agent. To be sure, inasmuch as legal or financial expertise
enters into anyone's deliberations about their interests, others may advise
them as to where their interests lie and even, in special circumstances, act
as guardians of their interests. Except in cases of this latter sort, where
someone's judgement is rendered corrigible by their lack of special knowl-
edge, the fact that anyone's wants are their own entails that their own
assessments of their interests, spelled out in terms of their basic preferences
for their own welfare, are indefeasible. Apart from cases where technical
expertise or suchlike is involved, this view of interests implies that, when-
ever anyone's basic preferences about their own welfare change, their
interests change too: an agent's basic preferences cannot be the subject of
a criticism in terms of his or her interests. It is true enough that an agent's
current wants and interests are never the same thing even on a subjectivist
view; an agent's interests even on that view of them must encompass all the
wants he or she is reasonably likely to conceive in a lifetime. Indeed, if (as
I have assumed) it is an agent's preferences and not the agent's mental
 Essential contestability 225
states which go to make up his or her interests, then such interests may be
protected or damaged even after he or she is dead. In this subjectivist view,
then, an agent's current preferences may be overridden so as to protect his or
her prospects of satisfying all his or her preferences, taken over a lifetime,
and the person's interests may thereby be promoted; but it cannot be the case
that such wants or preferences, when they are taken as a whole, could be
thwarted for the sake of the agent's interests. In an objective view, on the
contrary, the criterial link between wants and interests is severed, and a
person's interests are explicated in terms of some notion of the conditions
appropriate to this person's flourishing. On this objectivist view, a person
may well be in error as to what his or her interests are, for though the agent's
wants remain his or her own, these wants no longer decide his or her interests.
How does this, admittedly rather crude, distinction bear on the argu-
ments of Lukes and Connolly? I think it impales them on a fork. Lukes, at
any rate, cannot endorse the objectivist view of interests without abandon-
ing the value-relativism he has elsewhere avowed7 and without elaborating
upon the theory of moral knowledge and the philosophical anthropology
which links interests with conditions of human flourishing. Again, adopting
an objectivist view of interests creates problems for the claim that judge-
ments about a person's interests are liable to a straightforward empirical
and behavioural test - a claim to which Lukes wishes to hold.· If a sub-
jectivist view is adopted, on the other hand, problems about testability of a
different sort surface. Discussing Crenson's analysis of air pollution in
American cities. Lukes tells us that it is impressive because
there is good reason to expect that, other things being equal, people
would rather not be poisoned (assuming, in particular, that pollution
control does not necessarily mean unemployment) - even where they
may not even articulate this preference.9
On a subjective view, this 'good reason' of which Lukes speaks can only be
an inductive wager grounded in exception-ridden generalizations about
human behaviour. Where other things are not equal, where people must
make trade-offs and the influence on their choices of culture, circumstance
and temperament cannot be discounted, such inductive wagers are fraught
with uncertainty. Where an inductive wager of this kind is invoked counter-
factually to explain a historical development, it is hard to see how it can be
the subject of any sort of test.
Underlying this disjunction between objective and subjective views of
interests lies a divergence in the concept of autonomy itself. On one view,
which has perhaps a genuine Kantian ancestry, autonomy is primarily a
formal, procedural and open concept: it desiderates the general conditions
of any respectworthy choice. Such a choice, if it is to be autonomous, must
226 Critiques
be rational - it must, let us say, satisfy criteria of means-end efficiency,
embody a precept to which the agent can give a relevantly universal assent,
and so on. If autonomy is construed in this way as an open notion, then it
should be evident to us (even if it was not clear to Kant) that it will not
specify any very determinate range of ways of life as being in conformity
with human interests. It will in fact be pretty destitute of prescriptive
content. If autonomy is construed in a more Aristotelian (or Spinozistic)
way as a relatively closed concept, autonomous choice will be compatible
only with a fairly narrowly defined range of ways of life. In this conception,
the relation between autonomous choices and real interests intimated in
Lukes's account will actually have been reversed: the conditions of human
flourishing will now serve as providing criteria for the identification of
autonomous choice itself. It should be clear that, whatever else it might be,
this is an account in which the requirement that identification of real
interests have an 'empirical basis' has been abandoned.
In Connolly's more extended treatment lO of the concept of autonomy, an
attempt is made to resist criticisms of this sort. He believes that:
The idea of real interests is bounded by a set of core ideas we share, if
often imperfectly and to a large extent tacitly, about those characteristics
particularly distinctive of persons. II
This reference to certain defining features of persons acts as a restriction on
the use of the choice criterion. Notwithstanding this restriction, Connolly's
account of real interests remains within the subjectivist view; it differs from
the standard 'liberal' account only by virtue of the stringent conditions it
comprehends for the specification of the privileged choice situation in
which the relevant preferences (which express a man's real interests) are to
be displayed. Specifically, Connolly's account goes beyond the liberal
account, and deviates from the standards of ordinary thought and practice,
in allowing for the possibility of internal constraints on action. Like David
Riesman, Connolly suggests that an agent fails to be autonomous if he or
she is swayed by peer-group pressure or social convention or guided by
imprintings of socialization which he or she has not critically sifted.
Connolly's account fails to meet the criticisms I have advanced and
breaks down in much the same way that Mill's theory of the higher
pleasures does: it underestimates the variety of human nature. This is to say
that reference to the powers of deliberation and capacities for emotional
involvement which make persons what they are will not allow agents'
preferences to be overridden in the name of their interests in a variety of
dilemmas of the sort that concern Connolly. The contented lifelong heroin
addict, and the person who lives and dies in willing slavery, 12 will some-
times be able plausibly to defend their lifestyles as most efficacious in the
 Essential contestability 227
promotion of their interests. Certainly, invoking in an aprioristic fashion the
idea of the person cannot defeat this possibility. There is nothing to support
the presumption that an autonomous agent cannot autonomously and
reasonably act to abridge his future prospects of autonomy.
The variety of human nature creates difficulties for the supposition,
entertained by both Lukes and Connolly,l3 that there may be a power-free
mode of consensual authority resting on autonomous choices and involving
a hannony of real interests. Given the variety of human nature, and the fact
that autonomous choice is a necessary ingredient of some but not of all
ideals of life and is incompatible with some valuable involvements and
ways of living, we have no reason to suppose that the choices of auto-
nomous persons will converge on any single way of life, or even that they
will converge on an extended family of lifestyles, all of whose members
have autonomy as a necessary ingredient. It may be that Lukes and
Connolly, like Mill, imagine that, whereas there is no one way of life which
all autonomous persons will elect to live, yet there is for each autonomous
agent one sort of life that is appropriate to that agent. Here the idea of a real
interest or real will is affirmed but predicated in Romantic (or late-
Scholastic) fashion of the essence of the individual rather than of the
individual's species. Apart from the point that the choice criterion has
fallen away, this construal of real interests abandons the assumption that
they cannot conflict with one another. Once this assumption has been
abandoned or relaxed, however, we have no reason to suppose that a
community of autonomous agents will lack power relations, or even that
these latter will be less prominent that they would be if agents were more
heteronomous. Nor can the idea of a power-free community be saved by
arguing that judgements about interests cannot have a purely subjective
character. 14 It might well be the case that people's judgements about their
interests always appeal to shared standards of a worthwhile life, and it is
sometimes the case that the want-regarding and the ideal-regarding ele-
ments in interest-judgements are inseparable. This norm-dependency of
judgements about interests does not support the possibility of a general will,
if only because of the manifest diversity of moral practices in the societies
with which Connolly is concerned. That there are some common elements
in all judgements about interests, which go some way to make them
publicly defeasible, in no sense supports a Marxian (or classical liberal)
thesis of the universality or non-conflictability ofreal interests. And finally,
as I have already intimated, power relations might emerge from conflicts of
goals, even in a society of autonomous persons whose interests
hannonized.
Two comments emerge from this critical survey of Lukes and Connolly
on power, interests and autonomy. There is in the first place a deep
228 Critiques
obscurity in the concept of autonomy itself. Different conceptions of auto-
nomy and of human interests seem to be embedded in different views of
human nature whose status is only partly empirical. If the choice between
rival views of human nature is not empirically severely constrained, the
possibility arises that autonomy may itself be an essentially contestable
concept. If it is, then so will be the concept of power, but the empirical
usefulness of Lukes's theory of power will be impeached. If autonomy is
not an essentially contestable concept, and rationally consensual judge-
ments about autonomy and interests are feasible, then it is hard (on the other
hand) to see how power can be characterized as essentially contestable. The
possibility arises that the conception of autonomy implicit in Lukes's and
Connolly's accounts may even be incoherent. Both Lukes and Connolly
hold to a view of the self as a social construct: human individuality is not
for them a natural datum but a cultural achievement. Further, Lukes expli-
citly and Connolly tacitly endorse a value-relative stance in which the
possibility of cross-cultural, pan-human standards of flourishing is
excluded. It seems at least plausible that a substantive ideal of autonomy
can be combined with the view of the self as socially constructed only if a
strong conception of human flourishing together with a tough realism in
ethics is incorporated into the theory of autonomy; but this is a move neither
writer seems willing to make.
My second comment is in the nature of a summary of the foregoing
examination. Neither in Lukes's nor in Connolly's accounts are the criterial
links between the concepts of power, autonomy and interests postulated in
their theory satisfactorily upheld. There seems no reason to regard conflicts
of interest as especially salient to power relations, or to treat interests as
analysable into the preferences of autonomous agents for their own welfare.
The natural relation to be expected between autonomy and interests is
rather the reverse: different construalS of autonomous choice are likely to
be embedded in different conceptions of human interests. In part, the
differences between rival conceptions of power will be found to derive
from the different views of human nature by which they are sponsored. It is
most unlikely that any purely empirical test will enable us to select one
from among the available range of such conceptions of human nature. I will
tum to the question of how such conceptions of human nature, which lie at
the root of different conceptions of power, may yet be criticizable in the
third and last section ofthis essay.

POWER, RESPONSIBILITY AND SOCIAL STRUCTURE

Against the one-dimensional view, Lukes insists l3 that power may be
exercised by institutions, collectivities and social forces as well as by
 Essential contestability 229
natural persons. He preserves the link between attributions of power and the
ascription of responsibility by making the claim that power is definitionally
exercised by agents, whose omissions as well as actions may figure as
exercises of power. He goes so far as to claim that structuralist social
theory, in excluding all possibility of ascribing responsibility for social
events, actually eliminates the concept of power from social theory. It is
hard to see how this latter claim can be maintained. A structuralist per-
spective in which the exercise of power loses its connection with agency
and so its interstitial character and instead is identified with the operation
of social structures is a perspective in which the primitive notion of power
singled out by Lukes - that of significant affecting - is preserved. In
Marxist structuralist theory, the requirement that exercises of power occur
in a social situation of conflict of interests is also preserved. What we have
in structuralist social theory is unfairly characterized as the elimination of
the concept of power.
The nub of the three-dimensional conception advocated by Lukes and
shared with Connolly is in the thesis that the exercise of power includes the
determination of preferences by socialization. Note that this is a far stronger
claim than the claim that, because overwhelming power pre-empts any
challenge to it, overt conflict cannot be a precondition of the exercise of
power or of .the existence of power relations. Rather, it is the claim that
power relations may be most pervasive and most efficacious when they are
invisible to their victims. This is a claim easily recognizable in theories of
ideology for which the propagation of beliefs, the maintenance of social
norms and so on may all count as exercises of power. A difficulty here is
that the production of ideology is rarely effective when it is a matter of
conscious fabrication. If experience is any guide, moreover, those who are
recognized as exercising power in ordinary thought and practice are
typically no more autonomous than those over whom it is exercised:
typically, that is to say, the exercisers of power are imbued with values that
they have absorbed from their early social environment and which they
have never submitted to a critical assessment. Since those who exercise
power, no less than those upon whom power is exercised, have their
preferences determined by the culture in which they are immersed and the
institutions by which they are surrounded, how can it be justified to impute
responsibility to the former but not to the latter? I take it that the latter are
excused responsibility for their actions and for their omissions because
manipulation and socialization may be represented as the application of
force to people's minds and as constituting an excusing condition. It should
be obvious, though, that those we ordinarily think of as exercising power
(or as being in the upper levels of a power structure) are invariably also at
the receiving end of a three-dimensional power relationship. So invariably
230 Critiques
is this the case that the analogy of a chain of command, which is irresistibly
suggested by Lukes's insistence that the conditions of responsibility are
coterminous with those of the exercise of power, tends to break down. In
the real social world, what we tend to find are feedback loops rather than
causal chains which begin in a detenninate agent. H this is so, what reason
can there be for insisting that judgements about power always involve
fIXing responsibility? This insistence seems part of a vestigially indivi-
dualist, residually voluntarist aspect of the theory for which no real justifi-
cation has been given.
I have pointed out already that, since the primitive notion of power is
retained in structuralist social theory, it cannot be a ,conceptual truth within
Lukes's argument that power has no place in structuralist thought. Further,
I have observed that Lukes's combination of the claim that power may be
exercised by collectivities and social forces as well as by individuals with
the claim that its exercise always presupposes a responsible agent comes
under great pressure when we recognize the ubiquitousness of heteronomy.
I want now to suggest some positive reasons for dissociating the notions of
power and responsibility.16 Such an association yields highly counter-
intuitive results. Consider the mad dictator, who throws his weight about,
does great damage to others' interests, but is not responsible (morally or in
law) for his actions. Are we to deny that power is being exercised in a case
of this sort? This example illuminates important areas of obscurity in the
accounts of Lukes and Connolly. It is not wholly clear whether the res-
ponsibility always presupposed by the exercise of power is merely causal
or properly moral. Again, are we to accept that power waxes and wanes as
the conditions of responsibility are more or less satisfied?
Above all, treating power as an interstitial phenomenon always mani-
fested by responsible agents imposes an arbitrary tenninus in the regress of
explanation. In a structuralist approach, by contrast, a circumstance of
social change is incompletely understood unless the deliberations of the
agents it includes are incorporated (at a higher level of abstraction) into a
theoretical framework in which are elucidated the mechanisms of ideo-
logical production whereby social systems renew themselves. In this
structuralist account, revolutions occur not because people unaccountably
take advantage of opportunities history offers them, but in virtue of crises
within the mechanisms of ideological reproduction which are in principle
fully intelligible and whose necessity might conceivably be demonstrated.
The voluntarist perspective intimated in Lukes and Connolly has as much
difficulty in accommodating the idea of a systemic social contradiction as
it does that of a power structure. 17
The rigorous detenninism of the structuralist approach may usefully be
contrasted with the uncompromising voluntarism or actionism espoused by
 Essential contestability 231
ethnomethodologists and by writers in the tradition of Schutz. For these
theorists, social structures can never be more than the precipitates or
crystallizations of the actions of individuals. This is a stance that is also
displayed in Popperian writings and in the theories of interactionist socio-
logists for whom agents' identities are themselves superventient upon their
social activities. In this perspective, the terminal explanatory level in any
social theory is provided by the intentions, beliefs, decisions and activities
of individuals. I do not want here to comment on the large issues of
methodological individualism and determinism suggested by the contrast I
have sketched, except to say that the explanatory frameworks yielded by
the two approaches may not easily be subject to comparative critical
assessment. Indeed, the explanatory terminus imposed within Lukes's and
Connolly's account by the requirement that power relations be initiated by
responsible agents seems arbitrary given the fact that power exercisers are
ordinarily themselves the recipients of three-dimensional power.
If these are truly incommensurable frameworks of explanation, the
attempt made explicitly by Lukes to adopt a commonsensical or dialectical
middle way between them approaches the brink of incoherence. For the
structuralist, after all, the realm of agency - of teleological and inter-
subjective modes of understanding and explanation - must finally dis-
appear into a system of objective social coordinates. Any methodology
which retains a decisive explanatory role for people's hopes, regrets, inten-
tions and decisions remains wedded to an obsture, moralistic and
reactionary humanist anthropology. For the actionist, on the other hand, it
amounts to an error of reification, a mystifying and animistic superstition,
to regard social structures as more than the residues of the practical and
intellectual activities of human subjects. It is acknowledged in the actionist
account that social situations may acquire a logic (though not a necessity)
of their own, but this is always accounted for by reference to the unintended
consequences of human actions.
Thus, whereas structuralists seek to subsume discussions about human
agency into an explanatory framework in which it is altogether replaced by
reference to objective social coordinates, actionists try to account for the
constraints operative upon human agents in terms which always make
decisive reference to the acts of agents. Whereas structuralists want to
eliminate discourse about agency from the discursive practices of social
science, actionists aim to collapse social structures into the momentarily
uncontested products of the practice of individual subjects. Whereas the
actionist perspective incorporates a strong conception of the autonomy of
the individual subject, for the structuralists 'autonomy' refers rather to
modes of theoretical discourse or of social practice whose independence
must, in the last analysis, be undermined in a comprehensive social theory.
232 Critiques
We can now see, then, that since judgements about power and structure
are theory-dependent operations, actionists and structuralists will approach
their common subject-matter - what goes on in society - using divergent
paradigms in such a fashion that incompatible explanations (and descrip-
tions) will be produced. This much is indisputable: but are we to go one step
further, and say that each of these divergent perspectives generates its own
corroborating evidence? Hwe adopt this relativist position, then we will be
committed to the view that we have here incommensurable views of man
and society between which a rational choice is impossible. This radically
relativist position depends on the claim that as between such views of man
and society what we have is meaning-variance rather than theory-
incompatibility. Is what we have here, then, an instance of competition
between incommensurable paradigms? And, if so, must we accept that the
choice between them is rationally unconstrained? Are we here at last in an
area of genuine essential contestability?

SOCIAL THEORY, ESSENTIAL CONTESTABILIT¥ AND

COMPETING PARADIGMS OF POWER
Before I try to answer this question, it may be worth looking briefly at the
history of the idiom of essential contestability. The epithet 'essentially
contestable' was coined and first applied to concepts by W. B. Gallie in a
paper presented to the Aristotelian Society in 1956,18 but a closely similar
idiom was adopted by Stuart Hampshire when in his 11wught and Action
(1959) he referred to 'essentially questionable and corrigible concepts'.19
As a crude approximation, it may be hazarded that, however these writers
conceived the sources of a concept's essential contestability, its contest-
ability consisted in its being somehow inherently liable to rival inter-
pretations. Nor could this simply be passed off as a matter of the open
texture of many or most, perhaps all, of our concepts: what distinguished
essential contestability from ordinary open texture was precisely that
rivalry and dispute tended to break out, not at the edges of a consensus of
agreed uses, but in what each side regarded as the heart of the concept's
subject-matter. 20 Naturally, these claims tended to spawn a considerable
interpretive literature, which does not seem to have done much to clear up
their many obscurities. One very obvious area of obscurity is expressed in
the question whether the disputes mentioned by these early writers -
disputes about art, Christianity, democracy and morality, for example - are
really disputes about concepts at all. Might they not be just quarrels about
the uses of words? How, in any case, was an essentially contestable concept
to be distinguished from those that are essentially corrigible? (Might not the
identification of essentially contestable concepts as such take us into areas
 Essential contestability 233
of essential contestability?) Further, if some concepts or terms are
inherently liable to generate intractable dispute while others are merely
ambiguous or confused or just happen contingently to be matters of dispute,
what is it that accounts for this difference?
It is this last question that is particularly germane to my inquiry. It has
already been suggested that there are no very forceful reasons for tracing
the disputability of the terms of social thought back to their uses in promot-
ing rival moral and political commibnents. The result of my analysis of the
Luke~onnolly perspective on power, however, has been that two in-
commensurable perspectives on power are left in the field, each (the volun-
tarist and the structuralist) carrying with it a specific framework of
explanation. Given that it seems extremely implausible that any purely
empirical deliberation might settle the issue between these two per-
spectives, what kind of considerations could be decisive? The situation
seems even worse on further reflection. If these perspectives really are
incommensurable, it seems odd to say that they have any common subject-
matter: perhaps what we have is indeed meaning-variance rather than
competition in the use of a shared vocabulary. Perhaps, it has been sug-
gested,21 an incoherence in the idea of essential contestability reveals itself
at this point. For, apart from the fact that realist discourse about essences
seems at variance with the conventionalist presuppositions and implica-
tions of essential contestability, there seems to be a radical fault in the very
notion of a contest which cannot by its nature be won or lost. If the essential
contestability of a concept or of a term flows inexorably from some aspects
of its. subject-matter, how can useful argument proceed about the use of the
term in contention that is on balance to be adopted?
I do not see any of these obscurities as erecting insuperable obstacles to
criticizing the two perspective on power, or as detracting from what the
writers on power I have discussed call the essential contestability of ideas
and judgements about power. Whether these disputes are semantic or
conceptual in character, they involve a pragmatic competition in which
conflicting demands are made on ordinary thought and practice and on the
disposition of resources for research. Further, the point at which my
analysis of the Luke~onnolly approach was broken off itself suggests
how commitments standing at a terminal level of justification within the
theory of power may yet be criticizable at another level. Each of the
perspectives I have distinguished intimates a philosophical anthropology-
a moral psychology and a philosophy of mind and action - in terms of
which it may be criticized. Social theories invariably repose on some such
philosophical standpoints, and it is this dependency which accounts for one
aspect of their permanent vulnerability to contestation and openness to
questioning.
234 Critiques
Consider in this connection one variant of the structuralist approach to
power - that elaborated in structural-Marxist writings. The central area of
difficulty in all such accounts has been indicated obliquely by Hindess and
Hirst in the comment that
there is as yet no systematically elaborated theory of the political level
to compare with the Marxist analysis of the economic level of the
capitalist mode of production.22
The difficulty hinted at here is that, in developing a systematic account of
ideology and of superstructural phenomena such as the capitalist state, the
structuralist cannot avoid invoking some of the assumptions of a 'humanist'
anthropology which the terms of his account of social structure compel him
or her to reject. These anthropological assumptions, officially suppressed
by structuralist thought but presupposed by some of its main theses, are to
do chiefly with the complex capacities and attitudes necessarily possessed
by role-bearers in post-traditional (and, perhaps, also in traditional)
societies. A worker who enters into a wage contract, for example, must be
capable of understanding a whole set of internally complex background
notions, the possession of which gives him or her the capacity to form
alternative concepts and beliefs about his or her circumstance in society. If
a worker understands his or her situation well enough to discharge the roles
assigned to him or her in a capitalist (or a socialist) society, then the worker
has the capacity to imagine himself or herself situated in a very different
social landscape. 23
It will immediately be objected that it is precisely the function of
ideology so to constrain the political reflection of workers (and others) that
it remains at the level of ideology and false consciousness. To say this,
however, is entirely to miss the point. For it is precisely the human capacity
of critically transcending one's immediate social circumstances that
generates the necessity - which, according to Althusser, will remain a
feature even of Communist society - for ideological beliefs which restrict
the scope of such reflection. What I am claiming is that the myopically
deterministic account of human reflection encapsulated in structural-
Marxist thought disables its advocates from developing any remotely
plausible theory of social consciousness. Nor do I forget the self-criticisms
of Althusser's more recent writings, or the revisions made to structural
Marxism by his disciples Etienne Balibar and Jacques Ranciere. None of
these saves the structuralist-Marxist project from failure. To acknowledge
that human thought and social relations are reflexive, that is to say, that
reflection on the conditions of our thought changes those conditions in that
it modifies the beliefs and attitudes by which they are constituted, imposes
upon us the task of elaborating a philosophical anthropology in which these
 Essential contestability 235
distinctively human powers are explored. The metaphysical reasonings by
which such an anthropology will be supported are bound to have an
inherently controversial character, but this is not to say that criticism is
altogether out of place here. In this instance, it is to note the latent posi-
tivism of those Marxists in the structuralist tradition who are committed to
the self-defeating project of elaborating a rationalist science of society
which is supposed to be able to do without a metaphysic of human nature.
My argument is that structural-Marxist theory about power and social
structure incorporates a philosophical anthropology at odds with official
structuralist commitments. These later commitments, in tum, I suspect,
could be shown to be at odds with the anthropological postulates of our
ordinary thought and practice. At no point is my argument intended to be
conclusive or demonstrative. It is taken for granted that in discussing
questions in the philosophy of mind and action we are in an area of
metaphysical uncertainty where what are at stake are only more or less
well-defended statements of opinion. As Stuart Hampshire has put it:
No critical philosopher can not believe that an inquiry into the concept
of man, and therefore into that which constitutes a good man, is the
search for an immutable essence. He will rather think of any definition
or elucidation of the concept as a reasoned proposal that different types
of appraisal should be distinguished from each other in accordance with
disputable principles derived from a disputable philosophy of mind. He
will admit that this is the domain of philosophical opinion, and not of
demonstration. 24
One major source of intractable dispute in social theory (and in moral
theory) is its connection with areas of metaphysical uncertainty explored in
questions in the philosophy of mind and action. Dispute is interminable in
social theory, in part because it slides irresistibly into dispute in philosophy,
and there we have the defence of opinion and not demonstration. At any
moment in discussion, one stand in social theory may well be better
supported, better defended than others in the field: others (such as the
structural-Marxist view I have criticized) may do badly. This character-
ization of criticism in social theory shows that acknowledging the
inherently controversial aspects of social concepts may be combined with
affirming the criticizability of social theories.
I have not suggested that intractable dispute in social theory is always an
expression of conflicts in philosophy. It has another source in the under-
determination of social theory by evidence. That this is so ought to be clear
from my discussion of the methodological difficulties faced by the Lukes-
Connolly account. The decision procedure proposed for the identification
of power relations is not just counterfactual in method, but doubly so:
236 Critiques
power relations exist if people's interests would conflict under certain
circumstances, and their interests are a subclass of the preferences they
would display in a condition of relative autonomy. This approach commits
Lukes and Connolly to an extraordinarily arduous testing procedure. Apart
from all the difficulties involved in assessing degrees of autonomy where
this is regarded as a complex achievement having distinct and diverse
elements, the method seems to have an aspect of circularity: we test for
power by hypothesizing what an autonomous agent would do, but we know
that an agent is autonomous only if we first know the agent's power
situation. Lukes's suggestion2S that we study circumstances of social break-
down and disorder for evidence of people's latent interests is no help here:
such social conditions, typically conditions where demagogy and crowd
psychology are important factors, seem peculiarly ill-suited for the pur-
pose. The Lukes-Connolly approach seems wanting if its empirical useful-
ness and testability are seen as including its capacity to yield falsifiable
predictions and retrodictions.
These difficulties are plainly only one instance of a greater problem in
any project for a science of society. Many of the evidences going into any
such science will be matters of historical interpretation, and large-scale
social experiments will rarely, if ever, be conclusive, owing to the imposs-
ibility of isolating distorting variables. Most crucially, however, a science of
society comes up against a problem having to do with its public character.
Evidence tends to elude the social-scientific investigator, if only because the
latter's subjects tel'd to react to the categorial framework of interpretation
which the investigations impose on their activities. There is this contrast of
social science with natural science, that its subject-matter is people's notions
and beliefs and these are worked over reflexively, both by the investigator
and by the subjects in response to the investigator. I am far from attempting
to dismiss in aprioristic Winchian fashion any project for a scientific study
of society: at microsociallevel, as Milgram's study of authority and obe-
dience and Laing's studies of the family suggest,26 studies satisfying fairly
rigorous standards may be deeply instructive. But the reflexive relations
between social-scientific investigation, human beliefs and the objects of the
social world should confine the prospects of such study within very narrow
limits. In the natural sciences, as in the social studies, available evidence
may be compatible with a range of theories whose implications are them-
selves incompatible; but the social studies have the extra disability that
crucial experiments are rarely, if ever, practically feasible.
It may be time to try and tie together the loose ends of my inquiry and to
assess the bearing of its results on the prospects of social theory. I began by
making the observation that disputes in the theory of power have been
widely and not unreasonably interpreted as expressive of deeper differences
 Essential contestability 237
over the merits of a positivist philosophy of social science. We can now see
that this interpretation has validity only with important qualifications. If the
ideal of value-freedom in social thought is one test of positivism, our
conclusion must be that nothing in Lukes's or Connolly's writings on
power impeaches this tenet of positivism: so far as we have been able to
discern, this Weberian standard of scientific rigour in the social studies
remains uncompromised. Persisting relativity in social science has two
sources in the underdetermination of theory by evidence and the depen-
dency of theory on metaphysical commitment. These aspects of the limita-
tion of social thought should incline us to a Pyrrhonian humility about our
own social theorizing and encourage us in scepticism about any large
claims for social thought. I cannot see that these insights into the limitation
of social thought warrant description in the inflated idiom of essential
contestability.
The writings of Lukes and Connolly retain great heuristic (and peda-
gogic) interest in disclosing two radically divergent paths of development
for social thought. One, associated with an open conception of autonomy, a
non-cognitivist moral epistemology and a fallibilist and conventionalist
methodology, supports a recognizably Weberian conception of social
theory. It has no direct bearing on conduct, but in its commendation of
impartiality as an intellectual virtue and its frank recognition of the diver-
sity and internal conflicts of human values it may have some affinity with
a tradition of justifying argument in liberalism. The other, bound up with a
closed view of autonomy, a form of moral objectivism and a method of real
definition, is familiar from Marxist projects for a social science. Here the
scientific study of society, because it is supposed to reveal necessary laws
of social development, has a directly practical motive. It is such a concep-
tion which is supported by the constant emphasis in both Lukes and
Connolly on the supposed fraudulence of the distinction between fact and
value in social theory and by their attribution to social theory of an ineradic-
able political dimension. Nothing in recent writing justifies the attribution
to social theory of this political or practical character, and even the thesis of
the essential contestability of social concepts may be impoverished by it.
The current of writings on power and related concepts in which theses
about essential contestability or conceptual relativism are deployed has had
the virtue of strengthening the sceptical spirit in social thought and of
undermining dogmatic and absolutist claims. But, if social thought is not to
be lost in a no man's land of political controversy, its practical character
needs to be denied and its direct links with moral and political life severed.
We need to be able once again to assert with confidence that, however
meagre its result in increased insight, social theory has no warrant for
existence save in the pursuit of understanding.
16 An epitaph for liberalism

If there is a single characteristic that typifies liberal political philosophy in

the United States over the past quarter century, it is its domination by a
jurisprudential paradigm. Complex questions about restraint of liberty -
such questions as the control of pornography, and the termination of life in
abortion and euthanasia - that in other countries, and other traditions of
liberal discourse, are treated as issues in legislative policy, involving a
balance of interests and sometimes a compromise of ideals, have come to
be treated in the United States, primarily or exclusively, as questions of
fundamental rights. The model of reasoning presupposed in this tum to
legalism in recent American theory is that of the judicial interpretation of
constitutional rights rather than of the formulation of public policy in
political discourse. In its subtler articulations, such as the later works of
John Rawls, the juristic model for liberal theorizing has been substantially
qualified by the recognition that it is applicable only to a subclass of
constitutional democracies (perhaps comprising only the United States),
and still cannot decide all vital issues in political controversy about the
legal restraint of liberty. Yet, even the Rawlsian project of specifying a set
of fixed and determinate liberties exemplifies the legalist illusion that
animates recent liberal thought in America.
The legalist tum in recent American liberal thought has had predictably
deleterious consequences - ones that would not have surprised the authors
of the Federalist Papers. In political contexts, it has generated a series of
intractable conflicts, which portend deepening division, growing un-
governability and even a sort of chronic, low-intensity civil war in the
United States. Of these conflicts the abortion question is only the most
obvious, but it serves well enough to illustrate the fateful implication for
political practice of the hegemony of the paradigm of a rights-based juris-
prudence that has been inflated by the inordinate claims of liberal theor-
izing - that, since conflicts of fundamental rights cannot be moderated,
 An epitaph for liberalism 239
their judicial interpretation can mean only unconditional victory or com-
plete surrender for the protagonists to the dispute. This is not a recipe for
civic peace, or for a stable liberal civil society.
In intellectual contexts, the dominance oflegalism within liberal thought
has debauched standards of argument. Taken by themselves, neither the
terms of fundamental rights theory, such as they are, nor the provisions of
the American Constitution give any definite answer to current questions
about abortion, pornography or similar issues. In recent American liberal
theory on the legalist model, the vacuities of rights theory and the in-
determinacy of the Constitution have been remedied by the adoption of a
conventionalist stratagem in which the ephemeral banalities of liberal
academic culture - affirmative action, or the arcana of antisexism - are read
into the constitution, and then elevated to the status of universal verities. It
is probably an inevitable upshot of liberal legalism of this sort that it should
represent the local (or parochial) knowledge of its practitioners - shifting
and hardly very determinate as it is in the work of Ronald Dworkin and
Bruce Ackerman, for example - as a repository of universal truths.
Joel Feinberg's four-volume study The Moral limits of the Criminal
Law· breaks with this legalism in order to revert to an older, saner, if in the
end radically flawed tradition - that of John Stuart Mill. It is not an
extended exercise on the jurisprudential model, but instead a study of the
moral constraints on law-making, addressed to a hypothetical ideal legis-
lator. The fIrSt thing that must be said of Feinberg's work is that it is the
exemplary statement of Millian liberalism as it applies to the legallimita-
tion of liberty: it is comprehensive, systematic, argued with a rigour and
scrupulousness unmatched, let alone surpassed, in any comparable study.
These exemplary standards were set in Feinberg'S fIrSt volume, Harm to
Others. 2 There, he distinguished between a non-normative conception of
harm as setback to interest, and a normative conception of harm as wrong-
ful conduct in which a person's rights are violated; and went on to make a
host of important and useful distinctions in which moral harm, vicarious
harm, posthumous and prenatal harm, and other harm cases were analysed
and their content specified. Feinberg's minute and illuminating taxonomy
of the notions of harm and interests aims to give Mill's famous 'one very
simple principle' - the principle that no one's liberty may be restrained,
except to prevent harm to others - a definite sense sufficient at least for its
use in contexts of legal policy-making, but it is far from clear that it resolves
successfully the chief problems confronted by anyone who seeks to make
practical use of Mill's principle in legislation.
Feinberg interprets the harm principle as forbidding restraint of liberty,
save where there is harm to others' interests, and these latter are interests
that constitute a valid claim to a moral right. In this he is faithful to Mill
240 Critiques
himself, and to Mill's best twentieth-century interpreters, such as J. C.
Rees. No more than Mill or Rees, however, does Feinberg give any clear or
compelling account of the relations between having an interest, making a
valid claim, and possessing a moral right He tells us that any interest may
generate a valid claim, and so a moral right, except vicious and wicked
interests, 'if such there can be'. Aside from the question how we are to
determine the wickedness of interests, it is clear that the world of moral
rights will be a densely populated and quarrelling one, since the multitude
of interests, often making competing claims on action, which each of us
has, will spawn a host of conflicting rights. Whereas Feinberg recognizes
this to be a difficulty in the application of Mill's principle as he interprets
it, he seems to underestimate greatly the extent to which it diminishes the
utility of the harm principle as a guide to legislation. In Mill, the difficulty
is resolved, at least in theory, inasmuch as the deficiencies of the harm
principle as a guide to action are remedied by appeal to the demands of the
Principle of Utility itself. For, in both Mill and Feinberg, the harm principle
specifies only a necessary condition of justified liberty-limitation: it tells
us, not what we ought to do, but what we may do in restraining liberty.
Feinberg's case is, in truth, worse than Mill's, inasmuch as he lacks even
the fall-back position of appealing to the requirements of welfarist conse-
quentialism as a guide to action once the harm principle has been satisfied.
He avoids the illiberal possibilities of Mill's position, in which large and
inequitable restraints on the liberty of minorities could be sanctioned by the
prevention of small but widespread harms to large numbers of people, but
only at the price of leaving us with no way of weighing competing interests
or arbitrating the conflicting rights that they yield. Feinberg's account is not
saddled with the burden of a wholesale maximizing consequentialism, with
all the illiberal implications that creates in the application of the harm
principle. For that very reason, however, and because Feinberg deliberately
refrains from embedding the' harm principle in a more comprehensive
moral theory, his attempt to give a content to the principle in terms of the
redundant jargon of moral rights is thoroughly unpersuasive.
The larger project pursued in volumes two to four is that of defending
the Liberal Position, which is that 'The harm and offence principles, duly
clarified and qualified, between them exhaust the class of good reasons for
criminal prohibitions? The Liberal Position then states an exclusionary
principle, ruling out (as good reasons for legal coercion) such other prin-
ciple.s as legal paternalism, legal moralism, and perfectionism. Like Harm
to Others, Offence to Others argues that there are both normative and
non-normative uses of 'offence', with only the former, wrongful offence
being intended in the offence principle, which lays down that
 An epitaph/or liberalism 241
It is always a good reason in support of a proposed criminal prohibition
that it is necessary to prevent serious offence to persons other than the
actor and would be an effective means to that end if enacted.4
Feinberg develops a powerful argument to the conclusion that offences that
are harmless may nevertheless sometimes be subject to legal prohibition,
making extensive use of the model of nuisance law, but recognizing the
inadequacy of that model for what he calls 'profound offences', in which
the offended person experiences offence, not only or primarily to his own
feelings, but to values that are conceived to be independent of his feelings.
In the course of his analytical survey, Feinberg investigates the role of
consent in relation to offensive subjects, the place of the Volenti standard
('one cannot be wronged by that to which one consents') and the character
of pornography and obscenity. Hann to Self addresses the principle of legal
paternalism - that
It is always a good and relevant (though not necessarily decisive) reason
in support of a criminal prohibition that it will prevent harm (physical,
psychological or economic) to the actor himself.'
Feinberg defends this antipatemalist injunction by reference to the personal
sovereignty and right to autonomy of the individual - considerations at
once impeccably Millian in pedigree and at the same time eminently
questionable. He distinguishes between 'soft' and 'hard' paternalism,
defending the fonner, which mandates coercive state intervention with
dangerous self-regarding behaviour only when that behaviour is sub-
stantially non-voluntary, or when temporary intervention is necessary to
establish whether it is voluntary or not, as 'properly speaking, no kind of
paternalism at all' .
As well as this liberal argument for soft paternalism, Hann to Self
incorporates penetrating analyses of various riddles of voluntariness, such
as reasonable assumption of risk, failures of consent (as in coercive
pressures and offers), defective belief and incapacity. The final volume,
Hannless Wrongdoing, treats issues of legal moralism, moral conservatism,
the relations of autonomy and community, and legal perfectionism (legal
coercion for the promotion of excellence or virtue). Feinberg again tries to
argue for the Liberal Position that harm and offence to others are the only
considerations that are always good reasons for criminalization of
behaviour, while maintaining that the damage to tradition, community or
virtue flowing from adherence to this position has been exaggerated by
critics of liberalism, or is a cost that must be borne for the sake of pre-
serving personal autonomy and sovereignty.
242 Critiques
With admirable candour, Feinberg acknowledges that the theory that
emerged from his investigations is not at all the 'one very simple principle'
which Mill advanced in On Liberty, but 'a much more complicated con-
struction'. He goes so far, indeed as to admit in the conclusion to the final
volume that 'As things now stand, liberalism as strictly defined in this work
is not only not plausible, it is quite clearly not true'.6 At this point, Feinberg
is inclined to fall back on a view of liberalism as simply a presumption in
favour of liberty, from which it follows that 'Liberalism, however it is
further specified, is a matter of degree, depending on how great a surcharge
the liberal would impose on the reasons that can outweigh liberty."
In this last formulation, Feinberg allows to the Liberal Position an
indeterminacy he seeks to attenuate by the assertion that personal sove-
reignty has an absolute weight, trumping any interest that might support
legal coercion with the self-regarding domain. It is this stand on anti-
paternalism, as upon a last redoubt, that qualifies his definitive judgement
that, contrary to the original intent of his study, 'the most plausible
liberalism' allows that
All of the major coercion-Iegitimising principles (the harm and offence
principles, legal paternalism, legal moralism) and subprinciples (e.g.
moral conservatism, strict moralism, the exploitation principle, legal
perfectionism) state reasons that are at least sometimes relevant (to the
justification of restraint of liberty). I would now go so far as to say that
they all state reasons that are always relevant, though usually of very
slight weight. 8
It is evident from these statements that, in the course of writing this superb
series of volumes, Feinberg's intellectual virtues of rigour and honesty have
compelled him to abandon the Liberal Position with which he began. His
final position is worse than that. There is nothing in the final volume against
legal moralism, conservatism and perfectionism, that justifies the infinite or
absolute weight Feinberg assigns to personal sovereignty. After all, on any
view that accepts a diversity of ultimate values, or which accepts that the
human interest in sovereignty or autonomy will be only one interest among
many, personal sovereignty will not always trump all the rest. The only
argument Feinberg offers for the absolutist status of personal sovereignty is
that
without this supplementary feature liberalism will be fatally vulnerable,
that it will condone illiberal claims on ultimately liberal grounds.9
This argument is not only weak because it will appeal only to already
committed liberals or because it is dubiously consistent with Feinberg's
looser definition of liberalism in terms of a generally defeasible
 An epitaph for liberalism 243
presumption in favour of liberty. It weakens his position by suggesting that,
once liberal thought has reached the degree of sophistication it has in his
elaboration of the Millian paradigm, it has a self-defeating effect.
The self-effacement of liberalism that is evident in the closing chapters
of Feinberg's work, and which follows upon the ruin of his original project,
results from the pluralist insight - profoundly developed in the thought of
Isaiah Berlin - that a wide diversity of considerations is relevant to ques-
tions about the restraint of liberty, none of them always trumping the rest.
This pluralist standpoint is strengthened when it as acknowledged that the
promotion of personal autonomy, for example, may depend on the exis-
tence of a public or common culture, which presents to the agent an array
of worthwhile options. The preservation of such a culture may, however,
commit the law to moralist and perfectionist policies of sorts which Fein-
berg is bound to oppose. They will be not just non-coercive policies of
subsidy or education, which are beyond the scope of Feinberg's argument,
but (sometimes, at least) policies of legal prohibition. (Where they can be
made effective, and are not too costly in other terms, policies of criminal-
ization of extremely addictive drugs probably come into this category.)
Again, concern for the autonomy of the agent may well justify measures of
this kind, independently of their contribution to the preservation of
common culture within which worthwhile choices can be made. A pluralist
will accept, with a doctrinal liberal even a la Feinberg must resist, that
promotion of liberal values such as autonomy sometimes encompasses
adoption of policies that are illiberal in that they curb personal sovereignty.
Like any other ingredient in human well-being, autonomy is an interest that
is sometimes best enhanced when it is limited - and which has as its matrix
conditions that constrain it.
The self-effacement of liberalism, manifest in Feinberg's study, is
peculiar to the Millian tradition his book exemplifies. It is apparent in
Joseph Raz's great book The Morality of Freedom,lo in which the salience
to restraint of liberty of paternalist and perfectionist principles, the indis-
pensable contribution to personal autonomy made by a public culture, and
the impossibility of State neutrality in respect of the promotion of virtue
and the positive realization of human capacities are given a demonstrative
statement. In Raz's case, liberalism effaces itself by way of an immanent
critique in which liberal society emerges in the end as only one among a
variety of orders in which human beings may flourish. In both thinkers,
whose works together constitute the high-water mark of recent liberal
theorizing, the project of giving liberalism a universally compelling foun-
dation has been abandoned, and the attempt to formulate definitive and
uniquely determinate liberty-limiting principles given up or severely
qualified. Foundationless and virtually contentless, liberalism has then the
244 Critiques
aspect of the Cheshire Cat, becoming the vanishing spectre of a once-living
doctrine.
It is a tribute to Feinberg's study that, by contrast with those who seek in
the empty slogans of equality or fundamental rights to give liberalism a
content derived solely from the trivia of recent academic debate, he should
have returned liberal theory to a judicious and dispassionate consideration
of substantive issues. If his study should prove to be an epitaph on Millian
liberalism or yet another example of liberalism's self-effacement, it
remains a colossal achievement, definitive and magisterial, a book which
makes a permanent contribution to legal and political philosophy.
17 The end of history - or of
liberalism?

It is a truism that socialism is dead, and an irony that it survives most

robustly as a doctrine not in Paris, where it has suffered a fate worse than
falsification by becoming thoroughly unfashionable, nor in London, where
it has been abandoned by the Labour Party, but in the universities of
capitalist America, as the ideology of the American academic nomen-
klatura. But socialism is most obviously, and most irreversibly, defunct as
an ideology in the Communist bloc. There, glasnost has surpassed the
wildest hopes of Western anti-Communists in discrediting the institutions
of central planning and brilliantly illuminating the intractable problems of
the Soviet system.
But what does the collapse of socialism as a political faith portend for
the future of political life and thought? In a provocative and well-received
article, 'The End of History' ,I Francis Fulruyama announces in a quietly
apocalyptic voice that the failure of socialism means
an unabashed victory of economic and political liberalism [and he
promises] the end point of mankind's ideological evolution and the
universalization of Western liberal democracy as the final form of
human government.
The prophecy that human history is about to end and a new historical epoch
about to begin is of course a recurrent one in the history of Western thought.
It is probably an unintended irony that Fukuyama's article should stand as
a contribution to the project of a secular theodicy first undertaken by the
philosophes of the French Enlightenment, but most notably and ener-
getically pursued in the Marxian system of thought which Fulruyama
correctly perceives to be now in a terminal decline. But it is in any case
difficult to understand the basis for Fulruyama's confidence about the
historic role of liberal democracy in bringing history to a successful close.
His confidence cannot be a reflection of the state of liberal political philo-
sophy, since that is manifestly parlous. In my recent book, Liberalisms:
246 Critiques
Essays in Political Philosophy,2 and particularly in its Postscript, 'After
Liberalism', I have argued that despite its overwhelming dominance in
Anglo-American philosophy, liberalism has never succeeded in showing
that liberal democratic institutions are uniquely necessary to justice and the
human good. In all its varieties - utilitarian, contractarian, or as a theory of
rights - liberal political philosophy has failed to establish its fundamental
thesis: that liberal democracy is the only fonn of human government that
can be sanctioned by reason and morality. It therefore fails to give rational
support to the political religion of the contemporary intelligentsia, which
combines the sentimental cult of humanity with a sectarian passion for
political refonn.
The consequent debacle of liberal political philosophy is not something
we have any reason to lament. For liberals are committed to the heroic
enterprise of denying a very obvious truth - the truth that there is a
legitimate variety of fonns of government under which human beings have
flourished and may still hope to prosper. Who can doubt that human beings
flourished under the feudal institutions of medieval Christendom? Or under
the monarchical government of Elizabethan England? It is in virtue of its
repression of this evident truth that liberal discourse has acquired its
stridency and intolerance - indeed, its almost obsessional character. In
seeking to construct a liberal ideology, liberal theorists are attempting what
even they must sometimes see to be impossible. They are struggling to
confer the imprimatur of universal authority on the local practices they
have inherited. The absurdity of this project has, indeed, been tacitly
recognized by one of this century's subtler liberal thinkers, John Rawls,
when in his later work he revealed that he aims only to give a coherent
philosophical statement of the character and premises of a particular
historical tradition - the (American?) tradition of constitutional democracy.
If Fukuyama' s confident expectation of the End cannot be explained by
the state of liberal philosophy, from what does it derive? It is the expres-
sion, most likely, not of a political philosophy, but of a philosophy of
history, one dominated by the notion that liberal democracy is history'S
telos, other modes of government being recognized only as progressions
toward, or aberrations from, that end.
The grain of truth in this interpretation of history is that it is only through
the development of civil society - a society in which most institutions,
though protected by law, are independent of the state - that a modem
civilization can reproduce itself. Without those institutions - for example,
private property and contractual liberty under the rule of law - modem
societies invariably undergo a descent into poverty and barbarism. Civil
society is the matrix of the market economy, which both history and theory
show to be the precondition of prosperity and liberty in the modem world.
 The end of history - or of liberalism? 247
This is a truth that even the Soviet leadership may now be learning, after
having waged for over seventy years an incessant war on all the civil
societies that have come within its sphere of domination. It is one that the
Iranian fundamentalists are beginning to accept, however grudgingly, as
they retreat from the position that a modem state can be governed exclu-
sively through the precepts of the Islamic sharyah. And it is a truth that will
become painfully clear to the aged Stalinists of Communist China, when
they are forced by circumstances to perceive the economic ruin that flows
from trying to confine an emergent civil society in a newly re-sewn totali-
tarian straitjacket.
To say that no modem state can renew itself with a decent degree of
prosperity unless it contains the institutions of a civil society is, however,
very far from allowing that liberal democracy is 'the final form of human
government'. Civil societies come in many shapes and forms and thrive
under a variety of regimes. The authoritarian civil societies of East Asia -
South Korea, Taiwan, and Singapore - have combined an extraordinary
record of economic success with the protection of most individual liberties
under the rule of law without adopting all the elements of liberal democracy.
Or consider the case of Japan, which Fukuyama's mentor, the Hegelian
scholar Alexandre Kojeve, rightly recognized as the key exception to the
trend of global homogenization. To be sure, Japan has become a con-
sumerist culture, and its political institutions are liberal democratic. Yet the
crucial decades of modernization in Japan occurred in the late nineteenth
and early twentieth centuries; modernization was generated internally and
was not imposed from outside; and, uniquely, the Japanese succeeded in
grafting on to the unbroken stem of a traditional culture the institutions of
a modem civil society. As a result, Japan has in the last two decades
emerged as a global economic superpower, and must willy-nilly (whether
it wishes to do so or not) become a superpower tout court in the coming
century. This has been achieved without any deep commitment to the
constitution imposed upon Japan at the end of the Second World War, and
certainly without the support of the ideas and values that are supposed to
undergird market institutions in the West, such as individualism, univer-
salism, natural rights, Judeo-Christianity, or the idea of progress.
The East Asian examples show that Western achievements can be repro-
duced, and for that matter surpassed, without any acceptance of 'the
Western idea' of which Fukuyama speaks when he refers to 'the triumph of
the West'. The ongoing disintegration of Communism on the Soviet model
gives Fukuyama's argument no better support. The avowed aim ofthe twin
Soviet reform policies of perestroika and glasnost is to break the totali-
tarian mold and reconstitute a civil society. Even if it is successful, how-
ever, the Soviet reform policy is unlikely to result in a triumph of Western
248 Critiques
liberalism. To attempt to foretell the future cost of Gorbachev's reform
policy is idle. Already, however, glasnost has to its credit a considerable
achievement. It has revealed for all time the ruins of the totalitarian project
initiated by Lenin in 1917. This is the project, intimated in Lenin's horrible
saying, 'We must be engineers of souls', of destroying the traditional
identities of the human beings within its power and reconstructing them as
specimens of the new socialist humanity. Prosecuted relentlessly and with-
out mercy for over two generations in an incessant war against religion, the
family and nationality, this totalitarian project has been shown by glasnost
to have been a stupendous failure. As they emerge from the shadows of
totalitarianism, the peoples of the Soviet Union reveal themselves, not as
specimens of socialist (or liberal) humanity, but as Ukrainians or Baits,
Catholics or Muslims, bearing traditional identities in no way compromised
by decades of totalitarian indoctrination. The forms of national and reli-
gious life that are reasserting themselves in the Soviet Union give the lie to
the totalitarian idea (echoed by innumerable Western liberals) that human
beings can be remodelled according to the dictates of rationalist ideology.
If anything, the traditional identities of the peoples of the Soviet Union may
be healthier than those in many Western nations, where subtler forms of
indoctrination have had a more corrosive effect in rendering traditional
forms of life decadent.
It is precisely because the revelations of glasnost give the lie to the
totalitarian project of reshaping human nature that they also confound
Fukuyama's account. If the newly self-assertive peoples of the Soviet bloc
are not specimens of Homo sovieticus, neither do their political beliefs have
anything in common with the rationalist and egalitarian liberalism which
has dominated American life for fifty years. They are defined, and define
themselves, not primarily as buyers and sellers in markets, nor as abstract
bearers of rights and entitlements, but in terms of their membership in a
nation or a church. They may share a common longing for emancipation
from the Soviet system, but that is all they share. Each of the subject
peoples in the Soviet bloc harbours particular claims, territorial or other-
wise in character, which sets it in conflict with the rest. It is for this reason
that the waning of the Soviet system is bound to be accompanied by a
waxing of ethnic and nationalist conflicts - just the sort of stuff history has
always been made of. These conflicts are, in part, undoubtedly a legacy of
Stalinism, since it was Stalin who ruthlessly dislocated entire peoples and
relocated them without regard to their history or traditions. But these
conflicts also embody age-old enmities and loyalties, which are now
coming back to the surface after decades totalitarian suppression. What we
are witnessing in the Soviet Union is not, then, the end of history, but
instead its resumption - and on decidedly traditional lines.
 The end of history - or of liberalism? 249
All the evidence suggests that we are now moving back into an epoch
that is classically historical, and not forward into the empty, hallucinatory
post-historical era projected in Fukuyama's article. Ours is an era in which
political ideology, liberal as much as Marxist, has a rapidly dwindling
leverage on events, and more ancient, more primordial forces, nationalist
and religious, fundamentalist and soon, perhaps, Malthusian, are contesting
with each other. In retrospect, it may well appear that it was the static,
polarized period of ideology, the period between the end of the First World
War and the present, that was the aberration.
If the Soviet Union does indeed fall apart, that beneficent catastrophe will
not inaugurate a new era of post-historical harmony, but instead a return to
the classical terrain of history, a terrain of great-power rivalries, secret
diplomacies, and irredentist claims and wars. The vision of perpetual peace
among liberal states, which has haunted Western thought at least since it was
given systematic formulation by Immanuel Kant, will soon be seen for what
it always was - a mirage that serves only to distract us from the real business
of statesmanship in a permanently intractable and anarchic world.
Fukuyama's brilliant and thoughtful argument is a symptom of the
hegemonic power of liberalism in American thought. So ubiquitously per-
vasive are liberal ideas and assumptions in American intellectual life, and
such is their constraining power over public discourse, that it sometimes
seems barely possible to formulate a thought that is not liberal, let alone to
express it freely. The domination of the American mind by liberal ideology
has fostered blind spots in American perception of the real world that have
been immensely disabling for policy.
The fetish of open government, as symbolized in the Freedom of Infor-
mation Act, congressional oversight, and the respectability given to leak-
ing, prevents the United States from ever again engaging in any major
covert operation. The domination of public life by the power of the invasive
media calls into serious question the capacity of the United States to wage
any war larger, more protracted, or with significantly heavier casualties,
than the invasion of Grenada. Liberal egalitarianism in education. coupled
with absurd and counterproductive affirmative-action programmes. has
resulted in a de-skilling of America that is awesome in magnitude. (Con-
sider that. whereas at age six Japanese and American children have roughly
similar mathematical abilities. at age 18 the average Japanese child has the
mathematical competence of the top 1 per cent of American children.)
In many other areas, liberal ideology has in America proved itself to be
the enemy rather than the friend of civil society. In its expression in radical
feminism and in affirmative-action policy liberalism has sanctioned the
invasion of privacy, the curtailment of freedom of association and the
erosion of contractual liberty. Because of the ravages wreaked on civil
250 Critiques
society by liberal ideology America already has a more bureaucratized and
regulated, less tolerant, more divided, and more statist society than virtually
any other modem democracy, squandering the historical patrimony of civil
society on which American pre-eminence in the world rested. Liberal
ideology guarantees blindness to the dangers that liberalism has itself
brought about. In sum, the danger for America is that, confronted with
comparative and soon, perhaps, absolute economic decline, an un-
controllable crime epidemic, and weak or paralysed political institutions, it
will drift further and further into isolation and disorder. At the worst,
America faces a metamorphosis into a sort of proto-Brazil, with the status
of an ineffectual regional power rather than a global superpower.
In general, all speculations about the future are riddled with hazards.
Michael Oakeshott, the English conservative philosopher, has written that
we know as much about where history is leading us as we do about future
fashions in hats. There are, perhaps, only two things of which one may be
reasonably sure. The first is that the days of liberalism are numbered.
Especially as it governs policy in the United States, liberalism is ill-
equipped to deal with the new dilemmas of a world in which ancient
allegiances and enmities are reviving on a large scale.
We know this much at least: history will not end with the passing of
liberalism, any more than with the collapse of Communism. The second
thing we know for sure is that we have no reason whatever to expect that
our future will be markedly different from our past. As we have known it,
human history is a succession of contingencies, catastrophes and occasional
lapses into peace and civilization. If this is the case, there is at least one
misfortune that we will surely be spared - the melancholy and boredom that
is evoked by the prospect of the end of history.

October, 1989
Part III
Questions
18 The politics of cultural diversity

In 1946, the subtle and neglected Hungarian philosopher Aurel Kolnai

wrote of the Habsburg Empire:
Imperial Austria, like Switzerland, notwithstanding the numerical dis-
proportion of their different nationalities, did not have 'minorities'
because they had no ruling nation'. 1
I begin my reflections with this thought, since it seems to me to embody a
possible solution to a problem which besets much contemporary political
thought and practice. The problem I have in mind may be defined, pro-
visionally at least, by a question: What are the political forms best suited to
a condition of society marked by substantial cultural diversity? This ques-
tion ought to arise naturally, when we consider the novelty of our condition.
Almost all of us live in societies which are not unified by any single
cultural tradition, but which contain a variety of traditions and ways of life.
Our great cities shelter vast enclaves of traditional life, often introduced by
recent immigrants, at the same time as they nurture liberal bohemian
milieux where conduct is governed by taste and preference rather than by
any set of established mores. Within most modem states, we find forms of
family life based on romantic love and on arranged marriage. We meet,
each of us in his daily life, those for whom religious faith pervades their
sense of life, and those for whom it is a barely intelligible irrelevance. Our
societies encompass a kaleidoscopic diversity of attitudes to sexuality and
gender, death and the human condition, even as they harbour a prodigious
diversity of ethnic inheritances and styles of life. It. is this condition, in
which society intimates a diversity of possibly incommensurable values
and world-views, which is often characterized as cultural pluralism, which
I believe ought to be at the top of the political agenda of modem states. Yet,
though the fact of cultural diversity is noted often enough, its political
implications are rarely explored. We find the reason for this strange
neglect, I believe, in a doctrine that is held in common by most modem
254 Questions
political thought. It is this modem doctrine, or heresy - the heresy that
political orders oUght to embody or express the cultural identity of homo-
geneous moral communities - which I seek to attack and undermine in the
course of developing these reflections.
At the most abstract and fundamental level, the question I am addressing
has to do with how the human subject is to be conceived, and what are the
implications for political order of a well-founded conception of the subject
On the negative or critical side, my concern is to expose and overturn a
presupposition about the human subject which is held in common by the
political theories which dominate the contemporary scene. It is an assump-
tion about human identity or subjecthood which animates all the dominant
forms of liberalism, conservatism and socialism, which I wish to stigmatize
as erroneous in theory and pernicious in practice. This assumption - the
assumption that the varieties of human identity, if they are not altogether
constructions of political practice, ought nevertheless to be mirrored or
imprinted in the central institutions of political order - is made in varying
ways in the different political perspectives in which it is found.
In some perspectives, the role of political institutions is to shore up
tottering identities, to heal the scars inflicted on established traditions by
cultural change, whereas in others the project is that of constituting new
identities by the use of political power. The project of constituting new
forms of human identity through political practice is, indeed, the pre-
eminent contemporary form of this modem heresy, and it is found in many
current varieties of liberalism. It finds its starkest and crudest form in
Marxism, however, and, above all, in the species of Marxism contrived
with opportunistic genius by Lenin. For Marx, as for Lenin and all of
Marx's followers, the historical forms of human identity, as they are
expressed in inherited cultures' traditions and folkways, are in no way to be
taken as authentic expressions of human nature and creativity. Instead, all
the forms of life in which human beings have hitherto found their identities
are condemned as contingencies, shadows cast by underlying structures of
production. In this reductionist perspective, all the varied cultural achieve-
ments of mankind are epiphenomena, transitory reflections of changes in
the economic base of human society. With this much (or little) we are all
familiar. But the positive, and mythological aspect of Marxian historical
materialism, 2 is less obvious, and less commonly commented upon. It is the
Marxian conception of true or essential human nature as being hidden and
submerged, thwarted and mutilated by all the historic forms in which
human beings have ever lived.
Now in the real world of human history, men are not abstract instances
of the human species, but articulators of definite ways of life that reflect
their circumstances and at once express and confer on them a distinctive
 The politics of cultural diversity 255
identity. Marxian Communism, by contrast, seeks to expunge the cultural
inheritances of mankind, and thereby also to roll back the most powerful
achievements of the modem world in the interests of a mythic entity - that
of universal humanity. That this myth - the myth of a human essence shorn
of religion, family, locality and all the accidents of time, purged of conflict
and released from contingency - is a gnostic rather than a Christian or a
Jewish myth is unimportant here, save to illuminate the fact that the
character of Marxism-Leninism as a religion, remarked upon in a thousand
academic cliches, is extremely disputable. The peculiarity of Marxism, in
both its primitive form and its more developed Soviet version, is its fusion
of gnostic ambition with scientific pretension. If Marxism is a religion, it is
only in the sense that occultism and theosophy - modem forms of magical
thought which seek an abatement of tragedy and mystery through a sort of
rationalistic gnosis - are religions.
Marxian philosophical anthropology is committed to the view that all
the forms of human identity which have achieved concrete historic real-
ization are aberrations, unavoidable episodes (perhaps) in the growth of
human productive powers, but nevertheless episodes which conceal rather
than disclose the human essence. Men and women are not, as they suppose
themselves to be, Poles, Spaniards, Englishmen or Scots, Catholics, Jews or
Muslims. They are only men and women, instances of the human species
which have acquired an ersatz cultural identity through the accident of
belonging to a particular class in a given historical epoch. It follows from
this essentialist anthropology that, except in so far as they are instrumental
in promoting the expansion of production and so in building up the neces-
sary economic conditions of Communist society, all the historic
realizations of human identity - in religion, art and cultural tradition - are
alien and inimical to the realization of the human essence.
It must not be supposed for a moment that this Marxian philosophical
anthropology has been unimplicated in Communist practice. It is a central
element of Marxism-Leninism that, when it comes to pass, Communist
humanity will be unlike any that history has known. All the cultural
attachments which have granted real men their identities will have been
effaced, and only quintessential humanity will remain. It is one of my chief
arguments that, contrary to Marxism, the quintessential humanity which is
to appear on the historical stage has no existence and, in fact, designates
nothing. For this reason, the practical result of Communist policy has not
been to liberate men from ersatz and oppressive cultural identities, but
instead to make war on the only identities their subjects possess.
It is, perhaps, worth noting that the practical nullity of Marxian con-
ceptions of socialist man is evinced in the bizarre phenomenon of displaced
patriotism which may be observed in generations of the Western leftist
256 Questions
intelligentsia. These groups adopted the Soviet Union, and later China, as a
focus for the needs for allegiance which they denied in themselves by their
repudiation of their own inherited loyalties. The result was an ersatz patriot­
ism in which allegiance was transferred to a remote and largely imaginary
society and state. It is only by reference to this displaced or ersatz patriot­
ism, I believe, that we can begin to comprehend the many absurdities in
Western intellectual perceptions of the Soviet system. The grotesque repre­
sentations of the Soviet system evident in the work of Shaw and the Webbs,
Aragon and Sartre, are at least partly explicable in terms of the demands of
a repressed need for patriotic allegiance and collective identification. Nor
is this phenomenon only of historical interest. We need to remind ourselves
that it is probably only the relative brevity of the Pol Pot regime, and its
defeat by Vietnamese occupying forces backed by the Soviet Union, that
explains the fact that :t is not now routinely described by the Western
academic class in terms of iconoclasm and innovation in urban planning.
At this point, we may undertake a brief survey of Communist policy,
beginning by looking back at a neglected aspect of Marx’s system - its
antimodemist impulse. Marx inherited from Hegel a self-image of con­
summate modernity, but in Marx’s case this was only a self-deception.
Unlike Hegel, who accepted as an historical fate the complexity of con­
temporary society and its domination by a spirit of individuality, Marx
cherished an ideal - the ideal of Communism - whose content is given by
the denial of the most distinctive features of modem culture. Thus the very
things that best define a modem society - an extensive division of labour, a
complicated system of status stratification and a diversity of cultural tradi­
tions - are all set to disappear in Communism, along with the institutions of
money and market-pricing without which industrial society cannot (in fact)
reproduce itself. Most fundamentally, the central character of the modem
period - the autonomous individual with access to a variety of forms of life
and modes of thought - will vanish in a Communist society which has
suppressed the experience of moral conflict and intellectual suffering. For,
as we recall from the speculations of The German Ideology (written by
Marx with Engels in 1846 and first published in 1932), both art and philo­
sophy - the central activities in which modem men explore the ironies of
their condition and seek to plot its limits - will be redundant by virtue of the
dissolution of their historical locus, the human individual, and the absorp­
tion of that individual in to a new communal life.
Marxist theory has always been committed, then, to the elimination of
the modem expression of human identity as individuality. But it - and
therefore also Communist practice - has always aimed to destroy those
surviving forms of cultural identity which conserve a pre-modem sense of
selfhood. When the Bolsheviks came to power, they declared a war on the
 The politics of cultural diversity 257
native traditions of Russia that to this day has lost none of its urgency. Their
aim was to effect a radical rupture with the past, so as to force the emer-
gence of a new order in which national identity, along with other forms of
cultural identity that had existed in the old world, played no part. It is this
aim which probably best explains the economically counterproductive war
on the Russian and Ukrainian peasantries in the 1930s, and which accounts
in significant part for the Communist genocides of such peoples as the
Kalmucks.
It is vitally important to recognize the enmity of Soviet Communism to
Russian traditions, because this is what gives the lie to the dominant
Western (and East European nationalist) interpretation of the Soviet system
as a continuation of Muscovite traditions of despotism under a new flag. It
is this interpretation - according to which political cultures may be grouped
into pairs, civic and despotic, democratic and tyrannous, with that of Russia
being classified as a quasi-oriental despotism - that is invoked when there
is talk of 'the last empire' or 'the New Tsars'. This typology of regimes,
with its antique echoes of Aristotle and Montesquieu, wholly occludes our
vision of the Soviet system, which (like National Socialist Germany) is by
no means a traditional tyranny any more than it is an oriental despotism.
The conventional wisdom among Western scholars, in seeking to
explain away the horrors of the Soviet system as inheritances from a
barbarous Russian political tradition, neglects the role of Marxian theory in
constituting and reproducing the Soviet system3 and the relentless hostility
of both to the traditions and achievements of the Russian people. The
so-called Russian empire of our time has, in truth, few points of similarity
with the empires of nineteenth-century Europe. It benefits neither the
Russian people, who remain among the poorest nations in the Soviet
system, nor even the Soviet elite, whose most cherished privilege is the
opportunity to defect. In projecting into Soviet reality the concepts and
images of the past, Western observers fail to grasp the radical modernity of
the Soviet totalitarian system. At the same time, and paradoxically, the
Bolsheviks in laying waste the inherited traditions of Russian life failed to
lay down a solid foundation for genuinely modern development. In obli-
terating the civic culture that had prevailed during the last decades of
Tsarism, the Bolsheviks pushed Russia off the trajectory of real modernity
on to a bizarre and surreal path whereby some of the most distinctive
aspects of the modern world were realized in an uncompromising and
terrible form.
From the perspective of Marx and Lenin, the Soviet Union is a double
failure. It has not realized the antimodernist fantasy of a form of communal
life from which individuality has disappeared, but has rather brought about
a savage rebirth of a character from the early modern period, the Hobbesian
258 Questions
individual for whom every social relationship is an act of exchange. It is
this Hobbesian character who is the true Homo Sovieticus. At the same
time, the real social life which survives in the Soviet Union today is one in
which the broken traditions of the old world - traditions of religious
devotion, cultural piety and romantic love - are asserted against the immov-
able power of the Soviet state. As for Soviet man himself, he is uniquely a
creature of Soviet power and has no existence outside it.4 So it is that, even
in his Hobbesian reality, Soviet man is a lie - a vulgar forgery of the
desperate but not ignoble character depicted by the author of Leviathan. Yet
he is the only result of the project announced with Lenin's hideous [Link]
that the Bolsheviks would become engineers of souls - a result squalid and
pitiable enough to warrant Kolakowski's acid remark that 'thus Prometheus
awakens from his dream of power, as ignominiously as Gregor Samsa in
Kafka's Metamorphosis'.'
What do the catastrophic results of the Marxian project of constituting a
universal human identity, shorn of local and historical attachments, tell us
about the nature of human selfhood? In the simplest terms, the history of
Communist practice can be seen as the practical self-refutation of Marxian
philosophical anthropology. Seventy years after the Bolshevik Revolution,
most subjects of the Soviet system still define their identities in religious
and national, rather than class or political terms. They are, and remain,
Poles, Ukrainians, Muslims and Christians. The self-defeating effect of
Communist practice should lead us to jettison all theorizing about human
society which treats the historic cultural identities it contains as spurious or
merely contingent variations on the theme of generic humanity.
To a very considerable extent, human beings are self-defining animals:
they differ from the other animal species precisely in their ability partly to
constitute for themselves their own identities. This is to say that, for
humans, personal identity is not a natural fact but a cultural artifact. We are
what we are, not because nature has made us this way, but because we (and
'we' here includes our ancestors) have made us what we are. It is neglect or
suppression of this truth that goes far to account for the pernicious absur-
dities of racist thought and practice. If the identities of persons are arti-
factual and not natural, if they vary within the broad constraints imposed by
our biological endowment and exhibit a rich variety of kinds and forms,
then any essentialist or naturalistic conception of the human species is to be
rejected as false if not incoherent. We are left with human beings in all the
miscellaneous diversity with which they present themselves to each other.
This conception of human beings as partly self-defining creatures has a
critical leverage on contemporary theorizing which far extends from its
impact on Marxism. For it cuts against at least the dominant forms of
Western liberalism, which all turn on the philosophical fiction of abstract
 The politics of cultural diversity 259
individuality. All of them, that is to say, found their theorizings on a
conception of the human subject in which its identity and moral powers are
conferred on it, not by a moral inheritance or a culturahradition, but by the
attributes of universal humanity. But the idea of a universal human indivi-
duality, which is a feature of the currently dominant liberalisms, mis-
represents a distillation of our own experience as a universal truth. By
contrast with other and earlier societies, ours is one in which status hier-
archies are permeable, roles complex and often conflicting and social
monitoring of personal behaviour intermittent and weak. These features of
our culture are, indeed, ones we take a pride in, and which (in so far as they
are conditions of the values of personal independence and privacy that we
cherish) are parts of our freedom. The complexity of the social networks in
which each of us moves, and the fact that none of them, or their associated
moral communities, ever claims the entirety of our life or loyalty, is the
source in common experience of the liberal idea of the autonomous
individual.
Among us, and in some measure in every society touched by modernity,
the liberal individual is an historical reality, a cultural achievement perhaps
more precious than any other bequeathed to us by European civilization. To
deny the historical reality of liberal individuality is then absurd, but to tum
it into a universal theory - or, after the fashion of the Enlightenment, to
appoint it the ,elos of history - is to traffic in illusions. To acknowledge
liberal individuality as a distillation of modem experience, and thereby to
recognize the historicity of liberalism itself, is to relinquish the pretensions
of doctrinaire liberalism to universal validity. In coming to understand
liberal individuality as an appropriate rendition of our experience as
modems, however, we abandon the distinctively modem (or Enlighten-
ment) project of grounding liberalism on the universal maxims of auto-
nomous reason. That in so recognizing the time-bound modernity of the
liberal individual we do not (in the jargon or recent thought) deconstruct the
individual, but rather initiate a form of post-modem individualism that is
fully conscious of its own historical particularity, is a paradox to which I
shall shortly return.
It is in its insistence on recalling to us the local character of our experi-
ence of individuality that the strength of conservative philosophy lies. It
dispels the hallucinatory perspective of abstract humanity, and draws us
back to the narrower, but more substantial standpoints of real human
beings, in all their quiddities and miscellaneity. Conservative thought has
inestimable value in correcting the illusion that we are, or can ever be,
dispossessed or unencumbered selves, free-floating sovereign subjects,
distanced from all social convention and heirs to no tradition. Conservative
philosophy contains, then, an incisive critique, not only of the stupendous
260 Questions
delusions of Marxism, but also of the lesser, but no less mistaken errors of
the dominant liberalism. From conservative thought, we can glean a devas-
tating criticism of the liberalism of J. S. Mill's On Uberty, where un-
criticized and flawed conceptions of autonomy and individuality combine
with an obsessional enmity to tradition and convention to yield a liberalism
in which rationalist hubris, antinomian individualism and a sentimental
religion of humanity reinforce and strengthen each other.
The liberalism of On Liberty hinges on the neglect of cultural tradition
as the matrix of human individuality and its issues in the absurd proposal
that autonomy be theorized (and practised) as an independence of tradition
rather than as variations on particular traditions. The liberalism of On
Liberty - which may not be the only, and, as I have argued elsewhere,6 is
certainly not the most defensible liberalism to be found in Mill's writings-
distracts us from the truth that autonomy and individuality wax, not in a
post-traditional society, but rather in a society encompassing many tradi-
tions, among which individuals may move freely, but with each of them
exercising a constraint of opinion over its practitioners.
The form of society celebrated by Millian liberalism- the liberalism
from which all the dominant liberalisms of the present day are derived - is
not, in truth, any sort of pluralist society. It is instead a society ruled by an
elite of opinion-formers - Mill's secular version of Coleridge's clerisy -
which relentlessly propagates a narrow, partisan ideal of rationalistic indiv-
idualism and progressivism. In this blinkered view, traditions are encum-
brances and obstacles to individuality, rather than the media in which it
alone exists and finds expression, and the inherited variety of folkways and
lifestyles is a garden of weeds to be winnowed and levelled rather than
explored and enjoyed. We see this clearly in J. S. Mill himself, who writes
sneeringly that 'Nobody can suppose that it is not more beneficial to a
Breton, or a Basque . . . to be (French) than to sulk on his rocks, the
half-savage relic of past times, revolving in his own little mental orbit',
thereby giving expression to that contempt for small peoples and cultural
minorities which he held in common with Marx.
In theory as well as in practice, Millian liberalism is a force for cultural
homogeneity and against diversity, a political tendency for which progress
is more important thap liberty. And by progress is here meant, not the
uncalculated exfoliation of human energies, but the imposition of a plan of
life in which the prejudices and anxieties of the late nineteenth-century
European intelligentsia are made mandatory for all. In its political mani-
festations in our times, Millian liberalism has been a programme for
cultural conformity, with much of the inherited diversity of our society
being treated as an aberration from the cramped and suffocating norms of
Hamp- stead and Bloomsbury. The ambition of Millian liberalism - a social
 The politics of cultural diversity 261
world without Gypsies or Hasids, in which immigrant cultures are speedily
assi- milated to the mores of a bourgeoisie from which the redeeming
virtues of entrepreneurship and self-reliance have all but disappeared - is a
sort of universal bohemia who~e practical realization, mercifully, is
probably impossible.
The impossibility of the goal does not, however, make its pursuit less
harmful. By a dialectical twist in its own inner logic, the dominant liberal-
ism of our time has responded to the threat of cultural homogeneity which
is the nemesis of its efforts by adopting another form of cultural imperial-
ism, in which the remnants of overwhelmed traditions are preserved as
spectacles for public consumption in subsidized ghettoes. It is in this way,
I suspect, that we should understand the liberal enthusiasm for policies of
positive discrimination - as a belated and unwitting recognition of the
desolation which liberal policies of social engineering have wreaked on our
societies. (Examples that come to mind are the destruction of working-class
communities in Britain by municipal housing policies, and of black com-
munities by urban renewal programmes in the United States.) It is this un-
conscious recognition of the levelling and conformist effect of modem
liberalism which also partly explains the bizarre excesses of multi-
culturalism - that fashionable form of paternalism which aims to embalm
the dead or dying vestiges of submerged or occluded traditions and preserve
their remains as public spectacles. I shall argue later that this multicultural
mania is in truth the negation of respect for cultural diversity. Here I wish
to assert that the disasters of recent liberal policy are evils which flow
inexorably from any liberalism, such as that of John Stuart Mill, which
seeks to achieve the political domination of the liberal individual instead of
aiming for a political order which can be shared by a variety of forms of
life, including that of the liberal individual.
It is in undermining the excesses of liberal individuality, in insisting that
our identities are constituted and not encumbered by the forms of life that
we inherit, that the virtue and success of conservative thought lies. At the
same time, the conservative critique of the abstract individualism of liberal
theory often, if not typically, issues in a conception of the human self, and
a prescription for political order, which can be shown to be radically
flawed. The conservative theorist, like the communitarian critic of liberal-
ism from the Left, moves unreflectively from the truth that we are none of
us encumbered or disembodied selves to the very different, and indeed
manifestly false proposition that we are, or ought to be radically situated
selves - that is to say (I am here explicating the idiom of recent com-
munitarian theory') selves whose identity is contoured by membership in a
single moral community and mirrored in the institutions of a single political
order. This is the belief, shared by Leftist communitarian theorists such as
262 Questions
Sandel and by Hegelian conservatives such as Scruton,' that moral life
flourishes only when personal identities are constituted wholly by embed-
dedness in a single moral community, and, further, that such moral com-
munities demand and are entitled to the protection of a political order with
which they are or ought to be coextensive. It is to this belief that I referred
when I spoke of a modern heresy as the target of my thoughts. In its
simplest form, this belief is expressed in modern thought as nationalism,
which has been characterized with un surpassable clarity by Isaiah Berlin as
the claim that
the essential human unit in which man's nature is fully realised is not the
individual, or a voluntary association which can be dissolved or altered
or abandoned at will, but the nation; that it is to the creation and
maintenance of the nation that the lives of subordinate units, the family,
the tribe, the clan, the province must be due, for their nature and purpose,
what is often called their meaning, is derived from its nature and pur-
poses; and that these are revealed not by rational analysis, but by a
special awareness, which need not be fully conscious, of the unique
relationship that binds individual human beings into the indissoluble and
unanalysable organic whole which Burke identified with society,
Rousseau with the people, Hegel with the state.9
The belief that political order presupposes deep moral solidarity, and its
further presupposition, that the identities of persons are or ought to be
defined by membership of a single moral community, else they be
deformed or estranged, is not restricted to nationalist theorizing. Rather, it
animates most modern thought, and it is especially prominent in con-
temporary criticism of liberalism.
I have acknowledged an important truth in the conservative and com-
munitarian critique of liberalism which fastens on weaknesses in the
dominant liberal .conception of the human subject. Against the presently
dominant kinds of conservatism, however, I wish to argue that they exhibit
a remoteness from historical reality which is peculiarly damaging in a form
of thought which rightly seeks to elevate the particulars of concrete practice
over the delusive universals of abstract theorizing. For the communitarian-
conservative conception of the radically situated self can be shown to have
no application to modern experience and to model what is, at best, a
limiting case. We can see this, to begin with, when we reflect on our own
experience. We are none of us defined by membership in a single com-
munity or form of moral life. We are 'suckled on the milk of many nurses' ,
as Fulke Greville put it, heirs of many distinct, sometimes conflicting,
intellectual and moral traditions. Further, the traditions to which we are heirs
are not windowless monads, self-sufficient and fully individuated entities
 The politics of cultural diversity 263
which (like pebbles) may coexist without interacting. They are rather prisms,
each of which refracts the light cast by the other traditions which environ it, and
which together throw into each of us a shifting pattetn of colours. The com-
plexity and contradictions of our cultural inheritance give to our identities an
aspect of complexity and even of plurality which is not accidental, but (if we
may use such a term) essential to them. For us, at any rate, the power to
conceive of ourselves in different ways, to harbour dissonant projects and
perspectives, to inform our thoughts and lives with divergent categories and
concepts, is integral to our identity as reflective beings.
This same point can be put in other terms: the experience of marginality
is familiar to all of us as a dimension of our identity which is integral to it.
Because it is our condition to belong to many different, and often discrepant
networks and communities, we belong wholly to none of them. The power
of distancing ourselves from any relationship or attachment, of imagining
ourselves to have severed or altered anyone of the many involvements
which enter into our identities, is itself a central element in our identity.
Even as we are constituted by our attachments, we (perhaps not all men and
women) are constituted by the knowledge that we can alter or dissolve any
of them. To wish this to be otherwise, to seek to suppress the experience of
marginality, is to seek to alter our nature.
To wish to alter our nature, however, is also to seek to expunge our
history. For the condition of moral complexity to which I have alluded is
not a novelty of modern decadence, but instead a phenomenon that mani-
fests itself in European life at least since the late-medieval period. We find
an accurate reading of European history, I believe, in Michael OakeshoU's
observation that
The urge to impose upon a state the character of a solidarite commune is
certainly a notable disposition, but, so far from being the dominant
disposition of the modern European political imagination, it is easily
recognised as a relic of servility of which it is proper for European
peoples to be profoundly ashamed ... no European alive to his inherit-
ance of moral understanding has ever found it possible to deny the
superior desirability of civil association without a profound feeling of
guilt. IO
That Oakeshott's reading of European history is superior to any advanced
by communitarian conservatives is confirmed by even a cursory glance at
the rudiments of the history of the modern European state. It is a plain fact
that few, if any, modern European states coincide with any national
community. The sovereign state of the United Kingdom is precisely that, a
kingdom encompassing four nations, and not a nation-state. Again, what is
Spain but an artifact of monarchy? And, lest we neglect an apparent
264 Questions
counter-example, let us recall that as late as 1870, a survey of French school
children reported that the majority of them were unable to name the nation
in which they lived, but instead named their province or region. lJ The belief
that national consciousness is a long-established feature of European
peoples is, in fact, profoundly unhistorical, and for that reason deeply
unconservative. It is a chimera produced by a mistaken theory which
understands political ord'!r in a quasi-naturalistic fashion as an expression
of a pre-existent community, when it is properly understood as a matter of
strictly political allegiance to the artifact of sovereign authority. As Kolnai
again puts it well:
human society is not composed of nations ... in the same clearcut way
in which it is composed of individuals, or, for that matter, of sovereign
states. The spectrum of nationalities is full of interpretations, ambi-
guities, twilight zones. It follows that the conceptions of nationalism (as
a universal principle), the conception of a 'just' or 'natural' order of
nation-states is - in fact and in theory - pure utopia. There can be neither
an order of states nor of frontiers in which there does not enter to a large
extent the factor of arbitrariness, contingency and historical accident
Pretending to 'purify' the body of mankind -like other enterprises of a
naturalist pseudo-rationalist sort, purporting to lay down 'evident prin-
ciples' which generally prove illusory - means to push arbitrariness to
its extreme limit. 12
I have ascribed to conservative thought deep insights into human identity
which much liberalism neglects. It is a conservative insight that the Whig (and
Marxian) idea of historical folkways and concrete forms of life as aberrations
- as veils behind which is to be found (in Stuart Hampshire's words)
an abstract universal man dressed in neo-classical drapery, as in some
Reynolds paintings, to indicate that he belongs to no particular place or
time, (framed in) the unearthly light of the ideal, classical and timeless
scene (wherein) reason cannot tell him how he should be married or how
he should speak to his children ... or give one local loyalty precedence
over another13
- that this idea is a mirage.
I have found in conservatism, that is to say, the important truth that
liberal and Marxian conceptions of universal humanity are false or in-
coherent and harmful in practice. I want now to develop further a criticism
of the dominant forms of contemporary conservatism, which are com-
munitarian and often nationalist in temper, by trying to delineate more
positively the sort of political order which does not depend on a pre-existent
moral solidarity. My submission is that, in the subtle mosaic of traditions
 The politics of cultural diversity 265
which is modem society, government is ill-fitted to act as guardian or
protector of any of the traditions it shelters. It cannot claim to express any
deep, undergirding moral community in the society, since no such com-
munity exists. In this circumstance, the task of the state is to keep in good
repair what Oakeshott calls civil association - that structure of law in
which, having no purpose in common, practitioners of different traditions
may coexist in peace. Modem states are peculiarly apt to be distracted from
this vital task by the reflex of governmental hyperactivism which they all
exhibit. Every problem or evil, from drug abuse to family breakdown, is
perceived as a dire threat to an established order which it is the duty of
government to protect. Against this modem prejudice, I would cite Witt-
genstein's remark that trying by deliberate contrivance to shore up an ailing
tradition is like trying to repair a broken spider's web with bare hands.
There is a special absurdity in invoking the authority of government to
intervene in problems to which government has no solution. None of the
manifestations of contemporary decadence about which conventional con-
servative opinion is so alarmed are threats to the survival of society, and a
true conservative policy would rather be one that in most things left society
to look after itself. The resources of renewal and self-repair in society are
surely far greater than any that can be mobilized by the ramshackle action
of government.
The implication of this perspective for the character of the modem state
is that (contrary to the conventional wisdom, socialist and conservative)
modem government must be limited government. Where deep moral
solidarity is lacking, where (as in all modem societies) there is cultural
diversity rather than seamless community, the role of government is first
and last that of preserving liberty in civil association under the rule of law.
The liberty that is preserved is that ofthe liberal individual, but it is a liberty
that thereby guarantees cultural freedom - the condition in which indivi-
duals may opt to explore an inherited form of life, or migrate across
traditions to a chosen lifestyle, if they so wish. Cultural diversity is pro-
tected, in this perspective, by a political order which protects the old liberal
freedoms of conscience, occupation, contract and enterprise. The form of
government distinctive of such a political order is not the minimum state of
recent libertarian theory, but it is akin to the limited state of classical
liberalism. It is a form of government in which each person, and therefore
also each cultural tradition which the society contains, possesses an equal
freedom to develop themselves, compatibly with the equal freedom of the
others to do likewise.
Two points of clarification are worth making about this conception of
government in a society characterized by substantial cultural diversity. The
first is that, according to the view I am developing, the place of government
266 Questions
is to respect diversity in cultural traditions, and so allow for the emergence
of rival and perhaps incommensurable forms of human flourishing, but not
to attempt to institutionalize traditions or ways of life in state-subsidized
ghettoes. The political order which I am commending is accordingly at the
farthest extreme from that promoted by contemporary agitators for multi-
culturalism. For these latter, animated by a moral hatred of the dominant
traditions and a voyeuriftic fascination with submerged, overwhelmed or
oppositional cultures, seek to entrench minority traditions in protected
institutions and to confer on them arbitrary privileges.
This multiculturalism is bogus and pernicious, because it expresses a
racist conviction that minority cultures can never maintain themselves
without paternalistic support and because it seeks to replace one form of
cultural imperialism by another. (That minority cultures do not always need
political protection is shown by the history of the Indian immigrant popu-
lation in Britain and of the Chinese and Japanese in the United States,
where in each case governmental institutions have proved a hindrance
rather than an aid to minority advancement.) Further, as the tragic history
of the American black population shows, such paternalistic and welfarist
measures typically serve only to complete the devastation of traditional
family and cultural life, already damaged by social prejudice and economic
hardship. The final result of contemporary multiculturalism, if it could be
achieved, would in fact be a mirror image of the system of privilege and
separatism favoured by authoritarian conservatives - in short, the institu-
tionalization of a cultural apartheid.
That government should wash its hands of contemporary multiculturalist
manias does not mean that it should become a nightwatchman state.
Government may properly act to supply underproduced public goods and to
assist the needy in their struggles toward opportunity and independence. In
general, however, government should refrain from itself producing the
goods it seeks to generate in society. In the vital area of education, for
example, there is much to suggest that we would be better off with a system
of tax deductions and vouchers whereby families could provide for them-
selves privately, than we are with the bureaucratic monolith of state school-
ing. It has been too little noted by proponents of state schooling that public
provision of education generates intractable problems where, as with us,
society contains divergent ways of life with discrepant conceptions of the
family and of personal fulfilment. The inner-city British Moslem does not
take the same view of the role of schooling as does the secular humanist in
Kensington, and it is unavoidable that an educational system provided by
the state will become an arena for political struggle in which weak or
ill-organized minorities will lose out and educators will be distracted from
their central tasks.
 The politics of cultural diversity 267
That state provision of education is a recipe for social conflict and the
politicization of life in a multicultural society is an instance of the general
truth that governmental provision of social services is bound to yield a
pressure for cultural homogeneity which only the cost of protracted poli-
tical conflict curbs. The policy implication of this truth is that both social
stability and individual liberty are best served if government withdraws
from such activities and confines its intervention to funding voluntary
choice in education and elsewhere for families and individuals with slight
means. It is in such a direction that wise policy should move, and this is a
point especially relevant to conservatives, who are conspicuous for their
neglect of the fact that the subversive effects of the invasive state on liberty
and tradition are integral to its very existence and are not significantly
diminished when the state machine is captured for a while by parties with a
conservative temper.
It is a consequence of the view I have been developing that, if govern-
ment is to shelter the variety of traditions which our society contains, then it
should favour none of them particularly. This, in tum, implies that there is
no guarantee that long-established traditions and forms of life will succeed
in reproducing themselves. It may well be, in fact, that the withdrawal of
government to limited and definite functions would actually quicken the
pace of cultural change in some areas of social life. In particular, the form of
life of the liberal individual will in the liberal political order I am outlining
be only one way of life among many, and nothing (except its own resources
of vitality) guarantees that it will retain primacy among the cultural tradi-
tions which have found a home among us. We do not need to see every social
change as an improvement to accept this as an inevitable result of the proper
limitation of governmental power. Traditions and ways of life that require
for the successful reproduction more from government than protection from
the rule of law and the provision of opportunity for those with few resources
probably do not deserve to survive.
It is not the office of government to buttress failing identities, to confer
identities on persons afflicted with anomie, or, in general, to make the
world safe for cultural traditions endangered by the freedom that is pro-
tected by civil association. Nor is it the task of government to protect
waning traditions by confining their practitioners to tax-funded institu-
tional reservations. This has been the increasing tendency of governments
in their policy toward minorities in many Western states. But it is this
protectionist policy in respect of social movements and ways of life, I
believe, that John Anderson was criticizing when he wrote his remarkable
essay on the Servile State.
The argument I have developed goes against the grain of most con-
temporary thought - socialist, liberal and conservative - in that it denies to
268 Questions
government the task of conserving (or, in the case of all Communist and
many liberal states, destroying) cultural identity. Against Marxian social-
ism I have urged the catastrophic consequences of the attempt to construct
a new human identity by the political engineering of souls. Against com-
munitarian and authoritarian conservatism, I have argued that the long-
standing complexity of European traditions disqualifies any project of
using political power to create or conserve a radically situated subject
whose horizons are those of a single moral community. This reactionary
conservatism, I have argued, neglects the fluidity and elusiveness of tradi-
tions in general, and, more particularly, it ignores the modem reality, in
which traditions constantly individuate themselves anew, transforming
each other as they do.
I have criticized, also, that form of liberalism (the dominant con-
temporary form) which seeks through government to impose upon all the
form of life of an alienated fragment of the Western bourgeoisie. My
argument is nevertheless consonant with a liberal tradition, sounder and
truer than that which has been dominant since John Stuart Mill, which sees
government not as the pacemaker for any specific conception of progress,
but instead as providing the framework within which different ways of life
and styles of thought may compete in peaceful coexistence. It is this
conception which animates the Scottish Enlightenment and the writings of
Tocqueville and Constant and which inspires much in The Federalist
Papers. It is this conception of limited government, with its origins in
classical liberalism, which I believe to be most appropriate to our condition
as persons living in a society which is not unified by any single tradition. It
is a conception which should be studied with special care, I believe, by
reflective conservatives and by liberals who have not lost the skill in
self-criticism on which they pride themselves.
The conception I wish to commend here confronts serious difficulties,
all of which arise from the fact that it appears to go against the spirit of the
age. In the flI'St place, it is not clear what are the institutional forms
appropriate to a state which respects cultural diversity and does not seek to
bolster or embody any specific form of cultural identity. The example of the
Habsburg Monarchy, with which I began and which certainly did shelter a
myriad of nationalities and traditions, is not an encouraging one for us.
Contemporary sentiment is relentlessly inimical to imperial institutions,
despite the evident historical fact that their desuetude or destruction has
mostly been attended by dictatorship or tribalism. Further, the instance of
Switzerland, which Kolnai cites as evidence of the possibility of several
linguistic and religious communities being integrated within a single poli-
tical order based on extensive decentralization and localization of govern-
mental functions, does not seem easily exportable. Such success in
 The politics of cultural diversity 269
harbouring diverse communities within a single political order depends in
Switzerland on political traditions which are centuries old and which have
very distinctive historical origins. Moreover, the record of local or federal
schemes of governmental devolution and decentralization is not elsewhere
particularly impressive.
In Britain, local government has proved more vulnerable to capture by
extremists than has national government, and has been the scene of many
conflicts between ethnic and cultural majorities and minorities. Again, in
the United States the massive apparatus offederalism, separation of powers
and judicial review has not prevented (and may well have facilitated) the
extensive politicization of social life via the exploitation of litigation and
the invocation of constitutional rights, whose content is often not liable to
strict construction. There is, finally, a danger in all schemes of con-
sociational or confederal devolution of government powers, that (save in
exceptional circumstances, as in Switzerland) they leave the association of
states open to conquest or subversion by external aggressors. For these
reasons, we go up a blind alley if we suppose that we can forge the political
institutions appropriate to a circumstance of cultural diversity by recourse
to traditional devices of devolution and localization which have nearly
everywhere failed.
There is a deeper difficulty in the view I have been developing of the
relations between political order and cultural identity, and that is a dilemma
of allegiance. Political orders which have successfully transcended the need
for moral solidarity or community have, historically, either been empires
(as with the Romans and the Habsburgs), religious institutions (such as
those of medieval Christendom and Islam in their periods of greatest
toleration) or else monarchies, as in the splendidly anachronistic case of the
United Kingdom. None of these sources of allegiance is available on any
large scale today. There are no true empires left, few monarchies, and the
exemplars of religious political order with which we are, alas, becoming
increasingly familiar are examples of a rabid and barbarous funda-
mentalism, as in Iran. We may go yet further. Allegiance to liberal orders,
where it has not had its roots in religious faith, has typically been imposed
by secular myths which few of us any longer take seriously. Thus liberalism
in France and America was sustained by the historical theodicy of the
Enlightenment, with its mythologies of natural rights and global progress.
These are the myths which suffuse modernity and ground the modem
project of giving moral and political life a rational foundation. When we
cease to subscribe to them, we acquire the paradoxical character of post-
modems, heirs to all the achievements of modernity, but not to its seminal
myths. From the perspective of my problem here, the question naturally
arises, what might sustain a political order that encompasses many tradi-
270 Questions
tions and varieties of human identity, if both the traditional and the modem
sources of allegiance to such an order are unmistakably on the wane?
I address this question in the hope that the dilemma it marks is not of the
Polish variety - insoluble, but not serious - which it is the fate of those
subject to Communist domination to suffer. I hazard the guess that we may
go some way to resolving it by looking back to the early modem period,
when many of our current dilemmas were perceived with a drastic clarity
which we have lost. It is perhaps in the works of Hobbes, who wrote for an
age of religious wars and barbarous movements much like ours, that we
glimpse the outlines of a form of government suited to our circumstances.
It is a form of government devoted to securing the peace, and that first of
all, thereby leaving the largest space for liberty of thought and action. It is
a unitary form of government, without complicated devices for devolution,
in which all or most of the activities not essential to the primary peace-
keeping task of the state are left to private initiative. It is in this notion of a
form of government having such strictly limited functions that we may,
possibly, find a clue to our present dilemma. What it suggests is the salience
to our condition of a state that is strong but small, in which the little that is
not privatized is centralized, and in which practitioners of diverse traditions
are left at liberty, so long as they do not disturb the common peace, to refine
and develop their forms of life. This is a form of government devoted not
to truth, or to abstract rights, and still less to any conception of progress or
general welfare, but instead one which by securing a non-instrumental
peace creates the possibility of civil association.
It is in this possibility, and perhaps in this alone, that we may hope for a
resolution of the perplexity with which I began my reflections. It may still
be doubted, all the same, if the kind of state I have envisaged could
command allegiance when it lacks the support of an inspiring mythology.
Here I would like to suggest the possibility that, in recognizing that we lack
the moral solidarity sought by socialists and most conservatives, we may
yet be able to draw on the resource of another kind of solidarity - that of
civilized men and women, practitioners of different traditions, who never-
theless have in common a perception of enmity in regard to the totalitarian
states and rebarbarizing movements of our time. We - the heirs of European
traditions who have enjoyed the freedoms of civil association, or who have
come within its sphere by forced or chosen migration - are few, feeble and
not mankind. We know that (as Hume observed of liberty in England) our
freedom is a singularity, not the upshot of any inevitable source of events,
and that it has no future that is guaranteed. We do not pretend that our
identities express the essence of the species; we recognize them to be
products as much of chance as of choice. In short, we take ourselves as we
are, and we are not ashamed of our identities. It is this sensibility which I
 The politics of cultural diversity 271
defend here and which, despite its clumsiness, I am tempted to call post-
modern liberal conservatism. 14
The name we give to this sensibility is unimportant. What is important
is the possibility of a fonn of political solidarity that does not depend on
shared moral community, but only on the mutual recognition of civilized
men and women. Such solidarity may be, as Spinoza says of wisdom, 'as
rare as it is precious'; amt the freedom that it expresses and engenders may
well prove to be an exception in the life of the species. Yet, so long as we
can learn the skill of defending with inflexible resolution an order we know
to be underwritten neither by nature or history, we have reason to hope that
the idea of a society of many distinct but interpenetrating traditions, a
society in which men and women come to respect and cherish their
differences and are ready to act together to protect them, is more than an
idle dream. 15
19 Conservatism, individualism and
the political thought of the
New Right

Perhaps the most remarkable and among the least anticipated developments
in political thought and practice, throughout the Western world in the
19808, was the conquest of conservatism by the ideas and doctrines of the
New Right. This conquest was nowhere total, and it was not by any means
universal. It can be justly tenned hegemonic, nevertheless, in precisely the
Gramscian sense, in that conservative parties, governments and intellectual
journals came to be dominated by a discourse and an agenda of policy that
emanated from thinkers of the New Right and embodied a searching and
often harsh critique of conservative thought and practice during the thirty
years or so that followed the Second World War. Indeed, the hegemonic
character of the political thought of the New Right - a hegemony that may
prove to have been fragile and ephemeral, but which at present shows only
few and small signs of strain - was attested by the fact that during the 19808
the parties of the Left felt constrained to adopt much of the discourse, and
many of the policies, that conservative governments had imbibed from the
theorists of the New Right In one case, that of New Zealand, it was the
Labour Party that implemented the most radical and far-reaching pro-
gramme of economic refonn (on lines suggested by New Right thinkers)
that has been attempted in any Western democratic state - a refonn pro-
gramme that, despite its upheavals and difficulties, has compelled the
conservative National Party in New Zealand to adopt a rhetoric of eco-
nomic liberty and market competition that is very different from the
Keynesian and protectionist rhetoric and policy that are its historical
inheritances. It would be rash to essay the judgement that the policy
achievements of governments animated by the ideas of the New Right are
irreversible. It is reasonable, at the same time, to affinn that a sea change
has occurred in political life in the Western democracies, and more parti-
cularly in conservative thought and practice in the English-speaking
nations, that has many of the marks of irreversibility.
 The political thought of the New Right 273
The hegemony of the New Right within conservatism emerged in a very
definite historical context. Even if, as I have intimated, it will probably
have some consequences for both conservative and socialist policy and
rhetoric that are irreversible, it is sensible to suppose that the hegemony of
the political thought of the New Right, both within and outside con-
servatism, will be limited by the historical contingencies that brought it into
being. This is so, most "articularly, in regard to its domination of con-
servative thinking. For, though the relations of the New Right with the
received tradition of conservative thought and practice are highly complex
and controversial, there can be no doubt that the novel orthodoxy of New
Right thought within conservatism has initiated as much rupture with recent
practice as it has (on some accounts) restored conformity with an earlier
and supposedly a purer conservatism. My object in this exploration will be
to attempt to define the content, limits and varieties of doctrine on the New
Right, and to specify the affinities and tensions that have arisen between the
New Right and the received inheritance of conservative practice. Except by
way of an occasional comparison with conditions in the United States, I
shall confine myself to the British experience. And I shall focus my
inquiries around the central and fundamental question of the place and
character of individualism within conservative doctrine - a question that
has been answered very differently both by thinkers of the New Right and
by their conservative critics.
A rehearsal, not of my arguments, but of my conclusions, may be
apposite. I shall conclude that, whereas the New Right correctly perceived
that a generation of conservative practice had led modem Western demo-
cratic states into a dead end of corporatist stagnation in which a ratchet-
effect operated that moved the political centre steadily leftwards, its
theorists consistently neglected the cultural inheritance which is the matrix
of a stable capitalist order. In part because of their debts to the rationalist
tradition of classical liberalism, the principal theorists of the New Right
failed to perceive the dependency of individualist civil society on a
dwindling but real patrimony of common ideas, beliefs and values. In some
cases, they subscribed to the unrealizable and dangerous utopian project of
a minimal or neutral state enforcing a regime of common rules that is not
underwritten by a fund of common culture. In no case did the thinkers of
the New Right undertake the historical researches necessary in order to
illuminate the sources of the stability of the capitalist order in Britain or the
United States. The political thought of the New Right, even in its subtlest
expressions (as in Hayek), transmitted to conservatism an abstract rational-
ism and legalism that occludes serious theorizing of the conditions under
which market capitalist institutions have for centuries enjoyed an almost
unchallenged hegemony in Britain and the United States. In other words,
274 Questions
the New Right has in its theorizing failed to grasp the historical and cultural
presuppositions and limits of the kind of civil society they seek to maintain,
restore or enhance. And this theoretical neglect has disabled the policies of
governments animated by the thinking of the New Right, in that policy (as
distinct from rhetoric) has been concerned almost solely with securing the
legal and economic conditions of market competition and thereby of
general prosperity, and has only rarely and inadequately addressed the
cultural conditions that undergird and sustain a stable market order. In so
far as the New Right ever nurtured a hegemonic project, this has been
compromised in political practice by the blind-spot it has contained regard-
ing the importance of the culture of individualism of which a market
economy is only the visible part.
It is to this culture of individualism, ancient in England and resilient if
insulted and injured, even in our own times, that theoretical inquiry and
historical research should now tum. Here the argument I develop diverges
from that of conservative critics of the neo-liberal New Right, such as
Roger Scruton, who have perceptively and powerfully criticized the neglect
within neo-liberalism of the common culture that founds Western civil
society and enables it to reproduce itself across the generations. My argu-
ment diverges from this conservative critique, in as much as it maintains at
once that our common cultural inheritance is individualist, and that its
political embodiment cannot always, or even as a rule, coincide with that of
the modem nation-state. The character of this individualist inheritance, as
we receive it at present, and the forms in which it may enjoy a political
embodiment, are very difficult issues, about which I can here only offer
inconclusive speculations. The upshot of my reflections, however, is that,
since individualism as a form of life pre-dates liberalism as a political
doctrine by centuries, it is reasonable to expect (and to hope) that indivi-
dualist culture will survive the demise of liberalism. And, since indi-
vidualism and liberalism are not to be confused, it is a mistake on the part
of conservative critics of the New Right to suppose that civil society can do
without its individualist inheritance, or find a surrogate for that inheritance
in the modernist project of integral nationhood.
What, then, was the New Right?' At the level of policy, it is not difficult
to identify the salient themes of New Right thought. Policy animated by the
ideas of the New Right sought to dismantle the corporatist institutions built
up in Britain in the post-War period, to limit government and at the same
time to restrain the power of inordinate organized interests such as the trade
unions, to achieve a stable currency and abandon deficit financing and, in
general, to engineer a transfer of initiative and resources from government
to civil society that was massive and politically irreversible. In practical
terms, such a policy orientation expressed itself in measures for the
 The political thought o/the New Right 27S
privatization of state-owned industries, in the Medium Term Financial
Strategy, tax reduction, a curb on public expenditure, and a variety of
supply-side measures deregulating prices, wages, rents and some planning
controls. It is not such detailed policy measures that will interest the
historian of the New Right, however, but rather their philosophical and
theoretical inspiration. The intellectual perspective which infused these
policies was not that which had dominated post-War British conservatism;
it came from outside the Conservative Party, from the works ofF. A. Hayek
and Milton Friedman, and from the free market think-tanks, above all, from
the Institute of Economic Affairs. 2 The pedigree of this perspective on
policy and society was classical liberal, not conservative. This is to say that
it was strongly and sometimes stridently individualist, it sought to reduce
government to an indispensable minimum, and (except, to some extent, in
the work of Hayek]) it concerned itself very little with the cultural or social
conditions of a stable restoration of market institutions. In economic policy,
the impact of classical liberal thought, especially as developed and applied
(during decades of neglect by the mainstream of academic life) by the lEA,
was crucial and indispensable. The classical liberal inheritance, con-
sistently and judiciously implemented in the early years of the Thatcher
period, enabled government to act to dismantle, probably for good, a
significant portion of the detritus of corporatist institutions left by a genera-
tion of passive and reactive conservative government and policy-making.
As it was embodied in the early years of Thatcherism, the classical
liberal orientation of much policy could also legitimately claim a genuine
conservative pedigree. The allegation of the Tory 'wets' ,4 that Thatcherism
represented a rupture in an otherwise unbroken tradition of conservative
thought and practice, does not survive historical scrutiny. Nineteenth-
century conservative government, whether that of Disraeli or of Salisbury,
conceived itself as superintending the institutions of civil society, and
occasionally supplementing them, but not as being in the business of
managing the economy. This is to say that, at least until the 1930s, British
conservatives conceived government as the guardian of civil association (in
Oakeshott's terminologyS). The idea of government as an enterprise asso-
ciation - as a sponsor or pacemaker of economic growth, say - is in fact
quite alien to the British conservative tradition. In this respect, then,
Thatcherism can make a genuine claim to have revived or restored an
earlier tradition of conservative government.
There is another, and more fundamental reason in support of the thesis
that Thatcherism embodies a genuine British, or more precisely English
tradition, and is not an incursion into British political life of an alien liberal
or conservative doctrine. This is in the historical fact, profoundly explored
by Alan Macfarlane6 that English life, contrary to the conventional
276 Questions
historiography of PolanyP and of Marx, has had an individualist character
at least since the thirteenth century. The idea that there was, in the seven-
teenth century or earlier, a Great Transfonnation of English life from
communal to individualist forms, an idea that has entered political thought
from the work of C. B. Macpherson,' does not accord with demonstrable
facts about English history - facts to do with personal mobility, land use
and the pattern of family life. If, as MacFarlane has persuasively argued,
England was an individualist society immemorially, then the project that
animated Thatcherism - the project of limiting government and reviving
civil association - is one that seeks to reassert the most ancient and
fundamental English traditions.
Neo-liberalism and Thatcherite conservatism are, then, seeking to
restore and reproduce an English individualist culture that is our historical
inheritance. Where they appear to differ is in their recognition, or neglect,
of the historical conditions and inheritance which allowed this culture to
sustain itself for so long. As Jonathan Clark has argued' English indivi-
dualism - the culture that grounded English civil society - was not liberal
but instead authoritarian individualism. This is to say that it depended on a
nexus of beliefs, practices and inhibitions which conferred legitimacy on it
and constrained the corrosive tendencies about which the Scottish writers
and particularly Ferguson, were concerned in their theorizing about the
potentially self-destructive effects of the anonymity and moral laxity that
were latent in individualist life. These tendencies were inhibited in English
authoritarian individualism by the strength in England of a common moral
culture and, more particularly, by the authority of the Anglican church (and,
later, by the moral discipline imposed by non-confonnist Christianity).
Similar religious traditions in the United States for a century or more
constrained the tendencies toward anomie and hedonism which (as Tocque-
ville perceived) were otherwise present in the individualism of a culture
lacking in the hierarchies that had structured individualism in England.
Thatcherite conservatism has distinguished itself from neo-liberalism,
and has established its affinities with American neo-conservatism, by its
reiterated emphasis on the familiar and religious values that legitimated
capitalist institutions in their Victorian heyday. Thatcherite conservatism
has the advantage over neo-liberalism, for this reason, in that it explicitly
addresses the character of the cultural matrix of individualist civil society.
It is far from clear, however, that the invocation of Victorian values -
religious and familial - can do the job of legitimating market institutions
that is required of such values by the Thatcher project. This is so for several
reasons. In the first place, England is in the late-twentieth century a
massively secularized society. In this it resembles most closely the Scandi-
navian countries and Japan, and differs sharply fonn the United States
 The political thought of the New Right 277
(which in tum accounts for the non-existence in England of anything
resembling the American phenomenon of the Moral Majority). It is wholly
anachronistic to expect that Christianity, and more particularly the Church
of England, will ever have the political significance in England that it
possessed throughout the seventeenth, eighteenth and nineteenth centuries.
Indeed, in so far as the Anglican church has had influence on political life
in England, it has been to further the tendencies whereby market institu-
tions and individualist civil society are delegitimated, since the dominant
part of Anglican social theology has for decades been infused with the
anti-capitalist mentality that pervades intellectual life in England more
generally.
The second reason why the Thatcherite appeal to the common cultural
inheritance of individualism cannot work is the substantial enhancement of
various sorts of pluralism within English society. The mass immigration of
the 1960s has contributed to English life an ethnic and a religious pluralism
that is unalterable and irreversible in any foreseeable future. Given the
self-assertion of British Islam, and its demands for separate state-funded
schooling, it is likely that this pluralism will increase, rather than diminish,
over the coming decades. A similar, though more diffuse and less politic-
ally visible pluralism is widespread in English society in attitudes to
marriage, sexuality and other belief-dependent institutions and practices. If
the English common culture was ever a seamless garment, it is no longer so.
It is questionable whether a strong and resilient common culture any longer
prevails in England. It is beyond reasonable doubt that its content has been
attenuated, and its significance in performing a legitimating role in respect
of civil society in England weakened. Nor is there any reason to suppose
that Thatcherite rhetoric has had any transforming impact on the diverse
and indeed fragmented culture which now exists in England.
What might supplement the dwindling common culture as the legiti-
mating force that sustains a capitalist regime in Britain? In practice, both
Thatcherites and neo-conservatives in the United States have relied - thus
far successfully - on economic growth, and the manifest failure of both
post-War macroeconomic management and of socialist command eco-
nomies, to sustain the electoral support needed for neo-liberal economic
policies. Unless the most optimistic among the American supply-siders 1o
are right, and the business cycle has been abolished, it seems imprudent and
complacent to suppose that the hegemony of market institutions can rest
forever on increasing prosperity. (Environmental concerns may in any case
constrain the pursuit of constant economic growth.) In the context of the
mass democracies of contemporary Britain and the United States, there is
every reason to suppose that a period of protracted economic difficulty
would swing the political pendulum back to interventionist, and in the
278 Questions
United States protectionist, economic policies, and could well end the long
tenure in power of the Conservatives and the Republicans. This is not to say
that we need fear in England a return to full-blooded socialism. The intel-
lectual hegemony of the New Right in economic theory, and the collapse of
socialism as a doctrine in both Eastern and Western Europe, rule out any
doctrinal survival of socialism. Again, many of the policies implemented
under the incisive Thatcher government are politically irreversible. (Most
privatizations fall into this category.) What is to be feared in the wake of
protracted economic difficulty is not socialism, but a return to corporatist
and interventionist policies that not only injure the autonomous institutions
of civil society but also compound the economic difficulties that evoked
them. In the United States, more than in Britain, the end of the long boom
may well see a serious crisis in financial institutions weakened (as much of
corporate America is) by inordinate debt, and a policy response which
reinvigorates the American tradition of combining isolationism with protec-
tionism. In the British case, where fiscal prudence has been maintained, we
are likely to see a return to 'stop-go' economic management, against a
background of high inflation, high interest rates and revived union power.
Exclusive or primary reliance on economic growth as the legitimating
condition of market capitalist institutions in the context of mass democracy
is unwise and will predictably fail. An alternative, suggested by High Tory
critics of the neo-liberal components in Thatcherism such as Roger Scruton,
is little more promising. Scruton argues that it is nationhood that provides
the indispensable foundation of liberal civil society in our time. Scruton's
argument has considerable merit in exposing the insufficiency, and the
inconsistency, of those such as Parekhll who have maintained that a multi-
cultural society can be supported (as supposedly in America) solely by
liberal legalism and constitutionalism. Scruton's positive endorsement of
nationhood as the condition of a stable civil society is nevertheless un-
persuasive. In historical terms, what has sustained the most successful civil
societies has not been nationhood but monarchy (as in the British case) or
empire (as in the Hapsburg Empire which Scruton discusses.) If civil
society has a future in contemporary Spain, again, it is in virtue of
monarchy, since Spain as it now exists is little more than a composite state
which itself is an artefact of monarchy. Scruton also underestimates the
modernity of nationhood. As late as the 1870s, a majority of French school
children identified themselves by their provinces, and not as French. In
Europe, nationhood is in most countries the construction of political 61ites,
not the expression of the solidarity of peoples. Ironically, the modem
European nation-state is mostly the creation of nineteenth-century classical
liberalism, which was an agency for centralization and against localism. 12
Examples such as Poland and Japan, where nationhood and statehood
 The political thought of the New Right 279
coincide, are in fact highly exceptional, and not at all the rule.
Minogue and Kedourie 13 are closer to the mark than Scruton, when they
observe that nationalism is a modernist project inspired by romantic
doctrine. In the post-colonial world, it has acted as a destroyer of traditional
forms of life and as a pace-maker for Communism. Where it has expressed
a genuine, pre-political or social popular solidarity, it has typically been
one that was once embodied in statehood, and is preserved as such in
popular memory - as in the Baltic states: (Only the Kurds come to mind as
a possible exception to this generalization.) Finally, when nationalism has
arisen within a civil society governed by a rule of law, as in the Sikh
phenomenon in India, it has acted as a threat to civil society. In all these
respects, the modernist cult of nationalism is akin to the no less modernist
phenomenon of fundamentalism. With rare exceptions, such as Poland and
the Baltic states, it is characterized by enmity to civil society and not by
support for it.
Neither neo-liberalism nor its conservative critics, if my reasonings are
sound, recognize the cultural foundations and historical limits of indivi-
dualist civil society. In this they may be influenced by a conflation of
individualism with civil society that is valid in the English and American
contexts but not in a world context. The English and European theorists of
civil society - Hobbes, Locke, Hegel, Touqueville and others - represented
as an inherent and universal truth the connexion between individualist
culture and a civil society encompassing market institutions. The experi-
ence of the last decade or so, however, in which powerful and stable market
economies have emerged in the non-individualist cultures of Japan and
Korea, for example, suggests that the connection between individualism
and civil society is contingent and not necessary. This recent experience
supports the Scottish and neo-conservative fear that, in slowly consuming
its pre-individualist foundations, Western civil society may (as Schumpeter
speculated l4) be a self-limiting historical episode.
For the present, no doubt, such fears are exaggerated - although it
remains to be seen how American culture responds to the novel experience
of continuing, and probably inexorable, economic decline. The concerns I
have explored suggest, however, that there are no easy answers to the
question, what is now the content of the common culture that supports and
legitimates market capitalism? And this in tum suggests that the policy
agenda of post-Thatcherism will need to be judicious and balanced when it
comes, as it inevitably must, to seek to retard or reverse the further erosion
ofthe moral capital on which market institutions must rely. The crusade for
Victorian values is, as has already been intimated, an exercise in anachron-
ism. This is not to say that there cannot be serious policies aiming at
strengthening what remains of the common individualist culture in
280 Questions
England. In education, in social and welfare policy, there are desirable
refonns, well within the limits of what is politically possible, which could
well strengthen the family and reinforce the popular sense of individual
responsibility. Enforcing the obligations of the delinquent parent in one-
parent families, wherever this is feasible, is only one rather obvious sample.
In education, a Natural Curriculum which enforces literacy in English,
which promotes an understanding of British history, and which perhaps
incorporates a civic education, could do something to restore the fabric of
the common culture.
No government can create a culture where it is absent, or engineer the
renewal of a culture where it is moribund. As Wittgenstein put it, deliber-
ately continuing to renew damaged or weak traditions is as hopeless as
trying to repair a broken spiders web with one's bare hands. Government
has, nevertheless, and contrary to neo-liberalism, a vital role in regard to the
culture of individualism. Negatively, it can refrain from damaging it, by
abandoning policies which create dependency, overturn traditional disci-
plines and encourage irresponsibility. 15 Positively, it can provide a frame-
work of policy and institutions within which individualist culture is
protected and nurtured. In so doing, it is tending the pre-individualist fonns
of life - family and loc!,l life, for example - that are the soil on which
individuality grows and thrives. It is only by addressing these cultural
pre-conditions of individualism, and framing sensible policy in respect of
them, that the current hegemony of market institutions can be preserved,
and a secure future for civil society achieved.
The philosophical insight which such desirable reforms embody is that,
even where (as in England) individualist culture is immemorial, it is not
primordial, but depends on a background of beliefs, values and institutions
which both fonn and constrain individuality. It is this cultural patrimony of
individualism that neo-liberal thought ignores, and which Thatcherite
policy has insufficiently addressed. Individuality, as we know it, is not a
natural fact, but a cultural achievement. The persons who inhabit and
reproduce a liberal civil society are not biological organisms but historical
artefacts, products of a long and often arduous struggle to build up and
maintain specific sorts of institutions and practices. With us, at any rate,
individualism and civil society have always gone together, and civil society
has no future apart from the culture of individualism. Ironically, it is this
culture that is being threatened, especially in the United States, by a
left-liberalism which regards the undergirding institutions of liberal civil
society - marriage, private property and so forth - as constraints upon, and
not as conditions of individual freedom. The irony in intellectual history is
that it is liberalism, at least since Mill,16 that has transmitted the enmity to
civil society and to the moral traditions that support it that has long been
 The political thought o/the New Right 281
evident in modem Europe in the antinomian character of the anti-individual
(in Oakeshott's expression'7) - the dislocated, anomie character, typified
by Rousseau, who seeks a release from the burdens of freedom in the
recreation of a form of communal solidarity that, in England at any rate,
was always largely imaginary.
Individualism in England is not - as it is in Turkey, say - an imposition
from above, generated hy legality and military force. It is our common
cultural inheritance. But, as I have argued throughout, the renewal across
the generation of this individualist tradition cannot safely be left to chance.
Government has a vital role in preserving, or repairing, the framework of
practices and replenishing the fund of values, on which individualism
depends for its successful reproduction. If Thatcherism was ever an hege-
monic project - and that is how is should rightly have been conceived - it
condemned itself to failure by its faulty grasp of the cultural preconditions
of market liberalism. It may be that, despite its continuing economic
success, Western civil society is in an irreversible decline, which govern-
ment can retard but not reverse. Even in this worst case, it remains the
intellectual responsibility of conservatives in the era of post-Thatcherism to
theorize the conditions under which civil society in England may retain its
vitality. In this task they are distracted by neo-liberal and holistic High Tory
doctrine which fails to attend to the minute particulars of our individualist
inheritance and which does not illuminate the deeper tradition that it
expresses. Civil society - which is to say a society that, under a rule of law,
allows the great majority of transactions to be conducted in autonomous
institutions, and in which the bulk of economic life is realized in market
exchanges among legally recognized owners of private property - is the
only form of life in which civilization has been enjoyed by modern Euro-
peans. Oakeshott has characterized civil society in the English form with
un surpassable clarity:
What, then, are the characteristics of our society in respect of which we
consider ourselves to enjoy freedom and in default of which we would
not be free in our sense of the word? But first, it must be observed that
the freedom we enjoy is not composed of a number of independent
characteristics of our society which in aggregate make up our liberty.
Liberties, it is true, may be distinguished, and some may be more general
or more settled and mature than others, but the freedom which the
English libertarian knows and values lies in a coherence of mutually
supporting liberties, each of which amplifies the whole and none of
which stands alone. It springs neither from the separation of church and
state, nor from the rule of law, nor from private property, nor from
parliamentary government, nor from the writ of habeas corpus, nor from
282 Questions
the independence of the judiciary, nor from anyone of the thousand
other devices and arrangements characteristic of our society, but from
what each signifies and represents, namely, the absence from our society
of overwhelming concentration of power.
Similarly, the conduct of government in our society involves a sharing
of power, not only between the recognised organs of government, but
also between the Administration and the Opposition. In short, we con-
sider ourselves to be free because no one in our society is allowed
unlimited power - no leader, faction, party or 'class', no majority, no
government, church, corporation, trade or professional association or
trade union. The secret of its freedom is that it is composed of a
multitude of organisations in the constitution of the best of which is
reproduced that diffusion of power which is characteristic of the
whole. II
If conservatives are serious in the commitment to the project of achieving a
political hegemony for this form of life, they must return their theorizing to
history. For, in returning us to a self-understanding hitherto clouded by
defective history - liberal, Marxist or High Tory-holist - they will thereby
strengthen the intellectual hegemony which the thought of the New Right
has achieved in economic theory and policy. And, in so doing, they will
make a small, but vital contribution to the success of the political project of
securing hegemony for civil society - by creating a discourse in which its
most central institutions can be theorized as historical achievements.
20 What is dead and what is living in
liberalism?

This book is written for those who are in sympathy with the spirit in which
it is written. This is not, I believe, the spirit of the main current of European
and American civilisation. 'The spirit of this civilisation makes itself mani-
fest in the industry, architecture and music of our time, in its fascism and
socialism, and it is alien and uncongenial to the author. This is not a value
judgement. It is not, it is true, as though he accepted what nowadays passes
for architecture as architecture or did not approach what is called modem
music with the greatest suspicion (though without understanding its lan-
guage), but stiU, the disappearance of the arts does not justify judging
disparagingly the human beings who make up this civilisation. For in times
like these, genuine strong characters simply leave the arts aside and tum to
other things and somehow the worth of the individual person finds expres-
sion. Not, to be sure, in the way it would at a time of high culture. A culture
is like a big organisation which assigns each of its members a place where
he can work in the spirit of the whole; and it is perfectly fair for his power
to be measured by the contribution he succeeds in making to the whole
entetprise. In an age without culture, on the other hand, forces become
fragmented and the power of an individual man is used up in overcoming
friction. But energy is still energy and even if the spectacle which our age
affords us is not the formation of a great cultural work, with the best men
contributing to the same great end, so much as the unimpressive spectacle
of a crowd whose best members work for purely private ends, still we must
not forget that the spectacle is not what matters.
I realise then that the disappearance of a culture does not signify the
disappearance of human value, but simply of certain means of
expressing this value, yet the fact remains that I have no sympathy for
the current of European civilisation and do not understand its goals, if it
has any. So I am really writing for friends who are scattered throughout
the comers of the globe.
L. Wittgenstein l
284 Questions
INTRODUCTION
In this chapter of eight sections. I will consider what are the most essential
or fundamental constitutive elements of liberalism. In the first four sections
of the chapter. I will consider how these fare when subjected to the force of
radical value-pluralism. I conclude that none of the four constitutive ele-
ments of doctrinal liberalism - universalism. individualism. egalitarianism
and meliorism - survives the ordeal by value-pluralism. and that liberalism.
as a political philosophy. is therefore dead. In the last four sections of the
paper. I argue that what is living in liberalism is the historic inheritance. now
re-emerging in parts of the world in which it was suppressed. of a civil
society whose institutions protect liberty and permit civil peace. Arguing
that. over the long run of human history. civil societies are not the only
legitimate societies. I conclude nevertheless that a liberal civil social society
is the best one for cultures. such as all or virtually all contemporary cultures.
which harbour a diversity of incommensurable conceptions of the good.
The position defended here is post-liberal in that it rejects the founda-
tionalist claims of fundamentalist liberalism. This is to say that it denies that
liberal regimes are uniquely legitimate for all human beings. Human beings
have flourished in regimes that do not shelter a liberal civil society. and
there are fonns of human flourishing that are driven out in liberal regimes.
Liberal orders have. then. no universal or apodictic authority. contrary to
liberal political philosophy. At the same time. it will be my argument that
the four constitutive features of liberalism can be given a rational defence
when they are contextualized and historicized as features of late modem (or
early post-modem) societies and polities. Though there will not be (and
need not be) any convergence on a single fonn of government. nevertheless
all. or nearly all fonns of government that allow for commodious living will
in the foreseeable future be ones that shelter the institutions of civil society.
These institutions. in their tum. will be animated by the practice of liberty
that is our historical inheritance. Our task as theorists of our historic
inheritance of civil society is. on this view. to seek for a better under-
standing of the practice and culture of liberty that is our patrimony. and
which is now exfoliating in many varieties in several parts of the world. If
the argument of this paper is sound. the death of foundationalist liberal
political philosophy should not concern us greatly. since it enables us to
return to its historic source in the practice of liberty.

DOES LmERALISM HAVE A mSTORY?

What was liberalism? In order to answer this question. we need to ask
another: Does liberalism have a history? This latter question is not as
 What is dead and what is living in liberalism? 285
absurd as it seems, since it is the result of recent work, by Pocock and
others, to controvert the standard liberal self-interpretation. On this con-
ventionalliberal view, memorably expressed by J. S. Mill,:! modern liberal-
ism is merely the contemporary expression of a tradition of free-thinking
and antinomianism that is virtually immemorial, but extends back at least
as far as Socrates. The upshot of recent scholarship has been to deconstruct
this liberal self-interpremtion, and to theorize 'liberalism' as a much more
discrete episode in the history of thought. On the Pocockian view, 3 at any
rate, it is an error to view Locke, Kant, Smith and Mill, say, as expositors
of a single tradition of ideas, if only because the Scottish thinkers, for
example, who are often regarded as the prototypical classical liberals -
Smith and Ferguson, say - were themselves deeply influenced by traditions
of civic humanism and classical republicanism whose relations with
'liberalism' are, at best, ambivalent. On this view, if 'liberalism' has a
history, it begins sometime early in the nineteenth century, when the word
'liberal' acquired its contemporary meaning. For exponents of this view, it
is a sort of whiggish illusion that is fostered in respect of liberalism, when
it is theorized as existing before the French Revolution, and as having a
genealogy extending back to the ancient world.
It is, fortunately, unnecessary to attempt to settle the historical con-
troversy which this view has generated. For, on this view and also on the
standard liberal self-interpretation, modern liberalism received its para-
digmatic statement as a comprehensive ideology in the writings of J. S.
Mill. To be sure, other, earlier writers are on both views clearly identifiable
as expositors of a liberal intellectual tradition - Constant and Tocqueville,
for example, and other disillusionist liberals who wrote in the wake of the
French Revolution. It is in J. S. Mill, however, that (on all views) we find
the liberal syndrome of ideas most explicitly and recognizably articulated.
With Mill, one is in the same company as later, and latter-day liberals - a
company that includes Dworkin and Rawls, among others.
To affmn the identity of a liberal intellectual tradition spanning Mill and
Rawls is not to deny that it is a very complex tradition, containing recessive
and dialectical moments. It is not to pass over, or to seek to suppress, the
new liberalism of the late nineteenth-century English Hegelians, such as
Bosanquet and Green, or the somewhat holistic, even communitarian
liberdlism of such as Hobhouse. Nor is it to endorse the mistaken view,
earlier exposed in my own writings,4 that liberalism can be easily and
uncontroversially divided into classical and revisionist varieties, with the
classical liberals being devoted primarily to negative liberty, and the
revisionists being a species of welfarist or egalitarian liberal. One result of
Pocock's work will stand, even if his substantive claims are rejected or
qualified: a methodological scepticism about intellectual historiography
286 Questions
which commends suspicion about history that is written in terms of endur-
ing concepts, after the fashion of Lovejoy. This scepticism should caution
us against seeking too much continuity or coherence, even in a liberal
intellectual tradition that starts with Mill.
This methodological preamble aside, we can still identify a matrix of
ideas, recognizable by all or most liberals, and by their critics, as con-
stituting the liberal syndrome. Following earlier work of mine,' I shall
specify four key ideas, found in J. S. Mill. and echoed in virtually all
subsequent liberal writers. First. there is the idea of moral or normative
individualism - the idea that. since nothing has ultimate value except states
of mind or feeling, or aspects of the lives of human individuals, therefore
the claims of individuals will always defeat those of collectivities. institu-
tions or forms of life. This claim. as Raz has noted,6 is a variant on the thesis
of humanism - the modernist belief, which has nevertheless deep roots in
Judaeo-Christianity, that only human beings and their forms of life have
ultimate value. A second element in the liberal syndrome is universalism -
the idea that there are weighty duties and/or rights that are owed to all
human beings, regardless of their cultural inheritances or historical circum-
stances, just in virtue of their standing as human beings. This strong
anti-relativist claim is, in my view, essential to any liberal view that is
constitutively liberal - any view, such as Kant's, Mill's. or that of the early
Rawls. that regards a liberal regime as, at least potentially, the best and
uniquely right one for all mankind. even if in many contexts approxi-
mations of it are all that can be achieved. In other words, without denying
that other regimes may be necessary stages on the path to a liberal regime,
and perhaps even going so far as to recognize that, for some peoples, a
liberal regime will never be fully or stably achieved. constitutive, funda-
mentalist or doctrinal liberalism must still affirm that all political institu-
tions are to be assessed on the single scale that measures their
approximation to a liberal regime. This second idea leads one, naturally
enough. to the third element in the liberal syndrome. namely meliorism. By
this is meant the view that, even if human institutions are imperfectible,
they are nonetheless open to indefinite improvement by the judicious use of
critical reason. To say this is to say that, though no contemporary liberalism
can credibly presuppose historical laws guaranteeing inevitable human
improvement, equally, no contemporary liberalism can do without some
idea of progress. however attenuated. The fourth and final element of the
liberal syndrome issues intelligibly from the first three - liberal egali-
tarianism. By this is meant the denial of any natural moral or political
hierarchy among human beings, such as was theorized by Aristotle in
respect of slavery and by Filmer of absolute monarchy. For any liberal, in
other words, the human species is a single-status moral community, and
 What is dead and what is living in liberalism? 287
monarchy, hierarchy and subordination are practices that stand in need of
an ethical defence. The implications of this position for ideas of consent as
the basis of political obligation, and of democracy as an institution devised
to meet requirements of public justification of the use of political coercion,
are clear enough.
The claim made here is that, even if it be true that not all liberals
subscribe to every one Of these theses, together they add up to a system of
ideas that most contemporary liberals would find difficulty in rejecting. All
four ideas are clearly present in the thought of J .S. Mill, who for the
purposes of my argument I will treat as the paradigmatic liberal. It should
be noted, finally, that I have not included in the four ideas constitutive of
the liberal syndrome any specific prescriptive principle - say, a principle
prioritizing liberty or equality. I have not done so, precisely in order to
advance a conception of liberalism that is sufficiently copious, catholic and
eclectic as to be tolerant of the entire extended family of views that have
and do legitimately call themselves 'liberal'. It is an implication of this last
point, again, that I have not identified liberalism with any foundational
political morality - a rights-based morality, say, or a contractarian morality.
This too secures for the discussion of liberalism a methodological tolerance
of its many legitimate varieties, and aims to guarantee that whatever critical
judgements the argument yields regarding liberalism will be true for all its
instances.

INCOMMENSURABILITY AMONG VALUES AND THE

ILLUSIONS OF PROGRESS
How far do the ideas I have specified - ideas shared, I would say, by Paine,
Condorcet, Kant, Mill, Dworkin and the early Rawls - stand up to
criticism? Are they viable elements in a currently defensible theoretical
outlook? I have in earlier work exploited recent inquiry into indeterminacy
and incommensurability in fundamental values to argue that a doctrinal
liberalism aiming to specify a unique set of basic principles having uni-
versal prescriptive authority cannot be achieved. I want now to examine the
four elements of the liberal syndrome in order to see what, if anything, of
them survives an analogous critique. So as to anticipate my conclusion, it is
that none of them can withstand the force of strong indeterminacy and
radical incommensurability among values. Considered as a position in
political philosophy, accordingly ,liberalism is a failed project. Nothing can
be done, according to the argument here developed, to rescue it: as a
philosophical perspective, it is dead. What then is living in liberalism? The
aspect of liberalism that remains alive for us, I shall argue, is the conception
and the historic reality of civil society that has been bequeathed to us. This
288 Questions
conception and set of practices embodies or exemplifies, in historically
contextualized form, the four constitutive features of doctrinal liberalism.
The argument will be that, though it is not the case that a liberal civil society
possessing these features is the only, or necessarily the best society from the
standpoint of human flourishing, nevertheless it is the only sort of regime
in which we - in our historical circumstance as late modems - can live well.
There is an historical argument for liberalism, in other words, which main-
tains that a civil society constitutes the only sort of society through which a
modem civilization can reproduce itself. (It is not denied on this view that
there may be exceptions to the generalization just stated - Saudi Arabia
may be one - but it is asserted that they will be few, and perhaps short-
lived.) On this view, although it is true that there have been great civil-
izations that did not encompass anything resembling a civil society, it is
nevertheless true that civil society, and therefore the liberal inheritance it
carries with it, is for us an historical fate. The worth of civil society for us
as modems (or post-modems) is that it permits the peaceful coexistence in
a modus vivendi of incommensurable values and perspectives on the world.
For this reason, I shall conclude, the only forms of liberal theorizing that
command a claim on our allegiance in our historical circumstance are the
Oakeshottian account of civil association and the Berlinian liberalism
which is itself founded on radical value-pluralism.
In Berlin's theorizing, the pretensions of philosophy are radically
humbled, in that it is denied that philosophical inquiry (or indeed practical
reasoning) can arbitrate deep conflicts among ultimate values, issue in a
prescription for the good life for all men, or found any specific way of life.
Further, in Berlin's theorizing, the current liberal project, most trans-
parently pursued in the work of Rawls, of aiming to state and derive a
unique set of principles about liberty, is explicitly rejected by reference to
the thesis that conflicts among liberties, no less than conflicts among other
values, may be (and often are) conflicts among incommensurables that are
insusceptible to rational arbitration. In Berlin's liberal perspective, the
hubristic ambitions of liberal political philosophy are chastened, and a
humbler and more realistic mode of theorizing adopted.
In Oakeshott's theorizing, which several commentators have perceived
to have important affinities with liberal thought, 7 the task of philosophy is
restricted to that of illuminating our inheritance of practice and specifying
its postulates. More particularly, Oakeshott seeks to gather from the mis-
cellaneity of practice the most essential features of civil association - that
form of human association, by no means natural or universal among human
beings, but rather a distinctive historical achievement, in which human
beings are united not in any common purpose (as they may be in war, for
example) but instead by their subscription to an authoritative body of
 What is dead and what is living in liberalism? 289
non-instrumental rules. For Oakeshott, as for Berlin, despite their profound
differences in many other areas of thought, philosophy may elucidate
practice; it can never aspire (as perhaps it did among the Greeks) to govern
it. It is this fonn of inquiry, in which we aim for a better grasp of practice
in order then to return to it in all its vicissitudes, that I shall defend against
the foundationalist illusions of liberal philosophy.
This is to anticipate the upshot of a complex and at every point con-
testable argument. Let us begin by considering the impact for the liberal
idea of progress of the thesis of fundamental value-incommensurability.
Here an analogy from the arts may be helpful, in the first instance, to
overcome intuitive resistance to the very idea of incommensurability. Con-
sider the drama of Sophocles and of Shakespeare, or, closer to us and to
each other, the drama of Shakespeare on the one hand and of the French
classicists, Corneille and Racine, on the other. In what sense are the plays
of Shakespeare, say, better or worse than those of Sophocles or Racine? It
is true that these art objects belong to the same recognizable genre - that of
drama; yet it seems thoroughly absurd to try to rank them on any single
scale of excellence. Despite their membership of a common genre, they are
simply too different in style and theme to allow of any such assessment.
Within any such subgenre, no doubt, assessments of better and worse are
feasible and even commonplace, and we may sensibly talk of improvement
or decline in the tradition which is expressed in Greek tragedy or in French
classicism. Again, within any culture we may be able to distinguish great
art from art that is mediocre or worthless. Accordingly, though it makes no
sense to rank the poetry of Donne and Gerald Manley Hopkins on a single
scale, we can clearly discern the superiority of these incommensurables
over the doggerel of e. e. cummings, say. Further, there may be cross-
cultural considerations which enable us to distinguish the art objects of a
higher culture from those of a lower: this is a point I cannot pursue here, but
which it is important be made, since the perspective of incommensurability
that is being opened up here is confused, almost infallibly, with that of
cultural relativism.
These misunderstandings aside, the argument is that, in general, there
cannot be progress in the arts, if by that is meant improvement throughout
human history. There may be progress or decline even across traditions, as
when a tradition of high and great art - art in which the powers of human
imagination and creativity are maximally employed - is succeeded by a
tradition that is shallow or feeble. The theme of incommensurability among
artistic goods affinns, nevertheless, that there is no one fonn of great art
that is best, since there is no overarching standard whereby such a judge-
ment could be made. Examples of this can be found, both within our own
cultural history, and by comparison with other great civilizations. Even
290 Questions
within a context as culture-specific as that of Christian churches, it is
difficult to see how the excellences of a Baroque cathedral could be put in
the balance against those of a Gothic. It is even harder to imagine what
overarching standards could be invoked to weigh the merits of the
Parthenon against those of Notre Dame, or of either against the Shinto
Shrine at Isee.
The idea of global progress in the arts is incoherent. It is incoherent, if
only because the arts are not in the business of representation and of
measuring, with ever greater accuracy, a wholly independent subject
matter. Instead, we understand the arts rightly, if we see them as historic
creations of a highly inventive species, embedded in and emerging from
specific forms of social life with definite and diverse culbJral and historical
inheritances. The analogy in ethics and politics should be plain. As Isaiah
Berlin has put the case for incommensurability in ethics and politics, with
a clarity and power upon which I cannot hope to improve:
There are many objective ends, ultimate values, some incompatible with
others, pursued by different societies at different times, or by different
groups in the same society, by entire classes or churches or races, or by
particular individuals within them, anyone of which may find itself
subject to conflicting claims of uncombinable, yet equally ultimate and
objective, ends.8
Speaking of Vico and Herder, Berlin observes that:
For them, values are many; some of the most fascinating come to light
in the course of voyages, both in time and space; some among them
cannot, in principle, be harmonised with one another. This leads to the
conclusion, not explicitly formulated by either thinker, that the ancient
ideal, common to many cultures and especially to that of the Enlighten-
ment, of a perfect society in which all true ends are reconciled, is
conceptually incoherent. But this is not relativism .... At the heart of the
best-known type of modem historical relativism lies the conception of
men wholly bound by tradition or culture or class or generation to
particular attitudes or scales of value which cause other outlooks or
ideals to seem strange and, at times, even unintelligible; if the existence
of such outlooks is recognized, this inevitably leads to scepticism about
objective standards, since it becomes meaningless to ask which of them
is correct. This is not at all Vico's position, nor ... is it in general that of
Herder either.9
The pluralism which asserts a fundamental incommensurability of ultimate
values is not, then, any species of relativism, nor does it belong to any of
the many varieties of modem subjectivism or scepticism about values. It is
 What is dead and what is living in liberalism? 291
instead a realist or objectivist view. Objective pluralism of this sort affirms
that ultimate values are knowable, that they are many, that they often
conflict and are uncombinable, and that in many of such conflicts there is
no overarching standard whereby their claims are rationally arbitrable:
there are conflicts among the incommensurables. The diversity of ultimate
values, great as it is, is not infinite; it is bounded by the limits of human
nature. 'Incompatible th~se ends may be; but their variety cannot be un-
limited, for the nature of men, however various and subject to change, must
possess some generic character if it is to be called human at all'. \0 This
pluralism, bounded as it is, may come in several varieties, and may operate
at several levels. Within the moral code of a particular culture, there may
be lacunae that generate dilemmas which neither the code itself, nor the
practical reasonings of the individual, can resolve. Hence are generated the
radical or tragic choices among competing evils or rival excellences, in
which whatever is chosen entails some great loss or involves some irrepar-
able wrong, of which Berlin has often written. Also, there is the variety of
pluralism which illuminates value-conflict, not within cultures or indivi-
duals, but between cultures or whole forms of life having incommensurable
values as constitutive elements. These varieties of pluralism may inter-
penetrate one another, especially when (as in the late modem world)
cultures and forms of life have come to interact deeply with one another,
are no longer easily individuated, so that many individuals find themselves
(in Fulke Greville's phrase) 'suckled on the milk of many nurses', formed
by many distinct cultural traditions.
Objective pluralism carries with it the implication that there is a radical
moral scarcity which it is our fate as humans to endure. In the form in which
it is most subversive of the classical foundations of Western civilization,
this species of pluralism denies the coherence of the Form of the Good, as
theorized by Plato, and rejects the thesis of the unity of the virtues, as
advanced by Aristotle. In this pluralist view, not only are all genuine goods
not necessarily harmonious, but goods may depend for their existence upon
evils, virtues on vices. In this aspect, pluralism is a variation on the theme
of the imperfectibility of human life, traceable back to Jewish and Christian
sources. In its deepest implication, however, pluralism does not mark the
imperfection inherent in the nature of things: it destroys the very idea of
perfection. It thereby strikes a death-blow at the classical foundation of our
culture, expressed not only in Plato and Aristotle, but in the Stoic idea of
the logos and in Aquinas's conception of a world order that was rational and
moral in its essence, even as it was the creation of the Deity, one of the
central attributes of which was perfection.
Objective pluralism, because it destroys the idea of a perfect human life,
is fatal to this Greco-Roman, rationalist dimension of our cultural
292 Questions
inheritance. It also undermines a no less foundational element in our
civilization, derived from the Judeao-Christian tradition, which is the
notion of the meaningfullness of human history, conceived in the moral
tenns of redemption or of improvement. For it is evident that, if we lack
overarching standards whereby to assess history as a whole, then, as Herder
perceived, 'general, progressive amelioration of the world' is 'a fiction'.
True, there may be improvement within any particular culture, as judged by
its own standards; and there may, as limiting cases, be a few instances, in
which we can judge that, granted all relevant incommensurabilities, there
has been a betterment or a worsening in goods and evils that are not
culture-specific, but generically human. This latter limiting case gives no
support, however, to the whiggish conception of history as a narrative of
progress, as a moral drama with ourselves as its telos, which is one of the
many illegitimate offspring of Judaeo-Christianity that inform the senti-
mental and absurd religion of humanity which is the secular faith of the
Western intelligentsia. On the pluralist view, human history as a whole has,
and can have, no meaning; II it is, at best, a series of adventures in civil-
ization, each singular and discrete, leading nowhere, and at no point dis-
closing or approximating the features of 'man qua man'.
Objective pluralism, then, has radically subversive implications for the
two traditions that form the twin pillars of Western civilization - the
Greco-Roman and the Judaeo-Christian. It is also subversive of central
elements of other cultural traditions. In China, for example, there is a
Confucian tradition of historiography in which history is written in tenns of
moral advance and decline, which is subverted by pluralism, if such history
is supposed to have more than a local authority. And in India there is a
tradition of moralizing and rationalistic Buddhism, disallowing contra-
dictions in the nature of things, which is also threatened by in-
commensurability.
Objective pluralism of the radical sort advanced here recognizes in-
commensurabilities among generic human goods and evils as we)) as
incommensurabilities between (and within) specific cultures or fonns of
life. As to generic human goods, it may be that there are virtues or excel-
lences that, however various their expressions in different forms of life, are
not culture-specific. It may be that courage is a human virtue tout court,
though its varieties may be immense, and we are sometimes unsure whether
we have rightly identified it. So, also, with justice and mercy. But - as many
works of literature, such as Melville's Billy Budd, show us - the practical
demands of these generic human virtues may be conflicting, and such
conflicts may be among incommensurables. Again, contrary to the thesis of
the unity of the virtues advanced by Socrates and Aristotle, it may be
(according to the objective pluralist view) that some are uncombinable in a
 What is dead and what is living in liberalism? 293
single person. This is a thesis in moral psychology and anthropology to
which I shall return later. There may be uncombinabilities and in-
commensurabilities among other generic human goods - between a life
devoted to intense physical pleasure, and a life devoted to long-term pro-
jects, or the life of a family man who wishes to live long enough to see his
grandchildren grow. And, not least importantly, there may be conflicts
among incommensurabks, where there are generic human evils. Between
pain and death, violence and subtle coercion, and many other impediments
to human flourishing, there may on the objective pluralist view be radical
and tragic choices to be made in respect of which reason leaves us in the
lurch. (It is at this point that radical value-pluralism undermines traditional
conceptions of natural law.) None of these incommensurabilities among
universal or generic human goods and evils in any way supports un-
restricted cultural relativism about values, or endorses any of the modern
varieties of ethical subjectivism and scepticism.
Objective pluralism recognizes that many goods are specific to, and
dependent upon, particular forms of life. The life of a late medieval trouba-
dour could not be lived in ancient Egypt any more than the life of a temple
courtesan could be lived in medieval Christendom. (The term 'form of life'
is here being used with deliberate indeterminacy, so as to encompass whole
cultures as well as styles of life within specific cultures. At no point does
the argument presuppose that forms of life are easily or uncontroversially
individuated.) On the pluralist view, it will sometimes be the case that
goods or excellences that are specific to definite and uncombinable forms
of life are also incommensurables. Different forms of life may embody
virtues and excellences which, though they are each of them recognizably
great, cannot rationally be ranked or weighed against each other. As
between the life of a bushido warrior and a Renaissance scientist, say, there
is on the pluralist account no measuring rod whereby they could be
assessed, no scales whereby they could be put in the balance. Conflicts
among incommensurables will also occur within specific forms of life.
Within the form of life constituted by Catholic Christianity, the virtues of a
nun are not combinable with those of a mother; and it may be that this
incompatibility expresses an incommensurability, resolved in practice only
by the non-rational call of vocation. On the pluralist account, then, conflicts
among incommensurables, thought hardly Ubiquitous, are pervasive in
human life. This is a result that goes very much against received intellectual
tradition and against much commonsense intuition.
Pluralism goes against the grain of our cultural inheritance by also
offending the banal pieties of liberal humanism, and by being offensive to
the common stock of ordinary intuitions. How, then, is it to be defended?
What sort of theory of knowledge could support these claims about the
294 Questions
incommensurability of ultimate values? It has already been noted that a
value-pluralist ethical theory cannot avoid being a realist theory. But of
what species of realism are we speaking? And what are its epistemological
credentials?
We can approach the latter question, first. by observing that objective
pluralism is bound to be anti-intuitionist in its account of our knowledge of
ultimate values. In this we follow J. S. Mill, who saw that intuitionism has
an inherently conservative tendency in that it treats as finally authoritative
the pre-reflective judgements of persons in their local cultural environ-
ments. If, as with most academic moral philosophers in our time, these are
meagre, conformist and impoverished cultural environments (the environ-
ments of most contemporary universities), and if the lives and experiences
of those whose intuitions are consulted and considered authoritative are also
therefore shallow and narrow, then the moral judgements and theories they
eventuate in will likewise be shallow and narrow. Intuitionism in ethics is to
be rejected, then, because it sanctifies the deliverances of local knowledge,
however restrictive or distorted they may be. (In our day, the local knowl-
edge that is so sanctified is not that of ordinary folk, but instead that of an
inexperienced and culturally illiterate lumpenintelligentsia.) The most plau-
sible alternative to intuitionism is the position sketched by Mill, in which
value judgements are always tested at the bar of experience, but in which
they are exposed to all the evidences of anthropology, history, literature,
social science and of experiments in living. As Skorupsky well puts the nub
of Mill's methodological naturalism in the epistemology of ethics:
Any political philosophy which supposes that some forms of life are
objectively better in that they offer truer forms of happiness or more
emancipated modes of existence - and which does not ground that claim
in something transcendental or revealed - must defend itself by appeal
to naturally shared human dispositions .... Certain dispositions of feel-
ing must be shared by humans in such a way as to define normal
responses. Not necessarily statistically normal; but one to which the
organism, when not impeded from its natural path, or suffering from
some internal capacity or disease, tends, and in which it finds a resting
point from which it does not wish to escape, which it reflectively prefers.
Then to fail in the response is to suffer from an incapacity which
deprives one of access to a mode of experience which, for human beings,
is intrinsically worthwhile. By the same token, it must be possible to
distinguish between external conditions which impede or distort the
natural flowering of a human being's dispositions, and those which
facilitate it. A human being achieves fullest satisfaction, the life of
greatest well-being, when those natural dispositions are fulJy expressed.
 What is dead and what is living in liberalism? 295
It achieves a stable eudaemonic equilibrium - a mode of life it would
reflectively prefer even with an imaginative grasp of options. 12
The epistemology that Skorupski identifies as Mill's is a naturalistic one
with important affinities with the less metaphysical aspects of Aristotelian
ethics. It specifies as the subject matter of ethics human well-being or
flourishing, and it treats the content of well-being and flourishing, not as
consisting merely or mainly in subjective preferences or desires, but as a
matter that is at least partly objective, encompassing the use of human
capacities in a life that is, or may be, reflectively judged to be worthwhile.
This neo-Aristotelian epistemology denies what Raz has perceptively
called 'the transparency of values' J3 - the error that the content of our
well-being is clear to each of us. Though I cannot here develop the point, but
will return briefly to it later, this Millian epistemology is akin to Aristotle's
in representing human emotions as nearer to perceptions than to sensations
- in other words, in attributing to emotions a rationality (or irrationality) they
are denied on any purely or crudely Humean account of them.
An epistemology of value that acknowledges objective pluralism and
value-incommensurability will nevertheless differ from Aristotle's in
crucial and fundamental respects. It will be non-hierarchical, in that -in-
commensurability precludes objective rankings or weightings of ultimate
values. It will therefore refrain from prioritizing or privileging anyone
form of life - say, the life of rational inquiry, of contemplation or wealth-
creation, of prayer or selfless devotion to others - as the best for the human
species. Unlike the Millian moral epistemology, it will not suppose that
moral inquiry will eventuate in a Peircean ideal convergence. Rather, as a
species of objective pluralism in ethics, it will expect inquiry to issue in an
ultimate divergence of ethical perspectives - a point to which I shall later
return. As has already been observed, it will deny the unity of the virtues,
emphasizing instead conflicts among human excellences, some of which
will be incommensurables. It will, for that reason, deny that the good life is
for most people necessarily a mixed life: for, if the ingredients of a mixed
life are incommensurables, there will be no uniquely rational combination
of them, and rough equality will be no more rational than any other mixture.
(From a standpoint that acknowledges incommensurabilities among ulti-
mate values, the Aristotelian doctrines of the unity of the virtues and of the
mean may seem like little more than a prescription for mediocrity.) And
finally, since a pluralist will not take for granted the local context of the
polis that seemed natural to Aristotle, but will recognize with Aristotle the
dependency of central human individual excellences on various forms of
communal life, an objective pluralist will have good reason to favour a
variety of regimes, communal and political, in which a diversity of
296 Questions
uncombinable excellences may flourish. It will be seen later that this last
point has radical and subversive implications for liberalisms of a doctrinal
or fundamentalist sort.
Objective pluralist epistemologies of value may be characterized as a
species of plural realism in value theory. They are comparable with, per-
haps even applications of, the species of plural realism (or, as it might also
be called, perspectivism) in general epistemology that Dreyfus has dis-
covered in Heidegger's later thought. As Dreyfus convincingly argues,
Heidegger is not (contrary to common interpretations) a pragmatist, an
instrumentalist or a relativist, any more than he is an idealist, a physicalist
or a materialist. Instead
Where ultimate reality is concerned, later Heidegger could be called a
plural realist. For a plural realist there is no point of view from which
one can ask and answer the metaphysical questions concerning the one
true nature of ultimate reality. Given the dependence of the intelligibility
of all ways of being on Dasein's being, and the dependence of what
counts as elements of reality on our purposes, the question makes no
sense. Indeed, since reality is relative to finite Dasein, there can be many
true answers to the question, 'What is rea)'? Heidegger looks like an
idealist or a relativist only if one thinks that only one system of descrip-
tion could correspond to the way things really are. But for Heidegger,
different understandings of being reveal different sorts of entities, and
since no one way of revealing is exclusively true, accepting one does not
commit us to rejecting the others. There is a deep similarity between
Heidegger and Donald Davidson on this point. Both would agree that we
can make reality intelligible using various descriptions and that what our
claims are true of under a given description has whatever properties it
has even if these descriptions are not reducible to a single description,
and whether we describers and our ways of describing exist or not 14
Dreyfus notes the affinities of the later Heidegger's plural realism with
aspects of the thought of the later Wittgenstein. 15 He further specifies what
he means by plural realism when he says:
once we see that Newton's laws are true, we see that nature was already
the way they reveal it to be even at the time of Aristotle. Conversely, if
Aristotle's terms successfully picked out natural finds (relative to final
causes), his account is still true today. Again, instead of relativism, we
get plural realism . . . there can be many systems of practices that
support assertions that point our many different and incompatible
realities ... even if a system does succeed in pointing out things as they
are, that does not show that the kinds it discovers are the ultimate basis
 What is dead and woot is living in Uhemlism? 297
of reality. Reality can be revealed in many ways, and none is meta-
physically basic. . . . For Heidegger . . . as finite beings capable of dis-
closing, we work out many perspectives - many lexicons - and reveal
things as they are from many pCrspectives. And just because we can get
things right from many perspectives, no single perspective is the right one. 16
Plural realism in ethics is the view that, whereas there are definite limits on
the varieties of human flourishing, there are many forms of life, often
exhibiting divergent and uncombinable goods, in which human beings may
flourish, and none of them is the one right way of life for man. The
epistemology which supports plural realism in ethics is what Brouwer has
called dispositional ethical realism l7 - the view, attributed to Mill by
Skorupski, that values (like secondary properties) are discovered by us
when we fmd our responses coming to eqUilibrium under appropriate
circumstances of experience and reflection. Such realism is a form of
internal realism, in that it conceives of values or reasons for action, not as
external Platonistic entities of some sort, but as truths about our natures and
practices. It is distinguished from any sort of subjectivism or scepticism by
its affirmation that we may be mistaken about our well-being, which is
never constituted, wholly or simply, by our wants or preferences. Where
plural realism in ethics differs from many other realist positions in ethics -
say, Aristotle's - is in its denial that there is one best form of life for the
human species that is rationally discoverable. On the contrary, plural
realism recognizes that, just as it is natural for our species to speak a vast
diversity of languages, so it is natural for it to flourish in a great variety of
ways. To put the point in Heideggerian terms: plural realism recognizes
that, since the nature of human beings is a question for them, since it is only
partly determinate and is therefore partly self-defined, no single conception
of the good life can be founded on a conception of human nature. This
follows inexorably from the recognition that human beings, unlike other
animal species, transform their needs and are part self-creators, over time
and in history. For this reason, on the plural realist view, no form of the
good life can be final, any more than any can be said to be uniquely rational
or natural.
In this version of internal realism, it is acknowledged that our forms of
life are underdetermined by our natures. Unlike other animal species,
human beings can flourish in an immense variety of worlds, each of their
own making. The objectivity that is claimed for values in this view derives
in part from their roots in common human nature and in part from the
public, but diverse and sometimes incommensurable, practices that human
beings have invented for themselves. Whereas these different practices or
forms of life may be mutually intelligible, and so, in one sense, comparable
298 Questions
with one another, they are not for that reason necessarily commensurable
by reference to our common human nature. Limiting cases aside, very
different foons of life may be equally legitimate foons of human self-
creation, each carrying within it its own norms. The point of this variant of
realism is that, though at the margin some forms of life may be injurious to
human well-being, there are many, diverse and incompatible, forms of life
in which human well-being may be realized. We can recognize as genuine
foons of human flourishing a whole variety of foons of life, while at the
same time recognizing that none of them is the best form towards which all
others approximate.
In other words, in this neo-Aristotelian version of objective pluralism I
have borrowed (with radical revisions) from J. S. Mill,la the variety of
incommensurable forms of human flourishing disclosed to us in history and
ordinary experience, though creations of an inventive species, are radically
underdetermined by the generic powers and capacities of human beings.
They are possibilities realized by us, not demands imposed upon us by our
nature. Though it will readily be acknowledged that here are forms of social
life in which the human good is thoroughly compromised, so that the
variety of incommensurable human flourishings is bounded, it is not for
that reason determinate. In particular, it is not confined to forms of flourish-
ings that can be accommodated within a liberal regime.
What does plural realism portend for the liberal conception of progress?
Its implications are highly destructive. The empty pieties of a liberal
humanism that seek to theorize human history as a whole in teons of a
notion of progress have been effectively subverted by Hampshire, when he
observes:
Hegelianism, positivism, Marxism, constructed in the shadow of
Christianity with a view to its replacement, purported to give an account
of the development of mankind as a whole, an account of the destiny of
the species: this included an alienation or fall, followed by a political and
social redemption, leading to a final salvation of humanity. From the
standpoint of a naturalistic philosophy, looking only at the so far known
facts of human history, the gross implausibility of these accounts comes
from the false speciation and the false humanism. 'Humanity' is either
the name of a distinct animal species, with impressively distinct powers
of mind and with an uncertain future, or it is the name of a class of being
constituted as distinct by the intention of its Creator; and of course the
name may sometimes be used with both meanings in mind. If the
supernatural claims about the Creator's intentions are dismissed, there
remains no sufficient empirical reason to believe that there is such a
thing as the historical development of mankind as a whole, unless the
 What is dead and what is living in liberalism? 299
natural history of the evolution of the species is intended. What we see
in history is the ebb and flow of different populations at different stages
of social development. interacting with each other and exhibiting no
common pattern of development. Using older historical categories. we
can reasonably speak of the various populations flourishing and be-
coming powerful at some stage and then falling into decadence and
becoming comparatively weak; and historians can reasonably look for
some general causes of these rises and falls. Even if some such general
causes can be found. they will not by themselves point to a destiny. and
to an order of development. for mankind as a whole. 19
The upshot of these reasonings is that liberal meliorism - the position
which ranks societies or regimes as they approximate to a liberal order. and
which treats them as corrigible by that standard - is indefensible. It is felled
by a pluralism that denies the coherence of the idea of perfection. even as it
is rendered implausible by the risibility of the Enlightenment fantasy of the
evanescence of imperfection. 20 How do the other elements of the liberal
syndrome fare when subjected to the impact of objective pluralism?

OBJECTIVE PLURALISM, UNIVERSALISM AND LffiERAL

EQUALITY
According to the universalist component of the liberal syndrome. there is a
set of principles conferring on all human beings weighty claims in justice
or rights. These may be contextualized somewhat in their content. or they
may be grounded in an ethical theory (such as Millian utilitarianism) that is
not itself deontic. It is definitive of liberalism. nevertheless. that it should
confer upon all human beings certain basic equalities. For this reason. the
universalist and the egalitarian elements of liberalism are complementary
and mutually supportive. rather than competitive. How does value-
incommensurability affect this pair of liberal values? It is evident. in the
first place. that there are many forms of human flourishing that cannot
coexist with liberal equality. All excellences that depend upon inherited
hierarchy or involuntary subordination. that presuppose the embeddedness
of persons in roles and statuses that are constitutive of their identities and
from which they are unfree to exit. are bound to be crowded out from any
regime that is at all liberal. or else to exist within it in peripheral. marginal
or ersatz forms. This is. in part. only to restate a point already made here.
that the Aristotelian account of human flourishing is excessively. if not
obsessionally. monistic. As Stuart Hampshire has wisely observed: 21
The claim that many well-known assertions about human potentialities
are mistaken and not reasonably defensible does not entail that we can
300 Questions
hope to show that one, and only one, such claim is defensible and not
mistaken. At this point I leave Aristotle because he believed that the
essential human potentialities are fixed, once and for all. ... Any con-
ception of human potentialities has to represent a target which is not only
always moving, but is also moving in several dimensions. Any simple
conception is either too abstract and too general, too lacking in com-
plexity, to count as a representation useful for ethics; or the representa-
tion, faithful to the changing and moving target, is too evidently open-
ended, and admittedly provisional, for any final truth to be claimed.
He goes on to ask: 22
Why did Aristotle, writing about the distinctive features of human
beings, not mention the Babel of natural languages, the proliferation of
religions with their exclusive customs and prohibitions, the attachment
of popUlations to their separate and peculiar histories, the manifold
frontiers and barriers, with the aid of which social groups and popu-
lations try to maintain their separate identity? Why did he not see this
species-wide divisiveness, the drive to separateness and conflicting
identities, as at least one distinctive feature of human beings among all
dle animal species?
Aristotle's extraordinary selective blindness to the universal truth that it is
one of the most distinctive dispositions of generic humanity to constitute
for itself a myriad of local identities is a species of cognitive dissonance
that has been transmitted to modem liberalism, with its denatured fiction of
generic personhood. It also supports, within any culture, the modernist
resistance to any division of roles or statuses, any unchosen division of
labour, whereby virtues and excellences that could not coexist in a single
person may flourish in a single society. As Hampshire again puts it:
So great has been the influence within contemporary moral philosophy
of Hume, Kant and the Utilitarians that it has been possible to forget that
for centuries the warrior and the priest, the landowner and the peasant,
the merchant and the craftsman, the bishop and the monk, the musician
or poet who lives by his performances, have coexisted in society with
sharply distinct dispositions and virtues.... Varied social roles and
functions, each with its typical virtues and its peculiar obligations, have
been the normal situation in most societies throughout history. 23
The impact of value-pluralism for the egalitarian component of the liberal
syndrome is that a richer diversity of human goods, excellences and forms
of flourishing may be, or at least has been, achieved in societies that are
highly stratified and differentiated, than may be achieved in ones that are
 What is dead and what is living in liberalism? 301
highly mobile and in which roles and statuses are easily interchangeable. In
the latter sort of society, there will be an unavoidable depletion, and
eventual extinction, of all those virtues and excellences that depend for
their existence on hierarchy and deference and on men and women knowing
their station and its duties. It is important to note that the uncombinabilities
Hampshire here invokes are what Raz has usefully termed constitutive
incommensurabilites24 - they are conflicts of values occasioned, not by
contingencies such as the shortness of human lives or the scarcity of
resources, but by the very natures of the goods that are uncombinable. A
priest cannot, compatibly with his avocation, adopt the virtues of a soldier,
or a nun the excellences of a courtesan. A friend, as we understand one,
cannot demand payment for his conversation. A lover, among us, cannot
remain such, while charging a fee for his or her company. These are
incommensurabilities constitutive of the identities at stake in such relation-
ships. It is important to note here, briefly and parenthetically, a point I shall
develop more systematically when discussing Berlin's argument from
value-pluralism to liberalism - the point that, whereas the most radical form
of value-conflict is found among constitutive incommensables, in-
commensurability and uncombinability are nevertheless distinct moral
phenomena. Goods may be uncombinable, and yet readily subject to scalar
evaluation: we may have good, even sufficient reason to make a ranking of
them. At the same time, incommensurable goods may be combinable: we
may each of us embody in our lives a chosen mixture of incommensurable
goods, such as professional advancement and family life, strenuous activity
and leisurely contemplation, and so on. It is when uncombinability and
incommensurability come together, as in Raz's account of goods whose
very constitution or nature precludes ranking or commingling, that we have
the deepest sort of conflict among values.
Constitutive incommensurabilities of these sorts are akin to those con-
jectured in moral psychology, when we reject Aristotle's thesis of the unity
of the virtues. A person with the virtues of courage, resolve, resource-
fulness, intrepidity and indomitability is unlikely to possess the virtues of
modesty and humility. One may go further. A man who has the Machia-
vellian Renaissance excellences of virtu and superbia cannot possess the
Christian virtues of humility, any more than can Aristotle's great-souled
man. Both between and within different forms of moral life, these consti-
tutive incommensurabilities among the virtues shatter the complacency of
Aristotelian moral psychology. Consider again the salience of value-
pluralism for the Aristotelian account of the rationality of the emotions. A
person whose orientation towards his life is tranquil and hopeful is likely to
be incapable of despair. By despair is meant here, not the depression or
sadness by which one may be struck, in bereavement or in tragedy, but
302 Questions
instead the emotion which persons may have when they judge that their
prospects of ever flourishing have been finally extinguished. Such an
emotion expresses a rational belief, not an ephemeral mood; and, for that
reason, a person incapable of it, in the appropriate circumstances, stands
convicted of irrationality. With the incapacity for despair, as with other
emotions, it may be that this depends upon irrationalities and defects,
including lack of self-knowledge, in the person. It may be that if Van Gogh
had passed through a successful psychoanalysis, he would have been a
calmer soul; but it is hard to see how he could have painted as he did.
Incommensurability enters deeply into moral psychology here, suggesting
that some human powers may depend for their exercise on weaknesses,
lacks or disabilities. If this is so, incommensurabilities may arise even
among the powers that go into making a person autonomous: there may be
forms or exercises of autonomy that are constitutively uncombinable.
Within a society or culture, liberal equality irons out the hierarchies that
permit a differentiated exfoliation of otherwise uncombinable excellences.
Among societies and cultures, liberal universalism disallows all those
goods and virtues that depend upon an inherited inequality of status and
authority or upon exclusivity in communities that make entry into them and
exit from them hard or virtually impossible. Indeed, the liberal project of a
universal civilization on which particularistic cultures converge (and into
which they vanish) itself incorporates another, subsidiary project - that of
levelling down all the forms of stratification and exclusivity that are
distinctive or constitutive of particular cultures, and which give them their
peculiar identities. It is difficult for liberals to see in this process - un-
doubtedly real, though fortunately not without exceptions - of global
cultural homogenization any loss of value. So ingrained is the liberal
presumption in favour of equality and universality that liberal modernism
cannot perceive, or refuses to admit, the losses entailed in the destruction of
societies characterized by such a division of labour among the virtues. Or
else, if the loss is allowed to enter subversively into reflective awareness, it
is immediately repressed by an unreflective normative individualism which
invokes the interests, rights or well-being of the individual against the
excellences of the forms of life that cannot coexist with, and so are lost in,
liberal regimes. It is to the errors of normative individualism that I shall
shortly turn.
It is evident, if the reasonings advanced have any force, that the egali-
tarian component of liberalism is undermined by strong value-pluralism. It
is important to point out that it is not here being claimed that any and all
kinds of inequality are permissible, or that the variety of allowable in-
equalities is unlimited. Such cultural relativism is incompatible with the
realist orientation of objective pluralism. On the contrary, the generic
 What is dead and what is living in liberalism? 303
nature of the human species enables us to identify evils that are not
culture-specific, but universal, and which are therefore to be avoided,
suppressed or mitigated in any society in which a worthwhile life is livable.
In Hampshire's words:
it makes sense to speak of happiness, freedom, and pleasure as good
things as contrasted with unhappiness, imprisonment, enslavement and
pain as bad things: [Link] to be pursued for their own sake and states to
be prevented and avoided for their own sake. For such states there is no
need to appeal to any distinctive conception of the human good, though
a distinctive conception of the human good would be invoked when a
decision had to be made in some situation between these evils, giving
priority to the one over the avoidance of the other.25
For Hampshire, then, there is a minimum content to morality that is univer-
sal, albeit chiefly negative in spelling out certain evils which prevent or
inhibit a worthwhile human life, and possessing an inevitable indeter-
minacy both in the embodiment of these evils in different forms of moral
life and in resolutions of conflicts of priorities among them. Hampshire
goes further, and identifies the central element in the minimum content of
morality in the practice of procedural fairness or justice. In this he follows
a long line of recent liberal theorists in treating justice as the first virtue of
social institutions. While endorsing Hampshire's position that morality
contains elements that are not culture-specific, I wish to reject his view that
it is justice or fairness that is central among these elements. No purely
deontic (rights-based or justice-based) conception can be fundamental or
primordial in morality. As Raz has shown,26 political moralities are never
rights-based, since rights are themselves never foundational, but rather
intermediaries between claims about human interests that are vital to well-
being and claims about obligations it is reasonable to impose upon others in
respect of these interests. Rights, in other words, gain their content from the
requirements of human well-being - and they will be variable as the
demands of human well-being vary. So, analogously, with the broader idea
of justice. As Hampshire himself recognizes,21 practices of justice and of
procedural fairness will differ according to the underlying conceptions of
the good which animate them. For these reasons, it is extremely implausible
to suppose that justice is at the bottom of the universal dimension of
morality. Rather, it is the minimum requirements of human well-being that
identify the bottom line in political ethics. Accordingly, for example, the
evil of slavery - at least in its worst, American, variety of chattel slavery -
resided, not in any distributional inequality in liberty which that institution
incorporated, but in the injuries to the well-being of the slaves which it
constantly inflicted. It is by an assessment of the harms to vital human
304 Questions
interests done by different regimes that we will identify those that trans-
gress the universal and minimal requirements of political morality.
Except in limiting cases - in eur times, the regimes of Pol Pot, of Hitler,
Mao or Stalin - such assessments will never be easy or uncontroversial. In
part this is because incommensurabilities break out among generic human
evils, as well as among generically human goods. Is a regime in which there
is pervasive fear and invasive political control of individual life, but little
or no actual violence or threat to life (such as Czechoslovakia in the 19708
and 1980s, where totalitarianism was sustained chiefly by economic sanc-
tions) better or worse than a regime, such as Pinochet's Chile, which
incurred far more violence, including violent deaths, but in which the
invasive impact of political power on individual liberty was far smaller?
The matter is evidently undecidable by reason. Even if we could rationally
compare and evaluate such regimes, our evaluations would not be trans-
missible, by further comparisons, to other regimes. For, as Raz has shown,
the most distinctive mark of incommensurability is a breakdown of transi-
tivity in reasoning: 'The test of incommensurability is failure of transitivity.
Two valuable options are incommensurable if 1) neither is better than the
other, and 2) there is (or could be) another option which is better than one
but is not better than the other' .28 Raz's account of incommensurability in
terms of the intransitivity of reasonings about valuable options applies,
equally, to judgements about evils. Even where two bad regimes, in which
the minimum requirements of human well-being are lacking, can be ranked,
our ranking will in virtue of incommensurabilities among their constitutive
evils often be untransmissible to other bad regimes.
Incommensurability as a breakdown in transitivity occurs in many areas
of ethical life. Sometimes it is innocuous. It is a commonplace that in-
transitivities occur in friendships: you may be a friend of mine, but your
friends may not be my friends. Sometimes the breakdown of transitivity in
ethical judgement is more subversive in its results. Consider justice, the
ruling fetish of recent liberal thought. It is an unreflective supposition of
recent liberalism, stated explicitly by Nozick,29 that justice is akin to deduc-
tion in that injustice cannot arise from justice. But this is false on any view
of justice that is less parsimonious, and more in accordance with universal
practice and sentiment, than Nozick's. In practice, prescription arising from
long possession, say, over many generations, everywhere yields an entitle-
ment (not necessarily unqualified) to the property in question. If, however,
as is often the case, the original taking was unjust, we have a breakdown of
transitivity in justice. Unjust takings yield just entitlements. Analogously,
just takings may yield unjust entitlements. If the distribution of holdings
has any weight in judgements of justice, however slight, then the upshot of
many just transfers of justly taken holdings may be unjust. The implication
 What is dead and what is living in liberalism? 305
of the intransitivity of justice is that the demands of justice may be complex
rather than monolithic. If justice encompasses intransitivities, indeed, its
demands may be deeply conflicting: they may be conflicts among incom-
mensurables. As Berlin has noted,30 conflicts among incommensurables
may arise within liberty, even negative liberty, itself: we may have no
means, save in limiting cases, of assessing on-balance liberty in any disci-
plined fashion, and so no means of determining a uniquely or most rational
resolution of a conflict among liberties. Radical value-pluralism, accord-
ingly, affirms incommensurability not only among ultimate values, but also
within them.
There is an analogy here with incommensurability in the arts, as dis-
cussed earlier. As between different artistic traditions, we may be able to
make comparative judgements - we may be able to judge this a better
example of a Gothic church than that is of Baroque church architecture; yet
we may be unable to make a judgement between the Gothic church and an
example of Byzantine church architecture, which we can judge to be better
than the Baroque church. Transitivity has again broken down. It has broken
down within a recognizable artistic and cultural tradition, even as it breaks
down between traditions. And it does so, even when we can firmly identify
the traditions at issue as all of them traditions of high civilizations, superior
to those found in other cultures.
Liberal universalism and liberal egalitarianism founder on the incom-
mensurability of divergent forms of human flourishing and on incommen-
surabilities among obstacles to human flourishing. There are many forms of
human flourishing that depend upon inequalities that any liberal view must
reject, and for that reason, if for no other, the liberal universalist thesis that
all human beings are entitled to the same stock of rights or claims must be
false. It will have force, only in respect of a minimum content of morality
that massively underdetermines liberal values and which is met by many
societies and regimes that are not liberal - which do not acknowledge
fundamental equalities in liberty, and so forth. Of course, against the
argument that there are valuable ways of life which cannot flourish in
liberal regimes, there is an inevitable liberal reflex response: that, because
such forms of life may impose significant costs on individuals, and because
they deny or render costly individual exit from them, such forms of life do
not deserve to survive.
This knee-jerk liberal response may come in two forms, each of which
presupposes a deeper doctrine. It may be claimed that forms of life which
depend upon, or have among their necessary conditions, involuntary sub-
ordination, are somehow necessarily lesser (that is to say, less valuable)
forms of life than those that do not. It is difficult to see how such an absurd
claim could be rendered credible. Is the claim being made the claim that the
306 Questions
life of the Greek gentleman-scholar, whose contemplative leisure was made
possible by the institution of slavery, is necessarily worse than that of the
self-supporting sweatshop worker in the period of laisser-Ioire? Or is it the
(only slightly less incredible) claim that the best life is necessarily one that
does not depend upon enforced subordination of others? If the latter, then it
is merely the old Socratic claim, which nothing in actual experience has
ever supported, that the life depending on injustice cannot ever be as good
as the life of the just man. Or else, perhaps, and second, the liberal response
trades on the claim that, however valuable a form of life may be, its pursuit
is constrained by deontic limitations of justice or rights. But, as has already
been argued, the content and weight, together with the very justification of
all such deontic constraints, must be spelt out in terms of their contribution
to human well-being. In other words, invoking such deontic constraints
against forms of life involving involuntary subordination begs the question
as to the possibility of human well-being being promoted within (or by
virtue ot) such forms oflife. (Is it self-evident that the life of someone who
is the subject of involuntary subordination to another is always worse from
the standpoint of well-being than one who stands in a relation of equal
freedom to otheJ""'! It is, after all, an obvious objection to Rawls's theory of
justice as fairness that the principle of greater equal freedom and the
difference principle may conflict in practice: the serf in a medieval order
may be better off, ifhis lord discharges his duties, than a free labourer. And,
if the two principles conflict, why rank the greatest equal liberty principle
over the difference principle?) Both standard liberal responses derive their
real content from their presupposition in the deeper liberal dogma that only
individuals and their states of mind, feeling and action can have intrinsic
value. It is to this response, with its invocation of the fourth element in the
liberal syndrome, normative individualism, that I now tum.

NORMATIVE INDIVIDUALISM AND THE INTRINSIC VALUE

OF FORMS OF LIFE
By normative individualism, I shall intend (following Raz) any moral
theory that does not recognize any intrinsic value in any collective good.
Normative individualism holds that collective goods have instrumental
value only. Accordingly, moral theories which hold that all values are
agent-relative values will all be varieties of normative individualism. In a
more positive formulation, one may say that, for a normative individualist,
only states of individual human lives can possess intrinsic value, with
collective goods having value only in so far as they make a contribution to
the good life of the individual. Though this is a large and deep question I
cannot address here, it is worth noting that normative individualism is a
 What is dead and what is living in liberalism? 307
species of humanism - the theory that 'claims that the explanation and
justification of the goodness or badness of anything derives ultimately from
its contribution, actual or possible, to human life and its quality' .31 I men-
tion this fact, since I wish to make clear that my argument against norma-
tive individualism, although it borrows heavily from Raz's, is unlike Raz's
in that it does not presuppose humanism. Indeed, my view is that humanism
is indefensible without resort to the ludaeo-Christian tradition which
grounds the supreme value o/human personality in its kinship with a divine
personality which animates the universe. In the forms in which it is found
in secular liberalism, humanism is merely an atavistic anomaly, a curious
relic of theism that has suppressed awareness of its genealogy in ludaeo-
Christianity. (Note here that normative individualism is not ethical sub-
jectivism. An ethical subjectivist may be a normative individualist, in that
he affirms that only the preferences of individuals confer value on any-
thing; but a normative individualist may be an ethical realist, who believes
- as did Kant - that it is a moral truth that only persons have intrinsic value.
Normative individualism appears to be neutral on the issue of the objec-
tivity or otherwise of values.)
The thesis of this section of the paper is that normative individualism is
false. Individualism must be regarded as merely a particular form of life, in
which certain distinctive excellences are achieved, but in which others are
excluded. On the view defended here, individualism is one form of human
flourishing among a variety of others, and has no special privileges, even if
(as I shall suggest) it is our historical fate - at least for those of us who are
heirs to a Western inheritance. Why, then, are we to reject normative
individualism as a basic position in ethics? Let us follow Raz and consider
the value of autonomy. Among us - the inhabitants of modem Western
societies characterized by a high degree of social mobility, pluralism in
lifestyles and individualism in ethical culture - autonomy is a constitutive
ingredient in any form of the good life. If we lacked even a modicum of
autonomy, if we were not even part authors of our lives - if our jobs, our
marriages or sexual partners, our place of abode or our religion were
assigned to us or chosen for us - we would consider our individuality stifled
and the goodness of our lives diminished. This is not to say that maximal
autonomy, if there could be such a thing, would be a good thing. Perhaps
autonomy is a fungible good, consumed in the choices, often irreversible,
in which it is embodied. Perhaps it is a positional good, in that the array of
options available to an autonomous person in a culture of heteronomous or
less autonomous people, but containing enclaves of autonomous life, is
richer than any he might have in a society, like that envisaged in On Liberty,
in which all are highly autonomous. Perhaps, again, maximal autonomy is
as much of a mirage as the greatest liberty, since autonomy may have
308 Questions
incommensurable dimensions or forms such that comparative on-balance
assessments of the greatest autonomy are rendered impossible. (There is a
further possibility that different forms or exercises of autonomous choice
may be constitutively uncombinable.) Again, it is not here assumed that
autonomy is a necessary ingredient in any good human life, nor that the best
human lives are necessarily the most autonomous ones. As Raz has put it:
I think that there were, and there can be, non-repressive societies, and
ones which enable people to spend their lives in worthwhile pursuits,
even though their pursuits and the options open to them are not subject
to individual choice. Careers may be determined by custom, marriages
arranged by parents, childbearing and child rearing controlled only by
sexual passion and traditions, part-time activities few and traditional,
and engagement in them required rather than optional. In such societies,
with little mobility, even friends are not chosen. There are few people
one ever comes in contact with, they remain there from birth to death,
and one just has to get on with them. I do not see that the absence of
choice diminishes the value of human relations or the display of excel-
lence in technical skills, physical ability, spirit and enterprise, leader-
ship, scholarship, creativity or imaginativeness, which can all be
encompassed in such lives. 32
In the neo-Aristotelian view presented here, the virtue of autonomy is a
local affair. It is left open whether a form of life in which autonomy is
inconspicuous or lacking - the form of life of medieval Christendom, or of
feudal Japan in the Edo period - may be better from the standpoint of
human flourishing, or else simply incommensurably different, by com-
parison with the form of life of autonomous individuals. Finally, in contrast
with Raz, it is not even assumed that autonomous individuals will neces-
sarily, or even generally, flourish best in open, pluralist societies. Perhaps,
as the example of recent Asian immigrants in the United Kingdom sug-
gests, social or ethnic groups in which autonomy is not greatly prized will
do better on most dimensions of human flourishing, when compared with
autonomous individualists, in a society that contains both. On a global,
historical scale, it may well be that the linkage between an individualist
moral culture and a prosperous market economy is an historical accident,
which in its ultimate development proves to be self-limiting in virtue of its
comparative disadvantage against other market economies (such as those of
East Asia) whose moral inheritance is not individualist.
All that is required for my argument against normative individualism is
that autonomous lives be intrinsically valuable lives. Or, at the very least,
that autonomy be a necessary, constitutive ingredient of some valuable
lives in our society. The central argument against the truth of normative
 What is dead and what is living in liberalism? 309
individualism is that valuable autonomy presupposes as one of its con-
stitutive elements a rich public culture containing a diversity of worthwhile
options. By contrast, consider the exercise of autonomous choice in a
Hobbesian state of nature. What would its value be? The Hobbesian auto-
nomous chooser could not marry, adopt a religion or engage in any of the
arts of commodious living; the choices of such an agent would be made in
an impoverished cultural environment - indeed, an empty one. Conversely,
a multitude of conventions, institutions and social forms is necessary for
valuable autonomous choice. As these forms wax, so does the value of
autonomy; as they wane, so does autonomy, so that at the limiting point of
Hobbesian anarchy, autonomy, barely exists. The vital point is that a rich
public culture is not an instrumental condition of a valuable autonomous
life, but an essential element in it. The lives of autonomous choosers who
live in a culture without art, science, friendship, religion or romantic love
are poor lives, even if they do not know what they are lacking and so cannot
regret its absence. To say this is to restrict the agent-relativity of value and
to affirm that forms of lives may themselves have intrinsic (which is to say,
agent-neutral) value. To deny this by asserting the agent-relativity of all
value, by denying intrinsic value to forms of life, seems tantamount to an
endorsement of unrestricted cultural relativism.
If the elements of a rich cultural environment enter into autonomous
choice constitutively, not instrumentally, then the inherently public goods33
that they comprise will also be intrinsically valuable. It follows that art, and
scientific inquiry, are intrinsically valuable activities or practices, indepen-
dently of the contribution they make to the good lives of individuals. 34 A
world without them is the poorer, even if (or especially if) no one notices
their absence. This argument establishes that, whatever has ultimate value,
it is not only states of the lives of individuals. A further step in the argument
is needed to show that it is forms of life or activity that are ultimately and
intrinsically valuable, so that it is from these forms of life that the value of
the lives of the individuals in which they are instantiated is derived. This
step is provided by the argument that choices, autonomous choices in
particular, have value only if they range over options that are themselves
choiceworthy. Such options can be provided only by activities and forms of
life that are intrinsically valuable. Inasmuch as worthwhile forms of life are
what make valuable autonomy possible, it is they, and not the individual
lives in which they are instantiated, that are ultimately and primordially
valuable. Whatever variations individual choice adds to common forms of
life, it is these forms that are the sources of the values expressed in
autonomous individual choice. Without them, individuals cannot be even
part authors of their lives.
310 Questions
This suggests a perspective on rights that differs greatly from the stan-
dard, conventional liberal perspective. It suggests a theory of rights that is
non-individualist in character. If scientific inquiry, say, is an intrinsic
collective good, then we may seek to protect it by an entrenched right,
where the right protects the activity, not its individual practitioners. There
may be collective rights to collective goods, if these are intrinsically valu-
able themselves, and if they are elements of a common life or public culture
that also has intrinsic value.
It is important to note that, since it recognizes that intrinsically valuable
forms of life and activity are instantiated only in individual lives, this view
does not license the wanton sacrifice of individual well-being for collective
good. Forms of life are not Platonistic entities which exist independently of
their practitioners. If a form of life is a type that can have indefinitely many
tokens in the lives of individuals who subscribe to it, it nevertheless cannot
exist unless it is instantiated in individual lives. Further, on this view, since
most of what constitutes a man's well-being is constituted by elements in
the form of life in which he partakes, there will not be any general conflict
between his well-being and his form of life. Indeed, in the individualist
form of life we are here considering, in which autonomy is a central feature,
we will be protecting and enhancing the form of life when we protect and
promote the individual's well-being. It is not, however, asserted that there
will always be this coincidence. In some cases, the fonn of life an indivi-
dual inherits will be injurious to the vital interests he has as a human being.
To say this, on the view presented here, is to say that these generically
human interests, or components of human well-being, can only be protected
in forms of life to which the individual is denied access. The American
chattel slave's most vital interests were injured, inasmuch as he was denied
access to the form of life of American free men. To affirm that forms of life
may injure generic human interests is therefore compatible with the claim
that it is only in forms of life and activity that intrinsic value is to be found.
Since cultures vary as to the intrinsically valuable collective goods they
contain, so will the collective rights they may wish to recognize. These
rights may not be very highly determinate in their content, and they may
protect collective goods that are incommensurables. On the view presented
here, these are not difficulties that are fatal to this view, since (unlike
fundamentalist liberalism) it acknowledges and is founded upon the cul-
tural variability and diversity of the forms of the good life. In its application
to the theory of justice and rights, it is conventionalist (but not therefore
arbitrary). It implies that the requirements of justice and the content of
rights be contoured by reference to the collective goods that are manifested
variously in the different cultures. This view, while it affirms the objec-
tivity of human well-being, acknowledges that justice is special, in that
 What is dead and what is living in liberalism? 311
(like the institution of any right) it is artefactual and conventional in charac-
ter. A man or woman may live well, if he or she is ruthless and lucky,
without the virtue of justice, providing most of his ot her fellows possess it;
he or she will not live well if he or she lacks courage, self-command and
good judgement. If the account of well-being and flourishing advanced
here is Aristotelian in its realism, the view of justice presented is Humean
in its conventionalism. Justice is theorized as a convention depending for
its authority on general acceptance and reciprocity. In its applications to
collective goods, justice demands protection for those that are intrinsically
valuable. Freedom of religious worship will on this view be protected as a
central element in the good life for human beings, or, at the very least, as a
collective good in our culture. It will also, no doubt, receive protection by
way of the individual rights we protect - rights of voluntary association, for
example. Equally, both collective goods or forms of intrinsically valuable
activity and individual claims may be accorded respect and protection in
traditional societies that have not yet adopted (or succumbed to) the dis-
course and practice of rights. The argument as developed so far suggests
that, though rights may be explicated in terms of their contribution to
individual well-being, it is not the individual that is the bottom line in the
theory of rights. For the well-being of the individual is owed to his partici-
pation in a form of life or activity that is intrinsically valuable and worth-
while. It may be that, while in practice individual and collective aspects of
the practice of rights are mutually complementary rather than competitive,
the ultimate source of rights (if we are to engage in rights discourse at all)
is in forms of life, rather than in individuals and their claims.
Individualism is never the bottom line in what has value. What has value
(in the human world) is the form of life - even if, paradoxically, it is an
individualist form of life. This is true, even as forms of life interpenetrate,
and are not easily individuated. Few, if any, of us are Sandel's3S radically
situated selves, whose lives are constituted by membership of a single,
exclusive community. Yet, even in this post-modernist context, the ultimate
sources of value are the diverse forms of life in which we move, and which
introduce complexities and even incommensurabilities into our identities
and perspectives on the world, rather than the choices of individuals. Like
the first language we speak, the parents to whom we are born and the
religion we inherit, the post-modem condition of deep pluralism is for us
unchosen. (I shall argue later that there is a valid Berlinian argument from
pluralism to liberalism, but one that is historical and local, not universal, in
its application.) If Raz, Berlin and others36 are right, value-incommen-
surability is a universal feature of the human condition, suppressed or
veiled in most philosophies; it is far less easily deniable by us, for whom
the irreconcilable diversity of human inheritances is a daily fact of life.
312 Questions
Nonnative individualism, along with the other elements of the liberal
syndrome, runs aground on the reef of incommensurability. At the very
least, as between pluralism and liberalism, pluralism is the deeper truth.
(Our condition is a peculiar, and perhaps novel one, in that, recognizing that
our individual identities are constituted by participation in forms of
common life, we at the same time find ourselves in the interstices of many
such forms. Our situation is perhaps akin to that of the sceptic about
meaning in Kripke's Wittgenstein, who in speaking a language that is
necessarily public perceives that the attribution of meanings depends ulti-
mately on primordial judgements that are made by individuals.) This is to
say that, whereas there is a powerful Berlinian argument, which I shall later
develop, from the truth of incommensurability among ultimate values to the
liberal prioritization of liberty, that argument has no universal validity. It
applies to those historical contexts, such as our own, that are marked by a
diversity of incommensurable conceptions of the good and perspectives on
the world. The universal implication of radical value pluralism, on the other
hand, is that no one political regime can be privileged as having a claim on
reason. This is an inexorable result of the thesis of incommensurability
among ultimate values.
Incommensurabilities arise, not just between values that are elements in
a conception of the good, but between whole conceptions of the good,
suggesting that the Millian naturalistic epistemology of ethics, which is
grounded on the prospect of an ideal convergence on value-judgements,
must be substantially qualified. The Piercean prospect of ethical consensus
has but little leverage on experience or practice. Rather, the prospect
appears to be that of ultimate divergence among incommensurable value-
perspectives, both individual and cultural. Here it is Nietzsche, not Mill,
who has most contemporary resonance. As Skorupski has put it: 'The
NietzschelMill polarity goes to the anthropological root of ethical and
political life; its charge has the power to disorientate and disturb all but the
most dogmatically entrenched on either side'.J7 It is important to try to
specify more precisely the kind of breakdown of convergence that occurs
in an objective pluralist epistemology. The position is that, though con-
vergence is expected on judgements as to which forms of life embody
human flourishing and which do not, there will not be convergence as to
which among these is to be adopted, since the conceptions of the good they
express are incommensurables. Further, within any conception of the good
or the form of life it animates, there will be convergence on its component
goods and excellences; but not on what is to be done when they come into
conflict with one another. Objective pluralism, then, postulates an under-
determined world of values, which can within limits be construed and
shaped by human decisions, but whose components are a matter of human
 What is dead and what is living in liberalism? 313
knowledge. (It is not, of course, being suggested that there can be knowl-
edge of the varieties of human flourishing that is definitive and final; since,
contrary to Aristotle, there are indefinitely many forms of human life, as yet
uninvented by the species, in which the human good will be realized. Of
these we can have knowledge only when they are embodied in practice.) I
shall not here inquire whether, as I have already intimated, an analogous
failure of Peircean convergence may occur in the sciences.3I My argument
is neutral on this issue. What I have argued is that radical value-
incommensurability, together with the strong indeterminacy in liberal prin-
ciples I have argued for in other places,39 spells ruin for fundamentalist
liberalism. No determinate set of principles about liberty, rights or justice
has been stated which is not disabled by deep indeterminacies - such as
afflict both Mill's Principle of Liberty, Rawls's Greatest Equal Liberty
Principle and the set of basic liberties, and all theories of rights. (Such
indeterminacies are clearest in cases of conflicts of liberties or rights, but
are by no means confined to them.) The upshot of this present argument is
that recognition of a pluralism of forms of human flourishing, each objec-
tive, of which only some can exist in a liberal regime, destroys the authority
of liberalism as a universal, trans-historical and cross-cultural ideal. It is in
order to reject the inevitable counterclaim about the costs of non-liberal
forms of life to the individuals that instantiate them that I have argued for
the falsity of normative individualism.
A final point needs to be made as to the import of my argument for the
objectivity of human well-being for the evidences of revealed preference.
It is often, and sometimes with truth, held that the propensity of people to
seek exit from a non-liberal regime is evidence of its detrimental impact on
human well-being. This certainly applies to the totalitarian regimes of our
century. It is far less clear that exit from a traditional to a modem society
demonstrates the superiority of the latter: denizens of a traditional society
may be ignorant of the conditions if their well-being and their revealed
preferences may only express the untransparency of values in their lives. In
practice, again, traditional forms of life are often simply driven out by
modem modes of production: there is no question of their practitioners
choosing exit from a way of life that will otherwise remain viable. Even
where exit is chosen, it may entail unnoticed losses, incur benefits that
prove delusive, entail the extinction of a superior form of life and adoption
of a lesser one. It is an implication of my argument that the universal
privileging of individual choice that is a feature of fundamentalist liberal-
ism (and which entrenches the modernist prejudice that good lives cannot
be unchosen lives) may sanctify the erosion or destruction of forms of
common life from which individual choices and the lives they eventuate in
gain much, if not all, of their meaning and value.
314 Questions
THE LIVING KERNEL OF LmERALISM: CIVIL SOCIETY
The result of the argument thus far is that none of the four elements of the
liberal syndrome withstands the force of radical value-pluralism. If there is
an ultimate diversity of forms of human flourishing, embodied in ways of
life only some of which can be accommodated within a liberal regime, then
liberal orders have no general superiority over orders that are not liberal. In
short, value-pluralism of this radical sort dictates pluralism in political
regimes, and undermines the fundamentalist liberal claim that liberal
regimes alone are fully legitimate. We may say of liberalism as a doctrine
with aspirations for universal prescriptive authority, then, that it is dead - a
result tacitly acknowledged in the most recent theorizings of the later Rawls.
What is still living in liberalism, if its philosophical foundations have
collapsed? I shall maintain that what is living in liberalism is not any
doctrine or comprehensive theory, but instead the historic inheritance of
civil society that has now spread to most parts of the world. It is civil
society, with its key guarantees of liberty and the limits on government they
imply, that remains alive, not liberalism. In civil society, the four con-
stitutive elements of liberalism specified earlier are preserved, not as uni-
versal principles, but as constitutive aspects of civil life. It is civil society
that should be the object of theorizing, not 'liberalism' or any abstract
conception of liberty. I shall maintain, further, that the only formofliberal-
ism that remains defensible is the Berlinian form, which encompasses a full
acceptance of value-pluralism. For most, if not perhaps for all, modem
states, civil society is the only sort of society in which the prosperity and
values of a modem civilization can be reproduced, and it is the only form
of society in which incommensurable conceptions of the good life can
coexist in peace. What are the arguments for this view, and what is meant
by civil society?
By civil society is meant a number of things. Civil society may be
defined contrastively, by noting as one of its contraries the weltan-
schauung-states of both ancient and modem times. A civil society is one
which is tolerant of the diversity of views, religious and political, that it
contains, and in which the state does not seek to impose on all any com-
prehensive doctrine. Thus, Calvin's Geneva was not a civil society, and
none of the twentieth-century species of totalitarianism encompassed civil
societies. Indeed, as I have elsewhere argued,40 totalitarianism in its Com-
munist varieties is (or was) a Westernizing, modernist project of destroying
or transcending civil society, so that totalitarianism is in general to be
defined by reference to its enmity to civil society, not to liberal democracy.
In a civil society, then, diverse, incompatible and perhaps incom-
 What is dead and what is living in liberalism? 315
mensurable conceptions of the world and the good can coexist in a peaceful
modus vivendi.
A second feature of civil society is that, in it, both government and its
subjects are restrained in their conduct by a rule oflaw. A state in which the
will of the ruler is the law, and for whom therefore all things are per-
missible, cannot contain or shelter a civil society, or, if it does, it will be one
that is weak and disordered. (An example of the latter might be Duvalier's
Haiti.) One implication of this feature of civil society is that it presupposes
a government that is not omnicompetent, but limited. In any civil society,
most social and economic activities will take place in autonomous institu-
tions that are protected by the rule of law but independent of government.
Civil society and unlimited government are, for this reason, incompatible.
This is only another way of saying that civil society has as one of its
conditions the limitation of government by the same rule of law that civil
society is constituted and governed by.
A third feature of civil society is the institution of private or several
property. Societies in which property is vested in tribes, or in which most
assets are owned or controlled by governments, cannot be civil societies.
The importance of several property for civil society is that it acts as an
enabling device whereby rival and possibly incommensurable conceptions
of the good may be implemented and realized without recourse to any
collective decision-procedure - a recourse which, in any pluralist culture,
would inevitably occasion conflict. There is a Hobbesian argument for a
civil society containing the institutions of several property - the argument
that, given pervasive value-pluralism, it is such a society that is most likely
to enjoy civil peace. This Hobbesian argument depends on the recognition
that civil war or strife is a great evil, threatening all modes of commodious
living, so that to conceive of political life as the unending pursuit of a
provisional modus vivendi is a mark of humility and realism, not of
ignobility. The central institution of civil society - the institution of private
or several property - has its rationale as an enabling device whereby
persons with radically discrepant goals and values can pursue them without
recourse to a collective decision-procedure that would, of necessity, be
highly conflictual. One may even say of civil society that it is a device for
securing peace by reducing to a minimum the decisions on which recourse
to collective choice - the political or public choice that is binding on all -
is unavoidable. Such, I take it, is the Hobbesian argument for civil society,
often echoed in Oakeshott's account of civil association. This will not,
however, be the most fundamental argument that will be advanced for civil
society, which will be defended instead by reference to the varieties of
human flourishing it shelters.
316 Questions
A few words are apposite regarding the history and diversity of civil
society. Societies having the characteristics specified appear to have
emerged in the wake of the Wars of Religion - in the Low Countries, in
England and Scotland and in subsequent centuries throughout most of
Europe. There can be no doubt that Tsarist Russia was a civil society for the
last fifty or sixty years of its existence, just as was Bismarclcian Prussia.
Whig England had all the characteristics of a civil society. The ultimate
historical source of civil society is, most likely, to be found in the dis-
solution of the European medieval order that began in the thirteenth
century. The varieties of civil society so far enumerated should allow us to
perceive that civil societies may go along with a variety of political institu-
tions. Civil societies need not have the political and economic institutions
of liberal democracy; most, in historical terms, have not. Nor need they
contain the moral culture of individualism. The authoritarian civil societies
of modem East Asia - South Korea, Taiwan and Hong Kong, for example
- have not embraced Western individualism (a fact that may go far to
explain their extraordinary economic success).
Nor, finally, is civil society to be identified with market capitalism.
Several or private property may come in a variety of forms, each of them
artifacts of law, and the institution of the capitalist corporation is only one
species of the private or several property institution on which a civil society
rests. This last point is of no small contemporary significance. As James
Buchanan has pointed out in a seminal paper,41 the post-Communist
regimes, especially that which is emerging in Russia in the wake of the
coup of August 1991, will go badly astray if they seek only to replicate the
forms of Western capitalism, as advocated by most Western economists.
What is needed in Russia, according to Buchanan, is not the transposition
of Western economic institutions, but instead a radical deconcentration of
economic activity, to municipal, village and cooperative levels, in which
native Russian traditions of cooperation can be revived. Equally, the
Western, and especially the American, project of forcing liberalization on
Japanese economic life, though it is bound to fail, is thoroughly and
dangerously misconceived. It neglects the embeddedness of economic
transactions in an underlying culture that is not, and is most unlikely to
become, individualist. In both Russia and Japan, Westernization (or further
Westernization) would only involve injury to valuable social forms, with
few, if any, corresponding advantages.
It is important that these points be borne in mind, so as to distinguish the
argument here from that of those - most absurdly, the American writer
Fukuyama42 - who perceive in the collapse of the Communist regimes 'the
triumph of the Western idea', 'liberal democracy as the final form of human
government' and 'democratic capitalism' as the universal form of political
 What is dead and what is living in liberalism? 317
economy. On the view advanced here, though civil society may be an
invention of Western cultures, its adoption by other cultures need not, and
in all probability, should not, involve the adoption of Western values. The
record of Western cultural imperialism - the near-destruction by Com-
munist totalitarianism of Chinese and Russian traditional cultures, the
deracination of Africa by over-rapid modernization and the triggering of
fundamentalist movements in Iran and elsewhere by hubristic projects of
social engineering - suggests that the emergence of civil societies where
they had not hitherto existed, and their re-emergence where they had been
suppressed, is a process that should occur at its own pace, subject to the
constraints of local cultural traditions.
There is another argument for civil society, which I shall term the
Hegelian argument, which characterizes it as the institutional infrastructure
of modernity. This is an argument that remains forceful, even if, as I have
maintained, liberalism conceived as the political theory of modernity is no
longer credible. It is an argument which points to the division of labour, the
definition of property rights and their embodiment in private institutions
whose power and immunities are protected under a rule of law as necessary
conditions for the reproduction of a modem society. This Hegelian argu-
ment is supported both by theoretical reasonings and by the evidences of
modem history.
The evidence of the past decades indicates (what economic theory had
suggested) that without the institutions of a market economy a modem
industrial society cannot successfully reproduce itself. For this reason
alone, the institutions of private property (not necessarily in the form of the
Western capitalist corporation) and of contractual exchange are essential
elements of any viable civil society. They are possible, in tum, only if their
scope and content are defined and protected by a rule of law. This, in its
tum, confers upon the subjects of civil society a set of immunities and
liberties. These are not just the liberties of buying and selling, but the
liberties of voluntary association, of occupation, of travel and so forth.
Because by its very nature a civil society cannot coexist with a weltan-
schauung-state, civil societies will practice tolerance in respect of personal
belief-systems, even if (as in the United Kingdom) they retain an estab-
lished religion. The liberties guaranteed in civil societies are, on this view,
necessarily not very determinate, but they must at the least incorporate
those sketched above. In all their varieties, civil societies approximate to
the Oakeshottean ideal-typical conception of civil association43 - that form
of human association in which we are bound together in no common
enterprise, but perceive each other as convives, each pursuing their own
projects, and respecting those of others. Whatever their varieties, civil
societies are the only ones in the modem world which reproduce
318 Questions
themselves stably as prosperous and peaceful communities. Attempts to
reverse the modernist movement toward civil society - in Nazism, com-
munism and fundamentalism - have everywhere issued in barbarism and
poverty. It is therefore to the minute particulars of our inheritance of civil
society, and not to the delusive vistas of liberal philosophy, that the friend
of liberty should therefore address himself.
Civil societies come in many varieties. The political regimes which
shelter them may be liberal-democratic, or they may be authoritarian. Civil
society will, however, always be voluntarist. The moral culture that ani-
mates it may be individualist, or (as in East Asia) it may not. It will be true
of any civil society, however, that it will embody the historic inheritance
(perhaps exported elsewhere in the world) that 'liberalism' theorized - the
inheritance of the practice of liberty. In so far as any kind of civil society
exists, it will posses the constitutive institutions of civil society - such as
the rule of law, private or several property, and the civil liberties of
voluntary association, conscience, travel and expression. On the view pre-
sented here, a civil society shelters the practice of liberty, whether or not it
tolerates democratic freedoms. The only form of equality necessitated by a
liberal civil society is, not political or economic equality, but equality
before the law. For the essence of the practice of liberty is that, in a civil
society, individuals and voluntary associations are bound together by no
common end or enterprise but live in peaceful coexistence under a regime
of non-instrumental rules. The civil condition, so conceived, is fully com-
patible with a variety of political regimes, but presupposes always the rule
of law and equality before the law.
On the view presented here, civil societies, in all their legitimate varie-
ties, are the living kernel of what was 'liberalism'. Even when their political
institution are authoritarian, or their moral culture not individualist, civil
societies of all kinds embody a voluntarist conception of human asso-
ciation, and thereby express (or soon come to be animated by) a culture of
liberty - a culture in which individuals are free to come together in pursuit
of shared purposes, but need have no enterprise in common. In this sense,
whatever their differences, all civil societies are liberal civil societies. Once
the fundamental liberal project of achieving an Archimedean point of
leverage on practice is abandoned, theorizing can tum to the institutions,
conventions and traditions that comprise the practice of liberty. Criticism
and reform of existing civil societies may then proceed, but it will be
immanent criticism, internal to the life of each specific civil society, which
does not pretend to be governed by any abstract conception of freedom.
Freedom is on this account constituted by the practices of civil society,
which it is the task of the theorist to illuminate. Any such understanding of
the practice of liberty is bound to be an historical understanding - one that
 What is dead tJ1Ul what is living in liberalism? 319
conceives of civil society as an historical artefact - but one that. in the
context of the condition of late modernity (or early post-modernity) offers
the best. if not the only prospect for the reproduction of civilized life.
The four constitutive elements ofliberalism as a doctrine here re-emerge
as characteristics of civil society. The legal structure of a civil society is
bound to be individualist since none of us is (in the jargon of recent
communitarian theory) 1 radically situated self whose identity is con-
stituted by membership of a single community. On the contrary, we are,
each of us, members of a host of communities, sometimes overlapping but
often conflicting in their claims, roles and statuses. In this historical context
of deep cultural pluralism, in which traditions and fonns of life inter-
penetrate and are not easily individuated, and in which most of us have
access to incommensurable conceptions and perspectives of human life and
the world, the fundamental legal structure of a civil society is bound to be
individualist. (It will, no doubt, contain corporate legal persons; but these
will not typically possess the powers and immunities possessed by indivi-
duals.) It has been noted that the justification for rights of freedom of
religion or of scientific inquiry need not be, and in the end cannot be, an
individualist one; but this is not to deny that. in any modem pluralist
culture, the principal bearer of rights (or of the various immunities con-
tained in the English common law) will be the individual, and not any
collectivity. Here individualism is affmned, not as any set of universal
nonnative claims about the species, but instead as a necessary feature of a
modem civil society.
Similarly with egalitarianism. Though a civil society presupposes
neither political nor economic equality, it does require equality before the
law. For it is a necessary feature of a civil society that. just as no one in it
is above the law, so no one is denied the protection of the law. Again,
inasmuch as civil society is not a weltanschauung-state in which one set of
beliefs is enforced on all, but rather an association in which all subscribe to
common procedures in which no belief or objective need be shared, it
confers upon all the same basic immunities and entitlements. Accordingly,
every member of a civil society, unless disqualified by special circum-
stances, has the immunity from arbitrary arrest and the entitlement to a fair
trial, freedom to acquire and alienate property, liberty of occupation, travel,
conscience and so forth. Since they have the same immunities and entitle-
ments under the rule of law, they may be said to enjoy equality under the
law. Such equality need be neither absolute nor exceptionless: it may be
abridged by the institution of preventive detention for suspected terrorists,
or (as with Peers of the Realm in the United Kingdom) be qualified by
special privileges and disabilities. Legal equality will nevertheless be the
general condition in civil society. It is so, not in virtue of any doctrine of
320 Questions
universal moral equality, but because the absence of any shared hierarchy
of ends in a civil society precludes ranking or prioritizing of anyone's
freedom over that of anyone else, save in limiting cases.
What of meliorism'! We have seen that radical value pluralism under-
mines the idea of progress as a category applicable to human history
globally. Within the history of any particular civil society, however, it
makes sense to talk of improvement or decline and to frame projects of
reform. The standards whereby such projects are conceived and assessed
will be to a considerable extend dictated, not by the ideal-typical notion of
civil association, but by the specific circumstances and traditions of the
various civil societies. Thus, in a civil society defined by a written con-
stitution, such as the United States, reform of current practice may appro-
priately be conducted by reference to the interpretation of that constitution,
while in the United Kingdom discourse will be conducted by reference to
the tradition of the Common Law and other forms of precedent. It is not that
an ideal type of civil Sudety in the most abstract sense cannot be invoked
in any specific historic context; rather that such invocation will not (given
the different traditions and narratives of the various civil societies) result in
convergence on a single model. The Oakeshottian ideal-typical conception
of civil association has value in capturing the essential characteristics of
civil society shorn of its historical contingencies; but, for that reason, it will
not as a rule be invoked in practical political discourse. Discourse as to
amelioration or decline will in general be governed by standards that are
imminent in the specific histories and traditions of the diverse civil
societies.
The universalist element of liberalism survives, not by civil societies
converging on any single model, but in virtue of the universality, or
near-universality, of civil society itself as a condition of prosperity and
peace for any modern civilization. Without the constitutive institutions of
civil society specified earlier, no modern society can expect to enjoy peace
or commodious living. It may be that, nevertheless, few modern, or post-
modern states can achieve a stable civil society for long: there is nothing to
support the hope that civil society will be universal in fact. Its universality
is that of a necessary condition, in virtually all contemporary historical
contexts - which are contexts of cultural pluralism in varying degrees - of
a common life for those with divergent values and conceptions of the world.
The four constitutive features of doctrinal liberalism return in a contex-
tualized form in the institutions of civil society. It has been affirmed, also,
that any contemporary civil society will be animated by a culture of liberty.
It remains to connect the Oakeshottian idea of civil association with the
idea of liberty as it is supported by value-pluralism in Berlin's theorizing.
 What is dead and what is living in liberalism? 321
AFfER LIBERAL POLITICAL PHILOSOPHY: THEORIZING
THE PRACTICE OF LIBERTY
In the post-liberal theoretical perspective advanced here, the project of
giving the historic achievement of civil society a universal foundation has
been abandoned. Once the foundationalist project of liberal political philo-
sophy is abandoned, what then are the tasks of the theorist? It should be
noted, first of all, that the position defended here preopposes acceptance of
a thesis in axiology - the thesis of objective value pluralism and in-
commensurability - and that, since this thesis is one within philosophy, the
view I defend diverges from the currently fashionable one, advocated most
notably by Richard Rorty,44 in which philosophy itself seems to have been
abandoned. The view defended here, however, radically restricts the ambi-
tions of 'philosophy' by denying to it any prescriptive authority; it concurs
with Wittgenstein, in his remark on philosophies of mathematics, that
philosophy should not (and indeed cannot) engage in the 'bourgeois' pro-
ject of founding any specific practice. This is not to deny that philosophical
inquiry may have definite results, still less, absurdly, to assimilate philo-
sophical argument to conversation.
The task of philosophy, properly conceived, is to clear away the illusions
that obstruct a clear vision of practice. Its task is then principally prophy-
lactic - as exemplified in Berlin's work on liberty, which shows the
delusiveness of any fixed set of detenninate and harmonious liberties and
thereby the impossibility of a libertarian calculus. Once this prophycatic
against philosophical hubris is achieved, the theorist returns to his own
inheritance and tradition - in our case, to the practice of liberty. Once we
cease to be liberal political philosophers, we are free to become theorists of
liberty - and to return to a great tradition of such theorizing, which en-
compasses the thought of the Scottish School, of Constant and Tocqueville,
and of Maine and Dicey. The conception of theorizing advanced here has
affinities with that of Oakeshott,45 in that the theorist seeks to discover the
postulates of practice. But it has also a content derivable from social theory,
in that it will not forswear the illumination of our practices that comes from
well-developed explanatory theories. In this the theorist only renews a
tradition of discourse broken, in our times, by the fragmentation of the
disciplines of the humanities that is a symptom of the broader decom-
position of the common culture. For the theorist of liberty, the study of the
practice of liberty need not, and should not, be divided neatly into norma-
tive and explanatory disciplines: it is the task of theory at once to illuminate
the values that animate the practice of liberty and the structure of the
institutions in which it is embodied.
322 Questions
It is an important feature of this approach to theorizing liberty that it does
not. after the fashion of the anachronistic method of philosophical analysis,
seek to elucidate liberty as a concept or definition. nor does it (in funda-
mentalist or foundationalist style) try to state fixed and determinate 'prin-
ciples' about the restraint and distribution of liberty. Rather, it looks to the
practices and institutions which embody liberty in any particular historical
context and which, in truth, constitute 'liberty' for human beings. This mode
of theorizing involves a reversal of method in regard to that practised in the
dominant schools of post-War Anglo-American liberal political philosophy,
in which free-floating abstractions are deployed in a void, having no con-
crete institutional, cultural or historical embodiments. The process of
abstraction advocated here goes in the opposite direction, from the investi-
gation of the various autonomous institutions that go to make up any con-
crete civil society to the life of the autonomous agent that such institutions
facilitate. We move from the autonomy of institutions to autonomous indivi-
duals, where the former are constitutive conditions of the latter.
Liberty returns as the animating value of civil society, not in virtue of its
foundational place in any liberal doctrine, but as a characterization of a
form of life that can be realized fully only in a society constituted by
autonomous institutions and activities. Liberty is prioritized (as a charac-
terization of the status of autonomous agents) because it enables such
agents to chart a course among incommensurables which the various forms
of life by which they are surrounded present to them. On the view here
presented, 'liberty' understood in 'negative' terms as a condition, status or
sphere of action protected from interference or coercion derives its prin-
cipal value from its contribution to 'autonomy' - the partial self-creation of
human beings by choices among incommensurables that occur within and
between forms of life. Such negative liberty may be valuable for other
reasons, in historical contexts other than our own, in which autonomy is
itself not especially valuable. For us, negative liberty is an essential, con-
stitutive condition of autonomy, with the autonomous institutions of civil
society supplying the array of choiceworthy options which are autonomy's
other constitutive condition. Let us see how this connection among auto-
nomy's constitutive conditions is further worked out

FROM RADICAL VALUE-PLURALISM TO LIBERALISM VIA

AUTONOMY
As has already been noted, the historicist argument for civil society here
advanced has an early statement in Hegel. It differs from Hegel in denying
to history, or to civil society, any teleological character. There was nothing
inevitable in the emergence of civil society, and there is nothing that
 What is dead and what is living in liberalism? 323
guarantees its survival. Its potentially (but unlikely) universal spread (in its
many varieties) means an unavoidable loss of those excellences that civil
society drives out. On the value-pluralist view presented here, there can be
no universal argument to the superiority of civil society. There is never-
theless a powerful case that, for the family of societies having our historical
context, in which incommensurable forms of life are deeply inter-
penetrated, it is a liberal dvil society that best promotes human flourishing.
The argument is essentially Berlin's, but is reinforced by Raz's theoriz-
ings on autonomy. Berlin's argument is that, if there is an irreducible
plurality of objective values that are sometimes incommensurable, then
liberty may reasonably be privileged among these values, since when they
possess liberty men and women may freely choose between uncombinable
ends and make their own combinations of those conflicting values among
which a balance may be struck. The argument here has two prongs, nega-
tive and positive. Negatively, it trades on the truth that, since there is no one
rational ordering or combination of incommensurables, no one could ever
provide any reason for a particular ranking or combination of in-
commensurables. More positively, the argument invokes the value of auto-
nomy, both when we are choosing our own mixture of incommensurables
that are combinable, and when we are choosing between goods that are
constitutively uncombinable. In both cases we are part-authors of our lives,
and part-creators of our selves, as we chart our own course among in-
commensurabilities between which reason cannot arbitrate. And in both
cases, our status as autonomous individuals is made possible only by our
having available to us autonomous institutions and practices which
generate options among which we may freely choose.
Individual choice among incommensurables occurs, accordingly, at two
levels. As individuals, we choose (not always, or often, with full reflective
awareness) not to explore forms of life or modes of experience that cannot
coexist with others we prize: we thereby curtail, often irreparably, a whole
range of goods in our lives. If we adopt the life of the devoted family man,
we cannot (at least in our day) enjoy the goods of the life of a voluptuary:
we deny them to ourselves, as being constitutively incommensurable with
the form of life we have chosen. Once this radical choice has been made,
we have still many others, often no less difficult, to make, among the
incommensurable (but often variously combinable) options that remain.
How much of my life will I devote to my family, and how much to my
career, or to charitable works? The virtue of a liberal regime is that in it
individuals (and, also, groups, associations, cultural traditions and com-
munities) can, each of them, adopt different mixtures of incommensurable
values, thereby achieving across the society a variety of forms of human
flourishing that could not be achieved within a single human life.
324 Questions
This objective-pluralist argument for a civil society that is liberal in the
sense that it accords individuals a wide space for freedom in styles of life is
strengthened if one accepts Raz's view that autonomy is accepted by most
of us as a vital ingredient in the good life. IT, using the texts we find in the
forms of common life around us, we wish to be part-authors of our lives,
then we will have good reason to demand that choices among in-
commensurables be made by us, freely and on our own initiative. Even if
the life that issues from these choices is not itself an especially autonomous
one, the fact that it was autonomously chosen makes it a narrative that one
has chosen for himself. And a society that contains a diversity of such
self-chosen lives will have a more complex (but not for that reason neces-
sarily or always a more valuable) moral ecology than one dominated or
unified by a simple traditional way of life.
For us, then, for whom the practice of autonomy is an essential part of
the good life and for whom individualism and pluralism are an historical
fate, a liberal civil society is the one in which the richest diversity of forms
of flourishing is most likely to be achieved. This result is avowedly culture-
specific, and it is not meant to be universalized. Its culture-specificity is no
weakness in it, if with Oakeshott we acknowledge that:
A man's culture is an historic contingency, but since it is all he has he
would be foolish to ignore it because it is not composed of eternal
verities. It is itself a contingent flow of intellectual and emotional
adventures, a mixture of old and new where the new is often a backward
swerve to pick up what has been temporarily forgotten; a mixture of the
emergent and the recessive; of the substandard and the somewhat flimsy,
of the common-place, the refined and the magnificent.46
At the same time, the argument from value-pluralism to a liberal civil
society has in our time a force that is near-universal. For, given that no
modem culture is sealed off hermetically from others, and that in our time
the forms of life of different cultural tradition interact and interpenetrate
with one another, there will in every modem society be intimations of
incommensurable values and perspectives on the world. A liberal civil
society has the advantage, for virtually all modem peoples, that in it
epistemic freedoms are protected - freedoms of inquiry and expression -
whereby these intimations of incommensurability can be explored and
pursued. Societies that are overtaken by the project of reversing the late
modem tendency toward interaction among cultures - societies that are
overtaken by religious fundamentalism or fundamentalist political ideo-
logies, such as Marxism - infallibly fall into a barbarism in which all
prospects of commodious living, or of peaceful coexistence among men
 What is dead and what is living in liberalism? 325
and women, are lost. As Berlin has put it, with respect to the twentieth-
century political religions:
The conviction that, once the last obstacles - ignorance and irrationality,
alienation and exploitation, and their individual and social roots - have
been eliminated, true human history, that is, universal harmonious
cooperation, will at last begin, is a secular form of what is evidently a
permanent need of mankind. But if it is the case that not all ultimate
human ends are necessarily compatible, there may be no escape from
choices governed by no overriding principle, some among them painful,
both to the agent and to others. From this it would follow that the
creation of a social structure that would, at the least, avoid morally
intolerable alternatives, may be the best that human beings can be
expected to achieve, if too many varieties of positive action are not to be
repressed, too many equally valid human goals are not to be frustrated. 47
Berlin's restatement of liberalism is a humbler and far more realistic one
than could be made by any fundamentalist liberal, such as Mill. Unlike
Mill, Berlin does not suppose that there is, or could be, any single measur-
ing rod, on which the merits of different cultures or epochs could be ranked.
Nor does he indulge the fantasy that determinate principles are statable and
rationally demonstrable which arbitrate conflicts among liberties and
between liberty and other values. And he nowhere accords liberal regimes
the apodictic superiority over all others that they have in all doctrinal
liberalisms. Though it is not a relativistic argument for liberty, Berlin's
objective pluralism carries with it an historicist element, when it is applied
to societies, such as ours, in which incommensurabilities among values and
world-views are pervasive and undeniable. Unlike Mill's or other liberal-
isms,48 with their antique echoes of progress, personhood and the religion
of humanity, Berlin's is one that can stand the test of our time, which
includes radical and tragic choices among goods and evils of which Millian
liberalism knew nothing.
Nor does Berlinian liberalism (at least on my account of it) privilege
over all others that form of life in which individuals make choices among
incommensurable conceptions of the good. On my account, forms of life in
which conflicts among incommensurables are resolved by tradition or
precedent may be no less forms of human flourishing than ours. That our
way of life is ours gives us good enough reason to defend it; we do not need
the spurious and illusory support of a theory such as Mill's. The view
presented here, unlike Mill's, respects the diversity of modem civil
societies, each with its own forms of (perhaps partly incommensurable)
immanent criticism, and espouses no paradigm of civil society, to which all
326 Questions
must approximate. It does not suppose, as Mill did in his Considerations on
Representative Government, that there is a fonn of government that is
ideally best for all. Rather, as in the work of Hume as reinterpreted by
rectnt revisionary scholarship,49 it is assumed that each civil society will
understand itself by reference to a distinctive historical narrative, in tenns
of which criticism and refonn of its institutions will proceed. Civil societies
are most unlikely, then, to converge on anyone model; indeed on the
pluralist view developed here, the very notion of such a paradigm is
incoherent. This in no way weakens, but ought instead to strengthen, our
disposition to defend the civil society we have, each of us, inherited or
adopted. Our way of life will survive and prosper, not because we devise a
comprehensive theory of it - for, on Berlin's view as on Oakeshott's and
mine, moral and political life is in any case not fully theorizable - but only
if we are steadfast in our commitment to it. And that commitment is
manifested in the realm of practice, not that of discourse or philosophy.
The development of a comprehensive liberal ideology is dangerous, in
any case, for this reason. It obscures our perception of the virtues and
excellences that liberal society drives out, and so dulls our sense of the
undoubted losses that the transition from traditional to civil societies incurs.
At the same time, liberal ideology represses the (often pre-liberal) virtues
and excellences on which liberal culture itself depends - including fonns of
common life that liberal individualism tends to corrode. In both these
respects, liberal ideology blurs our sense of the historical singularity of
liberal civil society.

LIBERAL IDEOLOGY VERSUS LIBERAL CML SOCIETY

We recognize, then, that civil societies are not without losses and costs.
Many fonns of life that are intrinsically valuable are lost, or weakened, in
them. Perhaps the fonns of flourishing they allow are not the best our
species has achieved, or can achieve; or perhaps we are back in the realm
of incommensurables, when we ponder that hard question. We know,
however, how terrible are the costs of destroying civil society, of wrecking
it in a frenzy of ideological or religious fervor. For us, in any foreseeable
future, no fonn of the good life can be lived outside a civil society. The
subtle and humble approach suggested by Berlin, and, in a very different
mode, by Oakeshott, in which we seek a modus vivendi, never stable or
complete, in which we can pursue the intimations of our divergent and
incommensurable values without waging war on one another, may appear
a dull one to those whose perceptions have been distorted by the hallu-
cinatory horizons of hubristic philosophy -liberal and otherwise. For those
who have a sober appreciation of the likely course of the coming century -
 What is dead and what is living in liberalism? 327
a course that, virtually inevitably, will encompass fundamentalist con-
vulsions, near-apocalyptic ecological catastrophes, Malthusian wars and
the spread of technologies of mass desb'Uction in an increasingly anarchic
world - the defence of liberal civil society will appear a more attractive
(and demanding) commitment. Nothing in the reasonings developed here is
meant to imply the inevitability of liberal civil society in our times. Present
trends suggest the opposite, with the fragile institutions of civil society
being increasingly threatened by recent fundamentalisms and by the re-
emergence of atavistic ethnicities. The likely prospect, for most of the
world in the coming century, may well be terror and new forms of
barbarism, not any species of civil [Link] In this regard, Spengler may
prove to be a better guide to our likely future than the Panglossian fantasies
of Fukuyama. Nothing suggests, in other words, that we can take the
inheritance of civil society for granted. The argument is only that, for us,
the loss of civil society means the loss of civilization.
Further, we ought not to neglect the prospect of a slow decline in our
own civil societies, as traditional practices of law and civil association are
eroded by the inordinate demands of abstract conceptions of equality and
rights. Over the longer run of history, these assaults on civil societies may
yet prove as desb'Uctive of it as the more dramatic and explicit attacks of its
avowed enemies among religious and political fundamentalists. It is,
indeed, an implication of my argument that, if civil society is among us in
danger, it is in virtue precisely of the hubris of fundamentalist liberalism -
the liberalism of those (such as Ackerman,the early Nozick and Rawl, or
Dworkin51 ) who seek to use the hallucinatory perspectives of uncritical
philosophy to distract us from the practice of liberty. We are not far from
the point at which the mass availability of hubristic philosophy imposes
impossible strains on the institutions and practices from which it illicitly
derives all its genuine content.52 It is the argument of this essay that it is
only by abandoning such false philosophy and re-situating ourselves within
our inheritance of practice that that practice can be successfully renewed.
The danger we confront in the established civil societies of the West is
that of the corrosion of liberal practices by liberal ideology. This is a danger
that is ubiquitous in the West, but especially severe in the United States,
where liberal ideology (especially as it is found in academic institutions)
exhibits a hostility to virtually every aspect of civil society, particularly the
market economy. There is a further danger that it is the destructive hubris
of liberal ideology, rather than the inheritance of civil society, that will be
transmitted to the emergent post-Communist world. It is only by deflating
the pretensions of fundamentalist liberal ideology that we can hope to curb
or slow the frittering away of our patrimony of civil institutions that is
presently well under way in the West.
328 Questions
CONCLUSION
I once said, perhaps rightly: The earlier culture will become a heap of
rubble and finally a heap of ashes, but spirit will hover over the ashes.
L. WittgensteinS3
Liberalism was the political theory of modernity. As we enter the closing
phase of the modem age, we confront the spectre of renascent atavistic
barbarisms, which threaten to ruin the modem inheritance of civil society.
Our task, as post-modems no longer sustained by the modernist fictions of
progress, rights and a universal civilization or by classical conceptions of
natural law as embodied in Greco-Roman and Judaeo-Christian traditions,
is to preserve the practice of liberty that is transmitted to us by the inherited
institutions of civil society. The task may be a daunting one, but it is one
which our circumstances and prospects impose upon us, if we retain any
sense of the worth of our inheritance, and any perception of the nature of
the alternatives to it. In the past. there may have been viable alternatives. in
which civilization and civil society did not go together. Our argument has
been to that conclusion. At present. and in any foreseeable future. we have
no such options. In the contemporary Western world, a self-denigrating
gUilt inhibits the discourse of civilization and barbarism that informed the
Scottish Enlightenment (the only episode of enlightenment from which. if
I am not mistaken. we have anything to learn); yet it is the lesson of our
century that. if the claims of commissars or mullahs. Nazis or clerical
fascists are conceded. civilization is lost along with the civil societies they
seek to subjugate. and barbarism supervenes. We will best serve what
remains alive in liberalism. if we come to see liberal civil society as a
particularistic form of life, spreading throughout the world. but everywhere
threatened by modernist fundamentalisms and atavistic ideologies.
A commitment to the preservation of civil society is, for us. a commit-
ment to the maintenance of civilization. For. though it may be only one of
the diverse forms of flourishing our species has achieved. a liberal civil
society is the form of society in which we have made our contribution to the
human good; and. in defending it. we defend the best in our cultural
inheritance. and the best that the species can presently reasonably hope
for.54
Notes

1 HOBBES AND THE MODERN STATE

I Lniatlum, (New York and London: Dent and Sons, 1949) SO.
2 Michael Oakeshott, Hobbe, 011 Civil Ar,ociation.r, (Oxford: Basil Blackwell,
. 1975)73.
3 Lniatlum, 49.
4 Quoted in Hobbe" (London: Macmillan, 19(4) by Sir Leslie Stephen. 139.
5 Lniatlum, 79-80.
6 Lniatlum, 103-4.
7 Oakeshott, op. cit., 55.
8 Bernard Gert, in his Introduction to Man and Citilen by Thomas Hobbes, (New
York: Doubleday, 1972).
9 B. Gert (ed.), Man and Citizen by Thomas Hobbes, 48-9.
10 Elias Canetti, 71te Human Province, (London: Picador, 1986) 115-16.
lIOn this, see Donald Livingstone's Hume', Philosophy of Common Ufe,
(Chicago and London: University of Chicago Press, 1984).
12 B. Gert, op. cit., 42-3.
13 See Isaiah Berlin, Vieo and Hertkr: Two StlUlie, in the History of Ideal, (New
York: Viking Press, 1976); and 'Vieo's concept of knowledge' and 'Vieo and
the ideal of Enlightenment' , in Isaiah Berlin,Against the Current, (London: The
Hogarth Press, 1979).
14 Oakeshott, op. cit., 63.
15 A good sununary of the work of the neo-Hobbesian Virginia School of Public
Choice may be found in James Buchanan and Gordon Tullock, 71te Calculus of
Consent, (Ann Arbor: University of Michigan Press,I962).
16 For example, Gregory P. Kavka, Hobbe,ian Moral and Political 71teory,
(Princeton: Princeton University Press, 1986).
17 A brilliant explanation of the character of the Soviet system is to be found in
Alain Besancon's 71te Soviet Syndrome, (New York, 1976).
18 Oakeshott, op. cit., 74.

1 SANTAYANA AND mE CRITIQUE OF LIBERALISM

1 See T.L.S. Sprigge, Santayana: an Examination of his Philo,ophy, (London and
Boston: Routledge and Kegan Paul, 1974) 1.
330 Notes
2 rust published by Charles Scribner and Sons (New York), republished by Dover
Publications, Inc., (New York, 1955).
3 Ufe ofReason, or The Phoses ofHuman Progress, (New York: Charles Scribner
and Sons, 1905-6) 5 vols.
4 Realms of Being (New York: Charles Scribner and Sons, 1927-42) 4 vols.
5 Winds of Doctrine: Studies in Contemporary Opinion, (New York, Charles
Scribner and Son, 1913).
6 Soliloquies in England and Later Soliloquies, (New York: Charles Scribner and
Sons, 1922).
7 Winds of Doctrine, 146.
8 Soliloquies in England and Later Soliloquies, 165-6.
9 ibid., 165-6.
10 ibid., 207-8.
11 Dominations and Powers: Reflections on Uberty, Society and Government,
(New York: Charles Scribner and Sons, 1951) 340.
12 Soliloquies in England and Later Soliloquies, 184.
13 Dominations and Powers, 438.
14 ibid., vii.
15 ibid., 158.
16 'Alternatives to Liberalism' in Santayana's The Birth of Reason and Other
Essays, (New York: Columbia University Press, 1968) 114.
17 Dominations and Powers, 211-12.
18 ibid., 450.
19 ibid., 452.
20 ibid., see 440.
21 ibid., 454-6.
22 'Alternatives to Liberalism', op. cit., 108-9.
23 P.S. Schilpp, The Philosophy of George Santayana, (New York: Tudor Publish-
ing Company, 1951) 559.
24 'A long way round to Nirvana' in Santayana's Some Tums of Thought in
Modem Philosophy, (Cambridge: Cambridge University Press, 1935) 94.
25 'Alternatives to Liberalism', op. cit., 115.

3 HAYEK AS A CONSERVATIVE
1 F.A. Hayek, Constitution ofUberty, (Chicago: Henry Regnery Company, 1960).
2 F.A. Hayek, Law, Legislation and Uberty: a New Statement of the Uberal
Principles of Justice and Political Econamy, vol. 2: The Mirage of Social
Justice, (London: Routledge and Kegan Paul, 1976).
3 Hayek's epistemology is most systematically presented in his treatise on philo-
sophical psychology, The Sensory Order: an Inquiry into the Foundations of
Theoretical Psychology, (London: Routledge and Kegan Paul, 1952), and in !he
earlier essays collected in his Studies in Philosophy, Politics and Econamics,
(London: Routledge and Kegan Paul, 1967).
4 For the intriguing idea of a meta-conscious rule, see Hayek's Studies in Philo-
sophy, Politics and Economics, ibid., 60-3.
5 Saul Kirpke, Wittgenstein on Rules and Private Language, (Oxford: Basil
Blackwell, 1982) 88n.
 Notes 331
6 Hayek acknowledges the inevitability of legislation in the modern state in his
response to the most interesting and original criticism of his views in Bruno
Leoni's Freedom and the Law. (Princeton. New Jersey: D. Van Nostrand, 1961).
See Hayek, Law. Legislation and Liberty. vol. I: Rules and Order. 168. 35n.
7 F.A. Hayek. The Rood to Seifdom. (London: George Roudedge and Sons. 1944).
8 I refer. most especially. to Oakeshott's HUmlln Conduct. 274-8. for its masterly
evocation of the sources and character of the modem European sense of
individuality.
9 For Hayek's conceptioll of an unviable morality. see Law. Legislation and
Liberty. vol. 2: The Mirage of Social Justice. ch. 11.
10 On the Platonist and Christian roots of Marxism, see L. Kolakowski, Main
Currents of Marxism. vol. 1. ch. 1.
11 For a conservative criticism of Hayek's Mandevillain argument, see Irving
Kristol. 'When Virtue Loses all her Loveliness - Some Reflections on Capital-
ism and "The Free Society..•• The Public Interest. (Fall 1970) 3-15.
12 For a profound interpretation of contemporary moral inversion. see Michael
Polanyi's Personal Knowledge: Towards a Post-Critical Philosophy. (Chicago:
Chicago University Press, 1958) ch. 7. sections 9-16.

4 OAKESHOTr AS A LIBERAL
1 Rationalism in Politics and Other Essays. (Indianapolis: Liberty Press. 1991)
439-40.
2 The paper is 'Michael Oakeshott as Liberal Theorist' by Wendell John Coals, Jr.
Canadian Journal of Political Science. (December. 1985. xviii: 4) 773-87.
Oakeshott expressed his admiration of Coats's paper in a conversation with the
present writer.
3 The two books are: Oakeshon by Robert Grant. (London: The Claridge Press.
1990) and The Political Philosophy of Michael Oakeshon. By Franco. (New
Haven and London, 1990).
4 John Rawls. A Theory ofJustice. (Cambridge. Mass.: Harvard University Press.
1971); Ronald Dworkin. Taking Rights Seriously. (London: Duckworth, 1978).
5 Michael Oakeshott. Rationalism in Politics and Other Essays. new and
expanded edn. Timothy Fuller (ed.). (Indianapolis: Uberty Press. 1991).
6 See note 2. above.
7 Rationalism in Politics and Other Essays. 452.
8 Robert Grant, Oakeshott. 85.
9 See Rationalism in Politics and Other Essays. 384-406.
10 Henry C. Simon. Economic Policy for a Free Society. (Chicago and Cambridge:
Chicago University Press and Cambridge University Press. 1948).
11 Rationalism in Politics and Other Essays. 443.
12 See Oakeshott The Voice of Liberal Learning. (New Haven. 1989).
13 See Franco. note 3 above.

5 BUCHANAN ON LIBERTY
1 See my book. Liberalisms: Essays in Political Philosophy. (London and New
York: Roudedge. 1989) ch. 10. 'Contractarian method. private property and the
market economy'.
332 Notes
2 Michael Sandel, liberalism and the Limits of Justice, (Cambridge: Cambridge
University Press, 1982).
3 James Buchanan, 'The Constitution of Economic Policy' , Nobel Prize Lecture,
(1986) reprinted in Public Choice and Constitlltional Economics, J.D. Gwartney
and R.B. Wagner (cds), (Greenwich, Conn. and London: Jai Press, Inc.) 112.
4 ibid., 113.
5 Jan Narveson, The libertarian Idea, (Philadelphia: Temple University Press,
1988).
6 Such a thin veil of ignorance is also deploycd by G. Kavo in his exceDent
Hobbesian Moral and Political Theory, (Princeton, NJ: Princeton University
Press, 1986).
7 I am indebted to Zbigniew Rau for conversation on this topic, and for allowing
me to read his forthcoming paper, 'The Bast BuropcmcbaJlenge to dual con1ract
orthodoxy: replacing the state of nature with the state of enslavement', where
these questions are illuminatingly discussed.
8 Sec my 'Totalitarianism, reform and civil society,' ch. 12 of this collection.
9 Sec Nelson Goodman, Ways ofWorld-Malcing, (Indianapolis: Hackel Publishing
Co., 1975).
10 James Buchanan and Geoffrey Brennan, The Reason of Rilles: Constitlltionol
Political Economy, (Cambridge: Cambridge University Press, 1985).
11 Jeffrey Paul, 'Substantive Contracts and the legitimate basis of political
authority', The Monist, 66 (4), (October 1983).
12 For an exposition of actual-contract methodology see Gilbert Harman,
'Rationality in agreement: a commentaIy on Gauthier's Morality by Agreement',
Social Philosophy and Policy, vol. 5, issue 2, (Spring 1988).
13 Sec Michael Oakeshott, Hobbes on Civil Association, (Oxford: Basil Blackwell,
1975).
14 For the best account of Hume's social philosophy, sec D. livingston's Hume's
Philosophy of Common Life, (Chicago: University of Chieago Press, 1984).
15 For G.L.S. Shackle's account of invincible human ignorance in a kaleidic world,
see his masterpiece, Epistemics and Economics: A Critique of Economic Doc-
trines, (Cambridge: Cambridge University Press, 1972).
16 This paper bas benefitted from conversations over the years with James
Buchanan, Charles King, F.A. Hayek and G.L.S. Shackle. Responsibility for its
argument, including its interpretation of Buchanan's thought, remains mine
alone.

6 BERLIN'S AGONISTIC LmERALISM

1 Joseph Raz, The Morality of Freedom, (Oxford: Clarendon Press, 1986).
2 Isaiah Berlin, The Crooked Timber of Humanity: CluJpters in the History of
Ideas, (London: John Murray, 1990).
3 'Two Concepts of Liberty', in FOllr Essays on Liberty (Oxford: Oxford Uni-
versity Press, 1968) 118-72.
4 John Stuart Mill, On Liberty, in John Gray (cd.), On Liberty and Other Essays,
(Oxford: Oxford University Press, World's Classics Series, 1991).
5 Edna and Avishai Margalit (cds), Isaiah Berlin: A Celebration, (London: The
Hogarth Press).
 Notes 333
7 THE SYSTEM OF RUINS
1 Eugen von B6hm-Bawerk, KIlrl Marx tmd the Close o/his SySIlm, (Clifton, NJ:
A.M. Kelly Publishers, 1973).
2 Tom Bottomore (ed.), A Dictionary 0/ Marxist Thought, (Oxford: Blackwell,
1983).
3 ~rard Bekennan, Marx tmd Engels: A Conceptual Concordance trans. Terrell
Carver, (Oxford: Blackwell, 1983).
4 David Felix, Marx as Politician, (Carbondale: Southern IUinois University
Press, 1983).
5 Ernest Nolte, Marxism, Fascism, Cold War, (Asson, Netherlands: Van Gorcum,
1983).
6 Ernest Nolte, Three Faces 0/ Fascism, (London: Macmillan, 1965).
7 Nolte, op. cit
8 ibid., 84.
9 Alan MacFarlane, Origins 0/ English Individualism, (Cambridge: Cambridge
University Press, 1976).
10 G.A. Cohen, Karl Marx's Theory 0/ History: A De/ence, (Oxford: Clarendon
Press, 1978).
11 Alex Callinicos, The Revolutionary Ideas 0/ Karl Marx, (London: Bookmarks,
1983).
12 - - , Marxism and Philosophy, (Oxford: Oxford University Press, 1983).
13 David McLellan (cd.), Marx: The First Hundred Years, (Fontana, 1983).
14 Paul Craig Roberts, Alienation 0/ the Soviet Economy, (Albuquerque, New
Mexico: University of Mexico Press, 1971, 1st edn; New York and London:
Holmes and Meier Publishers, 1991, 2nd edn).
15 Norman Fischer, Louis Patsouras and N. Georgopoulos (eds), Continuity and
Change in Marxism, (Brighton: Harvester, 1983).
16 Susan M. Easton, Humanist Marxism and Wingensteinian Social Philosophy,
(Manchester: Manchester University Press, 1983).
17 Jorge Larrain, Marxism and Ideology, (Macmillan, 1983)..
18 Jean L. Cohen, Class and Civil Society: The limits 0/ Marxian Critical Theory,
(Oxford: Martin Robertson, 1983).
19 Michael Burawoy and Theda Skocpol, Marxist Inquiries: Studies o/labor, class,
and states, (Chicago: University of Chicago Press, 1983).
20 Barry Smart, Foucault, Marxism and Critique, (Routledge and Kegan Paul,
1983).
21 George C. Brenkert, Marx's Ethics 0/ Freedom, (Routledge and Kegan Paul,
1983).

8 THE DELUSION OF GLASNOST

1 Goldfarb, Jeffrey C., Beyond Glasnost: The Post-Totalitarian Mind, (Chicago:
University of Chicago Press, 1989).

9 THE ACADEMIC ROMANCE OF MARXISM

I Martin Jay, Fin-de-Siecle Socialism and Other Essays, (Routledge, 1989).
2 F. Laclau and C. Mouffe, Hegemony and Socialist Strategy, (Verso, 1985).
334 Notes
3 A.W. Gouldner, Against Fragmenlillion: Origins of Mamma and the Sociology
of Intellectuals, (Oxford University Press, 1985).
4 John E. Roemer, Free to Lose: An Introduction to Mamst Economic Philo-
sophy, (Century Hutchinson, 1989).

10 PHILOSOPHY, SCIENCE AND MYTH IN MARXISM

1 Theses on Feuerbach', VI, in Lewis S. Feuer (cd.), Marx and Engels: Basic
Writings on Politics and Philosophy, (London: Collins, Fontana Library, 1969)
285.
2 In Karl Marx: Early Writings, introduced by L. Colletti (Hannondsworth:
Penguin Books, 1975) 328-9.
3 S. Ryazanskaya (cd.), The Gemum Ideology, (Moscow: Progress Publishers,
1964) 31.
4 F. Engels (ed.), Irans. Samuel Moore and Edward Aveling Capital, I, (Moscow:
Progress Publishers, 1965) 179.
5 Economic-Philosophic Manuscripts, in Karl Marx: Early Writings, 328.
6 L. Kolakowski, Main Cu"ents of Marxism, Ill, The Breakdown, (Oxford:
Clarendon Press, 1978) 277.
7 Capital, I, 79-80.
8 On this and other points in my analysis I have learnt much from Marx's Theory
of Exchange, Alienation and Crisis by Paul Craig Roberts and Matthew A.
Stephenson, (Stanford: Hoover Institution Press, 1973).
9 The term 'catallaxy' I borrow from F.A. Hayek's recent use of it See his Studies
in Philosophy, Politics and Economics, (London: Routledge, 1967) 164.
10 'The Obsolescence of Marxism' , in N. Lobknvicz (cd.), Marx and the Western
World, (NOire Dame: University of Noire Dame Press, 1967) 411.
H ibid., 411.
12 ibid., 411.
13 ibid., 411.
14 ibid., 411.
15 'Repressive Tolerance', in A Critique of Pure Tolerance, (with Barrington
Moore and R.P. Wolff) (London: Cape, 1969) 93.
16 One-Dimensional Man, (London: Sphere Books, 1968) 200.
17 An Essay on Liberalism, (London: Allen Lane, 1969) 85.
18 One-Dimensional Man, 201. See also Marcuse's Counter-Revolution and
Revolt, (London: Allen Lane, 1972).
19 E. Kamenka, Marxima and Ethics, (London: Macmillan, 1969) 26.
20 L. Kolakowski, The Socialist Idea, (London: Weidenfeld and Nicholson, 1974)
ch. 2, 'The myth of human self-identity'.
21 G.A. Cohen, Karl Marx's Theory of History: A Defence, (Oxford: Clarendon
Press, 1978) lSI.
22 ibid., 151.
23 ibid., 134.
24 ibid., 134.
25 ibid., 135.
26 ibid., 156, 155, respectively.
27 ibid., 153.
28 ibid., 152.
 Notes 335
29 ibid., 153.
30 ibid., 153.
31 ibid., 151.
32 See on this R. Rhees, Without Answers, (London: Routledge, 1969) 23-49, for a
critique of Marxism from which I have learnt much.
33 lowe this example to D.Z. Philipps and H.O. Mounce, Moral Practices,
(London: Routledge, 1970).
34 I am indebted to the writings of John Anderson, and especially to his Marxist
Ethics, in Studies in Empirical Philosophy (Sydney, 1962), for these points.
35 Cohen, op. cit, 24.
36 All of the preceding quotations occur on p. SO of Cohen.
37 On p. 353 Cohen asserts that the theses of the labour theory of value are not
presupposed or entailed by any of the arguments he advances in the book. The
productivity criterion may presuppose some elements of the labour theory of
value; but I am not concerned to argue this there.
38 Cohen, op. cit, 248.
39 Cohen, op. cit., ISS.
40 Cohen, op. cit, 169.
41 Cohen, op. cit, 206.
42 Cohen, op. cit, 223.
43 Cohen, op. cit, 160-66.
44 Singer, New York Review of Boolc.r, (20 December, 1979) 46-7). In reply to
Singer, Cohen has insisted that Darwinian theory has a functionalist aspect. I am
not persuaded by his claims, but their cogency would not affect the main line of
my argument.
45 Cohen, op. cit, 159.
46 In Berlin's Against the Current: Essays in the History ofIdeas (London: Hogarth
Press, 1979), 2~322.
47 Lulcks, History and Class-Consciousness, (Cambridge, Mass.: Massachusetts
Institute of Technology Press, 1971).
48 Alasdair MacIntyre, A Short History of Ethics, (London: Routledge, 1968)
268-9.
49 For a systematic statement of Cohen's more recent views, see his History,
Labour and Freedom: Themes from Marx, (Oxford, 1987).
SO I have not forgotten those neo-Kantian Marxian thinkers who treat Marxist
social theory as purely explanatory. I would contend that their writings sacrifice
that unity of theory and practice which is distinctive of the Marxian standpoint.
51 I am particularly grateful to Gerry Cohen, David Miller, Bhikhu Parekh and Bill
Weinstein for their comments on previous versions of this paper.

11 AGAINST COHEN ON PROLETARIAN UNFREEDOM

1 For their comments on an earlier draft of this paper, I am indebted to Scott
Arnold, Gerry Cohen, Jen Elster,David Gordon, Andrew Melnyk, David Miller,
Ellen Paul, Jeffrey Paul, G. Pincione and Andrew Williams. Responsibility for
this paper, including its interpretation of Cohen's argument, remains mine.
2 'Capitalism, Freedom and the Proletariat' in Alan Ryan (ed.), The Idea of
Freedom, (Oxford: Oxford University Press, 1979); 'Illusions about Private
Property and Freedom', J. Mepham and D. Ruben, (eds), Issues in Marxist
336 Notes
Philosophy, vol. IV (Hassocks, Sussex: Harvester Press, 1981); 'Freedom, Jus-
tiee and Capitalism', New uft Review, vol. 125, (1981); 'The SIrUc:ture of
Proletarian Unfreedom', J. Roemer (cd.), AlIIIlylical MII1%Um, (Cambridge Uni-
versity Press, 1986); and 'Are Workers Fon:ed to Sell Their Labour-powcr?',
Philosophy tmd Public Affain, vol. 14, no. 1 (198S).
3 Coben, 'Capitalism', 12.
4 Cohen, 'Illusions', 227.
S ibid.,226-7.
6 Cohen, 'Capitalism', 11-12.
7 ibid., 11.
8 ibid., 16-17.
9 ibid., 17.
10 The 'negative' view of freedom as non-interference shifts easily into a view of
freedom as non-restriction of options. On this see John Gray, 'Negative and
Positive Liberty', John Gray and Z.A. Pelczynski, (cds), Conceptions of Liberty
in Political Philosophy, (London and New York: Athlone Press and St Martin's
Press, 1984) 321-48; reprinted in John Gray, Liberalisms: Essays in Political
philosophy, (London and New York: Routledge, 1989), ch. 4.
11 Cohen, 'Structure', 2S0, 21n.
12 On Berlin's conception of freedom, see Gray, 'Negative and Positive Library',
op. cit.
13 See H. Mareuse, 'Repressive Tolerance', inA Critique ofRepressive Tolerance,
(Boston: Beacon Press, 1968); and H. Arendt, 'The Revolutionary Tradition and
its Lost Treasure', M. Sandel (cd.), Liberalism tmd Its Critics, (New York: New
York University Press, 1984) 239-63.
14 George O. Brenkert, 'Cohen on Proletarian Unfreedom', Philosophy and Public
Affairs, vol. 14 (198S) 93-8; John Gray, 'Marxian Freedom, Individual Liberty
and the End of Alienation', Social Philosophy and Policy, vol. 3, no. 2 (1986)
170-4.
15 I do not mean to suggest that Cohen's is the best statement of a liberal negative
view of freedom, but only that it is Cohen's that I shall deploy in may argument
against him.
16 Cohen, 'Structure', 244.
17 ibid., 245.
18 ibid., 242, 7n.
19 ibid., 242.
20 ibid., 244.
21 ibid., 241.
22 ibid., 248.
23 ibid., 250.
24 ibid., 250.
25 ibid., 248.
26 ibid., 248.
27 C.B. Macpherson, Democratic Theory, (Oxford: Clarendon Press, 1973) 154.
28 Cohen, 'Structure', 248.
29 ibid., 248.
30 ibid., 249.
31 ibid., 251.
32 Cohen comments approvingly ('Structure', 245, IOn) on Elster's perceptive
observation that 'such structures [of collective unfreedomJ pervade social life.'
 Notes 337
33 Cohen, 'Structure" 244.
34 Cohen, 'Capitalism.. 18-19.
35 Cohen, 'Illusion', 228.
36 Cohen, 'Structure', 238.
37 Robert Nozick, Anarchy, State and Utopia, (Oxford: Basil Blackwell, 1974) 30.
38 Cohen, 'Structure', 243, 8n.
39 ibid., 243.
40 Cohen, 'Freedom', 10.
41 Cohen, 'Illusions', 224.
42 Cohen, 'Structure', 242.
43 ibid., 244.
44 ibid., 241.
45 For example, Felix Oppenheim. See his '''Constraints on Freedom" as a Descrip-
tive Concept', Ethics, vol. 95 (1985) 305-9, and Political Concepts: A Re-
construction (Chicago: University of Chicago Press, 1981).
46 Joseph Raz, The Morality of Freedom (Oxford: Clarendon Press, 1986), 14.
47 ibid., 16.
48 For an argument that power is best theorized value-neutrally, see John Gray,
'Political Power, Social Theory and Essential Contestability', D. Miller and L.
Siedentop (eds) The Nature of Political Theory, (Oxford: Clarendon Press,
1983); reprinted in this collection, ch. 15.
49 The idea of a moral notion is explored in J. Kovesi, Moral Notions, (London:
Routledge and Kegan Paul, 1971).
50 See, on this, Raz, The Morality of Freedom.
51 Cohen, 'Capitalist', IS.
52 Cohen, 'Illusions', 232.
53 ibid., 233.
54 On this see John Gray, 'Liberalism and the Choice of Liberties', T.A. Hig, D.
Callen, and J. Gray, (eds), The Restraint of Uberty, Bowling Green Studies in
Applied Philosophy, vol. VII (1985) 1-25; reprinted in John Gray, Uberalisms,
op. cit., ch. 9.
55 See J. Rawls, A Theory of Justice, (Oxford: Oxford University Press, 1972).
56 With reference to Rawls's later writings, I refer especially to 'The Basic Liber-
ties and Their Priority', Tanner Lectures on Human Values, (Salt Lalce City:
University of Utah Press, 1981).
57 See John Gray, 'Marxian Freedom, Individual Liberty, and the End of Aliena-
tion', Social Philosophy and Policy, vol. 3, no. 2, (1986) 180-5.
58 F.A. Hayek, The Constitution of Uberty, (Chicago: Henry Regnery, 1960) 121.
The central context of Hayek's argument is stated in somewhat Marxian fashion
by Jeffrey Reiman, Philosophy and Public Affairs, vol. 16, no. 2, (Winter, 1978),
41: 'The space between a plurality of centers of power may be just the space in
which freedom occurs, and conflicts between the centers may work to keep that
space open ... as a material fact, state ownership might ... represent a condition
in which people were more vulnerable to, or less able to resist or escape from,
force than they are in capitalism. It follows that, even if socialism ends capitalist
slavery, it remains possible, on materialist grounds, that some achievable form
of capitalism will be morally superior to any achievable form of socialism. '
59 Leon Trotsky, The Revolution Betrayed, (New York: Pathfinder Books, 1937)
76.
60 Cohen, 'Capitalism', 258.
338 Notes
61 Hayek, The Constitution o/Liberty, op. cit., 126.
62 For the argument that private property maximizes the liberty even of those who
have none, see John Gray, Liberalism, (Milton Keynes: Open University Press
and Minnesota: University of Minnesota Press, 1986) fi(Hl.
63 Cohen, 'Illusions', 224.
64 On positional goods, see John Gray, 'Classical Liberalism, Positional Goods and
the Politicization of Property', Adrian Ellis and Krishnan Kumar (eds),
Dilemmas 0/ Liberal Democracies, (London: Tavistock, 1983) 174-84.
65 A mass of evidence exists as to the extent of politically enforced social stratifi-
cations in the USSR. A useful survey of some of it is to be found in S. Simis,
'The Machinery of Corruption in the Soviet Union', Survey, vol. 23, DO. 4
(Autumn 1977-8).

12 TOTALITARIANISM, REFORM AND CIVIL SOCIETY

1 For the use of the word 'totalitarian' by Italian theorists of fascism, see L.
Schapiro, Totalitarianism, (London: Macmillan, 1983) 13-15; and E. Nolte,
Marxism, Cold War, (Assen, The Netherlands, 1982) 137-8.
2 A good compilation of the post-War literature may be found in C.l. Friedrich
(ed.), Totalitarianism, (Cambridge, Mass., 1954).
3 For a good sw vey of the literature that argues against the concept of totali-
tarianism in Soviet and East European context, see A. Gleason, 'Totali-
tarianism', Russian Review, vol. 45, (1984) 145-59.
4 For an argument that modern capitalist societies are totalitarian, see H. Mucuse,
One Dimensional Man, (London: Sphere Books, 1964) 105, et. seq.
5 See M. Heller, Cogs in the Soviet Wheel: The Formation 0/ Soviet Man,
(London: Collins Harvill, 1988).
6 R. Pipes, Russia under the Old Regime, (London: Weidenfeld and Nicolson,
1974) 317.
7 Barrington Moore, Jr, SociDl Origins 0/ Dictatorship and Democracy, (London:
Penguin Press, 1967) 206.
8 Leonard Schapiro, Totalitarianism, (London: Macmillan 1972) 23.
9 E. Nolte, Marxism, Fascism, Cold War, (Assen, The Netherlands: Van Gon::um,
1982) section B.
10 The economical exposition of the five-point totalitarian syndrome appears in
CJ. Friedrich, 'The Unique Character of Totalitarian Society' in Fredrich, op.
cit., 47-59.
11 Perhaps the best defence of this view may be found in Pipes, op. cit.
12 A. Walicki, Legal Philosophies 0/ Russian Liberalism, Oxford: Clarendon Press,
1987.
13 Norman Stone, Europe Trans/ormed, 1878-1918, (London: Fontana, 1983)200;
and John J. Dziak, Cheldsty: a History o/the KGB, (Lexington, Mass.: Lexing-
ton Books, D.C. Heath and Company, 1988) appendix, 173-6.
14 lowe this, and other statistics, to M. Heller and A. Nekrich's magnificent Utopia
in Power: The History o/the Soviet Unionfrom 1917 to the Present, (New York:
Summit Books, 1986) 15, et. seq.
15 N. Stone, ibid., 197.
16 N. Stone, ibid., 200, et. seq.
 Notes 339
17 D. Lieven, Russia's Rulers under the Old Regime, (New Haven and London:
Yale University Press, 1989) 290.
18 Heller and Nekrich, ibid., 15.
19 Dziak, ibid., 173, et. seq. The tenn NKYU refers to the People's Commariat of
Justice.
20 Dziak, ibid., 35.
21 Cited in Dziak, ibid., 33.
22 R. Pipes, op. cit., 301-2.
23 A. Besancon, The Sovi.:t Syndrome, (New York: Harcourt Brace Jovanovik.
1976) 56.
24 The significance of the Shakhty trial of engineers, see Heller and Nekrich, ibid.,
211-12.
25 Heller and Nekrich, ibid., 213.
26 Heller and Nekrich, 701.
27 New York Times, July 31,1989.
28 See E.H. Carr, The Russian Revolution/rom Lenin to Stalin, 1917-29, (London:
Macmillan, 1964) and M. Dobb, On Economic Theory and Socialism: Collected
Papers, (London, Routledge and Kegan Paul, 1955).
29 P. C. Roberts, Alienation and the Soviet economy, (University of New Mexico
Press, 1971); T. Remington, Building Socialism in Bolshevik Russia, (University
of Pittsburgh Press, 1984); S. Malle, The Organization 0/ War Communism,
1918-21, (Cambridge University Press, 1985).
30 Dziak, ibid., 32-3.
31 Dziak, ibid., 25.
32 Dziak, ibid., 29.
33.A. Nove, An Economic History o/the USSR, (London: Penguin Press, 1969, 198.
34 Nove, ibid., 266.
35 Nove, ibid., 180.
36 Nove, ibid., 379.
37 Heller and Nekrich, ibid., 264.
38 Heller and Nekrich, ibid., 235-6. It is not intended by Heller and Nekrich, I take
it, to equate the Communist genocides with the Nazi Holocaust, since the latter
clearly has dimensions that are incommensurable with other twentieth-century
genocides. The intention is simply to note that the Stalinist policy of collec-
tivation was genocidal in character.
39 On this, see Robert conquest's excellent book, The Nation-Killers, (London:
Macmillan, 1960).
40 Eugene Lyons, Assignment in Utopia, (New York: Harcourt Brace and Co.,
1937) 572-80.
41 On this, see N. Bethell, The Last Secret, (London: Fontaine, 1974).
42 See the path-breaking study by C. Andreyev, Vlasov and the Russian Liberation
Movement: Soviet reality and emigre theories, (Cambridge: Cambridge Uni-
versity Press, 1987).
43 The nuclear Gulag is that portion of the Soviet system of concentration camps in
which prisoners are required to service nuclear installations and devices.
44 Dziak, ibid., 52.
45 See Dziak, ibid., 47.
46 Ibid., 49. Further data about the 'Changing Landmarks' movement and the Trust
operation are to be found in Heller and Nekrich, Utopia in Power, 148-9; C.
340 Notes
Andreyev, VltJsov and the Russian liberation Movement, 175-17; A. Golitsyn,
New lies for Old, (London: Bodley Head, 1984) 12; Edward Jay Epstein,
Deception, (New York: Simon and Schuster, 1987) 25.
47 Epstein, Deception, 25.
48 Dziak, Chelcisty, 49-50.
49 A. Golitsyn, op. cit., 12. Golitsyn's predictions for the next stage of Soviet
development, made no later than 1983, are perhaps worth quoting in full:
Political 'liberalization' and 'democratization' ... [in the USSR] would be
spectacular and impressive. Fonnal pronouncements might be made about a
reduction in the Communist party's role; its monopoly would be apparently
curtailed. An ostensible separation of powers between the legislative, the
executive, and the judiciary might be introduced. The Supreme Soviet would
be given greater apparent power and the president and deputies greater
apparent independence. The posts of president of the Soviet Union and first
secretary of the party might well be separated. The KGB would be
'refonned'. Dissidents at home would be amnestied; those in exile abroad
would be allowed to return, and some would take up positions of leadership
in government. Sakharov might be included in some capacity in the govern-
ment or allowed to teach abroad. The creative arts and cultural and scientific
organizations, such as the writers' unions and Academy of Sciences, would
become apparently more independent, as would the trade unions. Political
clubs would be opened to nonmembers of the Communist party. Leading
dissidents might fonn one or more alternative political parties. Censorship
would be relaxed; controversial books, play, films, and art would be pub-
lished, performed, and exhibited. Many prominent Soviet performing artists
now abroad would return to the Soviet Union and resume their professional
careers. Constitutional amendments would be adopted to guarantee fulfill-
ment of the provisions of the Helsinki agreements and a semblance of
compliance would be maintained. There would be greater freedom for Soviet
citizens to travel. Western and United Nations observers would be invited to
the Soviet Union to witness the refonns in action ....
'Liberalization' in Eastern Europe would probably involve the return to
power in Czechoslovakia of Dubcek and his associates. If it should be
extended to East Germany, demolition of the Berlin Wall might even be
contemplated.
Western acceptance of the new 'liberalization' as genuine would create
favorable conditions for the fulfillment of Communist strategy for the United
States, Western Europe, and even, perhaps, Japan.
I do not intend to endorse Golitsyn's theories, but simply to note the respects in
which his expectations have been corroborated by events.
50 For an explanation ofthe term, 'the sixth glasnost', see Epstein, ibid., 279.
51 See, for data on the WIN operation, Dziak, ibid., 49-50.
52 Dziak, ibid., 5.
53 P.C. Roberts and M. Stephenson, Marx's Theory of Exchange. Alienation and
Crisis, (Stanford. Calif.: Hoover Institution Press, 1973) 94.
54 For an excellent account of the Austrian calculation debate, see D. Lavoie,
Rivalry and Central Planning: The Socialist Calculation Debate Reconsidered,
(Cambridge: Cambridge University Press, 1985).
 Notes 341
55 See Michael Polanyi, The Logic of Liberty, (Chicago: University of Chicago
Press, 1951).
56 On the Lange-Lerner model, see Lavoie, ibid., ch. 5.
57 I have criticized the market socialism of competing worker-cooperatives in my
book, Liberalisms: Essays in Political Philosophy, (London and New York,
1989) ch. 10.
58 James Sherr, Soviet Power: The continuing Challenge, (London: Macmillan,
1987) 27.
59 P.C. Roberts, Alienatio,. in the Soviet Economy, (Albuquerque, New Mexico:
University of New Mexico Press, 1971).
60 Peter Rutland, The Myth of the Plan, (London: Hutchinson, 1985, 183).
61 Sherr, ibid., 30.
62 Sherr, ibid., 31.
63 Simon Leys, The Burning Forest, (New York: Henry Hold and Co., 1985) 167.
64 Alexander Zinoviev, The Reality of Communism, (London: Paladin Books,
1985).
65 H. Arendt, The Origin of Totalitarianism, (London: Macmillan, 1958).
66 See V. Havel, The Power of the Powerless, (New York, M.E. Shase, Inc., 1985).
67 A. Zinoviev, Homo Sovieticus, (London: Paladin Books, 1985).
68 Francois Thom, Gorbachev, Glasnost and Lenin: Behind the New Thinking,
(London: Policy Research Publications, 1988) 12.
69 For an analysis that runs in parallel with mine, see L. Sirc, What Must Gorbachev
Do?, (London, Centre for Research into Communist Economies, Occasional
Paper Two, 1989).
70 As far as I know, the term 'Ottomanization' was first used in the context of
Soviet affairs by Timothy Gartan Ash. See his book, The Uses of Adversity,
(Cambridge: Grano Books, 1989) 188, 227-31. Details of the secessionist
movements in Lithuania, see Radio Free Europe Research, January 5, 1989; for
similar developments in Soviet Moldavia, see Radio Free Europe Research,
February 9,1989.
71 As reported in The New York Review of Books, (August 17, 1989) 24.
72 See Walicki's contribution to Totalitarianism at the Crossroads, E.F. Paul (ed.),
(New Brunswick and London: Transaction Books, 1990).
73 L. Kolakowski, Main Currents of Marxism, Volume One: The Founders,
(Oxford: the Clarendon Press, 1978) ch. 1.
74 See Eric Voegelin, The New Science of Politics, (Chicago: University of
Chicago Press, 1952) 107, et. seq.
75 Michael Polanyi, The Logic of Liberty, (Chicago: Chicago University Press,
1951) 93.
76 See N. Berdyaev, The Origin of Russian Communism, (London: The Centenery
Press, 1837).
77 See Alain Besancon, The Rise of the Gulag: Intellectual Origins of Leninism,
(New York: Continuum, 1981).
78 See A. Solzhenitzyn, The Red Wheel, (London and New York: 1989).
79 See, especially, Dostoyevsky's great novel, The Possessed (1871-2).
80 For a brilliant de mystification of the French Revolution, see Rene Sedillot, Le
Cout de la Revolution, (Paris, 1987).
81 See Besancon, ibid., ch. 12 as the Jacobin aspect of Leninism. J. Talmon's The
Regime of Totalitarian Democracy, (London: 1952), is also relevant here.
342 Notes
82 Theodore H. von Laue, Why Lenin? Why Stalin? A Reappraisal of the Russian
Revolution, 1900-1930, (New York, Lippincott Company, 1971) 213.
83 For their comments on this paper, I am indebted to Fred Mill, Zbigniew Rau,
Roger Scruton, Stefan Sencerz, Andrezej Walicki. I am particularly indebted to
Ellen Paul for her detailed written comments.

13 WESTERN MARXISM: A FICTIONALIST

DECONSTRUCTION
1 L. Revai, The Word as Deed: Studies in the Labour Theory ofMeaning, G. Olsen
and J. Kahn (eds), (Helsinki: Praxis Press, 1988).
2 L. Wittgenstein, Philosophical Investigations, (Oxford: Basil Blackwell, 1972)
5e, section 2.
3 Concerning Marxism in Linguistics, IV, 1,3.
4 G. Olsen, Sense and Reference in Marxian Semantics, (Oxford University Press,
1980).
5 P. Reimer, Analytical Foundations of Marxian Microlinguistics, (Berkeley,
1982).

14 POST-TOTALITARIANISM, CIVIL SOCIETY AND THE

LIMITS OF THE WESTERN MODEL
I Francis Fukuyama, 'The End of History?" in The National Interest, no. 16,
(Summer, 1989) 3-18.
2 See my 'Totalitarianism, Reform, and Civil Society' in Totalitarianism at the
Crossroads, Ellen Paul (ed.), (New Brunswick and London: Transaction Books,
1990) 97-142, reprinted in this volume, ch. 12.
3 Fukuyama, op. cit. 3.
4 On this, see Utopia in Power: The History of the Soviet Unionfrom 1917 to the
Present by M. Heller and A. Nekrich, (New York, summit Books, 1989) 15, et
seq.
5 See my 'Totalitarianism, Reform, and Civil Society', 102-5.
6 See F.A. Hayek, Individualism and Economic Order, (London: Routledge
1949); and P.C. Roberts, Alienation in the Soviet Economy, (Albuquerque:
University of New Mexico Press, 1971). Michael Polanyi's analysis (which
Roberts develops) may be found in his book The Logic of Liberty, (Chicago:
University of Chicago Press, 1951).
7 For an excellent introduction to the Public-Choice perspective, see James
Buchanan, The Limits of Liberty: Between Anarchy and Leviathan, (Chicago:
University of Chicago Press, 1975).
8 For a good account of the New Economic Policy, see Heller and Nekrich, op.
cit., 108, et. seq.
9 For the analysis of 'Z', see 'To the Stalin Mausoleum', Daedalus, (Winter 1990)
295-344.
10 I have set out the argument against market socialism in greater detail in my
Liberalisms: Essays in Political Philosophy, (London: Routledge, 1989) ch. 10.
For a powerful and scathing indictment of market socialism, see Antony de
 Notes 343
Jasay, Marlcet Socialism: a Scrutiny - the Square Circle, (London, Institute of
Economic Affairs, Occasional Paper 84, 1990).
11 The best defence of market socialism is to be found in David Miller's Market,
Community and State, (Oxford: Oxford University Press, 1989).
12 I have developed the idea of the New Hobbesian Dilemma in my Limited
Government: a Positive Agenda, (London, Institute for Economic Affairs,
1989).
13 The distinction between classical and revisionist liberalism is spelt out in my
Liberalism, (Minnesota and Milton Keynes: Open University Press, 1987).
14 I have discussed Buchanan's work in my paper, 'Buchanan on Liberty', in
Constitutional Political Economy, vol. I, no. 2, (SpringlSummer 1990) 149~8,
reprinted in this volume, ch. 5.
15 I have criticized Hayekian conceptions of cultural evolution in the Second
Edition of my Hayek on Liberty, (Oxford: Basil Blackwell, 1986).
16 The term civil association derives from Michael Oakeshott. See his On Human
Conduct, (Oxford: Clarendon Press, 1975) 108-84.
17 See Arendt's classic, The Origins of Totalitarianism, (New York, Harcourt
Brace Jovanovich, 1979).
18 For a good account of Lagowksi's neo-liberal views, see A. Walicki, 'Liberalism
in Poland', Critical Review, vol. 2, no. 1,8-38.
19 I criticize Rawls' account of the basic liberties in my Liberalisms, ibid., ch. 10.
20 See, especially, Buchanan's Limits of Liberty, ibid.

15 POLmCAL POWER, SOCIAL THEOR.Y AND ESSENTIAL

CONTESTABILITY
1 I refer especially to Steven Lukes, Power: a radical view, (London: Macmillan,
1974) and Lukes's Essays in Social Theory, (London: Macmillan, 1977); and to
W.E. Connolly, The Terms of Political Discourse, (Lexington: D.C. Heath,
1974). See W.E. Connolly, The Terms of Political Discourse, 2nd edn, (Oxford:
Basil Blackwell, 1983) for a response to the arguments of this chapter.
2 The distinction between a concept and its conceptions is made in John Rawls, A
Theory of Justice, (Oxford: Oxford University Press, 1972), 5~, and cited by
Lukes in Power, 27.
3 Lukes, Power, 27.
4 For his discussion of the contributions of Parsons and Arendt, see Lukes, Power,
27-31. For Connolly's discussion, see Connolly, Tenns of Political Discourse,
2nd edn, 114-15.
5 This suggestion is advanced in Martin Hollis, Models of Man (Cambridge,
Cambridge University Press, 1977) 173-80.
6 For this see B.M. Barry, Political Argument, (London: Routledge & Kegan Paul,
1965) ch. 10.
7 See Steven Lukes, 'Relativism: Cognitive and Moral', in Essays in Social
Theory, ch. 8.
8 Lukes, Essays in Social Theory, 2~, asserts that disputes about power and
responsibility may in some cases be resolvable empirically. He makes the same
claim in his 'On the Relativity of Power' in S.C. Brown (ed.), Philosophical
Disputes in the Social Sciences, (Brighton, Harvester, 1979) 261-74.
344 Notes
9 Lukes, Power 45.
10 Connolly, TM Terms 0/ Political Discoune, 2nd edn, cbs 2 and 4.
11 ibid., 68.
12 I have myself considered (vet)' inadequately as I now think) the question of the
lifelong contented slave in D. Robertson and M. Freeman (eds), TIte Frontiers 0/
Political 77Iought, (Brighton, Harvester, 1980), reprinted as ch. 5 of my LibertJl-
isms: Essays in Political Philosophy, (London: Routledge, 1990).
13 Lukes, Power, 32-3, and Connolly, Terms 0/ Political Discoune, 2nd edn,
107-16.
14 Such Wittgensteinian arguments are invoked in an otherwise very forceful
critique of Connolly by Grenville Wall, 'The Conc:epIs of Interest in Politics',
Politics and Society, 5 (1975), 487-510. See also John Plamenatz's earlier
treatment in 'Interests', PolitiCtlI Studies, 2 (1954) 1-8.
15 Lukes, Power, 52-3.
16 Some difficulties in the attempt to combine these claims are explored by Alan
Bradshaw, 'A Critique of Lukes's Power: A Radical View', Sociology, 10
(1976) 121-37.
17 The problems generated for their account by the notion of a power structure are
discussed, but not resolved, by Lukes, Essays in Social Tlteory, 9-10, and
Connolly, Tenns 0/ Political Discourse, 2nd edn, 116-26.
18 See W.B. Gallie, Philosophy and the Historical Understrmding, (London:
Chatto and Windus, 1964) ch. 8, for a slightly revised version of this article.
19 Stuart Hampshire, Thought and Action, (New York: Viking, 1959) 230-1.
20 On this see Alasdair MacIntyre, 'The Essential Contestability of Some Social
Concepts', Ethics, 8 (1973-4),1-9.
21 On this see my 'On Liberty, Liberalism and Essential Contestability', British
Joumcd 0/ Political Science, 8 (1978), 385-402, where I depart from some of the
formulations of an earlier and less satisfactory paper of mine, '00 the Essential
contestability of Some Social and Political Concepts', PolitiCtlI TIteory, 5 (1977)
331-48. A highly relevant and useful contribution to this area of debate is made
by Quentin Skinner in his 'The Idea of a Cultural Lexicon', in Essays in
Criticism, (July 1979) 205-24, especially in IOn on 224.
22 B. Hindess and P. Hirst, Pre-Capitalist Modes 0/ Production, (London:
Routledge and Kegan Paul, 1975) 33.
23 I am indebted here to W.E. Connolly, Appearance and Reality in Politics,
(Cambridge: Cambridge University Press, 1981).
24 Hampshire, Thought and Action, 233.
25 Lukes, Power, 47-9.
26 See S. Milgram, Obedience to Authority, (London: Tavistock, 1974) and R.D.
Laing and A. Esterson, Sanity, Madness and the Family, 2nd edn, (London:
Tavistock, 1971).

16 AN EPITAPH FOR LmERALISM

1 Feinberg, Joel, TM Moral Umits o/the Criminal Law, vol. 1: Harm to Others,
vol. 2: Offence to Others, vol 3: Hann to Self, vol. 4: Hannless Wrongdoing,
(Oxford: Oxford University Press, 1984).
2 Reviewed in the TLS of February 8, 1985.
3 Offence to Others, vol. 2, x.
 Notes 34S
4 ibid., xiii.
5 Harm to Self, vol. 3, 4.
6 Harmless Wrongdoing, vol. 4, 321.
7 ibid., 321.
8 ibid., 322.
9 ibid., 320.
10 Joseph Ru, The Morality 0/ Freedom, (Oxford: Clarendon Press, 1986).
Reviewed in the TLS of June 5, 1987.

17 THE END OF HISTORY - OR OF LIBERALISM?

1 Francis Fukuyama, National Interest, Summer 1989.
2 John Gray, Uberalisms: Essays in PolitiCtlI Philosophy, (London and New
York: Routledge, 1989).

18 THE POLmCS OF CULTURAL DIVERSITY

1 Aurel Kolnai, 'Les ambiguit6s nationales', La Nouvelle Releve, (Montreal,
1946/47_ 533-46, 644-55. Translated by W. Grassl and B. Smith in 'The
Politics of National Diversity', The Salisbury Review, vol. 5, no. 3, (April 1987)
33-7.
2 On this, see my 'Philosophy, Science and Myth in Marxism', in Marx and
Marxism; Royal Institute of Philosophy Lecture Series, vol. 14, G.H.R.
Parkinson (cd.), (Cambridge University Press, Summer 1982) 71-95; and my
'Marxian freedom, individual liberty and the end of alienation', Journal 0/
Social Philosophy and Policy, vol. 3, issue 2, (Spring 1986) 160-87.
3 For an excellent account of the crucial role of Marxist ideology in constituting
and reproducing the Soviet system, see M. Heller and A.M. Nekrich, Utopia in
Power: the History o/the Soviet Union/rom 1917 to the Present, (New York:
Summit Books, 1986); and Alain Besancon's brilliant The Soviet Syndrome,
(New York and London: Harcourt, Brace, Jovanovich, 1976).
4 On this, see Homo SOVieticllS by Alexander Zinoviev, (London and Toronto:
Paladin Books, 1985).
5 L. Kolakowski, Main Currents 0/ Marxism; vol. I, (Oxford: Clarendon Press,
1978) 420.
6 On this, see my paper 'Mill's aod other liberalisms', ch. 12 in my Uberalisms:
Essays in Political Philosophy, (London, Routledge, 1989). .
7 See M. Sandel, Liberalism and the Umits 0/Justice, (Cambridge and New York:
Cambridge University Press, 1982).
8 Roger Scruton, The Meaning o/Conservatism, (Harmondsworth and New York:
Penguin Books, 1980).
9 Isaiah Berlin, Against the Current, (London: Hogarth Press, 1979) 342.
10 Michael Oakesbott, On Human Conduct, (Oxford: Clarendon Press, 1985) 320-1.
11 For the late nineteenth-century transfonnation of France, see Nonnan Stone,
Europe Trans/ormed: 1878-1919, (Cambridge, Mass.: Harvard University
Press, 1984) ch. m, sec. 5.
12 Kolnai, op. cit., 536.
13 Stuart Hampshire, Morality and Conflict, (Oxford: Basil Blackwell, 1983)
138-9.
346 Notes
14 I intend to associate my conception with the rather different view offeml by
Richard Rotty in his 'Postmodem Bomgcois Liberalism'; TM Jamal ofPhilo-
sophy, vol. 10, (1983).
15 I am indebted to Isaiah Berlin and the late Michael Oakeshott for conversations
on some of the themes of this lecture. Responsibility for the thoughts it expresses
remains mine alone.

19 CONSERVATISM, INDIVIDUALISM AND THE POLmCAL

mOUGHT OF THE NEW RIGHT
1 For an excellent survey of the New Right. see Norman P. Barry, TM New Right,
(London: Croom Helm, 1987) and David G. Green, The New Conservatism: TM
Counter Revolution in Politica~ Economic and Social Thought, (New York: St
Martin's Press, 1987).
2 A good account of the role of the lEA in promoting successfully a counter
revolution in economic thought can be derived from several of the essays in A.
Seldon (cd.), TM 'New Right' Enlightenment, (Sevenoaks, B. and L. Booth,
1985).
3 So, especially, Hayek's essay, 'Individualism: true and falsc' in F.A. Hayek,
Individualism and Economic Order, (London: Routledge, 1976).
4 For an exposition of Tory 'wet' views, see Ian Gilmour, Inside Right: A Study of
Conservatism, (London: Quartet Books, 1978).
5 See Michael Oakeshott. On Human Conduct, (Oxford: Clarendon Press, 1975).
6 Alan Macfarlane, The Political TMory of English Individualism, (Oxford,
1978), and Alan Macfarlane, TM Culture of Copitalism, (Oxford. 1987).
7 Karl Polanyi, TM Great Transformation, (Boston: Beacon Editions, 1957).
8 C.B. Macpherson, TM Origin of Possessive Individualism.
9 See J.C.D. Clark's essay in Clark (cd.), Ideas and Politics in Modem Britain,
(London: Macmillan, 1990).
10 The writings of Paul Craig Roberts are among the best sources of the supply-side
argument for the eliminability, or radical reduction, in the business cycle.
11 See B. Parekh, 'The New Right and the Politics of Nationhood', in M. Deakin,
The Next Right: Image and Reality, (London: The Runymeade Trust. 1986). See,
also, R. Scruton's contribution to Clark, Ideas and Politics in Modem Britain,
op. cit
12 This point is well developed in Norman Stone's Europe Transformed 1879-
1919, (London: Fontane, 1984) 201, et seq.
13 See B. Kedourie, Nationalism, (London: Hutchinson, 1960) and K. Minogue,
Nationalism, (New York: Basic Books, 1967).
14 See Joseph Schumpeter's classic work Capitalism, Socialism and Democracy,
(London: Allen and Unwin, 1952).
15 I consider some such desirable measures in my Limited Government: A Positive
Agenda, (London: Institute for Economic Affairs, Hobart Paper 113, 1989).
16 On the failings of Millian individualism, see ch. 12 of my Liberalism: Essays in
Political Liberalisms: Essays in Political Philosophy (London and New York:
Routledge, 1989).
17 For Oakeshott's account of the emergence and character of the anti-individual,
see his On Human Conduct, op. cit.
18 M. Oakeshott, Rationalism in Politics, (London: Methuen, 1962).
 Notes 347
20 WHAT IS DEAD AND WHAT IS LIVING IN LIBERALISM
1 Culture and Value, G.H. von Wright (ed.), (Oxford: Basil Blackwell, 1980) 60.
2 On this, see my Liberalism, (Milton Keynes and Minneapolis: Open University
Press and University of Minnesota Press, 1986).
3 See Pocock's The Machiavellian Moment, (Princeton, N.J.: Princeton University
Press, 1976).
4 See my Liberalism, op. cit., chs 4 and 5.
5 See my Liberalism: Postscript, 'After liberalism', (London and New York:
Routledge, 1989) 239-66.
6 J. Raz, The Morality of Freedom, (Oxford: Oxford University Press, 1986) 194.
7 See 'Michael Oakeshott as a Liberal Theorist' by W.J. Coats Jr, Canadian
Journal of Political Science, XVIII (December, 1985) 773-87; and R.A.D.
Grant, Oakesholl, (London: Claridge Press, 199) 62-3.
8 Isaiah Berlin, The Crooked Timber of Humanity: Chapters in the History of
Ideas, (London: John Murray, 1990) 79-80.
9 Berlin, op. cit., 81-2.
10 Berlin, op. cit., 80.
11 As the Russian fideist L. Shestov puts it, 'History is one thing and meaning
another'. See his Athens and Jerusalem, (Athens, Ohio: Ohio University Press:
1966) 393.
12 John Skorupski, John Stuart Mill (London and New York: Routledge, 1989) 24.
13 J. Raz, The Morality of Freedom, 269, et. seq.
14 Hubert L. Dreyfus, Being-in-the-World: A Commentary on Heidegger's Being
and Time, Division I, (Cambridge, Mass.: The MIT Press, 1991) 262-3.
15 Dreyfus, op. cit., 357, 17n.
16 Dreyfus, op. cit., 280.
17 lowe the term to Bruce Brouwer, from whose unpublished writings on moral
epistemology I have much profited.
18 See my Mill on Liberty: A Defense (London and New York: Routledge and
Kegan Paul, 1983), ch. 3.
19 Stuart Hampshire, Innocence and Experiences, (London: The Penguin Press,
1989).
20 The term "the evanescence of imperfection" originates, so far as I can tell, in
Herbert Spencer's Social Statics: The Conditions Essential to Human Happiness
Specified, and the First of them Developed, (London, 1850, 1st edn; republished
New York: Robert Schaltenbach Foundation, 1970).
21 Stuart Hampshire, Innocence and Experience, 32.
22 Hampshire, op. cit., 33.
23 Hampshire, op. cit., 108.
24 Raz, op. cit., 345.
25 Hampshire, oj>. cit., 106.
26 Raz, op. cit., chs 7 and 8.
27 Hampshire, op. cit., 108.
28 Raz, op. cit., 325.
29 R. Nozick, Anarchy, State and Utopia, (New York: Basic Books, 1974) 151.
30 I. Berlin, Four Essays on Liberty, (Oxford: Oxford University Press, 1968) 130.
31 Raz, op. cit., 194.
32 J. Raz, Facing Up: A Reply, University of Southern California Law Review, vol.
62, March-May, 1989, nos 3 and 4, 1227.
348 Notes
33 lowe this expression to Raz. op. cit. 198.
34 This view was one held by John Anderson.
35 Sec Sandel. Uberalism and the Umits of Justice. (Cambridge: Cambridge Uni-
versity Press. 1982).
36 Sec. for example. Bernard Williams's Ethics and the Umils 0/ Philosophy.
(London: Fontana. 1985).
37 J. Skorupski. op. cit•• 37.
38 Sec H. Dreyfus. On being-in-the-worltl. op. cit. in which Hcidcggcr's version of
plural realism is both expounded and rendered alb'IIctive and plausible.
39 Sec my Uberalisms. last chapter.
40 Sec my 'Totalitarianism. reform and civil society', in Totalitarianism at the
Crossroads, Ellen Frankel Paul (cd.). 97-142.
41 James Buchanan, 'Tacit Presuppositions of Political Economy: implications for
societies in transition'. (unpublished).
42 F. Fukuyama. The End of History'. Natural Interests, (Summer. 1989).
43 See Michael Oakeshott, On Human Conflict. (Oxford: Clarendon Press. 1975).
44 See R. Rorty, Contingency, Irony and Solidarity. (Cambridge, 1989).
45 M. Oakeshott, On Human Conduct, op. cit.
46 M. Oakeshott, The Voice 0/ Uberal Learning, T. Fuller (cd.), (Yale University
Press, 1990) 28.
47 I. Berlin, The Crooked Timber 0/ Humanity: Chapters in the History o/Ideas.
op. cit., 235.
48 On this, sec my Uberalisms, op. cit., ch. II.
49 See, most especially, Donald Livingston, Hume's Philosophy o/Common U/e.
(Chicago: Chicago University Press. 1984); and Nicholas Capaldi. Hume's
Place in Moral Philosophy, (New York: Peter Lang, 1989).
50 For an astute exploration of this prospect, sec E.M. Cioran. History and Utopia.
(New York: Seaver Books, 1987); and Cioran's Anathemas and Admirations.
(New York: Arccde Publishing, Little. Brown and Co. 1991).
51 See, especially. Ronald Dworkin's Law Empire. (Cambridge. Mass.: Harvard
University Press. 1~g6).
52 Discussions with Donal~ Livingston have helped me to frame this point, but its
formulation is mine alone.
53 L. Wittgenstein, Culture and Valu~, (Oxford, Basil Blackwell. 1980).
54 Conversations with the late Michael Oakeshott, with Isaiah Berlin and Joseph
Raz have been of great importance in shaping the thoughts expressed in this
essay. Discussions with members of the seminar on the Discourse of Liberty at
the Murphy Institute of Political Economy at Tulane in early 1991 advanced my
thinking on several of the themes developed herein. I wish particularly to record
my indebtedness to conversations with B. Honig. Eric Mack. Jonathan Riley and
G.W. Smith. Comments by Tristram Engelhardt. James Fishkin. William
Galston, Chandran Kukathas, Kenneth Minogue. Emilio Pacheco and Andrew
Williams were extremely helpful in clarifying the thoughts expressed in the
paper. Conversations over several years with Nicholas Capaldi and Charles King
- often in the context of colloquia conducted by Liberty Fund of Indianapolis -
have helped to form some of the thoughts articulated in this paper. Discussions
with several people at the Social Philosophy and Policy Center - in particular
Eugene Heath, Loren Lomasky, Andrew Mclnyk, Jeffrey Paul and Fred Miller-
helped me frame my arguments more clearly. All the usual disclaimers apply.
Index

Ackennan,B.67,212,239,327 Belkin, D. 199-200

activist thesis 103, 105-6 Belloc, H. 12
Acton, J.E.E.D. 12, 115 Bentham, J. 56, 118, 130
Afghanistan 7, 179 Berdyaev, N. 194
agricultural collectivization 162, 173-5 Berlin, I. 64-9; 117, 130,243,262,
alienation 104-5, 106-8 288-91,301,305,311,320-1,323,
alleigence 269-70 325-6; The Crooked Timber of
Althusser 80, 99-100, 106, 234 Humanity 66; 'Two concepts of
Anderson, J. 268 Iiberty'66-7
Angola 162 Besancon, A. 168, 194
anti-autonomist thesis 105-6 Bismark 164
Aquinas, T. 4, 22,194,291 Bohm-Bawerk, E. von, Karl Marx and
Aragon 256 the Close of his System 74
Arendt, H. 130,214,220 Bohme 103
Argentina 13 Bosanquet 285
Aristotle 4, 21-2, 41, 65, 158,257, Bottomore, T., Dictionary of Marxist
286,291-2,295-7,300-1,313 Thought 74
arts, incommensurability 289-90, 305 Brazil 79
Austria 86, 253 Brendel, A. 69
Austrian School 96, 163, see also Brenkert, G.C. 82, Marx's Ethics of
Hayek, F.A. Freedom 82
autonomy 322-6; concept 225-6, 228; Brennan, G., The Reason of Rules 57
imd interests 223-4; right to 241-3;· Brezhnev, L. 179,187-8
value of 307-9; a variable British Empire 27-8
magnitude 222-3 Brittan, S. 40
Axelos, A. 80 Brodsky,J.69
Brouwer, B. 297
Bachrach 219 Buchanan,J.47-63,212,214,316;
Bacon, F. 34 The Reason of Rules 57
Bakunin, M. 73, 94 Bukovsky, V. 85
Balibar, E. 234 Bulgaria 192,203-4,206
Baltic States 52, 85, 89, 157, 160-1, Burawoy, M., Marxist Inquiries 81
189-90,204,207,209,279 Burke,E.3,13,38,44,262
Baranczak 87 Butler, J. 7
Baratz 219
Barry,B.224 Callinicos, A. Marxism and Philosophy
Becker 57 78, 80; The Revolutionary Ideas of
Bekennan, G., Marx and Engels 74 Marx 78
350 Index
Calvin 314 collectivism 61-2
Canetti, E. 8; Crowds and Power 8 communicative rationality 93
capital, scarcity of 135 communism 4, 156; anti-individualist
capitalism: Cohen's concept of 131-2; 75; common factors 162,206;
errors of interpretation 76-7; economic chaos and political
historical materialism, a theory of repression 82-3; failure of 257-8;
115; Marx's concept 131-2; modem and Fascism 75-6; lawlessness
106-7; on-balance freedom 148-54; under 15-16; universal 28; view of
production relations 115; human nature 255-8
proletarianless 133-4 competition 52-3
capitalist institutions 123-55, 211 concentration camps 160, 171-5, see
capitalist order, stability conditions 273 also Gulag; labour camps
capitalist production, impersonal 105 conceptual analysis 145-7
Carr, E.H. 170 Condorcet 287
Carver, G. 74 Connolly, W.E. 217-31, 233, 235-7
Cheka 167, 171-2 consensus, ethical 312
China 16,52,79,89-91,97,158, 160, conservatism 37-8, 259-65, 271; see
174,189-90,205,247,292,317 also New Right
Christianity 48 Constant, H.B. 23, 212, 268, 285, 321
Church of England 277 constitution political economy 59
Churchill, W. 176 constitutive incommensurabilities 65
civil association 42-6, 265-8, 270, constructivism: critical 48, 50-3;
288-9,315,317 civil society: uncritical 50-1
conception of 287-8; defined 159, constructivist rationalism 35
314-15; destroyed by totalitarianism contract, Hobbesian 13
157-60,202-3; diversity 316-18; contractarianism, indeterminate 47-50,
Hegelian argument for 317; historic 60-1
inheritance 284; history of 316-17; contractual exchange 317
and individualism 274, 276, 279, Comeille, P. 65,289
281-2,319; institutions of 12; Crenson225
liberalism 314-20, 326-7; liberties Critical Legal Studies Movement 92
under 10, 13-14; meliorist 320; Croce 101
objective-pluralist argument for Cuba 162,206
323-5; political framework: for 203-4; cultural: diversity 253-71, 277,
post-totalitarian 52-3, 88, 187-9, 192, 309-10; evolution 51-2; freedom
206-7, 209-10; precondition for 265-6; homogenization 302
prosperity and liberty 246-7; culture 283, 324; common 62, 86; of
prospects in Soviet Union 192, liberty 318-19; self-destructive 38-9
209-10; reconstruction of 62, 187-9, Cummings, E.E. 289
206-7; universalist 320; weakened in Czechoslovakia 86, 88, 91, 190-2,
Westem states 12-14,30,45,212 203-2,206-7
civilization, universal 68
Clarendon, E.H. 7 Dahl 219
Clark, J. 276 Darwin, C. 51, 53, 116
class struggle 101,117-21 Davidson, D. 296
class theory 81-2 de Maistre, J. 64, 177
Coats, W.J. 42 death avoidance 5-7
Cohen, G.A. 69, 83, 95, 100, 110-17, democracy: constitutional 49; mass 11;
122-55; Karl Marx's Theory of modem 4
History 78 democratic institutions, need for 210-14
Cohn, J., Class and Civil Society 81-2 Democritus 22
Coleridge, S.T. 260 Descartes, R. 9. 34
collective identities 7-8, 14-15 determinancy 49-50
 Index 351
Development Thesis 18, 110-15 exchange, voluntary 53
Dicey 321 exploitation 95, 91
disinformation, strategic 161-2, 115-80
dispute 235 fairness 303
Disraeli, B. 16, 215 false consciouness 106
Dobb,M.110 famine 111, 189
Donne, J. 289 Fascism 26, 15-6
Dostoyevsky, F.M. 194 Feinberg, J., The Moral Limits o/the
Dreyfus, H.L. 296 Criminal 239-44
Durkheim, E. 110 Felix, D., Marx as Politician 14-5
Duvalier 315 Ferguson 216, 285
Dworkin, R. 41, 69, 212, 239, 285, Feuerbach 99
281,321 Filmer, Sir R. 1, 286
Dzerzhinsky 112, 118 Fischer, N., Continuity and Change in
Dziak 165-1, 111, 116-8, 180 Marxism 80
force theory 115
Easton, S., Humtlnist Marxism and forceable restraint 140-2
Wittgensteinian Social Philosophy forms of life, and individual lives
80-1 309-10; intrinsic value of 306-13
economic growth, legitimating Foucault, M. 82
condition of market capitalism 211-8 Fourier, F.M.C. 24
economic imperialism 51 France 169,264,269,218
economic planning, see under planning Franco, P. 46
economic value 48, 53-5 Frankfurt School 94, 99, 105
economy, black 16 freedom: and capitalist institutions
Edgley, R. 18 123-55; Cohen's concept 150;
education 46, 266-1, 280 comparative 123; conceptual
EEC81 analysis 145-1; correlativity with
efficiency: in the market economy 48; unfreedom 132; cultural 265-6;
standard of 112 defined 124, 129-31; degrees of
Efimovsky 111 148; inequality and maximization
egalitarianism 286-1 152; and justice 140-2, 141; Marx's
Elster, J. 83,95 concept 130-1; a moral notion 141;
emigre movement 116-9 normative concept 130, 139, 141-2,
emotions 295 144, 146; on-balance between
empiricism, naive 51-8 capitalism and socialism 148-54;
Engels, F. 14, 16,102,198; on-balance judgements of 129; and
Anti-Duhring 115; The German private property 124-9; reducibility
Ideology 256 of collective to individual 131, see
Enlightenment 25, 68,195,268-9,328 also autonomy; liberty; unfreedom
enterprise association 43-4 Freidrich, C. 160
EOtvos, J. 81 French Revolution 194
Epicurus 5 Freud,S. 30, 108, 119
epistomological perspectivism 56-8 Friedman, M. 32,215
Epstein 119 Fukuyama,F.202-3,245-9,321
equality 299-306, 319-20 Fuller, T. 42, 46
ergoneme 198-9 functionalism 110-11
Erhardt 208 Furer, S. 199-200
essential contestability 232-1
Ethiopia 162, 114 Gallie, W.B. 232
Eurasion movement 111-8 Gauthier 48-9, 59-60
Europe 42-3; historic boundaries 86 general equilibrium models 95-6
evidence 236 genocide 162, 161, 114,251
352 Index
Genovese, E. 92 Heidegger, M. 34-5, 29Cr7
Gennany 86,174,190-2,208; aid for Heller, M. 157, 165, 169, 173-4
Soviet Union 168--9, 178-9; Herder, J.G. 64, 67-8, 290, 292
economic adjustment 206-7; Nazi Hindess, B. 234
15~; new boundaries 86, 88 Hirst, P. 234
ghunost85-9O, 172, 175-6, 179-80, historical materialism 77-8, 99-100,
204,245,247-8 106; a species of functionalism
Goethe, J. W. von 198 110-17; a theory of capitalism 115
Goldfarb, J.c., Beyond GlaJ'nost 87-8 history: end-state or telos 100, 114,
good 5, 21-2, 65, 292-3, 312 246; philosophy of 67-8; and
Goodman, N. 57 progress 292, 298--9
goods: fungible 307; incommensurable Hitler 159, 304
65; positional 185, 307; public 309 Hobbes, T. 3-17, 30, 48-50, 52-4,
Gorbachev, M. 16, 187, 191 60-2, 118, 185-6,270,279; Crowds
Gouldner, A. 81; Against and Power 8; De Homine 7-8;
Fragmentation 93-4 Leviathan 7, 13, 258
government: decentralization 269; Hobhouse 285
guardian of civil associ ~ion 275; Holbach 118
limited 30, 40-2, 59-63, 89, 265-8, Homo economic us 48, 5(r8, 61, 113, 185
270,274,314-15; variety of Homo sapiens: biological limitations
legitimate fonns 246, see also state; 122; condition 6, 30; economic
totalitarianism theory of behaviour of 5(r8; history
Grant, R. 44 38; human good 5,21-2,65,292-3,
Gray, J., Liberalisms 245-6 312; interests 112-13; nature 4-5, 8,
Greco-Roman tradition 118, 195, 20-1,29,62,100-10,254-9;
291-2,328 orthodoxy 19; rational Ill; reflexive
Gredeskul 177 thought 234; relation with nature
Green 285 102-3; self-defining 99, 118,258;
Greenleaf, W.H. 40 universal man 255-9, 264;
Greville, F. 262, 291 variability 227, 300-1
guaranteed income scheme 13Cr8 Homo sovieticus 187: 248, 258
Gulag 162, 170-1, 197, see also Hong Kong 203, 316
concentration camps; labour camps Hoover, H. 168
Hopkins, G.M. 289
Habennas, J. 93,201 humanism 286, 307
Habsburg Empire 88 Hume, 0.8-9,18,40,48,52,55.59.
Hampshire, S. 68-9, 232, 235, 264, 62,68--9,270,300,326
298-301,303 Hungary 52, 59. 79. 85-8, 91, 96. 157.
Haraszti 87 160-1,191-2,197,204,207
Hardy, H. 64
harm 241; and interests 239-40 ideal paganism 25
Harrington, J. 7 Idealism 103, 122
Hartwell, R.M. 7Cr7 Idealist constructivism 81
Haskell, F. 69 ideas, history of 64
Havel, V. 87, 187 ideology, theory of 81-2, 106, 119-20
Hayek, F.A. 12,24,32-9,51-3,58,67, Ignatieff, M. 66, 69
74,96,150-1, 181,207,212,214, incommensurability 321; in the arts
273,275; birth and influences 32-3; 289-90,305; and choice 323;
Constitution of Liberty 32; Road to constitutive 301-2; in ethics and
Serfdom 37 politics 290; in human goods and
Hegel, G.W.F. 9, 38,44,80-1,99, evil 292-3, 304; of values 65-7.
103-4,106,119,159,193,256,262, 287-306,311-12
279,322 India 279
 Index 353
individualism 24-5,29,37,254-9, Kripke,S.35,147,312
272-82,316,319; and civil society Kristol, I. 81
274,276,279,281-2; culture of kulaks 174
274-82; historical inheritance 276; Kundera, M. 87
moral 286-7; normative 286-7, Kutepov 178
306-7,312
individualist value theory 48-50 labour: camps 162, 171-5, see also
Industrial Revolution 77 concentration camps; Gulag; and
Institute of Economic Affairs 275 language 198-9; power 123; slave
institutions: critical appraisal 52; legal 162; see also workers
14, 42; role of 254; spontaneous 51, Laclau, F., Hegemony and Socialist
see also government; state Strategy 92
intellectual inversion 39 Lagowski 214
intelligentsia, .radical 92, 94 Laing, R.D. 236
interests: and autonomous choice Lange-Lerner model 182
223-4; and harm 239-40; objective language 32-3; and labour 198-9;
and subjective views of 225-6; philosphy of 197-201; private 35
public versus private 11-12 Larrain, 1., Marxism and Ideology 81
interventionism 36 law: common 36; equality under
intuitionism 294 319-20; politicization of 212; rule of
Iran 247, 317 163,182,204,209,213,315,
irrealism 57 317-18; spontaneous 36-7
Israel 68 legal and political means 14
Italy 26, 169 legal process, politicization of 14, 42
legalism 238-9
James, W. 57, 91 Lenin, V.I. 79, 83, 102, 159-60, 167,
Japan 58, 203, 247, 276, 279, 316 172, 193-4,248,254,257-8; State
Jay, M. 97; Fin-de-Sielce Socialism 91-4 and Revolution 170
Judeo-Christian tradition 24-5, 48, Leontiev, K. 177
118,193,195,286,292,307,328 Lewis, W. 23
justice 303-5, 310-11; concept of 49; Leys, S. 186
and freedom 140-2, 147 liberal age, transitional 24
liberal freedom 21
Kafka, F., Metamorphosis 258 liberalism: analysed 238-50, 283-328;
Kahn, J. 197,201 corporate-capitalist 24; force for
Kamenka, E. 109 cultural homogeneity 260-1 ;
Kant, I. 3, 10, 32-4, 40, 44-5, 54, 58, individuality 259; intellectual
60,147,226,249,285-7,300,307 tradition 284-6; self-effacement of
Kavka60 29,31,242-4; in United States
Kedourie, E. 279 249-50; versus civil society 326-7
Keynes, J .M. 36, 62 liberty 284, 314, 317, 323; civil or
Kierkegaard, S.A. 64 basic 148-9; in a civil society 10,
Klyuchnikov 177 13-14; classical 21; conflicting
Knight 63 66-7; constitutional constraints on
knowledge 81; and the market 181-2; 47; culture of 318-19; in
tacit 34-5,37, 181; theory of 33-5, 57 indeterminate contractarianism
Kojeve, A. 247 60-1; legal limitation of 239-44;
Kolakowski, L. 103, 109, 193,258 negativeJ22; personal or political
Kolchak 177 214; pluralist standpoint 243;
Kolnai, A. 253, 264, 269 theorizing the practice 321-2, see
Konrad 81-2, 87 also autonomy; freedom
Korea 279; North 92; South 203, 316 Lieberson, J. 69
Kraus, K. 32 Lieven, D. 164
354 Index
life forms 293, 298 ~1; as metaphysics 8~1, 101-10;
Locke,J.3,7, 10, 13,35,40,59-60, new 82-3; proletarian class struggle
159,279,285 117-21; self-refutation of 73-4, 83;
Lockhart, R.H. 176 as social science 11~17; in United
Lotze, R.H. 19 States~
Lovejoy 286 Marxism-Leninism: dogma 15; as
Lucretius 19,22 religion 255
Lukacs, History and mass manipulation 160
Class-Consciousness 12(' materialism 20
Lukes, S. 217-31,233,235-7 Mauthner, F. 32
Luxembourg, R. 90 meaning, materialist theory of 198
Lyons, E., Assignment in Utopia 175 media, Western 175
meliorism 286, 299, 320
Macfarlane, A. 275; Origins of English Melville, H., Billy Budd 292
Individuolism 17 Merelman 219
Mach,E.32 meta-conscious rules 34
Machajski, W. 81,94 Michels 12~ 1
MacIntyre, A. 121-2 Michnik87
McLellan, D., Marx: The first hundred Milgram, S. 236
years 78 military repression 189-90
Macpherson, C.B. 77,136,276 Mill, I.S. 35,40,104,212,226-7,
Madison, J. 23 239-42,261,268,280,285-7,
Maine, Sir HJ.S. 12,321 294-5,297-8,312,325-6;
majoritarian rule 129 Considerations on Representative
Malle, S. 170 Government 326; On Uberty 66,
man-worship 29 242,307
Mandel, E. 78-9 Minogue, ](. 279
Mandeville, B 39 Mises, L. von 35, 74, 79, 96, 181
Mao Tse Tung 92, 304 monetary reform 213
Marcuse, H. 80, 100, 105-10, 119, 130, Montesquieu, C. 158, 257
156; An Essay on liberalism 107; Moore, B. 158
One-Dimensional Man 107 Moore, O.E. 20
Margalit, E. and A. 68 moral culture, individualist 49
market economy 35-6, 39, 182,246; moral geometry 9
efficiency in 48; institutional moral inversion 39
framework 51 moral qualities, objective 20
market socialism 87, 15~2, 182, moral regeneration 117-21
21~11 morality: of individuality 61; minimum
markets: failure 54-5; impersonal 54; content 303-4; rational 68; unviable
justification 5~; and knowledge 38
181-2; parallel 184. 188; versus Morgenbesser, S. 69
planning 96-7 Mouffe, C., Hegemony and Socialist
Marr, N.Y. 199 Strategy 92
Marx,](.35,47,54, 73-84,~122, Mozambique 162, 169
129-33,139,170,172,193,254-7, multiculturalism 261,266
260,276; Capital 102; Muslims, in Britain 45
Economic-Philosophic Manuscripts Mussolini, B. 26,159
100; intellectural antecedents 193-4; myth 117-21
personality 75; The German Ideology
102, 256; unstable system of ideas Narveson,J.50,60,124
101; views on human beings 99-122 nation, and state 263-4, 278-9
Marxism: and capitalist self-criticism nation-state 25, 27
74; class theory 81-2; failure of National Socialism 156
 Index 355
nationalism 262, 265 180, 182
NATO 86,88 perfection 65, 291
Nekrich 165, 169, 173-4 personal sovereignty 241-3
Neo-Economic Policy (NEP) 176, 178, perspectivism 2%
180,208 perspectivist epistemology 48
neo-totaliarianism 157 Peterthe Great 187
New Right 272-82; hegemony 272-3; philosophical anthropology 106
policy 274-5, see also conservatism philosophical psychology 33
New Russia 177 philosophy: and practice 41,289; task
New Zealand 14,62,89,272 of321
Newton, I. 296 Pierce, C.S. 57, 91
Nietzsche, F.W. 57, 119,312 Pilsudski 176
Nolte, E. 160; Marxism, Fascism, Cold Pinochet 304
War 75-7; Three Faces of Fascism Pipes, R. 158, 167
75 Plamenatz 115
nomenklatura 162,183,185-6,205,208 planning: incentives 184; inherent
Nove, A. 173 difficulties 35-7, 181-3,207-8;
Nozick, R. 10,32,40-1,60,67,115, versus the market 9fr7
139-42, 304, 327 nuclear disaster Plato 65,291
175 Platonist ontology 109
plural realism 296-9
Oakeshott, M. 5,7,9,34-5,38,61, pluralism, see cultural diversity;
250,263,265,275,281,288-9,315, value-pluralism
320-1,324,326; a liberal 40-6; Pocock 285
Rationalism in Politics 42; 'The Pol Pot 304
Political economy of freedom' 44 Poland 16,52,59,85-9,91,157,160-1,
Oaslter76 189,191-2,203-4,206-7,270,279
objective pluralism 291-5, 299-306, Polanyi, K. 77
312-13,323-5 Polanyi, M. 34, 52,181,194,276
offence 240-1 political economy, classical 47
OGPU 172-7 political life, economics of 11
Olsen, G. 197; Sense and Reference in political regimes: and civil society 318;
Marxian Semantics 201 revealed preference 313
On the Eve 177 politics: a conversation 41-2;
order, spontaneous 48 incommensurability and ethics in 290
Orwell, G. 187 Polsby 219
ownership: communal 125-8, 145, Popper, Sir K. 33, 52, 58
151; and long-term plans 127-8; positional goods 16, 153
with majoritarian rule 129, see also positivism 216-18, 237
property, private post-industrial order 26
post-liberal age, Santanya on 24-31
Paine 287 post-modern liberal conservatism 271
Parekh, B. 278 post-totalitarianism 52-3,87-8, 156,
Pareto-optimality 59 187-9,192,203-4,206-7,209-10,
Parsons 220 248-9,316
paternalism 261, 266; autonomy- power 216-37; class struggle for 82;
maximizing 221; legal 241 collective 219-22, 228; conceptual
patriotism, displaced 255-6 analysis 219-22; essentially
Paul, J. 60 contestable 217-18; political 186;
peace 9-10, 51, 61, 270, 315 positive or normative 216-18; and
Pears, D. 69 responsibility 218, 229-31; and
peasantry, genocide 162,167,174,257 social structure 229-32; structuralist
perestroika 16,85,88-9, 160-1, 163, approach 230-1, 233-5
356 Index
power-free community 227 research programme, degenerating
preferences, and interests 234-41; for 109-10
political regimes 313 resouree allocation: market the only
prices,~ket35-6 rational system 79-80; political
primacy thesis III 153-4; without the market
Prisoner's Dilemma 13, 16, 163, mechanism 79
185-6,193 responsibility: individual 280; and
privatization 182,213,275,278 power 229-31
production: capitalist relaticns liS; Revai, L. 196-201; The Word as Deed
efficient 111-12, 116-17; planned 197
lOS; progress in III revolution 77,106-8,118,165, 194
productive power 100, 113-14 Ricardo, D. 47
productivist thesis 103-4, 106 Riesman, D. 226
progress 43, 286; in the arts 289-90; rights: autonomy 241-3; conflicting
belief in 22-4, 29; in history 292, 240; discourse 14-15; fundamental
298-9; productive III; versus 238; group 212; jargon of 42; moral
general welfare 24 240; theory 310-11; variable 303,
proletarian, class struggle 117-21; see also ownership; property
opportunities for escape from 134-5, risk49,60
144 Roberts, P.C. 170, 180,207; AliellDtion
property: private 123, 124-9, 145, 181, and the Soviet Economy 79-80, 183
183,204,211,317; rights 10, 12,51, Roemer, J. 83,94-7
149, see also ownership Romania 92, 162, 192,203,208,213
Proudhon, P.J. 24, 118 Rorty, R. 321
psychological egoism 7 Rousseau, U. 7, 110,262, 2g1
public choice theory II, 56, 57, 60, 94, Russell, B.A.W. 20
163,182-7,207-9 Russia 177
Russian Liberation Army (ROA) 175
Quine, W. 57, 147 Rutland, P., The Myth of the Plan 183
Ryle,G.34
Racine, J. 65, 289
Ranciere, J. 234 Sadler 76
rape 141-2 Salisbury 275
rationalism, critical 48 Sandel, M. 49, 262. 311
rationality 111-14 Santayana, G. 18-31; 'Alternatives to
Rau,Z.52 liberalism' 25-6, 28-9; DomillDtion
Rawls, J. 40, 48-50, 59-60, 66-7, 92, and Powers 20, 26-8; life and
142,152,214,219,238,246,285-8, philosophical development 19-20;
306, 313-14;A Theory ofJustice neglect of 18-19; Realms of Being
148-9 20; Soliloquies 20; The Life of
Raz, J. 146,286,295,301,303-4, Reason 20; The Sense of Beauty 19;
306-8, 311, 323-4; The Morality of Winds of Doctrine 20
Freedom 65, 243 Satre, J.-P. 256
Reagan, R. 17, 44 Saudi Arabia 288
realism: dispositional ethical 297; Savinkov, B. 176, 178
internal 297-8 Schapiro, L. 159
reason: critical 52; limits to 34-5 Schopenhauer, A. 119
redistribution 59-60 Schutz 231
Rees,J.C.240 Scottish School 47 , 54, see also
Reilly, S. 176 Buchanan; Hobbes; Smith
Reimer, P. 201 Scruton, R. 262, 274, 278
religion 3, 24-5,162,311 secret police 160, 162, 171
Remington, T. 170 self, a social construct 228
 Index 357
self-interest 7-8 Tsarist regime 158, 162, 165-8,203;
self-preservation 7 unstable 210, see also glasnost;
self-reference 120 perestroikil
self-refutation, of Marxism 73-4, 83 Spain 19,264,278
Shackle, G.L.S. 63 species-powers 106
Shakespeare, VV.65,289 species-relativism 103
Shakhty 169 speech 199-200
sharashkil 173-4 Spencer, H. 10, 60
Shaw, G.B. 256 Spender, S. 69
Sheffer, H.M. 64 Spengler, O. 46, 327
Sherr, J. 183, Soviet Power 183-4 Spinoza, B. 9, 34, 55, 271
Shulgin 177 Stalin, J. 90, 92,108,159,173-4,176,
significant affecting 219-22, 229 184,186,199-200,248,304
Simons, H., Economic Policy for a state: coercive power of 15; common
Free Society 44 cultural identity 44-5; as enterprise
Singapore 203 association 43-4; failure 58;
Singer, P. 116 Hobbesian 6-7, 9-10, 13; lex 44;
Skocpol, T., Marxist Inquiries 81 and nation 25, 27, 263-4, 278-9;
Skorupsky(i) 294-5, 297, 312 natural basis 8-9; redistributional
Slavic Dawn 177 role 11-13,60; totalitarian 4, 7, 12,
Smart, B. 82 43, 52-3, see also government;
Smith, A. 47, 285 totalitarianism state, modern 3-17;
social class, and speech 199-200 attempts to streamline 16-17;
social justice 33, 37 expansionist imperative 11-12,
social order, spontaneous 51 37-8; new Hobbesian dilemma 17;
social relations, reflexive 234 resources of 11; weakness of 4, 13
social science: and alienation 104-5; Stephenson,M.181
Marxism as 110-17; realist 101 Stirner 81
social structure, and power 229-31 Stoicism 48
socialism: an utopian prospect 107; Stone, N. 164
collapse of 245; competitive 182; Strauss, L. 65
defined 105; future for 91; subjectivity, ethical 307
impossibility of 35-6; market 87, subordination 138, 305-6
150-2, 182,210-11; on-balance subsistence 114
freedom 148-54; transition to 106-8 Sutton, A. 169
society: normalcy Ill; post-political 110 Sweden 211
Socrates 292 Switzerland 253, 269
Solzhenitzyn, A. 194 Szechenyi, I. 87
Sophocles 289 Szelenyi 81-2
Sorel, G. 64,100--1,117-21
Southey 76 Taiwan 203, 316
Soviet Union 7,16,28,52,68,83,86, Talmon, J. 194
88,149,156-96,247,256-8, Tamir, Y.69
316-17; aid 160-70, 183, 193; Taylor, C. 69
dependancy on the VVest 168-70; technological ethos 108
economy 79-80, 85, 164, 189-90; technological innovation 78, 115-16
future of 189-92; genocide of technology 169
peasantry 162, 167, 174,257; Thatcher, M.H. 89,16,44,275-6,278
independence movements 204, 210; The Change of Signposts 177
intelligence services 179; national Thery, E. 164
conflicts 248; October Revolution Thorn, F. 188
165; prospects for a civil society Tibet 91, 174
209-10; repressive activities 165-8; time-worship 23
358 Index
Tito, 192 value-pluralism 66-8, 284, 288-99,
Tocqueville, [Link] 23,212,268, 302,321--6
276,279,285,321 value-subjectivism 53-5
totalitarianism 156-95; chaos in 180-2; values: conflicts between 66, 291, 301;
conventional metbodolgy 161-3; conservative 37-8; convergence of
cultural origins of 157, 193-5; 93; incommensurable 65-7,
features of 202-3; history of idea of 287-306,311-12; spiritual 19-20;
156; incentive structures 182-7; tested against experience 294-5;
lawlessness 18~; Marxist-Leninist transparency of 295; ultimate
ideology 162-3, 170-2; nature of sources of 311; Victorian 276, 279
156-7; neo-l57; radical modernity Van Gogh, V. 302
of 157-8; reversion to 204-6; veil of ignorance 60
revolutionary character 158; stability Vico, O.B. 9, 64, 67-8, 290
of 187; suppression of civil society Virginia Public Choice School 11, 56,
157-60,202-3,314;transformation 94,163,182-7,207-9
to civil society 159; transition from virtues 65; conflicts 295
202-15; weltanschauung-state 160, Vlasov 175
162,171,190,203; see also Voegelin, E. 193
post-totalitarianism von Laue, T.H. 195
trade theory 117
transitivity 304-5 Walicki, A. 164, 190
Treaty of Rapallo 86, 168, 190 war: all against all 4, 16,212; human
Treaty of Versailles 168 condition 6; of redistribution 12; of
Trotsky, L. 90,150,172,174 religion 3
Trust (Monarchist Alliance of Central War Communism 170-1, 180
Russia) 176-9 Webbs, S. and B. 256
Weber, M. 110
uncertainty 48, I 86; metaphysical 235 welfare, general 24, 29
uncombinability 292-3, 301-2 welfare economics 54-5
under-determination 57 welfare-state 24
unfreedom: collective 123, 143-5; well-being, collective 53-4
defined 136; degrees of 134, 138; Wicksell, K. 47, 50
under different institutions 152-4 Wieseltier, L. 69
Unger,R.97 will 119-20
UnitedKingdom42,62,89, 151, 169, Williams, B. 69
171, 179,263,308; decline of civic Williams, R. 78
society 12-14,45; local government Wittgenstein, L. 8, 32, 35. 38, 46, 81,
269; New Right 273-82 196-7,265,280,283,296,312,321,
United States 23, 27-9, 89,151, 179, 328; Investigations 197
190,266; aid for Soviet Union 169; Wolfinger 219
decline of civic society 4, 12-14,41, Wollheim, R. 68-9
45; federalism 269; Harvard 19; workers: cooperative 151-2; forced
liberal ideology 212, 238, 249-50, selling oflabour 123, 139-40,
327; Marxism in 90-8; New Right 142-5; in socialist state 144-5;
273,276,278; philosophy 91-2 subordination 138-9; see also labour
universal empire 25, 27-8 working class, Marcuse on 108
universal humanity 255-9, 264 Wrangel 178
universalism 286, 299-306; in a civil
society 320 Yalta 86, 88
Ursprache 197-8 Yugoslavia 96
Ustryalov, N.V., Struggle/or Russia 177
Zinoviev, A. 186-7
value individualism 5~