The Political Philosophy of G. A. Cohen: Back to Socialist Basics 9781472528285, 9781474259668, 9781472532701

Citation preview

The Political Philosophy of G. A. Cohen

Bloomsbury Research in Political Philosophy Bloomsbury Research in Political Philosophy presents cutting-edge scholarship in the field of political philosophy. Making available the latest high-quality research from an international range of scholars working on key topics and controversies in political philosophy and political science, this series is an important and stimulating resource for students and academics working in the area. Titles include: The Concept of Justice, Thomas Patrick Burke Morality, Leadership, and Public Policy, Eric Thomas Weber Nozick’s Libertarian Project, Mark D. Friedman Perfecting Justice in Rawls, Habermas and Honneth, Miriam Bankovsky Public Reason and Political Community, Andrew Lister Rawls, Dewey, and Constructivism, Eric Thomas Weber Rousseau and Revolution, editors Holger Ross Lauritsen and Mikkel Thorup

The Political Philosophy of G. A. Cohen Back to Socialist Basics Nicholas Vrousalis

Bloomsbury Academic An imprint of Bloomsbury Publishing Plc

L ON DON • N E W DE L H I • N E W Y OR K • SY DN EY

Bloomsbury Academic An imprint of Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

1385 Broadway New York NY 10018 USA

www.bloomsbury.com BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2015 © Nicholas Vrousalis, 2015 Nicholas Vrousalis has asserted his right under the Copyright, Designs and Patents Act, 1988, to be identified as the Author of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. No responsibility for loss caused to any individual or organization acting on or refraining from action as a result of the material in this publication can be accepted by Bloomsbury or the author. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.



ISBN: HB: 978-1-4725-2828-5 ePDF: 978-1-4725-3270-1 ePub: 978-1-4725-3437-8 Library of Congress Cataloging-in-Publication Data Vrousalis, Nicholas, 1980The political philosophy of G.A. Cohen : back to socialist basics / Nicholas Vrousalis. pages cm. – (Bloomsbury research in political philosophy) Includes bibliographical references and index. ISBN 978-1-4725-2828-5 (hardback) – ISBN 978-1-4725-3437-8 (epub) – ISBN 978-1-4725-3270-1 (epdf) 1. Cohen, G. A. (Gerald Allan), 1941-2009. 2. Political science–Philosophy–History–20th century. 3. Socialism. 4. Equality. 5. Social justice. 6. Historical materialism. I. Title. JC257.C57V76 2015 320.53'1 – dc23 2015010282 Typeset by Integra Software Services Pvt. Ltd.

In memory of Jerry Cohen

Contents Preface Acknowledgements Introduction 1 Historical Materialism 2 Freedom 3 Equality 4 Exploitation 5 Community 6 Socialism and the Market

viii ix 1 17 41 61 83 99 115

Notes 135 Works Cited 155 Index161

Preface This book is the repayment of a debt – one gladly incurred, but a debt nevertheless. I first met Jerry Cohen as a graduate student in 2003. I was an economist-aspiring-to-be-philosopher, who had come down from Cambridge to Oxford to work with Jerry. He was the famous Chichele Professor of Social and Political Theory at All Souls College, then known to me only under his philosophical nom de guerre ‘G. A. Cohen’. At first we didn’t hit it off. Jerry probably thought I was too much of an economist, irreversibly steeped in confusion about normativity, preferences and agency. For my part, I initially thought Jerry was determined to demolish my every idea through analysis, just because he could. But we eventually became good friends and, by June 2009, when I defended my thesis under his supervision, we were singing ‘happy doctorate to you’ in Jerry’s backyard. I am very grateful to Jerry for long and intense discussions on almost every topic of philosophical interest (and therefore on almost every topic). I also greatly miss his political companionship and sense of humour. His death in August 2009 deprived me, and many others, of a comrade and teacher. This deprivation caused me to internalize his demanding philosophical voice as a philosophical superego of sorts. This book is the first full-length study on the unity of Jerry’s thought. Throughout this study I have tried to apply Jerry’s rigorous standards to his own work, as he himself relentlessly endeavoured to do. In this project I have been helped by three of Jerry’s former students: James Hall, Dan Halliday and Faik Kurtulmus. They all read the entire manuscript and made incisive criticisms. Hall was kind enough to provide me with excerpts of his excellent unpublished doctoral thesis on Cohen. I am also grateful to four anonymous reviewers for helpful written comments, audiences at Cambridge, Frankfurt, Leiden, Louvain-la-Neuve and University College London, where parts of this book were presented, and Colleen Coalter at Bloomsbury for her patient encouragement from the start of this project.

Acknowledgements

I thank the following publishers and editors for permission to draw upon or reproduce materials from: ‘G. A. Cohen’s Vision of Socialism’, Journal of Ethics 14 (2010): 185–216. Copyright 2010 by Springer. ‘Jazz Bands, Camping Trips, and Decommodification: G. A. Cohen on Community’, Socialist Studies 8 (2012): 141–163. Copyright 2012 by Society for Socialist Studies. ‘G. A. Cohen on Exploitation’, Politics, Philosophy and Economics 13 (2014): 151–164. Copyright 2014 by Sage.

Introduction The book at hand is a critical study on the unity of Cohen’s political thought. Its critical aspect consists in distilling the polemical crux out of the various debates Cohen was engaged in throughout his philosophical career. Its unifying aspect consists in reconstructing these polemics such that they are presented in the most coherent light possible. The text tracks Cohen’s philosophical development by using certain fundamental concepts as thematic signposts. It does not presuppose any acquaintance with Cohen’s work and assumes only an understanding of the basics of political philosophy. In what follows, I have resisted the temptation to offer a biographical sketch of Cohen. The reason is that others, including Cohen himself, have done this in a more informed and entertaining way than I could.1 In lieu of a biography, this chapter offers a broad-stroke introduction to Cohen’s lifeanimating philosophical project and the intellectual methods by which he pursued it.

Cohen’s theodicies An ancient theological quandary about the existence of God asks how it is possible that an omnipotent, omniscient and omnibenevolent creature allows the existence of evil. This is sometimes called the problem of evil. A theodicy is a putative response to this problem. It is an attempt to reconcile the existence of evil with the existence of God or, more weakly, to offer an explanation that provides rational consolation for all the evil in the world, thereby making life more bearable. Theologians and philosophers through the ages have developed and defended various forms of theodicy, none of which are very compelling.2 In the hands of the Enlightenment – mainly through the philosophies of Kant, Hegel, Feuerbach and Marx – the problem of evil became secularized. The question: ‘why does God allow evil?’ thus

2

The Political Philosophy of G. A. Cohen

became something like: ‘why, and until when, will humanity allow evil?’ With God stripped away from the original problem, consolation issues from a commitment to, and belief in, the eventual emancipation of humanity – one might almost say: humanity’s full appropriation of its own ‘species being’.3 Cohen’s political philosophy is animated by lifelong commitment to the feasibility and desirability of a fully emancipated humanity. It offers what is, in effect, a secular egalitarian theodicy. The theodicy motif manifests itself throughout Cohen’s work, albeit in different forms. Before categorizing these forms, I want to summarize the four major philosophical debates that Cohen was engaged in. The first, with Louis Althusser and the Althusserians, was on the theory of history – roughly, what explains what in Marx’s theory of epochal change. The second debate, with Robert Nozick, was on the nature of freedom and the justifiability of capitalist private property. The third debate, with Ronald Dworkin, was on the metric of justice: if we care about equality, in what respects should we aim to make people equal? And the fourth, with John Rawls, was on the justification of incentive-seeking in the marketplace. The most obviously theodicial motif among these debates recurs in Cohen’s particular brand of Marxism, the subject of the opening chapter. Cohen’s Marxism purports to map the trajectory of unemancipated humanity’s career, from the indignities of slavery and feudalism, ‘through the capitalist vale of tears’ (Cohen 2008, p. 176), to the full human self-emancipation consummated by socialism. This depiction of history as a kind of primordial relay race towards freedom can be non-prosaically characterized as a form of secular theodicy. But the theodicy motif reappears, in another form, in Cohen’s later work. It is especially striking in the context of the third and fourth debates: Cohen’s commitment to the universal realization of freedom and equality survives, well after his allegiance to historical materialism has waned. Michael Rosen goes to the heart of the matter when he writes: To the extent that Marxian socialism inherits the aspiration for the universal realization of justice in history (and Jerry [Cohen] certainly saw it that way) then the thought that Jerry’s ideas about justice had religious roots seems to me obviously true. (Rosen 2010, p. 17)

A missing premiss in Rosen’s argument is that Cohen’s ‘ideas about justice’ are a form of ‘Marxian socialism’. Whether or not Cohen ‘certainly saw it that way’, Marxian socialism and Cohenite socialism are different beasts. Cohen

Introduction

3

went to great pains to describe the former, and distinguish it from the latter.4 But the truth remains that late Cohen’s ‘radical egalitarian socialism’ entails an aspiration ‘for the universal realization of justice’ in history, which early Cohen’s Marxism merely predicts.5 A further methodological unity underlies Cohen’s philosophical inquiries, namely the desire to rethink his own socialist commitments by breaking them down to their constituent parts. If socialists are committed to a set of explanatory claims about history, then what are they? What are their necessary and sufficient conditions? If socialists are committed to a set of ethical claims about justice, or freedom, how are these claims conceptually related? And how do they differ from garden-variety liberal claims? This commitment to rethinking the very basis of his own fundamental convictions sets Cohen’s thought apart and gives it an existential urgency rare among contemporary political thinkers. Cohen’s obstinate desire to go back to socialist basics is in this significant sense reflexive. The rest of this introduction discusses Cohen’s method, his view of the role and significance of political philosophy, his critique of bullshit and his ‘small-c conservatism’. These topics do not explicitly occupy a substantial portion of the main part of the book, but they are crucial to understanding the means by which Cohen relentlessly persecuted his philosophical prey.

Method Cohen was a mastermind of guerrilla warfare: he was at his best locating foibles in an argument, tearing down defences erected to keep out unwanted inferences and breaking up resistance to conclusions he considered warranted. Throughout this book, I try to depict the philosophical theatre of war in which Cohen operates and to set out his moves within it. Cohen’s contribution to philosophy does not, however, consist in building a single army of mutually reinforcing ideas and have it parade over the dead bodies of its defeated opponents.6 Partly for this reason, one aim of this book is to reconstruct Cohen’s views, wherever possible, into a unified set of theses. It is, to borrow Lukacs’ (2009) succinct expression, a study on the unity of his thought. The difficulty with doing this – which is also, I think, the reason why no single work has

4

The Political Philosophy of G. A. Cohen

been written on the entirety of Cohen’s political philosophy – is that most of his ideas are forged in the midst of polemic: they are premisses or conclusions of immanent arguments. An argument is immanent, or ad hominem, in my special sense, when it involves affirmation of the putative opponent’s premisses and subsequent inference, from these premisses to conclusions the said opponent cannot (costlessly) accept.7 This Socratic way of doing philosophy8 is productive, but it is also piecemeal. If you are in the business of testing every claim, you understandably fail to examine the overall structure. The upshot is that any attempt at reconstruction of Cohen’s positive views has to rely substantially on conjecture and speculation about their content and connections. Cohen’s argumentative strategy, in other words, conceals positive commitment behind polemical fog. This work aims to dissipate the fog, in order to uncover the positive commitments that lie behind it. Cohen’s Socratic approach to philosophy is closely allied to his way of doing normative political philosophy, which he labels ‘intuitionist’ or ‘radical pluralist’: In this conception … we determine the principles that we are willing to endorse through an investigation of our individual normative judgments on particular issues, and while we allow that principles that are extensively supported by a wide range of individual judgments can override outlier judgments that contradict those principles, individual judgments retain a certain sovereignty. (Cohen 2008, p. 4)

This ‘Oxford’ approach is explicitly contrasted with the ‘Harvard’ approach of John Rawls, which offers ‘precisely, a theory, an organon for generating principles’ (Cohen 2008, p. 5), and which includes a set of priority rules for weighing principles and judgements against one another. These rules produce what Rawls calls a ‘reflective equilibrium’ between principles and considered judgements. One might be tempted to conclude that Cohen’s philosophically conservative approach9 is bound to also be politically conservative. But this does not follow, precisely because ‘individual judgments retain a certain sovereignty’: on Cohen’s approach, egalitarian beliefs will stand their ground even in the face of Rawlsian priority rules pushing for revision in an inegalitarian direction. Indeed, this is what allows Cohen to ‘rescue’ equality from liberal revisionism: his intuitionism protects egalitarian principle from the holist grinder of Rawlsian reflective equilibrium (Cohen 2008).10 This method has wide ramifications for Cohen’s views in, and of, moral and

Introduction

5

political philosophy, especially that branch of political philosophy concerned with normative principles. I turn to his understanding of the latter presently.

Metaethics As a result of a long-standing polemic with Rawlsian liberalism, Cohen came to defend the unfashionable view that normative principles, such as ‘you ought to keep your promises’ are grounded on – that is, made true by – fundamental normative principles that are ‘fact-insensitive’, that is, not dependent for their truth on empirical claims about the world. And he also claimed that this is a logical truth about the nature of principles. Cohen writes: Suppose someone affirms the principle that we should keep our promises (call that P) because only when promises are kept can promisees successfully pursue their projects (call that F) … Then she will surely agree that she believes that F supports P because she affirms P1, which says, to put it roughly, that we should help people to pursue their projects. It is P1, here, that makes F matter, which makes it support P, but the subject’s affirmation of P1, as opposed to whether or not that affirmation induces her to affirm P itself, has nothing to do, essentially, with whether or not she believes that F. (Cohen 2008, p. 234)

If some (set of) fact(s) F supports some principle P, then that sensitivity of P to F must be explained by some further principle P1, which is itself factinsensitive.11 Cohen is therefore a ‘Platonist’ not in the traditional sense that he is a moral realist12 but rather in the sense that he thinks there are principles which are true no matter how the dirty world of empirical facts turns out.13 This argument raises suspicions on two fronts: there is, first, the suspicion that the facts and principles thesis is hardly a plausible thesis for a Marxist, especially a Marxist of Cohen’s vintage, to defend. And then there is the suspicion that the thesis lacks polemical force against Rawls. I want to argue that the first suspicion, but not the second, is misguided. Regarding the first suspicion: what unites Marxists of all stripes is a commitment to the emancipation of humanity. This includes, indeed entails (for Marxists), an opposition to exploitation of all forms. But exploitation is unavoidable for some significant stretch of human history.14 It follows that the Marxist anti-exploitation commitment is fact-insensitive: no matter how the world turns out, whether one lives in ancient Rome or

The Political Philosophy of G. A. Cohen

6

contemporary London, there is reason to regret ensuing exploitative practices and struggle against them.15 Marxism is therefore compatible with – on some views entails – Platonism (of the Cohen variety). I turn now to the second suspicion: Cohen’s original impetus for defending the facts and principles thesis flows from an opposition to Rawls’ view that fundamental principles of justice should be sensitive to ‘general facts of economics and psychology’, including facts about the existence of a ‘competitive economy’ and largely unplanned production. [T]here is an unarticulated … background principle of equality (something like: ‘One ought not to cause too much inequality’) that explains why the stated fact [of economics] about a competitive economy supports the difference principle, and, for all that Rawls shows, that further principle either itself does not depend on any facts, or points to a still more ultimate principle behind it that does not do so. (Cohen 2008, p. 259)

Let’s assume, plausibly, that Cohen is right about the logical structure of principles. If so, then principle-statements have the general form: if X is possible, then you ought to do X. Facts about possibility are premisses extraneous to (fundamental) principle. Rawls has, I submit, nothing to lose by granting this pair of claims: his political philosophy, and the prominent role assigned by it to the original position, is perfectly compatible with the idea that ‘facts cast normative light only by reflecting the light that fact-free first principles shine on them’ (Cohen 2008, p. 267). Cohen’s metaethics therefore lacks polemical force. But it does have important implications for how Cohen conceives of the role and significance of political philosophy, more generally.

The role and significance of political philosophy Consider the following argument: (1) The sole purpose of political philosophy is to guide political practice. (2) Cohen’s fundamental principles do not guide political practice. Therefore: (3) Cohen’s fundamental principles are, quite literally, pointless.

Introduction

7

Cohen attributes this argument to Rodney Peffer and argues that both its premisses are false. In respect of (1), he writes: One may or may not care about practice, but one may also care about justice, as such, one may be interested in what it is, even if one does not care about practice at all. Political philosophy is, in my view, a branch of philosophy, not a branch of normative social technology. (Cohen 2008, p. 306)

It follows that ‘the question for political philosophy is not what we should do but what we should think, even when what we should think makes no practical difference’ (Cohen 2008, p. 268). In respect of premiss (2), Cohen claims that [F]undamental principles do bear on practice, since they are needed to justify the practice-guiding ruler of regulation. And it makes a big difference, moreover, whether we think the optimal rules of regulation by which we abide are in conformity with or in some tension with fundamental justice in particular. It affects the attitude we have when we obey them, and that can have further effects, such as, for example, on how we transform them when we find that we can do so, or on how we react to situations where the rules of regulation seem not to give us the right directives. (Cohen 2008, p. 307)

This passage enlists a distinction between principles of justice, on the one hand, and ‘optimal’ rules of regulation, on the other. Cohen charges Rawls with running the two together. This is a consequence of Rawls’ identification of justice with the outcome of a ‘constructivist’ procedure, which he calls the original position. That procedure, says Cohen, runs together principles of justice, efficiency and community to produce allthings-considered judgements as to how social institutions ought to be regulated. But in virtue of being all things considered, these judgements cannot be judgements of justice (or of efficiency, of community, etc.). For such judgements are pro tanto, and pro tanto judgements cannot be all things considered. It follows that ‘justifiable injustice’ is not an oxymoron. Moreover, the constructivist misidentification of optimal rules of regulation with principles of justice generates confusion of importantly distinct questions in political philosophy, such as: ‘what is justice?’, ‘what state of affairs ought to be brought about?’, ‘what should the state do?’

8

The Political Philosophy of G. A. Cohen

and so on.16 Such confusion in turn ‘generates disoriented practice: there are contexts where the ideal can be advanced, but is pushed forward less resolutely than it might be, because of a lack of clarity about what the ideal is’ (Cohen 2009a, p. 80). What is the connection between Cohen’s Platonism and his value pluralism? A lot of ink has been spilt criticizing the former view,17 but the connection between the two is rarely set out clearly. Now, one can affirm Platonism and deny pluralism: this is why, I argued, a Rawlsian can costlessly affirm the factinsensitivity thesis. But Cohen’s critics misunderstand him insofar as they claim that Platonism drives his metaethics. On the contrary, it is Cohen’s pluralism that leads him to affirm Platonism: Philosophers like me are not primarily, as philosophers, interested in what should be done in practice, all things considered. We are interested, instead, in what distinct things are worth considering. (Cohen 2004a, p. 4, emphasis added)

I add emphasis to ‘distinct’ because this seems to be at the heart of Cohen’s mature thought. Imagine that: A bunch of us are trying to decide which restaurant to choose. Suppose everybody talks a lot about how good the food is in various restaurants, how much it costs, and how long it takes to get there. Someone, hitherto silent, is uneasy. She feels that we are leaving something out of account. Then she realises what it is: ‘Like, nobody’, she says, ‘is considering the décor!’ This person has made a significant contribution to our practical discussion. (Cohen 2013, p. 146)

If Cohen is right that fact-insensitivity is merely a logical thesis about the nature of principles, then every pro tanto principle (such as: ‘note the décor!’ or ‘note the exploitation!’) either is fact-insensitive or presupposes a distinct fact-insensitive principle. There is, therefore, an important sense in which factinsensitivity is logically posterior to value pluralism. Fact-insensitivity protects the plurality of values, not just from the holist grinder of Rawlsian reflective equilibrium, but also from the juggernaut of empirical reality. Cohen’s retreat to principles therefore has a prophylactic role: his Platonism protects his pluralism, and not vice versa.

Introduction

9

Bullshit I have so far concentrated on Cohen’s methodological outlook in respect of normative political philosophy. But perhaps his way of doing philosophy, more generally, is best explained by outlining his conception, and critique, of bullshit. According to Cohen, bullshit is a pervasive social phenomenon, a study of whose nature is crucial to understanding its proliferation. He notes that academic Marxists tend to encounter a disproportionately large amount of bullshit in the course of their own research. This is no doubt the reason why: Before others taught me to call what they were doing ‘analytical Marxism’, it was my own practice to call it ‘non-bullshit Marxism’. That is a more aggressive phrase, since one might conceivably call what one practices analytical Marxism but also believe in the legitimacy of Marxisms of other kinds, whereas, when you call what you do non-bullshit Marxism, you seem to imply that all other Marxism is bullshit, and, therefore, that your own Marxism is uniquely legitimate. In fact, there exists Marxism which is neither analytical nor bullshit, but, once such … pre-analytical Marxism encounters analytical Marxism, then it must either become analytical or become bullshit. (Cohen 2000a, p. xxvi)

Bullshit, for Cohen, is ‘a product of an intellectually dishonest posture, one, more particularly, that includes an unwillingness to respond in an honest way to criticism’ (Cohen 2000a, p. xxvi). But what exactly is that product? Cohen’s answer issues forth from a critique of Harry Frankfurt. According to Frankfurt: The essence of bullshit is not that it is false but that it is phony. In order to appreciate this distinction, one must recognize that a fake or a phony need not be in any respect (apart from authenticity itself) inferior to the real thing … What is wrong with a counterfeit is not what it is like, but how it was made. This points to a similar and fundamental aspect of the essential nature of bullshit: although it is produced without concern with the truth, it need not be false. The bullshitter is faking things. But this does not mean that he necessarily gets them wrong. (Frankfurt 2005, pp. 47–48)

Unlike the liar, whose main purpose is to deceive, the bullshitter aims to deceive in a particular way, namely by bluffing (Frankfurt 2005, p. 48).

10

The Political Philosophy of G. A. Cohen

Cohen calls this an ‘activity-centred’ account of bullshit, one that derives the nature of bullshit from the nature of the activity of the bullshitter. He adds: It is a limitation of Frankfurt’s article that … he took for granted that the bull wears the semantic trousers … Bullshit as insincere talk or writing is indeed what it is because it is the product of something like bluffing, but talking nonsense is what it is because of the character of its output, and nonsense is not nonsense because of features of the nonsense-talker’s mental state. (Cohen 2013, p. 98)18

Unlike Frankfurt-bullshit, Cohen-bullshit is not primarily concerned with the bullshitter’s activity, but with the result of that activity, namely bullshit: it is the shit, not that bull, that ‘wears the semantic trousers’. It follows that the ‘proper polemical target is bullshit, and not bullshitters, or producers of bullshit, as such’ (Cohen 2013, p. 108). Now, Cohen offers a sufficient, though not necessary, condition for bullshit.19 Cohen-bullshit is said to consist in ‘a certain variety of nonsense, namely that which is found in discourse that is by nature unclarifiable, discourse, that is, that is not only obscure but which cannot be rendered unobscure … ’ (Cohen 2013, p. 104). Cohen-bullshit is unclarifiable unclarity.20 A text is unclear, says Cohen, if ‘adding or subtracting (if it has one) a negation sign from [the text] makes no difference to its level of plausibility: no force in a statement has been grasped if its putative grasper would react no differently to its negation from how he reacts to the original statement’ (Cohen 2013, p. 106). But, contra Cohen, this negation test cannot possibly furnish a sufficient condition for unclarity: statements that lack truth value, like ‘The king of France is bald’, or that express equi-probable, or equi-plausible, eventualities are not necessarily unclear, but must be characterized as such by Cohen.21 More generally, there is reason to think that Cohen’s account of bullshit is itself untenable. Consider statements like ‘the Thames is more than 200.52 miles long’. If this statement is unclear, then that must be in virtue of a certain indeterminacy in the object that ‘the Thames’ stands for (the predicate ‘more than 200.52 miles long’ is not unclear). It must be true, that is, of the Thames that it has fuzzy boundaries. Indeed, no matter how hard we try to precisify that particular fuzziness, we will never be able to fully eliminate it (as long as we are talking about the same thing, namely

Introduction

11

the Thames). The mooted statement is therefore an instance of irreducibly unclarifiable unclarity. But it is clearly not an instance of bullshit. The possibility of unclarity de re gives the lie to Cohen’s account of bullshit.22 What, then, is (or are) bullshit? Perhaps an account that better coheres with Cohen’s own characterization of it as the ‘product of intellectually dishonest posture’ (Cohen 2000a, p. xxvi) would keep the rough outlines of his ontology of bullshit, and his critique of Frankfurt, but define it differently. One such definition might take bullshit to be pretentious nonsense, that is, meaninglessness, triviality, blatant contradiction or patent absurdity presented or expressed as interesting truth.23 It is, I think, this kind of bullshit that mainly exercised Cohen and accounts for his widely confessed intolerance of posturing and academic dishonesty (see, for example, Cohen 2013, p. 95).

Conservatism What unites all of Cohen’s writings in political philosophy is a belief in the feasibility and desirability of a fully emancipated society, which Cohen identifies with socialism or communism. I now want to argue that, even when he tries to attenuate the strenuousness of that belief, he is implicitly affirming its implications. It is Cohen’s radicalism that explains his particular brand of conservatism, the ‘small-c conservatism’ he defended towards the end of his life. In what follows, I first introduce Cohen-conservatism and then try to show that it is explained by, indeed is a necessary concomitant of, his demanding brand of egalitarianism. According to Cohen, ‘conservatives like me want to conserve that which has intrinsic value, and injustice lacks intrinsic value – and has, indeed, intrinsic disvalue’ (Cohen 2013, p. 144). Indeed, justice has lexical priority over (conservative) values, and this is what enables someone to be both a socialist and a Cohen-conservative. Cohen-conservatism consists in a defence of personal value, particular value, and accepting the given. Most of Cohen’s argument centres on the first two modes of valuing, so I shall concentrate on these. In the first mode, ‘a person values something because of the special relation of the thing to the person’. In the second ‘a person values something as the particular valuable thing that it is, and not merely for the value that resides

12

The Political Philosophy of G. A. Cohen

in it, but not … because of her special relationship to the thing in question’. Unlike particular values, personal values do not necessarily possess intrinsic value, that is ‘value independently of consequences’. It follows that ‘both of these valuings … value a thing other than solely on account of the amount and type of value that resides in the thing’ (Cohen 2013, p. 148). I want to unpack this distinction further. Cohen distinguishes explicitly between conserving the valuable and the conservation of value. The former encapsulates the (small-c) conservative attitude of preserving ‘particular intrinsically valuable things, as such’. The latter encapsulates an ‘entirely unconservative’ attitude of indifference between destroying a valuable thing and replacing it with something of equal value (Cohen 2013, p. 152). The former attitude is said to be more apt, morally, for reasons related to non-replaceability: Sure, I want to be loved because of my qualities … but once my lover makes that move from my attractive qualities to me, I want it to be me that he loves, and not just the qualities … I do not want it to be inscribed in my lover’s love for me that I am a good illustration of my particular virtues, and if my qualities decline, then I want love not to alter when it such alteration finds. I want to be loved not as a good bearer of my virtues, but as the bearer of them that I am, or indeed was. (Cohen 2013, p. 154)

Similar considerations extend, says Cohen, to valuable, and therefore ‘lovable’, institutions: Just as you may love people because of who and what they are, rather than just for the value of what they produce and for the value of what they instantiate, so you may love a lovable institution because it is the institution that it is and it possesses the character that it has. (Cohen 2013, p. 148)

Particular values are therefore intrinsically valuable things whose value is not exhausted by their agent-independent intrinsic value.24 Personal values, on the other hand, need not have intrinsic value, but are still pro tanto worth preserving. By way of illustration, Cohen offers the example of an age-old pencil eraser that he would hate to lose: I would hate that even if I knew it could be readily replaced, the same offround shape and the same dingy colour that my eraser has now acquired.

Introduction

13

There is no feature that stands apart from its history that makes me want to keep this eraser. I want my eraser, with its history. What could be more human than that? (Cohen 2013, p. 167)

Cohen is quick to add that none of the standard ‘big-C Conservative’ notions about culture, identity and tradition follow from Cohen-conservatism: it does not, in particular, rule out, or condemn, revolutionary change, like the Russian Revolution (Cohen 2013, p. 170). This is due to the lexical priority assigned to justice. I now want to show that this priority, when set against the overall architectonic of Cohen’s political philosophy, has a role in explaining his attachment to small-c conservatism. In discussion of ‘value-maximizing consequentialism’, Cohen writes: ‘The bearers of value, as opposed to the value they bear, do not count as such, but matter only because of the value that they bear, and are therefore, in a deep sense, dispensable’ (Cohen 2013, p. 155). In Chapter 5, I argue that Cohen subscribed to this kind of ‘consequentialism’ with respect to justice. That is, he thought that justice is primarily a property of agent-independent states of affairs, from which it follows that the ‘bearers of value, as opposed to the value they bear’ are, in some sense, ‘dispensable’. Now, it is true that: This deference to justice does not trivialize the ‘we want to conserve what is valuable’ thought. For conservation of the valuable still sometimes trumps value maximization (and so forth) and it needs something as momentous as justice to trump the (nonjustice) value that we seek to conserve. (Cohen 2013, p. 172)

So the reason why justice can ‘trump’ the personal, or particular, values that ‘we seek to conserve’ is that it is too ‘momentous’, indeed too impartial, to be either a personal, or a particular, value (in Cohen’s sense). Such justice is not, for Cohen, primarily concerned with the ‘bearers of value’, as opposed to the states of affairs in which they feature. Furthermore, as I shall explain in Chapter 3, Cohenite justice is very rigorous, warranting a generalized ethos of equality – indeed, an egalitarian hegemony – conducive to the equal distribution of flourishing in society. Given the demandingness of Cohen’s egalitarianism, and its emphasis on agent-independent states of affairs, we are now in a position to offer an explanation for his small-c conservatism. The commitment to the particular and the personal is here to act as a normative

14

The Political Philosophy of G. A. Cohen

break to the demanding, agent-neutral egalitarianism of the rollercoaster of justice. This explanation resonates with the connection Cohen establishes between his conservatism and Hegel’s idea that freedom is achieved when ‘the subject finds itself in its own object [such that] … the subject is at peace with the object’ (Cohen 2013, p. 143).25 It is impossible to find such peace in the strivings, scuffles, brawls and struggles for justice, whether in the street or in the page. Thus the ‘bias towards existing value will render an injustice more tolerable (than it is for a nonconservative compromiser) when this injustice cannot be eliminated without eliminating intrinsically good things’ (Cohen 2013, p. 172). Cohen’s conservatism is a psychological compensating differential for his radicalism.

Outline of the book This introduction records rudiments of Cohen’s project and method. A better way to understand both the project and the method is through immersion in the polemics where his philosophy was forged. The structure of this book therefore mirrors Cohen’s lifelong attempt to rethink the basics of the socialist tradition: historical materialism, freedom, equality, exploitation, fraternity and the market. Chapter 1 studies Cohen’s defence of historical materialism. It maps his attempt to rethink the fundamental concepts of the Marxist tradition through defence of a ‘technological’ interpretation of Marx. According to this interpretation, social structures rise or fall depending on whether they facilitate development of the material productive forces. After setting out Cohen’s argument for assigning explanatory priority to the productive forces, the chapter discusses his mounting pessimism about the empirical validity of historical materialism and its prospects as a social theory throughout the 1980s. It then broaches some conceptual problems with the very idea of identifying materialism with technological determinism, and defends an alternative account that emphasizes the mode of production. Technological determinism is mistaken, but Cohen’s mistake may well have been necessary: his conceptual strictures have set the standards that any defensible theory of history must meet.

Introduction

15

Chapter 2 studies Cohen’s original forays into normative political philosophy in the late 1970s. It sets out by sketching his immanent critique of Isaiah Berlin. That critique shows that lack of money, that is, poverty, restricts freedom. Cohen’s critical theory of money leads naturally to a critique of private property, encapsulated in the idea that capitalism denies freedom to the propertyless. Freedom is the battleground where Cohen’s lifelong polemics with defenders of capitalist private property are fought. In this connection, the chapter takes up Robert Nozick’s programmatic libertarian argument for the minimal state. It then examines Cohen’s painstaking premiss-by-premiss rebuttal of that argument. It concludes by discussing one brainchild of the Cohen/Nozick debate, the idea of left-libertarianism. Chapter 3 discusses Cohen’s egalitarianism and its connections with Marxism. It kicks off by summarizing the debate between Cohen and Ronald Dworkin and demonstrates its bearing on the legitimacy of market transactions. The chapter then situates Cohen’s position within the ‘equality of what?’ debate, by drawing on his criticisms of Amartya Sen. It then turns to Cohen’s rebuttal of liberal arguments for inequality on grounds of efficiency. This criticism gives rise to his famous insistence that a consistent liberalism must embrace an egalitarian ethos in citizens. The chapter concludes by showing that Cohen’s egalitarianism, forged in the midst of these debates with liberal egalitarians during the 1980s, is in tension with certain socialist beliefs explicitly affirmed by Cohen throughout his career, such as the principle of distribution on the basis of need. Chapter 4 studies Cohen’s account of exploitation. It sets Cohen’s theory against the background of his interpretation of historical materialism and summarizes his critique of the labour theory of value. It then argues that Cohen initially accepted, but eventually repudiated, distributive justice as a basis from which to launch charges of exploitation. The chapter then sketches an alternative account that better coheres with Cohen’s considered view of the nature of exploitative transactions and their wrongmaking features. Chapter 5 studies Cohen’s theory of fraternity or community. It sketches Cohen’s ‘jazz band’ exegesis of Marxian community, and sets out his criticisms of the Marxian vision. It then takes up Cohen’s polemic against the liberalism of John Rawls, most of which took place in the 2000s. Cohen takes issue with Rawls’ contention that his principles of justice furnish a sufficiently robust conception

16

The Political Philosophy of G. A. Cohen

of community. Rawlsian liberals, according to Cohen, fail to live up to their own communitarian self-conceptions. To buttress this conclusion, Cohen takes up incentive-seeking in the marketplace: if liberal community presupposes a concern for the have-nots, then surely it must rule out disproportionately large rewards for those who have already have, namely the talented. Yet liberals tend to balk at this conclusion. The chapter concludes by arguing that Cohen’s conception of community turns out to be substantively in tension with his own theory of equality. Chapter 6 assembles disparate elements of Cohen’s critique of the market, in general, and of market socialism, in particular. It sets out his critique of David Miller’s defence of the market, and discusses Cohen’s eventual conversion to an argument from moral incentives due to Joseph Carens. The chapter then shows that Cohen retained a lifelong commitment to the desirability of decommodifying human relationships. This commitment sustains a withering but enduring allegiance to democratic planning, which the chapter attempts to reconstruct. By way of conclusion, this chapter offers a qualified defence of the principles enunciated in Cohen’s camping trip, one of the models of the free association of producers that socialism is supposed to reflect. The emerging overall picture is that of a remarkably tenacious radical thinker, whose thought may yet form the basis for a revival of socialist theory in the twenty-first century.

1

Historical Materialism According to G. A. Cohen: Marxism sees history as a protracted process of liberation – from the scarcity imposed on humanity by nature, and from the oppression imposed by some people on others … The growth across the ages of the human power to produce defeats scarcity and thereby enables humanity to free itself from both material and social adversity. But the freedom to be had in the productive maturity of the species is necessarily lacking in the process which achieves that maturity. (Cohen 1988, p. vii) It follows that ‘unfreedom, exploitation, and indignity are the price which the mass of humanity must pay for the part they play in creating the material wherewithal’ of their self-emancipation. This view of history bears a strong resemblance to the theodicies developed by Leibniz and Kant in the seventeenth and eighteenth centuries (see the section on ‘Cohen’s theodicies’ in the Introduction). Unlike them, Cohen’s theodicy explicitly repudiates the existence of God or God-like entities. It is a thoroughly secular theodicy, with material production at the centre of its explanatory ambitions. These ambitions are, moreover, irreducibly historical in their object. Marxists of Cohen’s vintage take social change to be not a sum-total of mechanical transformations of nature by human beings but rather a set of transitions from a certain developmental stage to another on the basis of an internal logic. Historical materialism offers a periodization of history and of the processes that propel it. Cohen thus defends: an old fashioned historical materialism … a traditional conception, in which history is, fundamentally, the growth of human productive power, and forms of society rise and fall according as they enable or impede that growth. (Cohen 1978, p. x)

18

The Political Philosophy of G. A. Cohen

History, in this view, just is the temporally extended period of scarcity-imposed unfreedom, that part of humanity’s career necessary to create a bountiful substitute for the meagre provision of nature. Indeed, Cohen explicitly contrasts actual human history with life in mythical Arcadia, where ‘the fruit falls from the tree into man’s lap and men make no history because they do not have to – history is a substitute for nature’ (Cohen 1978, p. 24).1 What makes Cohen’s ‘historical materialism’ historical, therefore, is that it studies the trajectory of unliberated humanity’s career. What makes it ‘materialist’ is that it does so by positing certain kinds of conceptual and explanatory constraints (I shall turn to these in a moment). But what makes it into ‘an old fashioned … traditional conception’? The answer has to do with the centrality attached to the distinction between forces and relations of production and the assignment of explanatory primacy to the former. This is the distinguishing mark of the Marxism of the Second International,2 which shaped Cohen’s intellectual path from childhood (see, for example, Cohen 1995a, pp. 245–249). This chapter studies his attempt to defend the theory lying at the heart of that tradition and much subsequent Marxism: technological materialism.

Definitions Cohen’s Marxism operates on a master distinction between forces and relations of production, on the one hand, and relations of production and superstructure, on the other. This tripartite schema is the brainchild of Karl Marx, who makes deft use of it in his famous Preface to A Contribution to the Critique of Political Economy (Marx 1977 [1859]).3 Its three terms of art need some disambiguation. According to Cohen, some facility counts as a force of production if and only if ‘production occurs (partly) as a result of its use, and it is someone’s purpose that the facility so contribute to production’ (Cohen 1978, p. 32). This definition brings means of production (raw materials and instruments of production, including machinery, physical space, etc.) and human labour power (including strength, skill and knowledge) under the extension of the productive forces. Cohen further notes that if this definition is to have any explanatory value, it has to distinguish between ‘productive activity’ and ‘activity which enables

Historical Materialism

19

or assists production but which is not itself productive’. A soldier supplying security ‘essential to uninterrupted agrarian labour’ engages in unproductive activity, for his service is not ‘materially necessary; it is not imposed by the nature of the soil and the technology available for working it’ (Cohen 1978, pp. 33–34). The distinction between those aspects of production which are materially necessary and those which are not presupposes a further distinction between social and material properties or between social form and material content. Cohen offers an unusually narrow construal of the latter, which excludes the technical division of labour and other forms of direct material relations between producers, from the productive forces. I shall return to this important distinction later. Having tentatively defined the productive forces, Cohen needs a definition of the economic structure, or, in the terms of Marx’s 1859 Preface, the economic base: The economic structure of society is the whole set of its production relations. Production relations are relations of effective power over persons and productive forces, not relations of legal ownership. (Cohen 1978, p. 63)

This definition permits Cohen to cut through the old theoretical knot of so-called juridical definitions. On numerous occasions Marx characterizes the economic structure by drawing on juridical notions, such as property, ownership and rights. Historical materialism, for Cohen, cannot rest on such an equivocation between law and the economy, for juridical properties must be causally explained by economic properties and not the other way around. It follows that these two categories must be conceptually distinct. Cohen’s solution consists in showing that economic properties are uniquely picked out by being constituted by production relations, that is, relations of effective power over human labour power and productive forces. Juridical relations, including legal rights, on the other hand, are uniquely picked out by their character as relations of ownership. What is an effective power? Cohen writes: A man has the power to φ if and only if he is able to φ, where ‘able’ is nonnormative. ‘Able’ is used normatively when ‘He is not able to φ’ may be true even though he is φ-ing … Where ‘able’ is non-normative, ‘He is φ-ing’ entails ‘He is able to φ’.4

20

The Political Philosophy of G. A. Cohen

This provides Cohen with enough conceptual leverage to construct a useful set of correspondences between effective powers, on the one hand, and legal rights, on the other: The power to φ is what you have in addition to the right to φ when your right to φ is effective, and … the right to φ is what you have in addition to the power to φ when your power to φ is legitimate. (Cohen 1978, p. 219)

This is the fundamental step in distinguishing between juridical and economic relations. For it is now ‘logically possible to have a right without the power you have when the right is effective, and to have a power without the right that would make the power legitimate’ (Cohen 1988, p. 34). The ‘legitimacy’ aspect of legitimate powers or the ‘rights’ aspect of effective rights properly belong to the set of juridical relations. This brings us to our final definiendum: the superstructure. Marxists before Cohen traditionally defined the superstructure as the sum total of the non-economic institutions of society, including legal systems, political institutions, ideological apparatuses, etc. This definition generates further indeterminacy: are universities or schools or hospitals parts of the economic structure or of the superstructure? Cohen’s response is that the term ‘institution’ confuses things and should drop out of the definition of the superstructure altogether. Indeed, not only is it impossible to distinguish between ‘basic’ and ‘superstructural’ phenomena on account of salient institutional features (such that one and the same institution performs both economic and superstructural work), it is also possible that one and the same action has both basic and superstructural features. To see this, imagine a community where exploiting landowners hire peasants to plough their fields and policemen to enforce discipline. To do these things, they issue instructions about ploughing to peasants and instructions about peasants to policemen. The novelty of their situation is that the promulgation of instructions takes place, quite literally, in one breath: The peasants and policemen speak different languages, and this enables the landowners to issue both sets of instructions in one and the same breath: they utter phonemes which, as it happens, have suitably different semantic interpretations in the two languages. (Cohen 1978, p. 179)

Historical Materialism

21

Here the landowners perform both ‘basic’ and ‘superstructural’ work in one and the same action. It follows that talk of superstructural institutions, as opposed to mere superstructural facts or properties, obstructs a proper understanding of the relationship between the base and the superstructure. One might ask why all this apparent nitpicking is deemed necessary. The answer is that any well-formed explanatory relation between basic and superstructural phenomena must involve well-formed, clearly individuated relata.5 Cohen adds that the powers constituting the economic structure ‘are usually legitimate because … legal protection, a covering of legal norm, is generally indispensable to the enjoyment of economic power’ (Cohen 1988, p. 34). In other words, ‘a base needs a superstructure’. Cohen therefore owes us an account of how such need translates into explanation. The next two sections elaborate on the explanatory structure of Cohen’s Marxism.

The argument for primacy Cohen’s historical materialism consists in an extensive defence of two claims: the first primacy thesis, according to which the productive forces, narrowly construed, have explanatory primacy over the economic structure, and the second primacy thesis, according to which the economic structure has explanatory primacy over the superstructure. Since most of the controversy and debate surrounding Cohen’s Marxism are concerned with the first thesis, I shall devote most of my attention to it. The argument for the first primacy thesis can be roughly summarized as follows: the productive forces have an autonomous tendency to develop on the basis of asocial facts about human nature. These forces are only compatible with certain economic structures. As the productive forces grow, there will be a tendency for the whole set of production relations to change in a direction facilitating that growth. Concomitant with growth in the productive forces, and the consequent change in economic structure, different classes will shrink and others grow, depending on the extent to which they are capable of presiding over the development of the productive forces. Ruling classes will therefore rise and fall depending on whether they can carry productive power to its optimal levels. It is, therefore, the productive forces that explain

22

The Political Philosophy of G. A. Cohen

economic structures and not vice versa. Here’s a more elaborate reconstruction of this argument: (1) A given type of economic structure is compatible only with a certain type, or certain types, of economic structure. (the compatibility thesis) (2) The productive forces have an autonomous tendency to develop throughout history. (the development thesis) (3) Given (1) and (2), either the economic relations will change ‘without lag along with productive development, or – the theoretically prescribed alternative – there will now be “a contradiction” between forces and relations … the relations become fetters’ (Cohen 1978, pp. 158–159). (the fettering thesis) (4) Given (3), rational economic actors looking to improve their material conditions will develop and exercise class-constituted capacities sufficient to revolutionize the economic structure in the direction of productive growth. (the capacity thesis) The first primacy thesis follows. The rest of this section offers Cohen’s considered views on claims (1) to (4) and refutes common misunderstandings. The next section reviews the main arguments against these claims. The compatibility thesis (1) is defended by Cohen on the intuitively compelling ground that: Slavery … could not be the general condition of producers in a society of computer technology, if only because the degree of culture needed in labourers who can work that technology would lead them to revolt, successfully, against slave status. (Cohen 1978, p. 158)

Consider, by extension of this reasoning, Cohen’s four ‘progressive epochs’ in economic structure. The lowest stage of productive development obtains when there is no surplus, a stage of development that corresponds to the classless structure of primitive communism.6 The next stage involves a small surplus. The very existence of that surplus, argues Cohen, generates a division between those who produce it and those who control it. Indeed, the existence of a surplus not only makes class division possible but also necessary. For, there is no way other than class division to develop the productive forces beyond their modest pre-class-division levels (Cohen 1978, pp. 209–215). Next comes a stage of

Historical Materialism

23

moderately high surplus, which corresponds to capitalist society. Capitalism requires a substantial surplus, for that is ‘necessary for repeated introduction of new productive forces, and thus for regular capitalist investment’ (Cohen 1978, p. 198). The ‘productivity precondition’ of socialism is therefore a ‘massive surplus’, large enough to ‘make it no longer true that most of life and time and energy must be spent joylessly producing means to imperative ends’ (Cohen 1978, pp. 198–199). The incompatibility thesis does not, by itself, entail the first primacy thesis. What is missing is a certain determinate directionality that prevents humanity from being bogged down into slavery, serfdom or indeed wage labour, for the whole of its historical career. This directionality is provided by the development thesis (2), which is importantly distinct from the following claim: (2ʹ) The productive forces have a tendency to develop autonomously throughout history. Cohen points out that (2) does not entail (2ʹ). Here is why: A child has an autonomous tendency to grow up. He is born with a disposition to do so which is not externally instilled in him by, for example, his parents. But it does not follow he has a tendency to grow up autonomously, where that means independently of parental and other assistance. (Cohen 1988, p. 90)

The distinction is of some moment, because all that Cohen’s argument requires is that there exist certain fixed points of human nature conducive to an autonomous tendency of the productive forces to develop. Those critics of Cohen who interpret the development thesis as entailing a mysterious capacity of the productive forces to develop despite what men do are therefore mistaken: it is precisely because of what historically situated agents (individuals, groups or classes) do, not despite what they do, that certain economic and social forms are historically bound to occur (Cohen 1988, pp. 51–83). Now, the argument for the development thesis has two parts: the first relies on an a priori claim about human nature, and the second on an empirical claim about actual historical development. The first claim is that human beings are, on the whole, rational, and that, when faced with scarcity, they will be inclined to take opportunities that expand productive power, and thus

24

The Political Philosophy of G. A. Cohen

reduce scarcity. The second claim seeks to fill in the lacuna in the argument for the development thesis by pointing out that productive progress has been much more common than productive regress in human history. Indeed, regress not only offends human reason but is often productively unfeasible: Once railways are in use, it is hard to go back to horse-drawn modes of transport, partly because after a time the population of horses dwindles and the crafts of coachmaking, grooming, etc. disappear, but also because it is hard to renounce the increased mobility the railways bring. (Cohen 1978, p. 155)

This concludes the argument for (2). And if (1) and (2) are true, then the fettering thesis (3) follows. The latter is of paramount significance for Marxists of Cohen’s stripe. For them, the question of fettering and the question of revolution are insolubly entangled: fettering is revolution written backwards, as it were. In Cohen’s own words: [W]hat circumstance leads to the revolutionary displacement of existing relations of production? … [F]ettering of some or other kind. (Cohen 1988, p. 121)

What exactly is it that putatively obsolete production relations fetter? Cohen’s early view, heavily reliant on Marx’s 1859 Preface, was that such relations fetter the development of the productive forces, where ‘development’ denotes the growth rate of output per worker. Output per worker is, in turn, defined as the ratio between the size of the product (as a mass of use values) and the ‘amount of direct labour required to produce it’7 (Cohen 1978, p. 56). In later works, Cohen came to the conclusion that this account of fettering was too strong and arbitrarily one-sided. For it overlooked a salient possibility distinct from development-fettering, namely use-fettering. An economic structure E development-fetters the productive forces P if and only if E hinders P’s optimal growth. This is Cohen’s original notion of fettering. By contrast, E use-fetters P if and only if E hinders the optimal use of P’s existing productive capacity (Cohen 1988, p. 114). On this notion, feudal relations of production fetter the productive forces when they obstruct worker migration to large factories, whether or not such migration induces optimal growth in the forces. According to Cohen, use-fettering must somehow be incorporated into our understanding of fettering simpliciter:

Historical Materialism

25

Call a system fettering if, given both the rate at which it develops the [productive] forces and how well it uses them, the amount of productive power it harnesses at given future times is less than what some alternative feasible system would harness. (Cohen 1988, p. 117, emphasis added)

Cohen calls this conjunctive account ‘net fettering’. The net fettering revision entails that historical materialism does not, on Cohen’s mature view, determine the viability of an economic system on the basis of indicators like (growth in) real GNP per capita. On the contrary, the extent to which an economic structure fetters the productive forces will depend, at least in part, on ‘how well it uses them’ – that is, on whether ‘certain strongly qualitative judgements of what sorts of uses of productive forces would be most worthwhile’ are fulfilled (Cohen 1988, p. 119). What explains this revision? Two reasons. First, Cohen discerned, very early on, what he called a ‘distinctive contradiction’ in advanced capitalism. That ‘contradiction’ consists in capitalism’s inherent bias in the direction of increasing consumption. For the massive increase in productivity that capitalism makes possible in principle allows for either an increase in consumption or a reduction in toil. But: Capitalism inherently tends to promote just one of these options, output expansion, since the other, toil reduction, threatens a sacrifice of the profit associated with increased output and sales, and hence a loss of competitive strength. (Cohen 1978, p. 304)

Now, although capitalism is both necessary and sufficient for this ‘distinctive contradiction’ (Cohen 1978, pp. 313–317), Cohen’s early account of fettering has no room to accommodate it as such, that is, as a contradiction (in the relevant sense). For in that account ‘contradictions’ arise when, and only when, the economic structure (in this case capitalism) fetters the development of the productive forces, meaning growth in output per worker. Even if capitalism is systematically biased in favour of output-expansion, the said bias is compatible with, indeed necessary for, productive development. As such it cannot be the fruit of contradiction. Thus Cohen’s 1978 emphasis on development-fettering is incompatible with his sapient 1978 diagnosis of advanced capitalism. This is one explanation for his later revisionist gestures in the direction of net-fettering.

26

The Political Philosophy of G. A. Cohen

A second reason for the revision is that Cohen became increasingly pessimistic about the productive prospects of the socialist mode of production, and therefore about the possibilities of social change on the basis of development-fettering. Although this pessimism is not fully made explicit until the mid-1990s, one of its corollaries is that: [S]ocialists who accept the superiority of capitalism as a developer of productive power can defend their preference for socialism on the ground of how it uses productive power only by affirming that socialism offers a better way of life. (Cohen 1988, p. 121)

I shall return to this conclusion in Chapter 2. For the moment note that a necessary final step in any defence of the primacy thesis must tell us something about the constitution of classes. That is, the fettering thesis requires some story about the type of agency which is purported to carry out the revolutionary transformation that fettering blocks. Wright et al. (1992) offer such a story for the case of capitalism: [T]he capacities of the working class are enhanced by capitalist development both because the increasingly universalistic quality of their class interests fosters class alliances and because the development of the forces of production directly enhances their organizational power. (Wright et al. 1992, p. 30)

This sounds like a promising way of vindicating Cohen’s claim of a ‘rough correspondence of interests between ruling classes and humanity at large’ (Cohen 1978, p. 153). But, as Cohen himself points out, a story like this one cannot be enlisted as a premiss in the argument for the primacy thesis. For the story seems to presuppose primacy, rendering an argument for the latter question-begging. More precisely: the Wright story offers a defence of the capacity thesis (4), which is a premiss in the argument for primacy. Any good argument for primacy must not presuppose primacy, but the Wright story does presuppose it. Therefore, the Wright story cannot be enlisted in any good argument for primacy. In light of all this, Cohen is forced to admit that: I do not have a good answer to the question how productive forces select economic structures which promote their development. To be sure, we can say that the adjustment of relations to forces occurs through class struggle.

Historical Materialism

27

But that is not a fully satisfying answer to the stated question, since it does not specify the filiation, or filiations, from contradiction between forces and relations of production to the class struggle which is supposed to resolve it. What activates the prospective new class? What ensures its victory? (Cohen 1988, p. 17)

These are important questions, I shall return to below.

Functional explanation I have now set out the argument for the first primacy thesis but have said little about the exact form explanation is supposed to take if that thesis is true. One novelty of Cohen’s materialism was its defence of functional explanation. Such explanation is commonplace in fields like evolutionary biology but not in social science. (Cohen explicitly distinguishes between his own functional-explanatory machinery and the sociological ‘doctrine of functionalism’, the former being compatible with rejection of the latter (Cohen 1978, p. 284).) Take the common-stock example of hollow bones in birds: the possession of hollow bones normally facilitates flight, and flight is normally a consequence of the possession of hollow bones. Hollow bones are the proximate cause of a healthy bird’s flying disposition. But we know that the main explanation why birds possess hollow bones consists in the fact that such bones tend to facilitate flight. If we take a sufficiently long timehorizon, the possession of hollow bones is explained by that consequence, namely the flying facility it furnishes. More precisely, let property P have consequence C. A consequence explanation provides an explanation of P in terms of P’s tendency to facilitate C, given environmental conditions. Consequence explanations in social theory introduce a similar explanatory structure.8 The defence of functional explanation is central to Cohen’s account of historical materialism because it offers him a way out of the ‘bicausality’ (or ‘overdetermination’) problem plaguing Marxist thought since Plekhanov (1947). The problem, roughly stated, is that the economic structure unquestionably impacts on the productive forces, and the superstructure on the economic structure. How then, can the first and second primacy

28

The Political Philosophy of G. A. Cohen

theses be true? According to Cohen, this quandary can be easily solved by enfranchizing functional explanation into Marxism. Thus (i) a particular set of production relations obtains, as it does, when and because it tends to develop the corresponding productive forces, and (ii) a particular superstructure obtains, as it does, when and because it tends to foster the corresponding set of production relations. (i) is compatible with the relations impinging on the forces and (ii) is compatible with superstructure impinging on the relations. This solution has the added benefit that it rids Marxism of the obscure Althusserian notion of ‘structural causality’ and ‘relative autonomy’ (see the section ‘Is historical materialism dead?’ for one interpretation of that notion).9 Naturally Cohen’s use of functional explanation was controversial and debated extensively in a series of exchanges with Jon Elster (Cohen 1980, 1982, Elster 1980, 1982). The crux of Elster’s argument is that Marxism is not analogous to biology, and that Cohen is therefore not entitled to make use of the kinds of explanations biologists use. According to Elster, the soundness of an explanation presupposes a sound mechanism that turns a consequence C into the (best) explanation for its proximate cause P. In biology, the relevant mechanism is provided by random variation and Darwinian natural selection. But Cohen’s theory of history fails to provide an equivalent mechanism, such that the mooted presupposition of sound explanations is not met. Cohen’s response is that: [B]efore Darwin discovered the chance variation/natural selection mechanism, the belief that species had the useful characters they did because they were useful was already justified. The belief was certainly widely held, by men who had no idea how to elaborate it, and by others, like Lamarck, who had what proved to be an unworkable idea of how to elaborate it … I do not think an innocent reader of Lamarck can avoid the conclusion that he knew that the utility of features explained their presence. It is implausible to deny that he knew that, and not implausible to suggest that historical materialism may be in its Lamarckian stage. (Cohen 1980, pp. 133–134)

This proved to be too optimistic an appraisal of the status of historical materialism. The next section reviews some reasons why many commentators, including Cohen himself, came to think so.

Historical Materialism

29

Problems with the argument for primacy Cohen’s defence of historical materialism attracted a lot of attention and generated numerous criticisms. Indeed, it was the founding document of what may non-presumptuously be called a ‘school of thought’, that of Analytical Marxism (originally dubbed ‘non-bullshit Marxism’).10 This section summarizes some of these criticisms. Many commentators focused their fire on the development thesis. Joshua Cohen (1982), for example, sought to attack a presupposition of that thesis, namely that the putative tendency of the forces to develop is founded on asocial facts about human beings and their environment. Without the asociality constraint the argument for primacy becomes viciously circular. That is, if the development thesis presupposes facts about social structures, including facts about the relations of production, then it cannot be enlisted in an explanation of these facts. Joshua Cohen writes: ‘it is not facts about individuals that are now in question, but facts about the structures that determine the outcomes of individual acts’ (J. Cohen 1982, pp. 264–265). Collective action problems, and the social nature of collective action, undermine the asociality constraint on the development thesis. G. A. Cohen’s response is that there are no a priori reasons to deny that ‘material’ (i.e. asocial) conditions constrain social structures and therefore no a priori reasons to reject his account of the development thesis. To buttress this claim he adduces an example from the field of human sociobiology, which ‘is surely not to be dismissed on the mere ground that complaisant social structures are required to realize the tendencies it posits’ (Cohen 1988, p. 106). In Cohen’s considered view, therefore, the asociality constraint ventures an empirical hypothesis, one subject to confirmation by historians and anthropologists. Cohen’s capacity thesis has also been extensively challenged. One of the claims supporting that thesis is that classes in the periphery of social change will tend to join forces with those classes that promise a better future. But why would such coalition-building occur under capitalism?11 The answer is especially difficult for Cohen, given that he foreswears a breakdown account of capitalism’s fate: Marx was not a breakdown theorist, but he did hold that, once capitalism is fully formed, then each crisis it undergoes is worse than its predecessor.

30

The Political Philosophy of G. A. Cohen

But the forces improve across periods which include crises in which they stagnate. Hence they are more powerful just before a given crisis than they were before any earlier one. It follows that the more severe a crisis is, the more developed are the forces whose progress it arrests. Therefore socialism grows more and more feasible as crises get worse and worse (but not because they get worse and worse). (Cohen 1978, p. 204)

This remarkable passage seeks to explain both how socialism becomes feasible and why it will eventually come to be (seen as) desirable by workers and their allies. Cohen has been criticized in this connection for ‘downplaying’ the significance of class struggle, or for ‘overplaying’ the significance of the productive forces. Such criticisms overlook the obvious fact that historical materialism, for Cohen, is not just a theory about the occurrence of class struggle, and the mechanisms that bring it about but also an account of who will win: Marx was not being untrue to what I claim was his theory when he called on workers, rather than scientists and technicians, to revolutionize society. In encouraging workers to bring about social change he was not asking them to bring about what would explain their doing so … (Cohen 1988, p. 17)

Historical materialism is therefore concerned with the deeper question, namely what enables a particular class to shoulder the historic mission of developing human productive power at any particular time. And if no general explanation of that sort is forthcoming, then ‘no theory of history emerges’ (Cohen 1983a, p. 207). It may be worthwhile, at this point, to point out the momentous significance of some of these questions for Marxists of Cohen’s generation. Those perplexed souls doubting their significance have only to look back to the debate among Marxists c. 1914. At that time, Lenin thought he discerned ripeness for revolution in the productive forces of Western Europe and in the class consciousness attendant to the inter-imperialist rivalries culminating in the First World War. Karl Kautsky, the chief theoretician of social democracy, disagreed. But that disagreement was not about the truth of the primacy thesis, for both Lenin and Kautsky affirmed that thesis.12 And both appositely inferred that, given a sufficiently developed set of productive forces, socialism becomes not merely historically possible but also necessary (see, for example,

Historical Materialism

31

Lenin 1974, Kautsky 1996). What they did disagree about was whether the fettering condition, globally applied, was satisfied in 1917. Cohen briefly addresses these landmark questions in an essay entitled ‘Marxism after the collapse of the USSR’. In that essay, he argues that Soviet decline does not retrospectively refute the primacy thesis. Indeed, it probably vindicates it: [H]istorical materialism does allow for the possibility of a premature revolution against capitalism, provided that it is not successful in the medium or long run … the restoration of capitalism in the Soviet Union confirms the truth of … historical materialism in general: the restoration shows that no social order perished here before all of its possibilities of productive development were exhausted. (Cohen 2000a, p. 392)

The argument that follows the passage just quoted amounts to a defence of technological materialism by Cohen in 2000. This is puzzling, because Cohen himself had severely criticized his early views in a series of essays in the mid1990s. More specifically, he raised two fundamental objections to these views. The rest of this section summarizes these objections. The first objection, which may be called the agency problem, can be seen as a further challenge to the class capacity thesis (4). It arises due to the fragmentation of the working class, which deprives history of the agent with the will and capacity to carry revolution forward. According to Cohen (1995a), the optimism of classical Marxism flowed from its joint ascription of certain properties to one class. That class would encompass the majority of people in society, produce the wealth of society, represent the exploited people in society, and comprise the needy people in society. The concomitance of these four properties in one class was deemed a necessary condition for socialist revolution. But these four properties, says Cohen, have now come apart. It follows that there is ‘no group with both (because of its exploitation, and its neediness) a compelling interest in, and (because of its productiveness and its numbers) a ready capacity to achieve, a socialist transformation’ (Cohen 1995a, p. 8). Indeed, he claims that ‘the old (partly real, partly imagined) agency of socialist transformation is gone and there is not, and never will be, another one like it’ (Cohen 2000b, p. 112). Cohen’s second challenge to his own early Marxist self-confidence may be called the scarcity problem. This can be seen as a challenge to the development thesis (2):

32

The Political Philosophy of G. A. Cohen

[T]he development of the productive forces now runs up against a resource barrier. Technical knowledge has not stopped, and will not stop growing, but productive power, which is the capacity (all things considered) to transform nature into use-value, cannot expand pari passu with the growth of technical knowledge, because the planet Earth rebels: its resources turn out to be not lavish enough for continuous growth in technical knowledge to generate unceasing expansion of use-value. (Cohen 1995a, p. 7)

It follows from these two rebuttals that today: Marxism has lost much or most of its carapace, its hard shell of supposed fact. Scarcely anybody defends it in the academy, and there are no more apparatchiki who believe that they are applying it in Party offices … Its shell is cracked and crumbling, its soft underbelly is exposed. (Cohen 1995a, p. 6)

The next section argues it is not just the empirical carapace of technological materialism that has cracked and is crumbling.

The content/form distinction I argued above that, despite increasing pessimism about the viability of technological materialism, Cohen never fully repudiated that theory. Indeed, he seemed to offer a partial defence of it in 2000. By the mid-2000s Cohen used to joke that ‘it’s not theory that’s mistaken, it’s history’. The quip seems to imply that technological materialism has been falsified, or lost its ‘shell of supposed fact’, although the theory is, and always was, conceptually coherent. The latter claim is not obvious. In this section, I argue that the conceptual apparatus Cohen mobilizes to defend historical materialism is not up to the task he set it. Historical materialism gets part of its name from the idea that certain properties, namely material kinds, explain economic and spiritual phenomena, namely social kinds. In a slogan: material content (functionally) explains social form. Cohen’s 1978 opus magnum develops that idea through an original reconstruction of Marx’s texts. It defines material properties as follows: M (or S) is not S in virtue of what is necessary and sufficient to make it M. A man or slave is not a slave in virtue of what is necessary and sufficient to

Historical Materialism

33

make him a man. A set of means of production or portion of constant capital is not constant capital in virtue of what is necessary and sufficient to make it a set of means of production. (Cohen 1978, p. 92)

The relevant contrast class is social properties: A description is social if and only if it entails an ascription to persons – specified or unspecified – of rights or powers vis-à-vis other men. (Cohen 1978, p. 94)

Cohen thinks these definitions illuminate numerous distinctions that Marx draws, for example, between value and use-value, social and technical division of labour, labour process and its social form, etc.13 He also thinks that the theory stands, or falls, on the basis of the soundness of that distinction. Thus, if it can be shown that some material properties are necessarily also social properties (in Cohen’s sense), then the distinction collapses, and the conceptual integrity of Cohen’s materialism is undermined. In an important paper, B.T. Coram (1989) sets out to refute the material/ social distinction by looking at cases where the two properties are coinstantiated. Coram’s argument has four steps: (i) Some production relations are productive forces. (ii)  Some production relations that are productive forces are also work relations. (iii) Such work relations are social properties. And since (iv) All productive forces are material properties. it follows that, (v) Some productive forces are both material and social. The technical division of labour is perhaps the most widely accepted instance of claims (i) to (iii). Cohen resists (i) on the grounds that ‘relations obtaining when tasks are divided as [a principle for allocating tasks] prescribes are neither used nor owned’ (Cohen 1978, p. 114). He infers that such relations cannot constitute productive forces. But, pace Cohen, the production relations established through the division of labour can come under effective control in

34

The Political Philosophy of G. A. Cohen

exactly the same way that machines and labour power do. They are therefore used and owned (in the relevant sense). Cohen also resists (ii), on the grounds that work relations, that is, material relations between direct producers, are never productive forces. He writes: [K]knowledge of ways of organizing labour is a productive force, part of managerial labour power, but the relations established when that knowledge is implemented are not productive forces. (Cohen 1978, p. 113)

This is inconsistent with Cohen’s definition of the forces. Operational factory machines count as productive forces, whether or not they count as instances of ‘implemented knowledge’. The crucial remaining step is (iii). Recalling the sufficient conditions for social properties mentioned four paragraphs back, Coram writes: Some [work relations] will meet these requirements because information costs and the division of labour entail organization and authority. The authority to make decisions entails the right to make them, and this right is only effective if it is accompanied by the power to have them carried out. If so, the authority entails the power. (Coram 1989, p. 223)

One might venture the further generalization that the task of abstracting material from social properties through redescription of the latter is philosophically misguided. An analogy with the metaphysical thesis of reductive physicalism readily suggests itself: the ambition to understand the dynamics of golf balls does not require that the physical universe be populated with properties more ‘fundamental’ than, or non-inferrable from, golf balls. Positing such a constraint would be superfluous, indeed incoherent. Similarly with historical materialism: the ambition to understand the dynamics of production relations does not require that the social universe be populated with properties more ‘fundamental’ than, and non-inferrable from, production relations.14 The distinction between material content and social form, as Cohen draws it, is conceptually flawed. What does this objection do to the argument for primacy? Its main upshot is that the compatibility thesis becomes trivially true. Suppose the productive forces P consist, in part, of work relations W. W is, in turn, constitutive of economic structure E. Then P can’t possibly be incompatible with E. Here’s

Historical Materialism

35

an example. Say the productive forces under advanced capitalism are based on computer technology. These productive forces consist, in part, of computerbased work relations (from claim (ii) above). And since such work relations have implications for economic rights and powers (from (iii) above), the productive forces under advanced capitalism turn out to be pro tanto incompatible with slavery by definition. The resulting circularity deprives the compatibility thesis of its teeth, and the primacy thesis of its explanatory bite.

Is historical materialism dead? In the previous section, I noted that Cohen’s definition of the productive forces is excessively narrow because it excludes work relations. Their inclusion makes the definition of the forces more compelling, but brings the whole edifice of technological materialism to the brink of incoherence. Indeed, it implies that Cohen’s hand-in-glove distinction15 between material and social properties fails. Does this mean that historical materialism is dead? A pessimistic conclusion along these lines is not unavoidable. Cohen’s Marxism can instead be seen as a fresh start, another self-conscious attempt to make socialism scientific, if you will. Its ambition to revive the grand explanatory ambitions of classical Marxism may still be within reach. The rest of this section discusses an interpretation based on critical engagement with Cohen’s writings, which is due to Richard Miller, and concludes with brief discussion of what Marxists have to lose if they grant a positive answer to the question heading this section. Miller (1984a) defends a non-technological interpretation of historical materialism, one based on a ‘mode of production reading’ of Marx. Miller wants to keep Cohen’s framework more or less intact, but drop the content/ form distinction, and therefore the explanatory primacy of content. Miller writes: In Marx’s view, both stable social structure and dramatic social change are ultimately based on the mode of production, the activities, facilities and relationships, material and social, through which material goods are produced. The mode of production consists of productive forces in the technological sense, productive forces such as work relations, in my broader sense, and relations of production. (Miller 1984b, p. 74)

36

The Political Philosophy of G. A. Cohen

In the Millerian schema, the productive forces can’t explain the work relations that constitute them (except in the trivial sense in which intersecting straight lines explain the existence of angles). The productive forces cannot, therefore, explain the economic structure of society. Figure 1.1 represents different accounts of the conceptual geography of historical materialism using Venn diagrams:

Figure 1.1  Conceptual map of materialisms

On the Cohenite interpretation, technology, work relations and the economic structure form three disjoint sets. Crucially, the productive forces do not intersect with work relations, although the union of these two sets carves out the set of material properties. On the Millerian interpretation, the productive forces and the economic structure form two distinct but intersecting sets. Work relations are their intersection. On this interpretation at least some work relations are both material and social properties, which is why the two types conceptually overlap. Part of Miller’s explicit motivation for this revision is that it makes room for the symmetrical explanatory relationship between forces and relations of

Historical Materialism

37

production that Cohen (1978, p. 138) disparagingly calls the ‘zig-zag dialectic’. An upshot of that ‘dialectic’ is that the relations of production, and more specifically (the nature of) control over economic surplus, take centre stage once again. This is not only more faithful to the writings of Marx, says Miller, but also makes actual history relevant to historical materialism. The main explanation for the rise of the medieval manor, or the decline of manufacturing, for example, need not be traced back to the productive forces, but can ultimately be explained by contradictions within the mode of production itself: In feudalism, the lords of the manor dominate, not because undominated peasants would have been less productive but because the lords’ and the peasants’ locations in the mode of production give the former so much more coordination, mobility, and concentrated leadership in the use of force. (Miller 1984a, p. 227)

The mode of production interpretation broaches a sort of explanatory pluralism that assigns explanatory priority to none of the entities constitutive of modes of production (technology, work relations, economic structure).16 This has numerous advantages. For one, it is now conceptually possible to defend the development thesis without recourse to asocial premisses, along the lines canvassed by Levine and Wright (1980). The Millerian interpretation also reduces the explanatory burden assigned to material properties in the first primacy thesis. For even if Elster is right, and mechanism is a necessary condition for the validity of a putative explanation (see the section ‘Functional explanation’ above), the mode of production interpretation makes sufficient room for mechanism-friendly intentional or quasi-intentional explanations.17 Now, the mode of production interpretation is materialist in that it preserves the second primacy thesis almost intact. It is the mode of production that has explanatory priority over the political, legal and juridical (in a word: spiritual) phenomena covered by the superstructure, and not vice versa. Cohen himself devoted considerable attention to spelling out exactly which phenomena such priority covers. In doing so, he distinguished between inclusive and restricted historical materialism. Unlike its inclusive variant, Restricted historical materialism does not say that the principal features of spiritual existence are materially or economically explained … [It] only

38

The Political Philosophy of G. A. Cohen

asserts that the material determines the spiritual to the extent necessary to prevent the spiritual from determining the material. (Cohen 1988, pp. 159–160)

On this view (shared by Richard Miller), materialism would be conclusively refuted if one could offer examples where religious charisma, political leadership, or military genius could be shown to effect epochal transformation, without recourse to claims about economic structure. This is why so much Marxist firepower has been concentrated against Weber’s (1905) explanation of the rise of capitalism in terms of Calvinist ideology, to take one prominent example.18 The pluralism of the mode of production interpretation flattens the explanatory hierarchy of Cohen’s materialism. Its obvious downside is that it lacks the analytical elegance of that hierarchy. It also leaves us in the dark as to which structure in the mode of production is actually doing (most of) the explanatory work. One might respond that Miller’s interpretation only provides us with a materialist armoury designed to catch spiritual prey. What weapon is used in each offensive will depend on the nature of the weapon and of the prey. But the mooted objection runs deeper. For if we stick to the positivist model of scientific explanation favoured by Cohen (1978),19 then the mode of production interpretation is a non-starter. The reason is that the latter offers many, perhaps too many, ways to explain away recalcitrant data, meaning ad hoc ways to reinterpret the theory such that it accommodates observations that appear to disconfirm it. This is not true of technological materialism, which seems more straightforwardly confirmable. Miller (1984a, p. 277 and passim) grants this, but argues that Cohen’s positivism presupposes an obsolete and untenable set of constraints on what counts as a scientific theory. In the light of all these difficulties, it may be worth entertaining a more radical possibility, namely that Marx himself had no theory of history, and – more strongly – that he did not need such a theory. These are strong claims, and I will not attempt to defend them here. I shall, however, offer rudiments of an argument in that direction. Marx’s justly celebrated and most important work, Capital, is subtitled A critique of political economy. In that work Marx mounts an immanent critique of the claims and categories of political economy. To accomplish this task he needs some rough-and-ready conceptual machinery

Historical Materialism

39

that serves the purpose of his critique. It does not follow, and it is false, that he also needs a theory of history, in the sense of a set of a priori categories that is to be applied transhistorically to explain social change from antiquity to the present day.20 Pace Wright et al. (1992), materialist categories need not be the social theory equivalent of the periodic table. If all this is true, then perhaps present-day Marxists should not be trying to find new ways to resuscitate the machinery of historical materialism. They should rather be getting down and dirty with the critique of political economy. And that almost certainly includes engaging with contemporary political philosophy. In so doing they will, no doubt, get to use of some of the tools Marx developed on the way, discard others, and develop new ones.21 Cohen did not go down the critique of political economy route after the publication of Karl Marx’s Theory of History. But he did go down the same conceptual neighbourhood, through life-long engagement with egalitarianism and systematic critique of liberalism. One can only conjecture as to the motives behind these moves. Increasing pessimism about the explanatory credentials of historical materialism, and the consequent ‘intellectual need to philosophize’ (Cohen 1995a, p. 8), must have played a role. Cohen’s improving acquaintance with the work of Robert Nozick, that ideal type of neoliberalism, also played its part: [O]ne day in 1972 … Jerry Dworkin nudged me. He began a process that, in time, roused me from what had been my dogmatic socialist slumber. He did that by hitting me with an outline of the anti-socialist Wilt Chamberlain argument, as it was to appear in Robert Nozick’s then forthcoming Anarchy, State and Utopia. (Cohen 1995a, p. 4)

The rest of this book deals with Cohen’s thought after this ‘rousing’, dating back to the mid-1970s.22 At about that time, Cohen became preoccupied with rescuing the idea of freedom from appropriation by the pro-capitalist Right.

2

Freedom Following the publication of Karl Marx’s Theory of History, Cohen experienced a ‘loosening’ of his original attachment to the main tenets of technological materialism that the book defended. He thus turned his attention to the three questions that [S]hould command the attention of those of us who work within the Marxist tradition today. They are the questions of design, justification, and strategy, in relation to the project of opposing and overcoming capitalism. The first question is, What do we want? What, in general, and even not so general terms, is the form of the socialist society that we seek? The second question is, Why do we want it? What exactly is wrong with capitalism, and what is right about socialism? And the third question is, How can we achieve it? What are the implications for practice of the fact that the working class in advanced capitalist society is not now what it was, or what it was once thought to be? (Cohen 1988, p. xii)

Cohen’s work on freedom addresses the second question. Indeed, it is the theory of freedom, along with the concept of exploitation, that form the conceptual bridge between Cohen’s work on Marx and his normative political philosophy. Some of the reasons for this transition have already been sketched in Chapter 1: It was not only my encounter with Nozick, but also my loss of confidence in the two large Marxist factual claims about the prospects of equality,1 which altered the direction of my professional research … In the past, there seemed to be no need to argue for the desirability of an egalitarian socialist society. Now I do little else. (Cohen 1995a, p. 7)

In the context of this series of overlapping arguments and polemics, Cohen’s practice of immanent critique is in full swing. This practice has two moments:

The Political Philosophy of G. A. Cohen

42

Cohen sets out by sharpening his opponent’s argument, by refuting cheap or unfair objections to it. This raises the stakes, by making the attempted refutation more difficult. But it also makes Cohen’s own arguments, if and where successful, stronger. The second moment consists in granting the said opponent’s premisses, and drawing from them conclusions the opponent cannot accept without intuitive cost. A good starting point for illustrating this method is Cohen’s critique of the Right’s pro-capitalist appeal to freedom. I will then illustrate it further, by setting out the main steps in Cohen’s critique of right-libertarianism. I will conclude by sketching a brainchild of the latter polemic, the idea of left-libertarianism.

Cohen against Berlin Some of the Right’s popular arguments defend inegalitarian conclusions by appeal to freedom. On one such argument: The problem the poor face is not that they lack freedom, but that they are not always able to exercise the freedom that they undoubtedly have. When the Left say that the poor, by virtue of being poor, lack freedom itself, the Left, so the Right claim, indulge in a tendentious use of language. (Cohen 2011, pp. 167–168)

Cohen’s refutation of this position is one of the most exhilarating exercises in recent political philosophy, so I propose to discuss it at some length. The argument of the Right can be reconstructed as follows: (1) Freedom is compromised by (liability to) interference (by other people), but not by lack of means. (2) To lack money is to suffer not (liability to) interference, but lack of means. So (3) Poverty (lack of money) does not carry with it lack of freedom. But (4) The primary task of government is to protect freedom.

Freedom

43

So (5) Relief of poverty is not part of the primary task of government. (Cohen 2011, p. 168) The ‘first movement’ of this argument, from premisses (1) and (2) to conclusion (3) attempts to establish a conceptual claim about the nature of poverty. The ‘second movement’, from premiss (4), adds a normative claim which, in conjunction with the antecedently established conceptual claim, entails the anti-egalitarian conclusion (5). People on the Left usually deny either (1) or (4). That is, they reject either the negative conception of freedom embodied in (1) wholesale or what they consider to be a one-sided understanding of the state’s proper goals in (4). So, for example, John Rawls (1971) and, according to Cohen, Isaiah Berlin,2 assert (1) and (2) but reject (4). A similar move is made by Eric Hobsbawm (1984) and, more recently, by David Harvey (2005). Hobsbawm’s argument reformulates (4) as: (4ʹ) There are only ‘negative’ rights, that is, rights protecting negative freedom. And since alleviation of poverty is no ‘negative’ right (from (3) and (4ʹ)), it follows that: (5ʹ) There is no right to poverty relief. Hobsbawm rejects (4ʹ) by asserting a distinction between rights to negative and positive freedom (he calls the former ‘individual rights’ and the latter ‘social rights’). It follows from the Hobsbawm position that lack of money and the relief of poverty can only be morally prioritized by asserting the existence of ‘positive’ rights. Indeed, this is how many understand the debate on freedom between Left and Right. They therefore tend, too quickly, to assume that this debate has reached a stalemate. Cohen thinks that the Hobsbawm/Berlin position grants too much to the Right. For it implicitly asserts (2), which is manifestly false: Suppose that an able-bodied woman is too poor to visit her sister in Glasgow … If she attempts to board the train, she is consequently without the means to overcome the conductor’s prospective interference. Whether or not this woman should be said to have the ability to go to Glasgow, there

44

The Political Philosophy of G. A. Cohen

is no deficiency in her ability to do so which restricts her independently of the interference she faces. (Cohen 2011, p. 176)

Similar hindrances obtain when a poor person attempts to pick up a sweater or an iPod from her local store. The moment she picks up the iPod and fails to produce the monetary equivalent at the counter, she becomes liable to having it removed from her possession by force (and/or by being arrested, imprisoned, and so on). It follows that ‘the only way you won’t be prevented from getting and using things that cost money in our society – which is to say: most things – is by offering money for them’ (Cohen 2011, p. 177). Cohen thus purports to undermine normative conclusion (5) by defeating its conceptual primitive (3). And (3) is defeated if (2) is false, even when the Right’s conception of freedom (1) is granted.

Money In this critique of the Right, Cohen constructs an important weapon in the critique of capitalism: a critical theory of money. He writes: [T]o lack money is to be liable to interference, and the assimilation of money to physical, or even mental, resources is a piece of unthinking fetishism, in the good old Marxist sense that it misrepresents social relations of constraint as people lacking things. In a word: money is no object. (Cohen 2011, p. 177)

This ‘misrepresentation’ issues from the fact that money appears to be, but is not, a mere thing. Rather, money is: As Karl Marx said, ‘social power in the form of a thing’, but it is not, like a screwdriver, or a cigarette lighter, itself a thing (meaning, here, by a ‘thing’, a physical object), for social power is not a thing. If you swap your ten onepound coins for a ten-pound note, you’ve got a different thing from what you had before, but the very same money. You’ve got the same license to travel, to acquire goods and services and so on, the same social entitlement, the same prospects of non-interference that you had before … (Cohen 2011, p. 185)

What, precisely, is the role of money? To answer this question, Cohen draws an analogy with a moneyless economy in which ‘the state owns everything, and in which courses of action available to people, courses they are free to follow

Freedom

45

without interference, are laid down by the law’ (Cohen 2011, p. 181). The state gives people tickets, which permit each person to perform a conjunction of actions (action X and action Y) or a disjunction of conjunctions (actions U and V, or actions W and X, or actions Y and Z, etc.). The state thus determines who can get what when. This is similar to what money does: ‘A sum of money is tantamount to (≠ is) a license to perform a disjunction of conjunctions of actions, actions like, for example, visiting one’s sister in Glasgow, or taking home, and wearing, the sweater on the counter at Selfridge’s’ (Cohen 2011, p. 182). But there are disanalogies: Whereas it is the government that restricts a person’s freedom in the moneyless society, it is not, standardly, the government, but the owner of the good to which a person desires access, who, in the first instance, restricts her freedom in the money case. What the government in a money economy does is to enforce the asset-holder’s will, inter alia when that will is a will to deny access except in return for money. And the strategic role of the asset holder’s will means both that money does not absolutely ensure access … and that lack of money does not absolutely ensure lack of access. (Cohen 2011, pp. 182–183)

In other words, unlike the government in the state ticket scenario, money is neither necessary nor sufficient for the performance of a particular action (or a disjunction of conjunctions thereof). It is not necessary because one might simply gift the good in question without recourse to monetary exchange. It is not sufficient because one has to choose to engage in some particular act of exchange for it to be deemed socially valid. Money is, rather, an INUS condition (that is, an Insufficient but Necessary part of a condition which is itself Unnecessary but Sufficient) for that act, or the conjunction of acts thereof (Cohen 2011, p. 177). But now, note that the state ticket case can be amended to also reflect an INUS condition: Imagine, then, that, like money, the state tickets are neither always necessary nor always sufficient to secure goods, because state-appointed asset administrators are free, to some small extent, to grant access to ticketless people and to withhold it from people with tickets: this is an officially recognized perk of office. The administrators, let us further suppose, exercise bias in favour of some citizens and against others to precisely the same extent that private asset-holders do in the money economy … tickets

46

The Political Philosophy of G. A. Cohen

establish what you are free and not free to do, not, now, to be sure, as we originally supposed, tout court, but within the feasible set established by asset administrators’ spheres of discretion. (Cohen 2011, p. 183)

The analogy is extraordinary because it lays bare, with unprecedented clarity, the parallels between a despotism of bureaucrats and a despotism of money. To see this, imagine that the administrators in Cohen’s original example are systematically and irreversibly biased in favour of possessors of green eyes, who happen to be 1 per cent of the population. The 1 per cent thus ends up with the bulk of freedom tickets. The remaining 99 per cent get some, but the average non-green-eyed person is significantly less free than the average green-eyed person. Now suppose that the administrators are biased in favour of those who possess green pieces of paper, perhaps featuring images of dead people at their centre. Imagine – by no great stretch of the imagination – that 1 per cent of the population owns about a third of the total stock of such green papers. The 1 per cent thus ends up with the bulk of freedom tickets. Indeed, if there is a fixed amount of total tickets, and each gets an amount proportional to their green-holdings, then the 1 per cent gets exactly one-third of the total tickets. The remaining 99 per cent gets the rest, but the average green-poor person is significantly less free than the average green-rich person. There is a name for this kind of despotism, and it is ‘market capitalism’.3 Cohen himself does not draw this conclusion. He writes: I have not claimed that either [the state-based or the money-based] economy is more attractive than the other. Many will prefer the private money economy in which freedom does not depend so immediately on the state, but on decisions of other people that the state endorses. But that does not touch the present point, which is that what depends on those decisions in the money economy precisely is my freedom. (Cohen 2011, p. 186)

But he immediately adds: Capitalist economies are often thought superior to state-controlled economies, from the point of view of freedom, just in that there is a wider dispersion of property in the former. But, by that token, a market socialist society, with far wider dispersion of property, and, consequently, of the freedom that goes with it, is even better. (Cohen 2011, p. 186)

Freedom

47

This argument has polemical significance. In Capitalism and Freedom, Milton Friedman emphasizes ‘the role that inequality plays in providing independent foci of power to offset the centralization of political power’ (Friedman 2002, p. 168). Friedman’s argument is invalid. His (plausible) major premiss is the desirability of centralization-attenuating foci of political power. Friedman invalidly – but conveniently – infers the desirability of inequality. What he should have inferred from the mooted major premiss is the desirability of equality. And this is exactly what Cohen argues for: if our goal is the widest possible ‘dispersal’ of freedom, then we had better opt of socialism.4 For capitalism tends to reproduce centralization of political power, not in the hands of the bureaucrats of a monolithic state à la Friedman but rather in the hands of a moneyed few: ‘To think of capitalism as a realm of freedom is to overlook half of its nature’ (Cohen 2011, p. 152). Both state capitalism and market capitalism are sources of unfreedom.

Unfreedom and private property If Cohen is right, then lack of money implies lack of freedom, on the Right’s conception of freedom. Having hoisted one ‘bourgeois defence of capitalism with its own petard’ (Cohen 1995a, p. 162), Cohen employs a similar line of reasoning to argue that private property also restricts freedom. The Right claims that private property protects freedom. But, so long as we are talking about negative freedom, that is, freedom as absence of (liability to) interference, that claim is, rather obviously, false. The enforcement of private property does imply that owners are guaranteed certain robust negative freedoms. These freedoms are, however, thereby denied to non-owners. I cannot, for example, pitch my tent in your backyard without (being liable to) interference from you, or your bodyguards or the police in a system where private property is universally enforced. Indeed, private property is simply a distribution of freedom and unfreedom or, more precisely: a distribution of rights of interference. If A owns P and B does not, then A may use P without interference and B will, standardly, suffer interference if he attempts to use P. (Cohen 2011, p. 176)

48

The Political Philosophy of G. A. Cohen

Capitalism requires vast amounts of (threats of) interference in the service of private property: next to the invisible hand of the market, the visible fist of the capitalist state. It follows that one ‘cannot both deny that justice restricts freedom and claim that private property is just’ (Cohen 1988, p. 252). For private property manifestly restricts freedom, and if it is just – as proponents of capitalist private property claim – then justice restricts freedom. This is one reason why these proponents often eschew talk about freedom and instead choose to talk about liberty: the latter is more amenable to moralized, rightsbased, definition than the former, in a way that does not lock them into the anti-property dilemma just rehearsed. This is precisely the strategy pursued, as we shall see, by right-libertarians such as Robert Nozick. Now, the defender of capitalism, and therefore of private property, might respond that under a well-enforced system of private property rights, total negative freedom is maximized. But this move, as Cohen points out, is unwise, for it turns the capitalism vs. socialism debate into a mere empirical question – one whose answer might not satisfy defenders of capitalist private property (Cohen 2011, p. 188). He goes on to illustrate some ways in which total negative freedom might increase, relative to private property, through socialist common property and/or distribution according to need: There is an overwhelming case, from the point of view of freedom, in favour of our actual practice of public ownership of street pavements. Denationalising the pavements in favour of private ownership of each piece by the residents adjacent to it would be bad for freedom of movement. (Cohen 2011, p. 156)

One reason why pavements should be public is that public pavements entail greater freedom of movement than do private pavements. Under a system of private pavements, I am liable to interference at every possible juncture of my walking life: when I walk to work, to university, to the pub.5 We might add that a system of free public transport analogously removes interference for longrange movement: trains, buses, and other forms of mass transit are, in that sense, just pavements on wheels (compare: seat-equipped, spacious, kilometrelong conveyor belts like the ones used at airports). Remarkably, these mobile pavements have been almost completely privatized in the United Kingdom and the United States over recent years. It follows that consecutive UK and US

Freedom

49

governments have recently contrived to reduce people’s freedom of movement through privatization.6 And all that on the Right’s definition of freedom. I tentatively conclude that negative freedom, according to Cohen, cannot do the work the Right assigns to it and may, indeed, serve to justify anti-Right political goals. Moreover, people on the Left, like Hobsbawm, construct an obsolete theoretical conflict – that is, a conflict between ‘individual’ and ‘social’ rights that misunderstands the nature of negative freedom. The real conflict arises earlier than those leftists envisage, within the sphere of freedom itself. Cohen concluded from all this that we need a theory of the just distribution of freedom. In constructing one, he thought, socialists must inevitably try to make good on the old but ambiguous socialist slogan that capitalist private property is theft, that is, a species of unjust taking. I shall come back to this in Chapter 3.

Proletarian unfreedom In the mid-1980s, Cohen turned his attention to the conceptual connections between freedom and exploitation. Part of his aim was to precipitate revision of the traditional Marxist characterization of the relations of production,7 invariably premissed, as it was, on the idea of force. In revising the traditional Marxist position, Cohen attacked some of the fundamental premisses on which the debate on exploitation is commonly conducted. According to a common-stock argument about freedom: (6) If A is forced to φ, then A is unfree to φ. According to Cohen, the Left asserts: (7) Workers are forced to work. and infers, from (6) and (7), that workers lack freedom. The Right, on the other hand, claims that: (8) Workers are not unfree to work. and infers, from (6) and (8), that workers are not forced to work. Cohen dissects the dialectic of this putative argument by pointing out that both Left and Right typically enlist a premiss like (6). And he points out that

50

The Political Philosophy of G. A. Cohen

(6) is false: if A is forced to φ, then all that follows is that A is not free not to φ. It does not follow that A is unfree to φ. Indeed, Cohen argues that if A is forced to φ, then A must be free to φ: ‘for how can someone be forced to do what he is not free to do?’ Therefore both Left and Right are mistaken, and (6) is false (Cohen 1988, pp. 239–255). Workers are forced to work under capitalism precisely because they are free to work.8 Having made this move, Cohen goes further, by arguing that the term ‘workers’ in (7) involves a fundamental equivocation. According to Cohen, the vast majority of able-bodied workers under capitalism are not forced to work for any capitalist. This directly contradicts Marxist lore. Indeed, most workers under capitalism are individually free, in the sense that each can make her way into the petty bourgeoisie and escape her proletarian predicament without (liability to) interference from others. But, Cohen argues, it does not follow, and it is false, that workers are collectively free, – that is, free as a class. To infer collective freedom from individual freedom is to commit a fallacy of composition. For, suppose all the workers were to attempt selfemancipation together. They would then find themselves incapable of jointly squeezing through the petty bourgeois loophole. Indeed, successful collective self-emancipation entails the end of capitalism (Cohen 1988, pp. 261–265). To illustrate this set of apparently incompatible claims, Cohen asks us to imagine that there are ten people in a cave and only one key. Once a single person exits, the door is automatically sealed forever. Prior to someone using the key, each of the people in the room is free to exit. And if, for whatever reason, nobody attempts to escape, it is true of each of the denizens of the cave that she is free to exit. Cave denizens are not, however, collectively free to do so, for they cannot escape collectively. They are therefore unfree as a class. By the same token, workers are free to exit the working class in sensu diviso but not in sensu composito. This is why, according to Cohen, socialism promises ‘a better liberation: not just from the working class, but from class society’ (Cohen 1988, p. 265). Proletarian unfreedom thus consists in a complex configuration of collective unfreedoms and individual freedoms (and unfreedoms)9 for the members of the working class. Cohen’s thesis, if true, leads to a further significant point. Take contemporary ideological beliefs, such as the American Dream. Its content ostensibly is not: ‘everyone can become rich’ but rather ‘I can become

Freedom

51

rich.’ This has ideological value only if the stated disposition is sometimes realized for some working class ‘I’s. That is, part of the allure of servitudeperpetuating ideologies issues from the fact that their content is sometimes realized for working-class people. And that fact in turn vindicates the notion that some of these people are, in fact, free: It is part of the genius of capitalist exploitation that, by contrast with exploitation which proceeds by ‘extra-economic compulsion’, it does not require the unfreedom of specified individuals. There is an ideologically valuable anonymity on both sides of the relationship of exploitation. (Cohen 1988, p. 265)

For all its sophistication, it is unclear where Cohen’s argument leaves us. If one thinks, as many Marxists do, that exploitation is about the extraction by capitalists of forced, surplus, unpaid labour from workers, then the falsehood of (7) for at least some workers would seem to entail that these workers are not exploited. And this goes against the grain of traditional Marxist commitment. In Chapter 4, I will argue that Cohen believed the traditional Marxist commitment to be flawed: Marxists should be interested in exploitation not because it is a species of forced transfer (which is, after all, what redistributive taxation also entails) but rather because it is a form of injustice. And conversely, it is injustice, not unfreedom, that the procapitalist Right is concerned with when it makes ‘promiscuous … use of the rhetoric of freedom’ (Cohen 1995a, p. 67). For these and related reasons, the theory of distributive justice moves to the centre of Cohen’s philosophical vision, both as part of the polemic against right-libertarianism and as part of the defence of socialism. The rest of this chapter is concerned with the former polemic.

The programmatic libertarian argument We have seen that Cohen’s engagement with liberal political philosophy began in the mid-1970s. This was a time when right-libertarianism, one of neoliberalism’s closest ancestors, was beginning to stand on its own intellectual feet. Its highest philosophical expression was to be found in

52

The Political Philosophy of G. A. Cohen

Nozick’s Anarchy, State and Utopia (Nozick 1974). Here is a reconstruction of Nozick’s argument, in the form that may well have awakened Cohen from his ‘dogmatic socialist slumber’: (9) Persons are morally rightful owners of their person and its powers. (the self-ownership thesis) (10) Persons are entitled, by moral right, to appropriate as much of the world as they like, as long as they do not thereby worsen the position of others. (the justice in acquisition thesis) (11) Persons are entitled, by moral right, to alienate or acquire any title to worldly resources appropriated in accordance with (9) and (10), through the market. (the justice in transfer thesis) But, (12) Coercive redistributive taxation violates (some of) these moral rights. Therefore, (13) A state that includes redistributive taxation is unjust. This argument establishes Nozick’s case against redistribution. Cohen’s first extensive engagement with liberal political philosophy is an attempt to refute this argument premiss by premiss. In the following sections, I look at some of the considerations Cohen adduces to achieve this.

Against Nozick … on transfer Nozick defends the justice in transfer thesis (11) by recourse to his Wilt Chamberlain example, focusing on the putative fortunes of the famous 1960s basketball player. Suppose we start from a ‘patterned’ distribution D1, in which everyone’s holdings abide by some just and/or socially desirable pattern. Chamberlain signs a contract with his team, or with the basketball association, which will allow him to collect 10 per cent of all revenue per game. Thus when a million people go watch him play, paying $0.25 each, Chamberlain makes $250,000.10 Call the new distribution D2. Now, the movement from D1 to D2 is fully justified, according to Nozick, because it flows from the fully voluntary choices of all parties. And, assuming the original distribution D1 is just, D2

Freedom

53

must also be just. This is how markets are incompatible with equality or how ‘liberty upsets patterns’ (Nozick 1974, p. 160). Cohen proposes to take a closer look at this argument. According to Nozick, a transition is morally untainted if and only if it is voluntary. The transition from D1 to D2 is voluntary. Therefore, the transition from D1 to D2 is morally untainted. Cohen points out two problems with Nozick’s account of voluntariness. The first problem is that it is incoherent and the second is that it equivocates between importantly distinct objects. I look at each of these objections in turn. What constitutes a voluntary transfer? One possibility might be to define voluntariness in terms of (the absence of) force. We could say that A is forced to φ if and only if there is no acceptable or reasonable alternative to φ-ing. Hence, A φs voluntarily if and only if A is not forced to φ. But Nozick cannot accept this definition of voluntariness. For, he appreciates that the poor are forced, on pain of starvation, to sell their labour power to their employers. Nozick cannot assert that they thereby transfer their labour power involuntarily, for that would emasculate his argument for the minimal state. He asserts, instead, that ‘voluntary’ is to be defined in terms of rights, that is, the set of rights that flow from theses (9) to (11). Cohen points out that this renders the whole notion of voluntariness circular: a transition is just if and only if it is just. When we now ask what voluntary performance is, we are told that someone’s action is voluntary, no matter how limited his opportunities were, if and only if there was no injustice in the production of the limitation of his opportunities. And that creates a circle: justice in transfer is defined in terms of voluntariness and voluntariness is defined in terms of justice. (Cohen 1995a, p. 61)

Nozick’s account of voluntariness is therefore incoherent, and premiss (11) in the argument for the minimal state is illegitimately derived. A second problem with Nozick’s account is that it equivocates between different objects of voluntary choice. Consider the following set of possibilities: (i) I choose to pay $0.25 to watch Wilt play. (ii) I choose to pay Wilt $0.25 to watch Wilt play. (iii) I, and another 999,999 people, choose to pay Wilt a total of $250,000 to watch him play.

54

The Political Philosophy of G. A. Cohen

According to Cohen, Nozick’s argument slides quietly from (i) to (ii) and then from (ii) to (iii): Nozick tacitly supposes that a person willing to play twenty-five cents to watch Wilt play, is ipso facto a person willing to pay Wilt twenty-five cents to watch him play. It is no doubt true that in our society people rarely care who gets the money they forgo to obtain goods. But the tacit supposition is false, and the common unconcern is irrational. Nozick exploits our familiarity with this unconcern. Yet a person might welcome a world in which he and a million others watch Wilt play, at a cost of twenty-five cents to each, and consistently disfavour one in which, in addition, Wilt rakes in a cool quarter million. (Cohen 1995a, p. 26)

Unconcern for the move from (i) to (ii) and for the move from (ii) to (iii) are both irrational, because they fail to consider the ‘effect on Wilt’s power’ (Cohen 1995a, p. 26). The move from (ii) to (iii) embodies, in addition, a collective action problem: even if it were individually rational to accede to (ii), acceding to (iii) is individually (and collectively) irrational. Thus: [A] convention might evolve not to make such payments, or more simply, there could be a democratically authorized taxation system which maintains wealth differentials within acceptable limits. (Cohen 1995a, p. 26)

One of the functions of coercive state redistribution is precisely to coordinate all of these rational desires: Wilt should not ‘rake in’ an extra $0.25 from me (over and above what I and everyone else gets) and Wilt should not ‘rake in a cool quarter million’ from us. ‘Democratically authorized taxation’ just is the upshot of the generalized institutionalization of a ‘Pattern Maintenance Association’ (Cohen 1995a, p. 30) designed to correct the mooted set of collective action problems. This is, Cohen says, how ‘patterns preserve liberty’.

 … on acquisition Nozick gets his account of acquisition, presupposed by claim (10), from John Locke. In his Second Treatise of Government, Locke sets out from the claim that God has given the world to ‘mankind’ in ‘common’, adding that appropriation of worldly resources is permitted if and only if it leaves ‘enough, and as good, in

Freedom

55

common for others’. Nozick interprets this ‘Lockean proviso’ as follows: A’s act of appropriation is legitimate if, and only if, it leaves B at least as well off as she was before A’s act. And he goes on to reassure his readers that private property, especially in the light of the rise of capitalism, will make (almost) everyone better off, such that there is no cause for concern that his specification of the Lockean proviso will, in general, be violated.11 Cohen points out that Nozick’s theory of appropriation, even if loyal to Locke,12 is untenable on its own terms. For, his (interpretation of Locke’s) proviso is both arbitrary and implausibly weak. To see this, imagine there are two people, A and B, who live in common arable land, and that each enjoys 5 units of well-being. I shall denote A’s and B’s respective levels of well-being as: (5, 5). A says to B: ‘I’ll appropriate all the land we now hold in common. You’ll then come and work for me, and I’ll keep all the proceeds of your work. The resulting distribution will be (50, 6).’ This outcome is consistent with Nozick’s proviso. Table 2.1 represents the three options in the feasible set: a pre-appropriation situation, a situation with private appropriation of the land by A, and a situation with joint appropriation by A and B. Cohen asks: why is the only relevant baseline of acquisition the first one? That is, why compare the state of affairs in which the most ruthless or greedy person privately appropriates most of the world for herself with pre-appropriation, as opposed to, say, the state of affairs in which A and B own the world jointly and share its fruits equally? In his defence of the proviso in (10), Nozick arbitrarily privileges pre-appropriation. And he illegitimately concludes that the world is unlike manna from heaven, for which the just distribution might be one of equality (Nozick 1974, p. 198). This explains why ‘Nozick signally and consequentially failed to observe that the raw resources of the planet Earth are manna from heaven’ (Cohen 2011, p. 229). Nozick’s theory of appropriation, says Cohen, rests on a non-sequitur and is implausibly weak.

Table 2.1  Appropriation baselines Pre-appropriation situation

Private appropriation by A

Joint appropriation

5, 5

50, 6

28, 28

56

The Political Philosophy of G. A. Cohen

 … on self-ownership A large part of Cohen’s polemic against Nozick is devoted to understanding, and refining, the self-ownership thesis (9). Self-ownership claims purport to answer the question: ‘Who is the rightful owner of the person and its powers?’ World-ownership claims purport to answer the question: ‘who is the rightful owner of such-and-such worldly resource (including produced machinery, physical space etc.)?’. But how encompassing is this notion of ‘self-ownership’? Assume a system of joint world-ownership, such that each person has a veto right over the use, possession and disposal over every single part of the world by others. In such a system, everyone needs everyone else’s permission to do, or not do, something (including to occupy physical space). Is this system compatible with self-ownership? According to Cohen, both an affirmative and a negative answer have significant ramifications for Nozick’s argument against redistributive taxation. Let us start with the first horn of the dilemma, that self-ownership is compatible with joint world-ownership. This seems to make self-ownership merely formal. That is, it endows you with a right to demand compliance with duties of non-interference with your person but little more than that. Indeed, one might well conclude that this is a reductio of the formal reading of selfownership. Furthermore, if Nozick were to accede to compatibility in this case, he would thereby grant that self-ownership is compatible with equality of condition. For, as joint owners we might well opt for an egalitarian distribution of resources: we are entitled, after all, to jointly decide that we are equally entitled to the sandwiches in your backpack, to the book in my pocket, etc. On the second horn of the dilemma, Nozick must maintain that selfownership is incompatible with joint world ownership. The reason, he must say, is that people in the jointly owned world can do nothing with their selfownership. It is freedom that we should be concerned with, and joint world ownership deprives people of that freedom. This horn is equally detrimental to Nozick’s account: if he grants that self-ownership is not merely about formal rights of non-interference with one’s person, then minimal-state capitalism does not deliver self-ownership. If you are unemployed, or lack money to get shelter, food, etc. then you are in a position exactly analogous to that of someone

Freedom

57

Table 2.2  The self-ownership dilemma (i) Self-ownership is compatible with joint world ownership.

(iii) Self-ownership is not compatible with joint world ownership.

(ii) But joint ownership generates equality of condition, that is, a morethan-minimal-state capitalism.

(iv) This lack of compatibility is due to the disempowerment of persons under joint ownership.

Therefore self-ownership is compatible (v) The poor under minimal-state with more-than-minimal-state capitalism capitalism are similarly disempowered. (from (i) and (ii)). Therefore self-ownership is incompatible with minimal-state capitalism (from (iii), (iv) and (v)).

in the jointly-owned world. Your self-ownership is therefore undermined (see Cohen 1995a, pp. 92–115). Nozick’s theory thus pulls in two contradictory directions: on the one hand, he would like to deny that one is a full self-owner under joint world ownership and, on the other, he would like to assert that the poor under minimalstate capitalism are full self-owners. Cohen claims that these two beliefs are incompatible. Table 2.2 depicts the argument schematically.

Libertarianism and the Left Cohen’s labours to defeat Nozick gave birth to a new theory of justice, sometimes called left-libertarianism. Although Cohen never explicitly refuted paternity of the left-libertarian child, he explicitly distanced himself from its distinguishing birthmark, the idea of self-ownership (see, for example, Cohen 1995a, pp. 116–143). He did, however, intimate the implications of the seed he had sown. Left-libertarians believe that self-ownership is compatible with substantive equality of condition. The way to argue this enlists a discrepancy in the stringency of rights to one’s own person, on the one hand, and rights to the world, on the other. To see this, compare a case where someone makes a sweater from her own hair and a case where she makes it from wool she found on an abandoned island. There is a sense in which, if you appropriate the person’s sweater in the first case, you are, effectively, trimming her hair against

58

The Political Philosophy of G. A. Cohen

her will. Hair-trimming of that sort is evidently less morally acceptable than merely appropriating previously unowned wool you found in an uninhabited island (and which was originally available for all to take).13 Here is one way in which self-ownership and equality can be rendered consistent. Imagine that two people, Able and Infirm, live on an abandoned island with three coconut trees (for the original example, see Cohen 1995a, p. 95). Coconuts translate directly into units of welfare: once picked up, they are automatically consumed. Able and Infirm have identical utility functions, but Able can pick up several coconuts per hour. Infirm can only pick a few, because he is too weak, or too slow. Owing to productivity differentials, the cardinal utility distribution between Able and Infirm is (9, 5). There is an inequality here, indeed, an inequality that seems unjust. For Infirm is worse off than Able through no fault of her own. What is an egalitarian planner to do?14 One thing she can do is hand over control of two coconut trees to Infirm and let Able pay Infirm a rent for access to these trees. Given a sufficient number of trees, and sufficient demand for them, we can achieve any desired distribution, such as the egalitarian one (7, 7). But we should first check that doing so does not violate the selfownership rights of Able and Infirm. Beginning with Infirm, the mooted form of predistribution does not seem to violate her self-ownership. For all that happens is that she gets a resource windfall of some coconut trees. Now turn to Able: egalitarian predistribution does not seem to violate her self-ownership, either. For the original distribution of trees in favour of Infirm is relevantly like a caveat emptor condition on Able’s acts of appropriation. That is, in giving only one tree to Able and instructing her to pay Infirm a rent if she wants access to more, we say to her: ‘given your superior talents, you can produce and consume as much as you like, as long as you give X% to Infirm’ (where ‘X’ is determined by our theory of equality). This does not violate Able’s selfownership, any more than renting a flat on condition one does not burn it, or return a proportion of the flowers that grow in it, violates their self-ownership. It follows that there is a system of self-ownership-consistent distributions of worldly resources ensuring substantive equality of condition. All this may sound artificial, or pedantic or both. But it has significant ideological ramifications. One implication of left-libertarianism is that, even if one starts from stringent self-ownership premisses, one can still justify

Freedom

59

equality, indeed an extensive welfare state, by combining self-ownership with certain egalitarian premisses about world-ownership. It can be shown, moreover, that stringent rights to self-ownership are compatible with socialist joint ownership of the means of production. For consider: socialists believe that equality, community and economic democracy can only be achieved by a system of joint ownership in, and control over, the means of production.15 This set of property rights does not, as such, tell us what rights individuals have to their own person. Libertarians believe that individual liberty and autonomy are only coextensive with a set of stringent rights to the person and its powers. This set of property rights does not, as such, tell us what rights individuals have to the external world. Bringing libertarianism and socialism together is therefore, in principle, possible. One of the most important implications of Cohen’s critique of Nozick is that ‘libertarian socialism’ is not an oxymoron.16 It does not follow that socialists should embrace self-ownership, as Cohen was wont to point out. But neither should they grant anti-socialists the claim that self-ownership is incompatible with equality.

3

Equality In Chapter 2, I argued that Cohen’s polemic with the defenders of capitalist private property led him directly to – what he took to be – a central question in political philosophy, that of the proper distribution of freedom. The justification of socialism, as opposed to capitalism, of democratic planning, as opposed to the market, the very integrity of his ethical commitments as a Marxist, seemed to turn, at least in part, on the fundamental question of justice.1 Having mapped the putative trajectory of human liberation through the ‘capitalist vale of tears’, Cohen’s life-long commitment to the desirability and feasibility of human emancipation now assumes the form of a radical egalitarianism. This chapter deals extensively with its specification and theoretical ramifications. At the outset, I must distinguish between four important questions that have a direct bearing on Cohen’s understanding of justice and of his brand of distributive justice in particular: they are the questions of pattern, metric, site and scope. The question of pattern is: how should we distribute whatever it is that justice is concerned with? Should we just give everyone enough? Should we give priority to the worse off? Or should we give everyone an equal amount of whatever it is that matters or an equal opportunity to achieve it? The question of metric (or currency) of justice is: what is it that we should distribute? Should we be concerned to distribute – in whatever pattern – welfare, resources, capabilities, primary goods or some combination thereof? The question of site is: what is the purview of justice? Is it coercive social institutions, or informal social institutions, or individual actions and behaviour or some inclusive disjunction thereof? The question of scope is: over whom does justice apply? Do we owe duties of justice to those distant in space, say, those residing on the other side of the world? Do we owe duties of justice to those distant in time, say, those who will live in a thousand years?

62

The Political Philosophy of G. A. Cohen

Cohen’s egalitarianism is almost exclusively concerned with the first three of these questions. In what follows I discuss pattern, metric and site in turn. I then rehearse Cohen’s argument against the Pareto argument for inequality and offer a sketch of the overall architectonic of Cohen’s theory of justice, and the place of equality within it. I conclude with a brief retrospective evaluation of the tensions between Cohen’s Marxism and his egalitarianism.

Patterns of justice Cohen’s understanding of distributive justice stems from his critique of the market and is directly opposed to the idea that markets are somehow constitutive of justice. In 1989 Cohen is, once again, engaged in an extended polemic, this time with the liberal legal philosopher Ronald Dworkin. During the early 1980s, Dworkin developed and defended a theory of equality operating on a master distinction between ambition and endowment (see Dworkin 2000). Ambitions are desires intimately connected with the kind of life their possessor would, upon reflection, like to lead. Simply put, they are the desires and attitudes one would not take a pill to get rid of. Endowments, on the other hand, are those social and natural attributes of one’s circumstances that are necessary means for leading the good life. The wealth of my parents, their social connections and education, my own wealth and education, in addition to my own genetic makeup, intelligence, native talents, etc. are all parts of my endowment (for the parallel distinction between social and natural primary goods, see Rawls 1999, p. 93). According to Dworkin, justice in distribution is ambition-sensitive but endowment-insensitive: how things turn out for me, at the bar of justice, should only depend on my choices and ambitions rather than contingent facts related to my social or natural endowment. Dworkin argues that this theory of justice presupposes a further distinction between brute and option luck. Option luck is ‘a matter of how deliberate and calculated gambles turn out – whether someone gains or loses through accepting an isolated risk he or she should have anticipated and might have declined’ (Dworkin, 2000, p. 73). Brute luck is, by implication, about gambles one undertakes and which one could not (reasonably) have anticipated or which one could not have declined. Dworkin’s claim is that a distribution is just if, and only if, it

Equality

63

reflects nothing but ambition. How exactly ambition and luck are connected is a matter of considerable controversy, but for our purposes it suffices that an ambition-sensitive distribution at least rules out forms of brute luck that fail the pill test.2 Cohen’s critique of Dworkin seeks to do two things, both intimately related to the socialist project: on the one hand, Cohen seeks to realign Dworkin’s ‘cut’ between ambition sensitivity and endowment insensitivity in a way that severs Dworkin’s putative link between justice and the market. The right cut is not in terms of ambition and endowment but rather in terms of choice and circumstance. On the other hand, Cohen seeks to re-ally egalitarianism with traditional socialist concerns, such as labour burden and economic power, into the proper equalisandum of justice. In this section, I discuss Cohen’s cut. Dworkin holds that justice is equality constrained by the costs individual choices impose on others. To buttress this claim, he asks us to imagine a group of ship-wrecked survivors, washed ashore on an abandoned island. The problem they face is how to divide the island’s resources fairly. He argues that they should each receive an equal share of clamshells and use these clamshells to bid for the island’s resources. The resulting auction is repeated until no one envies the bundle of another. Now, under Dworkin’s auction, each part of the world accrues to the highest bidder. If we rule out strategic bidding, this will be one who most desires the good in question or deems it necessary for the good life. More precisely: say A values G at 10 and B at 5. Then A will bid 5 and obtain G, at a price reflecting the opportunity cost to B.3 Opportunity costs therefore reflect the desires of others or the (strength of) their preferences for the good in question. Dworkin adds that equality requires4 some sort of market mechanism: (1) Justice is equality constrained by the costs individual choices impose on others. (2) The competitive market is the only institution that appropriately reflects these costs. Therefore (3) The only appropriate institutionalization of justice is achieved through the competitive market.

64

The Political Philosophy of G. A. Cohen

Cohen takes up both premisses of this argument. He questions (1), by arguing that sometimes justice requires that A compensate B for her choices even if they reflect a taste for things that happen to be expensive. That is, sometimes cost-imposing choices reflect unchosen facts about the person that mitigate her responsibility for these choices. Say A and B have an (expensive) taste for caviar, but A has cultivated it for snobbish reasons, whereas B needs caviar (and only caviar) to avoid constipation. Cohen claims that B should have caviar, and A should not, at the bar of justice. This contradicts (1). Cohen’s critique of Dworkin thus inaugurates what has come to be called the ‘expensive taste’ debate. This debate is centrally about whether individual tastes or preferences come under ambition or endowment (in Dworkin’s sense), or, more generally, about whether they merit satisfaction or compensation at the bar of justice. On Cohen’s view, some tastes are not chosen, and therefore merit satisfaction, or ground claims for compensation if they go unsatisfied. But Cohen goes further, in maintaining that the truth or falsehood of (2) is irrelevant to egalitarianism. That is, once the falsehood of (1) is granted, then putatively cost-reflecting market prices become irrelevant to justice in distribution: Paul loves photography, while Fred loves fishing. Prices are such that Fred pursues his pastime with ease while Paul cannot afford to. Paul’s life is a lot less pleasant as a result: it might even be true that it has less meaning than Fred’s does. I think the egalitarian thing to do is to subsidize Paul’s photography … He has a genuinely involuntary expensive taste, and I think that a commitment to equality implies that he should be helped in the way that people like Paul are indeed helped by subsidized community leisure facilities. As this example suggests there is between Dworkin’s account of egalitarian justice and mine the difference that my account mandates less market pricing than his does. (Cohen 1989a, p. 923)

Cohen returns to this question in a later essay, making even more transparent his main motivation for involvement in this debate: it is of some moment ‘because it bears deeply on the justice of the market process’ (Cohen 2004b, p. 4). For: it is unjust if I have to pay more for figs than you do for apples simply because few people like figs and many like apples – always assuming that

Equality

65

you get from apples more or less what I get from figs. In my view, markets can ‘produce’ justice only in the Pickwickian sense that they do so when in some unattainable possible world they are so comprehensively rigged that they induce a distribution that qualifies as just for reasons that have nothing to do with how market prices form. (Cohen 2004b, p. 18)

At this point Cohen professes commitment to the socialist principle of distribution: ‘Egalitarians like me see the market as, at best, a mere brute luck machine … the distributive norm I favour takes part of its inspiration from the socialist slogan “To each according to their needs”, according that is, to what they need for fulfilment in life’ (Cohen 2004b, p. 17). This is mysterious. For Cohen’s theory of equality – ‘luck egalitarianism’, as it came to be called – is neither necessary nor sufficient for the principle of distribution enunciated in the socialist slogan. It is not sufficient because it leaves the needs of the ‘undeserving’ poor unsatisfied.5 It is not necessary because it remains open to the non-luck-egalitarian advocate of the socialist principle to accept, or to reject, claims of compensation for those with cultivated expensive tastes, or indeed for those with uncultivated expensive tastes. Needs are, after all, determined by general requirements for human flourishing. Whether champagne, or plover’s eggs, form part of these requirements can be determined independently of luck egalitarian premisses. To be sure, one might try to weaken the argument that luck egalitarianism does not suffice for the socialist principle of distribution by pointing out that, under normal circumstances, justice is much more exacting of the undeserving rich than of the undeserving poor: [T]here remains in contemporary society typically much more offensive inequality than offensive equality: there are reasons for objecting more strongly to the corporate welfare bum than to the able-bodied plain welfare bum who gets as much as the working stiff does. (Cohen 2011, p. 121)

‘Corporate welfare bum’ might include welfare recipients who are, or were, in handsomely-paid corporate employment, or executives who receive payment for doing no or little work, or, more generally, capitalists: in her capacity as owner, the capitalist is, quite literally, someone who lives off the work of others in virtue of that ownership (see the discussion of the labour theory of value in Chapter 4). But the fact remains that the sort of ‘responsibility-constrained

66

The Political Philosophy of G. A. Cohen

equality’ that luck egalitarians advocate plainly neither entails, nor is entailed by, the socialist principle of distribution. Indeed, the two are substantively in tension with one another.

The problem of luck A lot of ink has been spilt in recent political philosophy about how exactly to specify luck egalitarianism. Part of the difficulty is inherited from the problem of defining moral responsibility. Let’s assume that A can be held morally responsible for intentionally φ-ing to bring about x, or for φ-ing unintentionally with the knowledge or foresight that x would occur as a result of her φ-ing, or for unintentionally φ-ing without foresight that x would occur, but where foresight would have been reasonably required.6 In Dworkin’s original formulation, luck egalitarianism is promulgated in opposition to Rawls’ difference principle, and framed partly in terms of the distinction between brute and option luck. According to Cohen, the original argument for luck egalitarianism is something along these lines: (4) Getting what one is responsible for is just. (5) To ‘extinguish the influence of brute luck on distribution’ (Cohen 1989a, p. 931) is to give people what they are responsible for. Therefore (6) To extinguish the effects of brute luck on distribution is just. Susan Hurley shows that premiss (5) cannot possibly be true. For ‘the fact that people are not responsible for difference does not entail that they are responsible for non-difference’ (Hurley 2005, pp. 151–152). By the same token, the fact that people are not responsible for inequality does not entail that they are responsible for equality. Say I have 5, which I’m not morally responsible for, and you have 7, which you’re not responsible for. It does not follow that you and I are responsible for 6 each. Cohen accepts this conclusion. He grants, by implication, that brute luck plays no role in an argument for luck egalitarianism, since it is impossible to appeal to a premiss like (5) to defend such an argument. But he goes on to argue that:

Equality

67

[L]uck extinction cannot be an argument for egalitarianism precisely because, pace Hurley, it does specify egalitarianism. The specification of this egalitarianism involves essential reference to luck, where ‘luck’ contrasts, constitutively, with the responsibility for outcomes that comes with genuine choice. Since luck egalitarianism accounts it an unfairness when some are better off than others through no fault or choice of their own, the relevant contrast with ‘luck’ is ‘choice’, complexly understood. (Cohen 2011, p. 119)

The complexity issues from the fact that ‘brute luck’ in Cohen has an idiosyncratic meaning: it denotes an irreducibly comparative property. To see this, imagine that the original distribution of well-being between A and B is (5, 5). The wind then throws open one of B’s coconuts over A’s randomly yawning mouth, such that A’s well-being increases at the expense of B’s, and the distribution becomes (6, 4). Cohen would say that there is unfairness here, since B is made worse off than A through no fault of her own. But now imagine the wind destroys two coconuts figuring in the original distribution, such that the new distribution becomes (4, 4). Cohen would now not want to assert that the resulting distribution, juxtaposed to the original (5, 5), involves brute luck. For nobody is worse off than another in (4, 4) and a fortiori there is no unfairness. This shows, I think, that it is the notion of differential responsibility (for differences in flourishing), and not the notion of luck, that drives the specification of Cohen’s egalitarianism.

Equality of what? One way to understand the kind of problem a thinker is trying to address is to study its history. The intellectual history of distributive justice over the past century can be seen as a series of critiques, rebuttals and responses to utilitarianism. It is no coincidence that some of the most influential political thinkers of the twentieth century, such as John Rawls and Amartya Sen, set themselves exactly this task. Part of the allure of utilitarianism consists in the simplicity of the injunction to maximize utility. The metric of justice, according to utilitarians, is some sort of simple, unidimensional, monolithic entity, such as pleasure, or preference satisfaction. Amartya Sen’s most important contribution to political philosophy consists in asking: ‘what is the

68

The Political Philosophy of G. A. Cohen

proper equalizandum of justice?’, offering a brilliant non-utilitarian answer to this question. Cohen made an important contribution to this debate, through painstaking critique of both Rawls and Sen. But before discussing Cohen’s own answer to the equalizandum question, I want to offer a dialectical story of the development of egalitarian political philosophy, which might help situate Cohen’s views on the philosophical landscape. Most economists and philosophers before the 1970s tended to assume that some form of preference satisfaction is both an adequate, and an exhaustive, characterization of the metric of justice. Call their view preferentialism. Preferentialism is sometimes characterized with reference to the satisfaction of actual preferences, crucial indicators of which – it is sometimes argued – can be obtained by estimates of a consumer’s willingness to pay. But actual preference satisfaction is plainly an absurd measure of welfare, for reasons well known to both economists and philosophers: actual preferences may be irrational (e.g. when the axioms of rational choice are violated), woefully uninformed or objectionably adaptive (e.g. when someone has been brainwashed into wanting something). We may therefore want to define preferentialism in terms of some informed preference account which overcomes these difficulties.7 This reformulation has the added benefit of severing the (highly dubious) link between welfare and willingness to pay, albeit at the cost of reintroducing the problem of comparability across utilities. Now, if (informed) preference satisfaction is what matters, then two people with equal amounts of preference satisfaction are equally well off, regardless of how much they differ in their objective conditions. Thus a first case against preferentialism can be made on grounds of cheap tastes. Consider the case of Tiny Tim, a young boy who lives in a slum without the use of his legs and without a wheelchair or people to move him around at will. He does, however, have a very sunny disposition, and enjoys a high level of utility (see Cohen 1989a, pp. 918–919). Egalitarians must supply Tiny Tim with the wheelchair he needs. But doing so cannot be justified on preferentialist grounds alone. For by assumption Tiny Tim, thanks to his prodigious capacity to transform utility out of resources, enjoys at least as much preference satisfaction (informed or otherwise) as other people. That capacity should not deprive him of access to important dimensions of human flourishing. At a minimum, he should have access to the wheelchair, adequate levels of nutrition, education, etc.8

Equality

69

Now consider one cognate, but more nightmarish, scenario. People in the present can produce two types of worlds for future people, both at no cost. The first world is one in which the natural environment is almost totally degraded, but still provides for the basic necessities of human beings. Future people, we may assume, have been genetically modified to withstand most forms of environmental degradation. The second world is one without such degradation, and without genetic modification. Present people produce the first world. Most would consider the actions of present people in this scenario objectionable – whether intentional or negligent – because intergenerationally unjust. Producing such worlds seems prima facie wrongful even if future people are genuinely indifferent between a ‘normal’ and a genetically highgrade world. But there is no reason, in principle, as to why preferentialism would deem either as unjust: if – barring brainwashing or manipulation – future people are indifferent between the two worlds, there is nothing wrong in producing either. Of course the preferentialist could conjure all sorts of restrictions to block this conclusion. He could, for instance, call forth future peoples’ rights to ‘an open future’, or something along these lines. But the mere fact that a preferentialist must conjure ad hoc restrictions of the sort, that is, the mere fact that the restrictionless metric she affirms does not, in principle, rule out nightmarish possibilities, may count as prima facie reason against it.9 A half-way alternative to preferentialism is resourcism.10 On Dworkin’s account, the proper equalizandum of justice consists in appropriately bundled sets of resources available for sale in the context of a hypothetical auction.11 This account suffers from a rather general problem lucidly identified by Amartya Sen. Sen attacks resourcisms of all stripes for failing to attach importance to certain morally significant lacks in peoples’ lives pertaining to human functionings and capabilities: ‘ … the rather limited notion of equality of resources, which overlooks the interpersonal differences in the mapping from resources to capabilities, remains unjustified … ’ He claims, in elaboration, that: The translation of resources into the ability to do things does vary substantially from person to person and from community to community, and to ignore that is to miss out on an important general dimension of moral concern. (Sen 1984, pp. 322–323)12

70

The Political Philosophy of G. A. Cohen

If it matters that people can do and be certain things, then resourcism offers, at best, an incomplete account of human flourishing.13 This is, incidentally, why utilitarianism-inspired preferentialism also fails, in Sen’s view. For what matters is not just what goods do to people (through the mediation of their mental states), but what people (can) do with these goods. The versatility of the Sen’ own capability approach makes it a serious contender for a rich characterization of human flourishing. On that approach, equality requires that all people, present and future, have equal access to certain capabilities, or ‘positive freedoms’. Capabilitarianism’s major advantages are that it avoids the pitfalls of adaptive preference formation afflicting preferentialism (Sen explicitly opposes his view to the latter), offers intuitive responses to cheap taste problems (such as the Tiny Tim case) and can deal intuitively with nightmarish scenarios of environmental or genetic degradation, such as the one advertised two paragraphs back. The question remains, however, whether Sen’s monist equalizandum exhausts human flourishing. Proponents of the capability approach seem to think it does. The answer hinges on how exactly capabilities are individuated. Sen has been somewhat evasive on this topic. Consider the case a man who is chained to a wall in a rat-infested prison. The rats cause severe disadvantage to the man, by transmitting disease, causing pain, etc. A decontamination company removes the infestation. The man’s environment is now free of rats, and therefore of disease, rat-induced pain, etc. Let’s assume, further, that the man is chained to the wall of his own free will (a priest who is into selfpunishment) and thus has no preference for being unchained. Now, it is clearly desirable that people do not live in rat-infested environments. Post-decontamination, the priest obtains a functioning he has reason to value, namely being free of a rat-infested environment, being free of rat-induced pain, etc. But does he thereby also obtain some capability, say, the capability to live in a rat-free environment? In much of his work, Sen has stressed the ‘active’ role capabilities play in viewing people as autonomous agents, rather than the ‘passive’ role of a theory that emphasizes (only?) achievement and well-being. Hence his normative emphasis on capabilities and not mere functionings (see, for example, Sen 1984). But if capabilities are thus defined, then clearly the priest in the mooted example lacks them. For he is chained to the wall, subdued, with neither will nor capacity to make use of

Equality

71

his autonomy. Yet, despite the emphasis on capability, it seems desirable that he live in a rat-free environment. This brings us to Cohen’s critique of Sen. What seems implausible about Sen’s understanding of the place of capability in well-being is: the associated athleticism, which comes when Sen adds that the ‘central feature of well-being is the ability to achieve valuable functionings.’ That overestimates the place of freedom and activity in well-being. As Sen writes elsewhere ‘freedom is concerned with what one can do’ and ‘with what one can do’: midfare fails, on both counts, as a representation of freedom. (Cohen 2011, pp. 56–57)

Cohen introduces the notion of ‘midfare’, lying ‘midway between goods and utility’, as a dimension of well-being that includes Senian functionings and capabilities, but is not exhausted by them: Midfare is a heterogeneous collocation, because goods do categorically various things for people: (1) they endow them with capabilities properly so called, which they may or may not use; (2) through people’s exercise of those capabilities, goods contribute to the performance of valuable activities and the achievement of desirable states; and (3) goods cause further desirable states directly, without any exercise of capability on the part of their beneficiary: an example would be the goods which destroy the insects that cause malaria. Capability (properly so called) is, then, a part of midfare, for it certainly cannot be excluded from the range of things that goods confer on people yet, equally certainly, it does not exhaust that range. (Cohen 2011, pp. 48–49)

That capability does not suffice for human flourishing is perhaps clearest in the case of infants or severely disabled people: the desirability that they attain certain states of being is independent of their attainment of corresponding capabilities which are, ex hypothesi, out of reach. To the extent, therefore, that states of achievement, rather than capability for achievement, is necessary for flourishing, capability does not exhaust flourishing. The attentive reader will have noted that Cohen’s arguments do not entail rejection of any of the theories of well-being examined so far. Rather, they are aimed against their underlying monism, the assumption that each exhausts the most compelling theory of what egalitarians should aim to equalize.14

72

The Political Philosophy of G. A. Cohen

According to Cohen, we should instead opt for a mixed metric. A mixed metric lends voice to the idea that human flourishing – which Cohen (1989a) mysteriously dubs ‘advantage’15 – consists in the satisfaction of irreducibly heterogeneous dimensions of well-being. Preference satisfaction alone, he argues, cannot tell us what it means for people to flourish: it is, in more than one sense, too ‘subjective’. Capabilitarianism, on the other hand, is too ‘objective’: it fails to incorporate preferentialist considerations that are of prudential import and tends to overplay the centrality of agency to well-being.16 A mixed metric that includes midfare is better, says Cohen. Thanks to its ‘objectivist’ elements, it avoids nightmarish conclusions and is immune to cheap taste objections. Thanks to its ‘subjectivist’ elements, it is well placed to respond to scenarios that involve suffering and to instances of preference-disappointing, but fully functional, activity. The contents of this list, along with the prioritization of its elements, may vary across individuals, and may also vary for the same individual across time and space. Distributive justice, then, consists in ‘equal access to advantage’, that is, responsibility-constrained equal flourishing. Cohen’s metric, or ‘currency’, of justice naturally raises problems of commensurability, comparison and trade-offs across the heterogeneous dimensions of egalitarian concern. At what rate are we to trade-off preference satisfaction against valuable functionings? How does the rate vary throughout an individual’s lifetime? How does it vary across different individuals throughout their lifetimes? These index-related questions are unavoidable.17 The difficulties are compounded by conceptual issues pertaining to moral responsibility, by normative complications in its ascription to agents, and by inevitable ‘competition’ between justice, and on the one hand, and other values, on the other (see Cohen 1989a, p. 906). I shall return to this kind of competition at the end of this chapter.

Where the action is Sections ‘Patterns of justice’ and ‘Equality of what?’ dealt with the question of pattern and metric, respectively. One of Cohen’s most celebrated contributions to egalitarianism deals with the site of justice, and takes its cue from a critique

Equality

73

of John Rawls’ theory of justice. The bottom line of that critique is that an egalitarian ethos in citizens is a necessary condition for the just society on Rawls’ own terms. Cohen develops his critique by first appropriating the feminist slogan ‘the personal is political’ and then generalizing it. The rest of this section elaborates on this line of attack. Rawlsian liberalism offers a famous interpretation of the public/private distinction. Rawls argues that the ‘primary subject’ of justice has to be restricted to the ‘basic structure’ of society, construed as its fundamental, publicly observable, institutional features. Illustrations: legal institutions protecting liberty of thought and conscience, ‘competitive markets, private property in the means of production and the monogamous family’ (Rawls 1971, p. 7). In general, Rawls also includes the institution of the family under the basic structure (see Munoz-Darde 1998, Rawls 2001), but some feminists, such as Okin (1991), criticize his ambivalence on the matter. For them, Rawls is too lenient on the inequalities that inevitably result from an unfair division of labour within the household: The substance of the feminist critique is that standard liberal theories of justice, and the theory of Rawls in particular, unjustifiably ignore an unjust division of labor, and unjust power relations, within families (whose legal structure may show no sexism at all). That is the key point of the feminist critique, from a political point of view. But the (often merely implicit) form of the feminist critique, which we get when we abstract from its gendercentered content, is that choices not regulated by the law fall within the primary purview of justice, and that is the key lesson of the critique, from a theoretical point of view. (Cohen 2000b, p. 123)

Does inclusion of the family in the basic structure render Rawlsian liberalism coherent? Those feminists who respond in the affirmative ‘fail to grasp the form of the feminist critique of Rawls’, says Cohen. For if non-coercive institutions, such as the family, come under the purview of justice, then – largely for the same reasons – so does (systematically) unjust personal behaviour. Rawls therefore finds himself on a slippery slope, for he cannot exclude from the basic structure what he ostensibly wants to exclude, namely personal behaviour. Cohen here subjects the Rawlsian conceptual apparatus to immanent critique, by confronting it with a dilemma. Rawls:

74

The Political Philosophy of G. A. Cohen

must either admit application of the principles of justice to (legally optional) social practices, and, indeed, to patterns of personal choice that not legally prescribed … or, if he restricts his concern to the coercive structure only, then he saddles himself with a purely arbitrary delineation of his subject matter. (Cohen 2000b, p. 139)

Indeed, if Rawls chooses to opt, in all consistency, for the first horn of the dilemma, then his theory aligns itself with Cohen’s own putative view that ‘social justice requires a social ethos which inspires uncoerced equalitysupporting choice’ (Cohen 2000b, p. 131). One can here discern two argumentative strands surfacing in Cohen’s critique of Rawls: the first relates to the wrong-making consequences of market behaviour as such, and the second to the content of such behaviour. That is, Cohen objects both to the practice of the talented self-maximizers (i.e. those who, by dint of superior talent, can reap great rewards for themselves in the marketplace) and to the motivating reasons for such practice (which is to enrich oneself at the expense of others). The second line of criticism gains prominence in Cohen’s last major work (Cohen 2008), and I shall return to it in Chapters 5 and 6. The first strand of Cohen’s critique of Rawls, on the other hand, focuses on the consequences of the behaviour of talented self-maximizers as such. He concludes that: if we care about social justice, we have to look at four things: the coercive structure, other structures, the social ethos, and the choices of individuals; and judgement on the last of these must be informed by awareness of the power of the others. (Cohen 2000b, p. 143)

What is the political significance of the ethos discussion? Cohen writes: In 1988, the ratio of top-executive salaries to production-worker wages was 6.5 to 1 in West Germany and 17.5 to 1 in the United States. Since it is not plausible to think that Germany’s lesser inequality was a disincentive to productivity, since it is plausible to think that an ethos which was relatively friendly to equality protected German productivity in the face of relatively modest material incentives, we can conclude that the said ethos caused the worst paid to be better paid than they would have been under a different culture of reward. It follows, on my view of the matter, that the difference principle was better realized in Germany in 1988 than it would have been if its culture of reward had been more similar to that of the United States.

Equality

75

But Rawls cannot say that, since the smaller inequality that benefited the less well off in Germany was a matter not of law but of ethos. I think that Rawls’s inability to regard Germany as having done comparatively well with respect to the difference principle is a grave defect in his conception of the site of distributive justice. (Cohen 2000b, p. 145)

Cohen’s conclusion, if true, has important implications for how we should conceive of our moral ecology, and thus for the strategies required to transform it in an egalitarian direction. It is, moreover, significant for the longterm sustainability of egalitarian institutions, such as the system of taxation that Cohen favours, which ‘ensures that top incomes are no more than three times as high as bottom ones’ (Cohen 1989b, p. 11). In effect, Cohen has recast Gramsci’s notion of hegemony in terms more congenial to the liberal vocabulary. Its most significant corollary pertains to education and social pedagogy: a truly egalitarian society will educate its citizens in mutual respect and reciprocal giving, such that its least well-off members are as well-off as possible.18 Suppose that Cohen’s critique of Rawls is granted. Suppose, that is, that justice in citizens is a necessary condition for social justice. A further question arises: how can there be rich egalitarians? This is the question that forms the title of Cohen’s fourth major work, If You’re An Egalitarian, How Come You’re So Rich? The question becomes interesting only if we distinguish between three possible responses, only one of which makes the question worth asking. Take any well-off person, who professes a commitment to equality – and is, by dint of that commitment, an egalitarian – but does not give a significant proportion of her income to the worse-off. She might display: (i) hypocrisy, or (ii) akrasia, or (iii) lack of – what we may call – integrity. Hypocrisy presupposes insincerity, and akrasia presupposes failure to do what one knows she ought to do. Neither (i) nor (ii) exercise Cohen: his question is not about rich pseudo-leftists. Nor is it an instance of Donald Davidson’s (1970) more general question (how is weakness of the will possible?). Rather, what Cohen is interested in lies in the vicinity of (iii), that is, whether a non-weakwilled person may be sincerely committed to certain ideals and justifiably, or excusably, not strive to make them true. He advances a number of possible answers to this question, including the assurance problem (where both A and B want to contribute X to some cause, but A will only contribute if B does) and

76

The Political Philosophy of G. A. Cohen

draws a distinction between unjust and blameworthy behaviour. Cohen argues that the former does not entail the latter. Significantly, this loosens somewhat the moralistic rigour of the ethos he advocates.19

The Pareto argument for inequality One aspect of Cohen’s work that makes it so interesting is the uncompromisingness with which he persecuted his opponents’ views, without ever being pedantic or uncharitable. This applies as much to his critique of Rawls as to his critique of Nozick. This section summarizes a further development of the dialectic within the context of the former critique. According to a famous liberal argument, inequality is a concession to efficiency. Here’s a prominent version of that argument: (7) Justice is compatible with the Pareto principle.20 (8) The Pareto principle mandates inequality, rather than equality. Therefore (9) Justice is compatible with inequality.21 Cohen claims that Rawls mobilizes just such an argument in justification of his difference principle, and the same is true of influential Rawlsian liberals (such as Brian Barry). Cohen’s aim is to refute the Rawlsian justification for claim (8), by showing it to be incoherent. His critique picks up from what be called Rawls’ trialectic of equality: the transition from the laissez-faire of ‘natural liberty’ to the liberalism of ‘careers open to talents’, to Rawlsian ‘democratic equality’ (Rawls 1999, pp. 57–73). Cohen claims that the rationale for moving from the second form of equality to the third contradicts Rawls’ Pareto-inspired endorsement of inequality. Consider: The vertical axis of Figure 3.1 represents well-being (in the relevant metric). The horizontal axis represents a continuous ordering of individuals, such that if Z finds herself to the right of Y, then Z’s well-being is at least as high as Y’s. D1 and D3 are fully egalitarian distributions. D2 is inegalitarian (the individual on the leftmost side of the horizontal axis is significantly worse off than the one on the rightmost side). Cohen writes:

Equality

W

77

D2

D3

D1

i Figure 3.1  The Pareto argument

If D3 is feasible and talented people are willing to produce [at an equal rather than unequal wage – NV], then Rawls’ claim about the irrationality of insisting on equality in the face of the possibility of a Pareto-superior inequality would lose its force, since a Pareto-improving equality-preserving move, in which no one is as badly off as some are in D2, would now also be available. (Cohen 2008, p. 101)

One cannot say, in response, that ‘D3 is just not feasible’. For ‘in normal circumstances nothing but the unwillingness (be it justified or not) of the talented to share equally the greater produce produced in D2 could make D3 impossible when D2 is possible’ (Cohen 2008, p. 102). D3 just is the distributive upshot of voluntary, uncoerced, equality-friendly choices by the talented. Nor can Rawls now insist that D3 is illegitimate because it constitutes unwarranted restriction on incentive-seeking. To do so would amount to begging the question. Thus Cohen is in a position to conclude that ‘there is no reason for an egalitarian to regard D2 as acceptable, and every reason for him to recommend D3’ (Cohen 2008, p. 105). Indeed,

78

The Political Philosophy of G. A. Cohen

the original (Rawlsian) rationale for democratic equality issues from a belief that the unequalizing effect on distribution of natural and social contingencies is unjust. For ‘it permits distributive shares … arbitrary from a moral point of view’ (Rawls 1999, p. 63). To be talented is therefore to be privileged in the offending sense: it is to be so ‘positioned’ or ‘gifted’ as to (be able to) reap a higher ‘distributive share’ in a ‘morally arbitrary’ way. So how could D2 ever be deemed inferior to D3 from that point of view? The question of the justification of market-generated inequality thus turns crucially on the question of the justification of incentives (which I discuss extensively in Chapter 5).

The structure of justice Cohen’s egalitarianism was forged, I have argued, in the midst of polemic, first with Dworkin, and then with Rawls. This accounts for the piecemeal fashion in which it is defended, and for the ostensibly non-committal way in which that defence is articulated. One common objection against Cohen’s egalitarianism (closely related to the Pareto argument just rehearsed) is that it implies levelling down. The levelling down objection, due to Parfit (1997), says that ‘it is a mistake to favour equality, because favouring equality commits you to levelling down when the alternative to equality is a Pareto-superior inequality’ (Cohen 2011, p. 228). Given the significance of this objection for Cohen’s egalitarianism, I propose to pursue it at some length. Suppose that the original distribution between A and B, in the relevant metric, is D1: (8, 6), such that B is worse off than A through no fault of her own, and that distribution D2: (5, 5) is also in the feasible set. According to the levelling down objection, the egalitarian is committed to moving from D1 to D2 because D1 is unfair to B. Cohen’s response is that the objection confuses different levels of agency. He distinguishes between four questions: ‘(i) What is justice?, (ii) What should the state do?, (iii) Which social states of affairs ought to be brought about?, (iv) Which social states of affairs are better than others, whether or not they should be brought about?’ (Cohen 2011, p. 226). This set of distinctions has important implications for equality:

Equality

79

While it might well be grotesque for the state to mandate a levelling down in circumstances where the equal 5/5 distribution and the unequal 8/6 distribution … exhaust the feasible set, whereas, that is, it may be grotesque for the state to make everybody worse off, it does not follow that there is no injustice in the 8/6 inequality, and, partly for that reason, it does not follow that no one should seek to bring the 5/5 world about. (Cohen 2011, p. 229)

Cohen here echoes Larry Temkin: [D]o I really think that there is some respect in which only some being blind is worse than all being blind? Yes. Does this mean that I think it would be better if everyone else was blind? No. As noted previously, equality is not all that matters. But it does matter some. (Temkin 2003, p. 780)22

This kind of response exemplifies the radical pluralism of Cohen’s moral philosophy. Cohen thought that there exists a plurality of different values that are mutually irreducible, or imperfectly reducible to one another, and, in most cases, potentially in conflict with each other. Temkin’s example purports to show that nobody ought to bring it about that everyone is blind. And yet the subsisting inequality is regrettable, partly because it is unfair. Indeed, Cohen’s pluralism goes further: it implies that different forms, or ‘aspects’ (Cohen 2008, p. 323) of a singular value, such as justice, are themselves potentially at odds with one another.23 In light of his explicit commitment to the socialist tetrad of equality, freedom, community and democracy (Cohen 1995a, pp. 260–261), each of which possesses its own aspects, the upshot is a proliferation of possible values, and therefore of possible value conflicts. Figure 3.2 sketches what Cohen took to be the structure of justice.

Figure 3.2  Aspects of justice

80

The Political Philosophy of G. A. Cohen

The species distributive justice and the species desert all fall under the genus of justice. I have posited desert as a distinct species since Cohen has, on occasion, resisted the inference from ‘in accordance with desert’ to ‘in accordance with distributive justice’ (see, for example, Cohen 1995a, p. 257). Both distributive justice and desert have comparative and non-comparative aspects. An instance of non-comparative distributive justice (node I) compares not what a person gets with what others get, but ‘with what she might otherwise have got’ (Cohen 2008, p. 322). It is also presumably concerned with intrapersonal distribution across lifetimes. Non-comparative desert (node III) is similar in that its basis (individual achievement) is specified independently of what others do, or how they fare. Thus non-comparative desert conflicts with comparative desert when everyone, say, gets less, or more, than they non-comparatively deserve, but in stable comparative-desert-compatible ratios.24 What is more interesting, for Cohen’s purposes, are instances of comparative distributive justice (node II) and of comparative desert (node IV). Cohen thinks that these conceptual neighbours are key to specifying luck egalitarianism, since they are both closely associated with the notion of responsibility: [M]y own animating conviction in political philosophy with respect to justice is a conviction about distributive justice in particular. It is that an unequal distribution whose inequality cannot be vindicated by some choice or fault or desert on the part of (some of) the relevant affected agents is unfair, and therefore, pro tanto, unjust, and that nothing can remove that particular injustice. (Cohen 2008, p. 7)25

It follows that comparative desert and comparative distributive justice (the domain of ‘choice or fault’) are concerned with fairness.

Cohen’s egalitarianism and Cohen’s Marxism I want to conclude this chapter by raising some doubts about the compatibility between Cohen’s Marxism and his considered view of equality. For it seems to follow from Cohen’s position on the egalitarian ethos that some of his fundamental earlier commitments must either be revised or abandoned. In Chapter 1, I argued that Cohen’s Marxism takes history to be ‘a substitute for nature’, that is, it takes history to be the career of humanity struggling

Equality

81

for liberation, first through slavery and feudalism, and then through ‘the capitalist vale of tears’. By some grand dialectical reversal of sorts, that early claim is completely inverted in Cohen’s discussion of the egalitarian ethos: (human) nature now becomes a substitute for history. This is, of course, wholly incompatible with Cohen’s Marx. Cohen, after all, thought it was because Marx ‘was so uncompromisingly pessimistic about the social consequences of anything less than limitless abundance that [he] needed to be so optimistic about the possibility of that abundance’ (Cohen 2000b, p. 114). It follows that Cohen never repudiated technological materialism as adequate exegesis of Marx. Whether he remained committed to that form of materialism is another matter, to which I now turn. Any commitment to the feasibility and desirability of equality that does not avail itself of a ‘technological fix’ – an expectation of material abundance – must shed any ‘uncompromising pessimism about … anything less than limitless abundance’. This gives rise to another kind of optimism, an optimism about the prospects of harnessing the benevolent aspects of human nature as a substitute for history. Cohen’s lingering belief in the feasibility of full human emancipation embraces this kind of optimism. It follows that, in Cohen’s later political philosophy, an invigorated spiritual picks up the slack from an emasculated material.26 But this explanatory inversion is rather obviously in tension with technological materialism. For on the latter theory ‘the material determines the spiritual to the extent necessary to prevent the spiritual from determining the material’ (Cohen 1988, p. 160). How, then, can a mere psychological revolution in the direction of an egalitarian hegemony effect the epochal transformations in production relations required for universal equality of flourishing? More starkly: Cohen cannot both assert the primacy of the material (however broadly construed) and the feasibility of an ethosled egalitarianism. He must therefore either significantly revise his particular brand of Marxism, or his particular brand of egalitarianism.27 I do not know whether Cohen was himself aware of the starkness of this dilemma, or of its ramifications for materialism quite generally. His reluctant defence of the latter in Cohen (2000a), well after he had developed and defended his critique of Rawls, implies that he was not. Be that is it may, any critical engagement with Cohen’s philosophy as a whole must somehow come to terms with this apparent incompatibility between his materialism and his egalitarianism.

4

Exploitation Two ideas forged the conceptual bridge between Cohen’s Marxism and his normative political philosophy: freedom and exploitation. This chapter deals with the latter idea. What makes the task of tracking down Cohen’s considered views on exploitation difficult is that they figure virtually in every one of the debates he was engaged in, albeit in different ways. His early defence of historical materialism, for example, is concerned in part with the etiology of exploitation, whereas his later, normative work with its individuation and ethical implications.1 I proceed as follows. First, I track down Cohen’s early views on exploitation, with emphasis on his discussion (and rejection) of the labour theory of value. I show that Cohen subscribed to a distributive paradigm, the view, that is, that exploitation obtains if and only if distributive injustice obtains. I then connect Cohen’s espousal of that paradigm with his defence of luck egalitarianism. I argue that the distributive paradigm is false and then sketch a better alternative, one that resonates better with Cohen’s critique of capitalism. I conclude by showing that Cohen himself had doubts about the distributive paradigm and, indeed, repudiated it towards the end of his career.

Exploitation and historical materialism Cohen’s materialism, discussed in Chapter 1, recommends individuation of social forms in terms of the dominant relations of production ‘binding immediate producers’ (Cohen 1978, p. 78). Each social form, moreover, entails a particular form of surplus labour:

84

The Political Philosophy of G. A. Cohen

The form of surplus labour is the way it manifests itself in the society in question. Under slavery it appears in the portion of the slave’s product retained by the master after he has provisioned the slave. Under capitalism it manifests itself as a quantity of exchange-value: surplus labour is revealed only in the disguised form of profit on investment of capital. (Cohen 1978, p. 82)

The form of surplus labour, says Cohen, is distinct from the mode of exploitation, which is only ‘the means whereby the producer is made to perform surplus labour (whatever may be the form of surplus)’. A particular form of surplus labour is consistent with two modes of extraction: direct coercion (an ‘extraeconomic’ means) or exchange (an ‘economic’ means). Under capitalism, ‘the wage worker owns his labour power [so] he cannot be threatened with violent reprisal if he withholds it, but because he lacks means of production, no such threat is needed: he must, on pain of starvation, enter the labour contract’. Thus the dominant mode of exploitation under capitalism is ‘mediated by labour contract with surplus in value form’ (Cohen 1978, p. 83). What is it that explains the persistence and ubiquity of the exchange-mediated mode of exploitation, or, less prosaically, what is it that renders the vast majority of people exploitable under capitalism? Cohen’s answer: the economic base, that is, the sum total of effective-power relationships, which implies a monopoly over the means of production. The nature of the capitalist economic base is, in turn, explained by the extent to which it facilitates the development of the productive forces. By the same token, the division of violence between coercer and exploiter under capitalism, long ago noted by Marx, is to be functionally explained.2 Capitalism’s official coercer and propagandist is the state, charged both with enforcing capitalist property rights (rendering them effective)3 and with disseminating ideology on a large scale. Generations of Marxists have held that these social mechanics of exploitation are intimately connected with a set of propositions called ‘the labour theory of value’. Cohen devotes considerable attention to understanding that putative connection.

The labour theory of value In a justly celebrated essay, first published in 1979, and entitled ‘The Labour Theory of Value and the Concept of Exploitation’ (Cohen 1988, pp. 209–238),

Exploitation

85

Cohen sets out to show that ‘the relationship between the labour theory of value and the concept of exploitation is one of mutual irrelevance’ (Cohen 1988, p. 209). In effect, Cohen sets out to refute the following set of traditional Marxist commitments: (1) If capitalists exploit workers, then the labour theory of value is true. (2) If the labour theory of value is true, then capitalists exploit workers. Given the significance of this refutation for Cohen’s later forays into the territory of normative political philosophy, I shall study it at some length. I begin with (1). Cohen construes the labour theory of value as the idea that ‘equilibrium price is determined by socially necessary labour time’, and attributes it to Marx. He then argues that the labour theory entails that labour does not, indeed cannot, create value, as some Marxists insist. For socially necessary labour time and the cognate notion of ‘embodied labour’ are counterfactual magnitudes. That is, the value of some use-value u (a chair or a car) may be positive even when zero labour time is actually expended in making it, and the value of u may be zero even when a positive amount of labour time is expended in making it (see Cohen 1988, pp. 214–219). The reason why Marxists insist on the labour theory of value as basis for a charge of exploitation, says Cohen, is that they tend to confuse different objects of labour’s creation, namely value with that which has it (Cohen 1988, p. 226). What matters for exploitation is not that someone (the capitalist) appropriates value v created by another (the worker). Rather, what matters is that someone (the capitalist) appropriates (a part of) the value v of u, where u is created by another (the worker), regardless of how v is itself determined or created.4 It follows that (1) is false: capitalists exploit workers because they take without giving, which is to say nothing about the source of the value of their takings. I now turn to (2). Cohen argues that, even if the labour theory of value is true (such that necessary labour time determines value), that truth is irrelevant to a charge of exploitation. Imagine that A ‘creates’ value merely by desiring usevalue u. Cohen is entitled to ask us to imagine this, as it does not contradict what he is trying to show, namely that labour being the source of value is compatible with the absence of exploitation. It does not follow, indeed, it would be absurd to think that A is exploited when B appropriates (part of) u just because A

86

The Political Philosophy of G. A. Cohen

‘created’ u’s value (in this case, by desiring u). Mere value-creation is no legitimate basis for a charge of exploitation. So (2) is false. Cohen’s argument is important: it makes clear the counterfactual status of propositions involving necessary labour time; refutes common misunderstandings of the distinction between value and exchange value; and shows that the ‘traditional’ construal of the labour theory of value does not take us very far vis-à-vis exploitation. But why take the labour theory of value to be a theory of equilibrium prices in the first place? It is true that many Marxists, following Sweezy (1946), have taken Marx to be offering some sort of economic theory (as opposed to a critique of such theory). However, this traditional reading of Marx misunderstands the role the labour theory plays in the architectonic of his argument, its critical value and its role in the deduction of abstract labour and related categories. The labour theory could, instead, be conceived as a sociological theory of (the constitution of) social relations (see, for example, Rubin 1972). Under capitalism, the producer’s concrete labouring activity (her ‘labour’) only achieves social expression in exchange through the realization of the value of its product. This value may be proportional to cost of production, or, in terms closer to the neoclassical economist’s heart, to marginal value product. But why, as Marx memorably put it, is the labour of the producer represented in the (marginal) value of its product (Marx 1976, p. 174)? The answer is that a producer’s labour under capitalism gets its social significance from, and only from, the producer’s objective success in realizing value. Success in exchange is the one and only site where the producer’s life activity can express itself socially. It follows that capitalism institutionalizes the exclusion of forms of social expression that do not abide by the market norm.5

The plain argument Having refuted a traditional reading of the labour theory of value, Cohen endorses what he calls the ‘Plain Marxian Argument’ (Cohen 1988, p. 228): (3) The labourer is the person who creates the product, that which has value. (4) The capitalist appropriates some of the value of what the labourer creates. (5) The labourer receives less value than the value of what he creates.

Exploitation

87

Therefore, (6) The labourer is exploited by the capitalist. The Plain Argument does not presuppose, nor imply, the labour theory of value. Indeed, it is explicitly intended to buttress part of Cohen’s claim to the effect that the relationship between that theory and the concept of exploitation is one of ‘mutual irrelevance’. Now (6) patently does not follow from the conjunction of (3), (4) and (5). Cohen admits as much: ‘the crucial lacuna in the Plain Argument is a statement about the distributive background against which the labour contract is concluded’ (Cohen 1988, p. 233). And here is what – Cohen thinks – is needed to fill that lacuna: Capitalists obtain some of the value of what workers produce because capitalists do and workers do not own means of production: that is why workers accept wage offers which generate profit for capitalists. The crucial question for exploitation is, therefore, whether or not it is fair that capitalists have the bargaining power they do. If it is morally all right that capitalists do and workers do not own means of production, then capitalist profit need not be the fruit of exploitation; and, if the pre-contractual distributive position is morally wrong, then the case for exploitation is made. The question of exploitation therefore resolves itself into the question of the moral status of capitalist private property. (Cohen 1988, pp. 233–234)

The ‘because’ and ‘why’ in the first sentence are ambiguous: are they meant to express causal or noncausal conditions for exploitation? The expression ‘resolves itself ’ in the last sentence, however, seems to imply a noncausal reading: exploitation supervenes if and only if capitalist private property is unjustified. ‘Exploiting’, Cohen says, is a ‘kind of taking without giving’. But not all such unreciprocated takings are unfair, unjust or illegitimate (as when you give me back my coat). Thus: A flow of value … constitutes exploitation only if the contract it fulfils arises out of an unfair bargaining situation, and regardless, moreover, of whether or not that situation precisely forces the worker to sell his labour to the capitalist. Once the truisms of the Plain Argument are to hand, the crucial question for exploitation concerns the justice of the distribution of the means of production. (Cohen 1988, p. 234)

88

The Political Philosophy of G. A. Cohen

Cohen’s early view of exploitation, therefore, is that it constitutes a form of unreciprocated, though not necessarily forced,6 exchange, whose justice or fairness depends upon the justice or fairness of the distribution in the means of production: (7) The labourer is exploited, and the capitalist exploits, if and only if the exchange in which they are engaged occurs against the background of an unjust distribution of alienable resources.7 And since: (8) Capitalist ownership of the means of production is unjust. Claim (6) follows, and the ‘crucial lacuna’ is filled. Call premiss (7) the distributive paradigm. The distributive paradigm posits a necessary and sufficient condition for exploitation. It is not, moreover, a causal thesis: it does not say that (capitalist) exploitation must issue causally (or tendentially) from distributive injustice, although Cohen seems to also have held this view (see Cohen 1995a, pp. 197–205). Rather, it distinguishes just unreciprocated takings (which do not constitute exploitation) from unjust unreciprocated takings (which do), by appealing to the moral legitimacy of the background asset distribution. Now, one might object to my ascription of the distributive paradigm to Cohen. For, he says that ‘[i]f it is morally all right that capitalists do and workers do not own means of production, then capitalist profit need not be the fruit of exploitation’, where ‘need not be’ denotes ‘does not entail’. Thus exploitation and distributive justice could, conceivably, coexist: Cohen does not say that distributive injustice is necessary for exploitation but only that it is sufficient. Therefore, Cohen does not assert (7). This objection to my exegesis fails. For one, Cohen adds, almost in the same breath: ‘[a] flow of value … constitutes exploitation only if the contract it fulfils arises out of an unfair bargaining situation … ’ (Cohen 1988, p. 234). In further defence of the ascription of the distributive paradigm to Cohen, I must briefly consider his debate with John Roemer, which is instructive for what follows. Roemer attacked the concept of exploitation, as employed by Marxists, in ‘Should Marxists be interested in Exploitation?’ (see Roemer 1996). He returned a negative answer to this question, effectively retracing Cohen’s early steps:

Exploitation

89

 … for the sake of clarity and consistency, I think exploitation conceived of as the unequal exchange of labour should be replaced with exploitation conceived of as the distributional consequences of an unjust inequality in the distribution of productive assets and resources. (Roemer 1996, p. 96)

The emphasis should be on stocks, not flows, says Roemer. Cohen’s response to Roemer turns on pointing out a confusion between different levels of emphasis: that of exploitation’s etiology (‘causal fundamentality’, as Cohen calls it) and its individuation (‘normative fundamentality’). According to Cohen: John Roemer has argued … that not all unequal product flow on the market is unjust, and, indeed, that such a flow is unjust only if it reflects an unjust initial asset distribution. Roemer infers that the issue of flow is uninteresting and that Marxists are therefore wrong to focus on it. But I think that, although Roemer’s premiss (the italicized statement) is substantially true, it follows neither that unreciprocated flow is normatively uninteresting nor that (as Roemer also thinks) the normatively fundamental injustice is the asset maldistribution. (Cohen 1995a, p. 204)

Cohen and Roemer agree that some condition other than unreciprocation renders flow of value unjust (Cohen 1995a, p. 205). Roemer finds this condition in the resource distribution and infers that only the resource distribution matters. Cohen merely rejects that inference: The reason why Roemer’s conclusion does not follow is that it remains possible, and plausible, for all that his premiss (unequal flow is not, as such, unjust) is true, that it is the unequal flow that is unjust, when it reflects an unjustly unequal asset distribution, which distribution is unjust precisely because it enables an unjust equal flow so that, pace Roemer, the latter injustice is normatively fundamental. (Cohen 1995a, pp. 205–206)

Cohen’s response to Roemer demonstrates, I think, Cohen’s commitment to the distributive paradigm. That is, there is flow injustice just when there is stock injustice, and exploitation is a species of flow injustice. It follows that exploitation occurs just when there is stock injustice, and the absence of such injustice entails the absence of exploitation. On balance, therefore, it is safe to say that Cohen asserts the distributive paradigm. But he was ambivalent about such an assertion, and this shows in a peculiar argument promulgated in Cohen (1995a), which is worth mentioning.

The Political Philosophy of G. A. Cohen

90

In the midst of polemic against Roemer, Cohen gives a summary of what he takes to be the standard Marxian argument for exploitation, relying in part on ‘Plain’ premisses: The Marxian position is that, since labour and labour alone creates the product [premiss (3) – NV], and since (differential) ownership of means of production enables non-labourers to obtain some of what labour creates, just because they own means of production, their ownership of means of production is morally illegitimate … Marxists say, in effect, that since labour produces the product and private owners of capital appropriate part of it, private capital is morally illegitimate and workers are exploited. (Cohen 1995a, p. 206)

This passage is puzzling for the following reason: unlike what comes before it and after it in ‘Exploitation in Marx’ (Cohen 1995a, pp. 195–208), namely the insertion of the distributive paradigm as a premiss in the argument for exploitation, distributive illegitimacy features here as the relevant conclusion. To put it in the language of wrongmakers: the passage just quoted makes it look as if distributive injustice in the means of production is the moral explanandum here, whereas Cohen seems to have intended it to be some sort of explanans, that is, the (one) feature that makes (or breaks) the case for moral censure on grounds of exploitation. More concretely, the italicized part of the passage looks like an argument against capitalist private property from an exploitation premiss: (3) The labourer is the person who creates the product, that which has value. (9) Differential ownership of means of production enables non-labourers to obtain some of what labour creates, just because they own means of production. Therefore, (10) Capitalist ownership in the means of production is morally illegitimate. Cohen’s argument from the conjunction of (3) and (9) to (10) is incomplete, but something like the following would make it valid: (i) The relationship described in (9) is exploitative; (ii) Other things equal, exploitation renders whatever makes it possible morally illegitimate; (10) follows from (3), (9), (i) and (ii). But Cohen disavows this move through his commitment to the distributive paradigm, which makes distribution the (one) wrongmaking feature of exploitation.8

Exploitation

91

This concludes my parenthetical discussion of Cohen’s ambivalence vis-à-vis the relationship between exploitation and distribution. The nuances of the connection between the two concepts virtually disappear in Cohen’s work after 1988 (the year his second and last book on Marx was published) and are gradually replaced by an interest in distributive justice. That interest seems, if anything, to strengthen Cohen’s commitment to the distributive paradigm.9 I now turn to a discussion of his account of distributive justice and the sparse references to exploitation within it.

Exploitation as injustice Cohen’s mature view of exploitation takes its cue from the theory of distributive justice. In Chapter 3, I discussed Cohen’s novel critique of Ronald Dworkin’s resource egalitarianism and his defence of the theory that came to be known as luck egalitarianism. Part of Cohen’s original impetus for developing it pertains to exploitation: the ‘primary egalitarian impulse is to extinguish the influence on distribution of both exploitation and brute luck’ (Cohen 1989a, p. 908).10 In further elaboration of this claim, he says that: (11) Exploitation is unfair advantage-taking (Cohen 1989a, p. 908, 1995a, p. 151). Where: (12) Unfairness involves ‘involuntary disadvantage’. Cohen construes ‘involuntary disadvantage’ as any morally significant lack ‘for which the sufferer cannot be held responsible, since it does not appropriately reflect choices that he has made or is making or would make’ (Cohen 1989a, p. 916). (11) and (12) together yield the luck egalitarian definition of exploitation. This definition is the exact mirror image of the distributive paradigm in wellbeing, rather than asset, space: (7ʹ) The labourer is exploited, and the capitalist exploits, if and only if the exchange in which they are engaged occurs against the background of an unjust distribution, that is, a distribution involving involuntary disadvantage.

92

The Political Philosophy of G. A. Cohen

Let us call (7ʹ) the broad distributive paradigm. The broad paradigm replaces the original reference to ‘unjust distribution in alienable resources’ with ‘involuntary disadvantage’. For, according to the luck egalitarian definition, A can exploit B even when they are exactly equal in terms of (access to) alienable resources or means of production. If A is very intelligent, for example, she can take unfair advantage of B, who is not (equally) intelligent, by tricking the latter into working long hours for A. Cohen’s mature allegiance is to the broad distributive paradigm. I now want to argue that the distributive paradigm (restricted or broad) is false. This rebuttal is indicative of a deep tension within Cohen’s egalitarianism, so I shall pursue it at some length.

Against injustice One way to refute the distributive paradigm is to show that injustice furnishes no sufficient condition for exploitation. Clearly (7ʹ) needs some refinement on that front, for A can’t exploit B if they live in separate islands, never to interact, even if they own unequal amounts of resources. Crucially, the set of sufficient conditions for exploitation must be completed by some clause referring to the exploitee’s labour or effort, and this has proven very difficult to do.11 Another way to refute the distributive paradigm is to show that distributive injustice furnishes no necessary condition for exploitation. It is to such a demonstration that this section is devoted. Suppose A finds B in a pit. A can get B out at little cost or difficulty. A offers to get B out, but only if B agrees to sign a sweatshop contract with A. This exchange is paradigmatic of exploitation: if this is not an instance of exploitation, then nothing is. Yet thinking that B’s acceptance of the offer results in A exploiting B is inconsistent with the distributive paradigm (i.e. it contradicts (7) and (7ʹ)). For A’s offer can be forthcoming even in the absence of involuntary disadvantage, or indeed any sort of distributive injustice: B may have found herself there through just steps, from a just starting position. B may, for example, have chosen not to purchase cheap insurance against falling in pits or against being rescued afterward. Alternatively, the pit may be the bottom of a huge vanilla ice cream cone, and B may have ended up there

Exploitation

93

by intentionally licking all of the vanilla. It does not follow, and it is false, that the offer is not exploitative: A can rescue B without asking for anything but instead uses B’s vulnerability in order to improve her own lot. It follows that asset injustice furnishes no necessary condition for exploitation and therefore no proper basis for a charge of exploitation. Some proponents of the luck egalitarian definition, like Cohen, will be inclined to grant that the pit case involves exploitation. But if they do, then they must defend the distributive paradigm, which is a corollary of the luck egalitarian definition and which is presently under the threat of counterexample. One response the defender of the luck egalitarian definition could offer is that falling in the pit was not reasonably foreseeable by B. She therefore could not have been expected to purchase insurance against it, which makes the supervening luck brute. The resulting distribution (of welfare, goods, capabilities, etc.) is unjust, and therefore (7ʹ) holds. But suppose it is reasonably foreseeable, although far from certain, that B will fall. Can we still square the intuitive idea that A’s offer is exploitative with the luck egalitarian definition? The luck egalitarian may – rather desperately – reply that, if there exists a third party, C, who starts out as well-off as A and B, then it is bad brute luck for C if A ends up much better off as a result of a rescue offer. After all, A just happened to be passing by when B needed help. There is, therefore, an unjust distribution between A and C, for which C must be compensated. (7ʹ) therefore still holds. This response solves nothing: if C does not exist or shares in the spoils with A, no one seems to have a grievance as a result of A’s actions (on the basis of the luck egalitarian definition). By way of summary of the argument so far, consider the following triad of statements, which are not jointly assertible: (11) Exploitation is the unfair taking of advantage. (12) Unfairness consists in the presence of involuntary disadvantage. (13) Exploitation obtains even in the absence of involuntary disadvantage. The pit case shows that (13) is true. And if unfairness is what Cohen says it is, that is, if (12) is true, then exploitation cannot be a species of unfairness. It follows that (11) is false. This conundrum is, I submit, a consequence of the forced marriage Cohen has effected between justice and exploitation through the distributive paradigm. I am therefore in agreement with Michael Otsuka, who argues that:

94

The Political Philosophy of G. A. Cohen

Cohen could vindicate the claim that [a price gouger who sells necessities at high prices against a background of distributive justice – NV] exploits by taking unfair advantage of the disaster victim only at the cost of the abandonment (or trivialization) of his claim that ‘unjust exploitative flow requires an initial unjust distribution.’ (Otsuka 2010, p. 222)

The distributive paradigm is very likely false. This has significant ramifications for how we should conceive exploitation-based complaints in general and against capitalism, in particular.

Exploitation and vulnerability One can exploit objects and things, such as raw materials, just as one can exploit human attributes or circumstances, such as an opponent’s dribbling weakness in football. Neither form of exploitation is necessarily wrongful. One can also exploit (whole) persons. Such exploiting entails wrong- or ill-doing. In other words, if A says to B ‘you are exploiting me’, and what A says is true, then B has eo ipso committed a wrong. Here is an alternative definition of exploitation, one that does not appeal to the distributive paradigm:12 A exploits B if and only if (i) A instrumentalizes, (ii) B’s vulnerability v, (iii) to extract a net benefit b from B, (iv) where v corresponds to a power A has over B. Clauses (i) and (ii) underline the fact that exploitation is pro tanto disrespectful;13 (ii) and (iii) distinguish exploitings from other forms of mere usings; while (iv) adds a structural, but largely unnoticed, implication of exploitation-claims (indeed, the conjunction (i)–(iv) shows why exploitation is a form of domination). I therefore suggest that (7) and (7ʹ) should be replaced by (something like): (14) The labourer is exploited, and the capitalist exploits, if and only if the capitalist instrumentalizes the worker’s economic vulnerability to appropriate (the fruits of) her labour. Let us call (14) the vulnerability paradigm. On any view of capitalism, capitalists use workers (by extracting labour time from them) to obtain a benefit (profit) by taking advantage of their vulnerability (their lack of free access to the means of production). The only controversial aspect of the demonstration

Exploitation

95

that all actual capitalists come under (14) consists in showing that they instrumentalize the vulnerability of workers. But all capitalists are constrained, on pain of survival qua capitalists, to treat their workers merely as sources of profit, just as they treat their machinery.14 Therefore capitalists exploit workers, and (6) is established without recourse to distributive premisses. This demonstration is no doubt incomplete. It also raises more questions than it answers, and I try to address these elsewhere.15 Unlike its vulnerabilitybased rival, Cohen’s distributive paradigm emphasizes what is sometimes called expropriation. Expropriation is the (forced?) separation of the labourer from the means of production. According to the vulnerability paradigm, exploitation may coexist with, and be exacerbated by, expropriation. But it remains a (morally and empirically) distinct phenomenon within capitalist relations of production. One important reason why the vulnerability paradigm captures intuitions surplus to the distributive paradigm (and is therefore superior as a theory of exploitation) is that it takes seriously the possibility of what Cohen calls a ‘cleanly generated capitalism’ (Cohen 1995a, p. 120). Here is an extreme instance of such capitalism: Grasshopper spends the summer months singing, whereas Ant spends all her time working. When the winter comes, Grasshopper needs shelter, which he presently lacks. Ant has three options: she can do nothing to help Grasshopper, she can offer him shelter that costs Ant nothing or she can offer him costless shelter on the condition that he signs a sweatshop contract. The possibilities, in the relevant metric, are: (Ant, Grasshopper) (i) Do nothing: (10, 1) (ii) Sweatshop: (12, 2) (iii) Shelter: (10, 3) Now, it is plausible to think that Ant has an obligation to help Grasshopper. But one need not have a view on that to believe that (ii) is morally worse than (iii). That is, if Ant decides to help (decides to do something other than (i)), then she must not opt for (ii), in part because doing so constitutes exploitation. I take the widely shared intuition expressed in this consequent as evidence that the vulnerability-based definition of exploitation is plausible. But now note that if (ii) involves exploitation, then asset injustice furnishes no necessary condition for it. For, according to (7ʹ), all of (i), (ii) and (iii) are equally acceptable at

96

The Political Philosophy of G. A. Cohen

the bar of asset justice. Cohen, Roemer and other Marxist advocates of (7ʹ) have nothing to oppose to such a cleanly generated capitalism on exploitation grounds, and this is an embarrassment to their claims to holding a socialist, or a socialism-friendly, theory of equality. This result generalizes, I believe, to any theory of exploitation that posits some defensible account of asset injustice as a necessary condition for exploitation.

Cohen’s change of mind I now want to argue that, although Cohen never advocated the vulnerability paradigm, he nevertheless disavowed the distributive paradigm towards the end of his life. This disavowal implies that exploitation is no longer taken to categorize injustice (or unfairness), as the early Cohen maintained. This gradual change of heart vis-à-vis the distributive paradigm begins with an abandonment of the luck egalitarian definition. Suppose Werther is deeply in love with Charlotte, whom he leaves wholly indifferent. He gives away all of his livelihood to impress her, and she is flattered and amused by his courtship. Charlotte takes Werther’s gifts without ever reciprocating. In this case Werther and Charlotte in effect agree that Werther be impoverished and Charlotte be enriched (in the relevant metric). In an essay published as part of a Festschrift for Hillel Steiner, Cohen (2011, pp. 124–146) calls examples like this the unanimity cases. These are instances of Dworkinian option luck.16 In the light of the unanimity cases, Cohen argues that option luck does not always preserve justice or fairness and hints that it never does.17 He merely ‘hints’ that option luck never preserves justice, because he is unsure about certain option luck cases which are unlike the lover case, in that they involve a gamble rather than a direct assignment of benefits to particular people. Hence Cohen: In the unanimity cases, everyone agrees to what would otherwise be unquestionably unjust. But what the gambler agrees to is a 50–50 chance, an equal chance, and that’s certainly not unquestionably unjust … The gambler is not agreeing directly, but only indirectly, to an outcome. (Cohen 2011, p. 135)

Michael Otsuka defends this ‘discriminating position’, which dubs unanimity cases as unjust but (outcome-equivalent) gambles not so. The discriminating

Exploitation

97

position relies on a distinction between willing an inequality (this is what Werther wills or foresees) and willing a procedure that is going to produce an inequality (such as a fair gamble). Thus in the former cases everyone opts ‘for what all know will be a distribution in favour of known people’, whereas in the latter cases ‘the gamblers do know that the gamble will produce an inequality, but neither knows who will benefit from that inequality’ (Cohen 2011, p. 138). I do not believe the discriminating position is justifiable. For Werther can choose to impoverish himself either to Charlotte or Maria, the choice between them decided by coin toss. Distribution is, then, not unequal ‘in favour of known people’ but is still not relevantly different from the unanimity cases.18 If I am right, the answer to Cohen’s puzzle is: both unanimity and gamble cases involve unfairness, and if this result generalizes, then option luck never preserves justice (Cohen asserts that it so generalizes in Cohen 2011, p. 132). What does all this have to do with exploitation? Cohen wants to defend the intuitive response to lover-like cases, namely that Charlotte-types exploit Werther-types. But if that is true, then it follows that: (15) Exploitation is consistent with voluntary (i.e. option-luck-generated) disadvantage. (15) clearly contradicts (7ʹ), the broad distributive paradigm,19 since the paradigm says that exploitation requires involuntary disadvantage, and voluntary disadvantage is not a species of involuntary disadvantage. Cohen’s own resolution of this contradictory stance consists in abandoning the distributive paradigm:20 I began by being against inequality that is not due to relevant choices by relevant agents. I thought such were the results of luck. But then it occurred to me that exploitation wasn’t bad luck, because it was a deliberate act. So I added a further source of injustice. But that little train of thought involved an equivocation on ‘bad luck’. Exploitation is bad luck for me in one sense, namely that it doesn’t reflect my choice, but in another sense it clearly isn’t: it happened not by accident but by the exploiter’s choice. I should have stuck with the choice formula, and restricted ‘bad luck’ simply to absence of relevant choice (on the part of its sufferer?). Then exploitation would be a distinct form of injustice, and not particularly a matter for egalitarianism. I was not pluralistic enough when I crafted or cobbled the conjunctive formulation. (Cohen 2009b, emphasis added)

98

The Political Philosophy of G. A. Cohen

The ‘conjuctive formulation’ Cohen is referring to is his early view that ‘the primary egalitarian impulse is to extinguish the influence on distribution of both exploitation and brute luck’ (Cohen 1989a, p. 908). The reference to ‘egalitarianism’ in the italicized part of the passage refers to the distributive paradigm: the reason why Cohen was ‘not pluralistic enough’ is that he tried to subsume claims of exploitation under distributive injustice, where they do not belong. The distributive paradigm is therefore, according to the late Cohen, false. I must now summarize the argument of this chapter. The early Cohen construed the labour theory of value as a theory of equilibrium price and dismissed it as an ‘incubus on progressive reflection about exploitation’. What matters, he thought, is not the labour theory of value, or unequal exchange of labour, as some Marxists maintain. It is, rather, who produces what and, crucially, who is morally entitled to what product of labour (or its material preconditions). This led him to what I have called the distributive paradigm, according to which the ‘question of exploitation resolves itself into the question of the moral status of capitalist private property’ (Cohen 1988, p. 234, 1995a, p. 204). Cohen refined this paradigm by pioneering a theory of distributive equality (which I discussed in Chapter 3). But he eventually realized that injustice in distribution is neither sufficient nor, crucially, necessary for exploitation, thereby disowning the distributive paradigm. An interesting and important corollary of this revision – if it is sound – is that exploitation no longer needs to wear the straitjacket of distributive justice or, at least, the straitjacket of distributive justice as luck egalitarians and their intellectual progeny conceive it.

5

Community Marx’s philosophical trajectory began, as is well known, with a ‘humanist’ critique of Hegel and the Young Hegelians (with whom Marx was originally associated), culminating into a ‘materialist’ theory of society and history.1 Cohen’s philosophical trajectory was, in some ways, the reverse of Marx’s: from his early elaboration, and defence, of historical materialism in the 1970s and 1980s, Cohen gradually moved towards normative political philosophy, which figures prominently in all his writings from the 1990s until his death in 2009. This chapter studies an important but neglected strand of Cohen’s humanist thought, his account of fraternity or community.2 Community is significant not only because it may have intrinsic value (see Wolff 1968) but also because it was, and remains, a significant point of difference between liberals and socialists or, to the extent that liberals can be socialists, between liberals and communists. What I will try to do is offer a more unified account of Cohen’s views on community than he himself provided, set against the background of his critique of (Rawlsian) liberalism. I begin by sketching some parallels between Marx’s thought and Cohen’s normative political philosophy. I then outline Cohen’s community critique of Rawls and describe Cohen’s view of the connections between Rawlsian liberalism and community. This will give rise to a general tension immanent to Cohen’s normative political philosophy between the values of justice and community. I argue that it may well go deeper than he thought and conclude with a discussion of the textual avenues available for a reconciliation of these two values.

100

The Political Philosophy of G. A. Cohen

Marx and Cohen on community Marx’s writings are replete with allusions to community, its forms in precapitalist society, its gradual dissolution under capitalism and its culmination under communism.3 Moreover, the dissolution ushered in by capitalism is posited as necessary for moving beyond feudal ‘idiocy’ and parochialism.4 As part of this process, large-scale enclosures, privatization of the means of production and the separation of men from their means of subsistence (their ‘radical chains’), all conspire towards the establishment of a classdivided society based on a capitalist division of social labour. In that society, the dominant form of relationship between human beings is one of ‘mutual indifference’. Marx does not deny that capitalism exhibits a certain form of reciprocity. Indeed, the market form of reciprocity is, he says, a ‘natural precondition of exchange’ (Marx 1973, p. 244). But that is not the rich form of reciprocity Marx envisages for communism:5 [Market] reciprocity interests [the subject to an exchange] only in so far as it satisfies his interest to the exclusion of, without reference to, that of the other. That is, the common interest which appears as the motive of the whole is recognised as a fact by both sides; but, as such, it is not the motive, but rather it proceeds, as it were, behind the back of these self-reflected particular interests, behind the back of one individual’s interest in opposition to that of the other. (Marx 1973, p. 244)

The reciprocity congealed in exchange relationships is said by Marx to have an ‘abstract universality’ (Marx and Engels 1975, vol. 28, pp. 40–41). It is a vehicle for the creation, and satisfaction, of more and richer human needs, a process which releases humanity from the limited and parochial relationships of feudalism. The abstract universality of capitalist exchange relationships is transcended under communist production, where free cooperative labour forms the basis of ‘truly human’ relationships, that is, relationships in which ‘man’s need has become a human need’ and where ‘the other person as a person has become for him a need’ (Marx 1986, pp. 31–34). Crucially, Marx does not infer that communist social relationships will be dominated by altruism or love: ‘the individuals’ consciousness of their mutual relations will, of course … no more be the ‘principle of love’ or dévouement than it will be egoism’ (Marx and Engels 1975, vol. 5, p. 439).

Community

101

As we shall see in the discussion of liberalism and justice that follows, Cohen’s work in normative political philosophy echoes Marx’s critique of capitalist community. But first I want to discuss Cohen’s characterization of Marxian communism. He pictures it as follows: [I]magine a jazz band in which each player seeks his own fulfilment as a musician. Though basically interested in his own fulfilment, and not in that of the band as a whole, or of his fellow musicians taken severally, he nevertheless fulfils himself only to the extent that each of the others also does so, and the same holds for each of them. There are, additionally, some less talented people around who obtain high satisfaction not from playing but from listening, and their presence further enhances the fulfilment of the band’s members. (Cohen 1995a, p. 122)

This is Cohen’s elaboration of the analogy: [A]s I understand Marx’s communism, it is a concert of mutually supporting self-fulfilments, in which no one takes promoting the fulfilment of others as any kind of obligation. I am not, of course, denying that each delights in the fulfilment of others. Unless they are crabby people, they probably do so. But no such delight is required: it is not something in the dimension of affect which is supposed to make communism possible. (Cohen 1995a, p. 123)

The emphasis on free, spontaneous, cooperative activity, largely in the absence of categorical interpersonal obligations, makes the jazz band analogy particularly apt.6 In a ‘well-ordered’ jazz band, improvisation takes place such that all members realize to the full their capacities and talents: each band member leads and is in turn led by the others. The emphasis is therefore not on altruism or duty but rather on the development of free individuality: ‘Each joins [the band] to fulfil himself, and not because he wants it to flourish for any independent reason’ (Cohen 1995a, p. 137). Now a question immediately arises: how is it possible that everyone can partake of this ‘concert of mutually supporting self-fulfilments’ in a world where the ‘circumstances of justice’, that is, non-unlimited abundance and non-unlimited altruism, obtain? How can Marxian communism, in other words, avoid the ‘struggle for necessities’, the concomitant ‘generalized want’ and the ‘old filthy business’ which characterizes class-divided societies (Marx

102

The Political Philosophy of G. A. Cohen

and Engels 1975, vol. 5, p. 48)? Cohen interprets Marx as saying that, under communism, the circumstances of justice no longer obtain: [I]n Marx’s good society, productive resources are available, gratis, to all, but  the individual remains effectively sovereign over himself … An overflowing abundance renders it unnecessary to press the talent of the naturally better endowed into the service of the poorly endowed for the sake of establishing equality of condition … . (Cohen 1995a, p. 122)7

It is Marxian optimism about communist abundance that makes possible the ‘truly human’, ‘mutual indifference’-transcending relationships that Marx favours (in the absence of a massive change in human attitudes, which Marx disavows). Having attributed this view to Marx, Cohen rejects such optimism about material possibility (Cohen 1978, p. 207, 1995a, pp. 5–12, 118–131, 135). He argues that it must be replaced by ‘voluntary equality’, that is, equality of condition motivated by widespread social and individual commitment to that value (see the discussion of the egalitarian ethos in section ‘Where the action is’ of Chapter 3). Although Cohen rejects the jazz band conception of communism,8 he does take on board Marx’s critique of the distinction between the droits de l’homme and the droits du citoyen: Above all, we note the fact that the so-called rights of man, the droits de l’homme as distinct from the droits du citoyen, are nothing but the rights of a member of civil society – i.e., the rights of egoistic man, of man separated from other men and from the community … Only when the real, individual man re-absorbs in himself the abstract citizen, and as an individual human being has become a species-being in his everyday life, in his particular work, and in his particular situation, only when man has recognized and organized his ‘own powers’ as social powers, and, consequently, no longer separates social power from himself in the shape of political power, only then will human emancipation have been accomplished. (Marx and Engels 1975, vol. 3, p. 168)

Marx’s complaint entails that ‘political emancipation’ preserves, indeed implies, the insulation of one area of life, civil society, from democratic accountability and control.9 As we shall see, Cohen gives renewed expression to this complaint against liberalism by mounting a comprehensive attack against its most prominent contemporary defender, John Rawls. Cohen’s attack, if successful, afflicts not only liberals but all defenders of capitalist markets.10

Community

103

Community and liberalism Having tread the path from historical materialism to normative political philosophy, Cohen inevitably has to face up to the massive edifice of Rawlsian liberalism.11 According to Rawls, the primary subject of social justice is the basic structure. He argues that a ‘well-ordered’ society must regulate this structure to conform with his two principles of justice, which are derived from an original position of free and equal rational choosers.12 Rawls’ favoured principle of distribution, the difference principle, is derived from this ‘constructivist’ device. According to the difference principle, ‘social and economic inequalities are to … be to the greatest benefit of the least advantaged members of society’ (Rawls 1999, p. 72). Between 1992 and 2009, Cohen largely occupied himself with criticizing the conclusions Rawls draws from the original position argument and, eventually, with attacking the constructivist reading of that position itself.13 Now, one of Rawls’ famous conclusions was that the difference principle represents a plausible interpretation of the principle of fraternity … The difference principle … does seem to correspond to a natural meaning of fraternity: namely, to the idea of not wanting to have greater advantages unless this is to the benefit of others who are less well off … Those better circumstanced are willing to have their greater advantages only under a scheme in which this works out of the benefit of the less fortunate. (Rawls 1999, p. 90)

Cohen’s strategy consists in pointing out a series of structural ambiguities in Rawls’ reading of the subject of justice, and therefore in his understanding of community. The argument has two steps. In the first, Cohen purports to show that there exist individual choices which fall under the purview of social justice and which Rawls cannot exclude, on pain of arbitrariness. He then argues that, were a case to be made that these individual choices manifest Rawlsian fraternity, then that fraternity would be too ‘thin’ to merit the name. If these two steps are successful, then Cohen will have shown that the most influential contemporary defence of liberalism is inadequate on at least one ground, that of community. The rest of this section discusses those two steps. The first step in Cohen’s argument involves an immanent critique of Rawls’ basic structure restriction, which I have already sketched in section ‘Where the

104

The Political Philosophy of G. A. Cohen

action is’ of Chapter 3. That critique is indispensable to Cohen’s community argument, for community applies irreducibly to relations between individuals within structures. The conclusion of Cohen’s critique is that there is nothing about personal behaviour within coercive institutions to rule out the possibility of oppression, exploitation and injustice supervening on such behaviour alone. Therefore, the basic structure must include more than coercive institutions. Rawls could respond by offering a more expansive account of the basic structure such that it includes personal behaviour. But that, Cohen argues, would make him into: a radical egalitarian socialist, whose outlook is very different from that of a liberal who holds that ‘deep inequalities’ are ‘inevitable in the basic structure of any society’. (Cohen 2008, p. 129)14

For Cohen, a ‘radical egalitarian socialist’ is someone who believes that equality applies not only to structures but also to individual behaviour. Radical egalitarian socialists affirm the feminist slogan ‘the personal is political’ because they are radical socialists and endow that slogan with egalitarian content (perhaps through an egalitarian ethos) because they are egalitarians. This completes the first movement of Cohen’s critique of Rawls: social justice does not exempt personal behaviour from the demands of justification. The second movement requires him to show that certain forms of personal behaviour, now calling for justification, cannot meet the demands of a form of community worth its salt. The kind of personal behaviour that exercises Cohen is that of talented incentive-seekers. He argues that there exists no compelling ‘I-thou’ Rawlsian justification that the talented can offer to the nontalented for benefitting from market-generated inequalities (such as: ‘the inequality between you and me was necessary to make you better off ’).15 The inadmissibility of the mooted form of justification turns on an ambiguity in the notion of ‘necessity’. Given the centrality of this démarche for Cohen’s account of community, I will study it at some length. In his Tanner lectures, entitled ‘Incentives, Inequality and Community’,16 Cohen attacks Rawlsian liberalism for its tolerance of inequality. As in his basic structure critique, Cohen insists that, since justice judges personal behaviour, and since the talented in Rawlsian society will get more than the

Community

105

nontalented, Rawlsian liberals are unjustifiably committed to ‘giving to those who have’ (Cohen 2008, p. 86). What justifies the fact that the talented will, in a Rawlsian society, fail to contribute, say, the same amount of work at a higher rate of taxation (for the benefit of the nontalented less well-off)? According to Cohen, such behaviour is normally unjust. The normality qualification is important: a refusal to work the same number of hours at a higher tax rate (or more hours at the same tax rate) can be caused by an agent’s unwillingness to work longer hours or by the practical impossibility of doing so without impugning one’s own life projects.17 That is, sometimes the talented can’t produce the same at a higher tax rate, and that’s a perfectly good excuse, indeed a justification, for not doing so. But sometimes they won’t produce the same at a higher tax rate. To the distinction between ‘can’t’ and ‘won’t’, there correspond, roughly, two readings of the difference principle. The strict reading of the difference principle ‘counts inequalities as necessary only when they are, strictly, necessary, that is, apart from people’s chosen intentions.’18 The lax reading, on the other hand, countenances intention-relative necessities as well. So, for example, if an inequality is needed to make the badly off better off but only given that talented producers operate as self-interested market maximizers, then that inequality is endorsed by the lax, but not by the strict, reading of the difference principle. (Cohen 2008, p. 69)

If the lax difference principle entails ‘giving to those who have’, says Cohen, then we should favour the strict difference principle on justice grounds. But Cohen also develops a (justice-independent) argument in favour of the strict difference principle, on community grounds. I turn to the latter presently.

Against liberalism Cohen’s community-based argument for the strict difference principle goes as follows: I believe that the idea that an inequality is justified if, through the familiar incentive mechanism, it benefits the badly off, is more problematic than

106

The Political Philosophy of G. A. Cohen

Rawlsians suppose; that (at least) when the incentive consideration is isolated from all reference to desert and entitlement, it generates an argument for inequality that requires a model of society in breach of an elementary condition of community. The difference principle can be used to justify paying incentives that induce inequalities only when the attitude of talented people runs counter to the spirit of the difference principle itself: they would not need special incentives if they were themselves unambivalently committed to the principle. Accordingly, they must be thought of as outside community upholding the principle when it is used to justify incentive payments to them. (Cohen 2008, p. 32)

Cohen’s conception of community takes as its point of departure the interpersonal test: This tests how robust a policy argument is by subjecting it to variation with respect to who is speaking and/or who is listening when the argument is presented. The test asks whether the argument could serve as a justification of a mooted policy when uttered by any member of society to any other member. (Cohen 2008, p. 42)19

The interpersonal test in turn furnishes a necessary condition20 for what Cohen calls justificatory community: A justificatory community is a set of people among whom there prevails a norm (which need not always be satisfied) of comprehensive justification. If what certain people are disposed to do when a policy is in force is part of the justification of that policy, it is considered appropriate to ask them to justify the relevant behaviour, and it detracts from justificatory community when they cannot do so. It follows that an argument for a policy satisfies the requirement of justificatory community, with respect to the people it mentions, only if it passes the interpersonal test. (Cohen 2008, pp. 42–43)

Justificatory community is said to ‘contribute to’ or be ‘indicative of ’ community tout court, a moral notion Cohen compares to friendship (Cohen 2008, p. 43). In effect, justificatory community restricts the class of admissible inequalitytypes to those that pass the interpersonal test. Moreover, Cohen claims that this class of inequalities rules out lax specifications of the difference principle. A well-ordered society must therefore affirm and practice the strict difference principle, thereby enforcing (some measure of) his egalitarian ethos. Here is Cohen’s poignant challenge to Rawls:21

Community

107

Two brothers, A and B, are at benefit levels 6 and 5, respectively, in New York, where they live. If they moved to Chicago, their levels would rise to 10 and 5.1, respectively. If they moved to Boston, they would rise to 8 and 7. Is fraternity, as Rawls means to characterize it, consistent with A proposing that they move to Chicago? If so, it is a thin thing. Or is Rawlsian fraternity strictly maximinizing? (Cohen 2008, p. 78)

Rawlsian fraternity is a ‘thin thing’, unless ‘strictly maximinizing’. In further defence of strict maximinizing, that is, of enforcing a distribution strictly maximizing the well-being of the least well-off, Cohen develops an analogy between a kidnapper and a talented egoist. He asks us to imagine a kidnapper directly addressing the following ultimatum to the kidnapped child’s parents: Children should be with their parents. Unless you pay me, I shall not return your child. So you should pay me. (Cohen 2008, p. 39)

The justification given by Rawls for paying higher salaries to the talented is similar in structure: Economic inequalities are justified when they make the worst off people materially better off. (Major premiss) When the top marginal rate is 40 per cent, (a) the talented rich produce more than they do when it is 60 per cent, and (b) the worst-off are, as a result, materially better off. (Minor premiss) Therefore, the top tax should not be raised above 40 per cent to 60 per cent. (Cohen 2008, pp. 34)

Cohen claims that the two arguments are relevantly analogous.22 To see this, all we have to do is imagine the latter argument articulated by the talented and addressed directly to the (nontalented) poor. If we do that, we will observe that both the kidnapper and the talented egoist make the minor premiss true. That is, the kidnapper makes it true that if he does not get the money, he will not return the child. Likewise, the talented who insist on high salaries or profits make it true that if they do not get the high reward, they will work less hard, to the detriment of the poor. The kidnapper’s conditional therefore violates community because it fails the interpersonal test: his illocutionary acts place him beyond the pale of an embarrassmentfree dialogical relationship with the child’s parents (Cohen 2008, pp. 41–46).

108

The Political Philosophy of G. A. Cohen

If the analogy with the kidnapper is sound, and if non-‘thin’ community requires justificatory community, as Cohen claims, then the posture of the talented is unacceptable on community grounds. To summarize the argument so far: I have tried to sketch Cohen’s twopronged attack against Rawlsian liberalism, set in the context of a broader argument for an egalitarian ethos (strictly speaking, for an intention-inclusive subject of justice). On the first prong, Cohen advances intention-inclusion on justice grounds: allowing the talented to be better off than the less talented is giving to those who have. On the second prong, Cohen advances intentioninclusion on community grounds, by arguing that in Rawlsian society the dialogical relationship between the talented and the untalented can never be embarrassment-free.23 How are these two grounds related? I address this question presently.

Against justice Cohen’s normative political philosophy is deeply ambivalent on the connection, and indeed the compossibility, between justice (roughly: equality)24 and community (roughly: justificatory community). Consider, first, Cohen’s characterization of the kidnapper: although what is (mainly) bad about the kidnapper is not his voicing the argument, but his making its minor premise true, he should still be ashamed to voice the argument, just because he makes that premise true. The fact that in some cases he would do further ill not to voice the argument does not falsify the claim that in all cases he reveals himself to be ghastly when he does voice it. (Cohen 2008, p. 40)

This passage implicitly distinguishes between makings (of the minor premise true) and voicings (of the conditional ransom demand). The supervening dialogical relationship between parent and kidnapper clearly puts the kidnapper beyond the pale of community. But the passage does not make sufficiently clear whether the making is a necessary condition for the voicing being bad or indeed whether the making is a necessary condition for the justifiability or aptness of feeling shame. In other words, we are not told whether injustice is a necessary condition for lack of fraternity. To grant that

Community

109

both the kidnapper and the talented egoist do injustice is not to say that justice is necessary for community. Cohen’s examples do, however, seem to gesture in this direction on occasion, for they furnish a plausible explanation as to why the voicings are wrong. This ambivalence becomes ever more transparent in a short book published a month after Cohen’s death, entitled Why not Socialism? (Cohen 2009a). I am rich, you are poor, because of regrettable bad choices … and not therefore because of lack of equality of opportunity. You have to ride the crowded bus every day, whereas I pass you by in my comfortable car. One day, however, I must take the bus, because my wife needs the car. I can reasonably complain about that to a fellow car-driver, but not to you. I can’t say to you: ‘It’s awful that I have to take the bus today.’ There’s a lack of community between us of just the sort that naturally obtains between me and the fellow car driver … I believe that certain inequalities that cannot be forbidden in the name of socialist equality of opportunity should nevertheless be forbidden in the name of community. But is it an injustice to forbid the transactions that generate those inequalities? Do the relevant prohibitions merely define the terms within which justice will operate, or do they sometimes (justifiably?) contradict justice? I do not know the answer to that question (Cohen 2009a, pp. 35–36, emphasis added).

Cohen is here asking not whether justice (sometimes) conflicts with community. Rather the question is: given that the realization of distributive justice will sometimes conflict with the realization of community,25 is it unjust to forbid the transaction-types that generate this conflict? The example Cohen gives is of a lottery, in which all can (freely) enter, beginning from a state of equality. The lottery generates massive inequalities of condition, which are inapposite to embarrassment-free dialogical relationships: ‘even though there is no injustice here, your luck cuts you off from our common life, and the ideal of community condemns that, and therefore also condemns the running of any such lottery’ (Cohen 2009a, p. 38). The conflict runs deeper than Cohen thinks. Let’s assume, plausibly, that the idea of community at work in the bus passage quoted two paragraphs back, entails, or partakes of, justificatory community.26 Cohen’s rationale for introducing this particular account of community is that its absence makes

110

The Political Philosophy of G. A. Cohen

room for too much inequality. Now, both the talented egoist case and the bus case involve too much inequality. Of course, the former case is unlike the latter in that the talented egoist is, by assumption, benefitting from injustice, whereas the car-owning bus-commuter is not. But both cases are said to involve transgressions of community. Why, then, dismiss the theory that endorses incentive-seeking but retain the theory that endorses inequality of car-ownership? In other words, if Cohenite community pro tanto undermines Rawlsian egalitarianism because that egalitarianism allows communityundermining inequality, why does it not also pro tanto undermine Cohenite (luck) egalitarianism – instead of (merely) making that egalitarianism ‘contradict’ community? Both theories are too inapposite to community. Cohen’s anti-Rawls strictures thus indirectly afflict his own account. The fox bites its own tail. Cohen has two responses. He can say that the relevant form of community tentatively broached in Cohen (2009a) does not imply justificatory community. Alternatively, he can insist that his original, antiRawls argument was purely immanent and that he himself does not accept justificatory community. Both responses purchase consistency at the cost of emptiness, for each leaves us in the dark as to what Cohen actually thinks about community.27 Cohen’s ‘somewhat concocted’ notion of justificatory community is deep and interesting but turns out to be in tension with his own theory of justice.28 I now turn to another important notion in Cohen’s discussion of community, one that looms large in his critique of markets, the idea of communal reciprocity.

Communal reciprocity Cohen’s (2009a) community-based argument against market socialism enlists two ‘modes of communal caring’.29 The ‘first mode’ is something like justificatory community and is illustrated by the bus case (see previous section). The ‘second mode’, dubbed ‘communal reciprocity’, is: [T]he antimarket principle according to which I serve you not because of what I can get in return by doing so but because you need or want my service, and you, for the same reason, serve me. Communal reciprocity is not the same thing as market reciprocity, since the market motivates

Community

111

productive contribution not on the basis of commitment to one’s fellow human beings and a desire to serve them while being served by them, but on the basis of cash reward. The immediate motive to productive activity in a market society is (not always but) typically some mixture of greed and fear, in proportions that vary with the details of a person’s market position and personal character … the market posture is greedy and fearful in that one’s opposite-number marketers are predominantly seen as possible sources of enrichment, and as threats to one’s success. (Cohen 2009a, pp. 38–41)

Cohen’s last book is full of passages reminiscent of Marx’s critique of market society (see the section ‘Marx and Cohen on community’ above, and the striking do ut facias passage in Marx 1976, p. 1047): the ‘abstract’ but ‘selfish’ form of market reciprocity; the alienation of producer from consumer and product; the semblance of independence ‘founded on dependence’; and so on. By contrast, in a society where human relationships are regulated by communal reciprocity, people can properly be said to constitute each other’s need, and the emphasis is not just on how they fare but also on how they approach and treat one another. Fear and greed thus cease to be the dominant motives in their everyday lives and are replaced by dialogically acceptable mutual concern. A question naturally arises about the exact relationship between equality and this second mode of community. Both notions are at the heart of Cohen’s political philosophy. Consider the equality-community biconditional formed by the conjunction of: (1) If community obtains, then equality obtains. and (2) If equality obtains, then community obtains. Unfortunately, Cohen rarely discusses the biconditional. Indeed, he never really says what equality tout court is – that is, what it means to treat people as equals. Of course, part of the impetus behind his defence of luck egalitarianism was precisely to sketch the distributive dimension of such treatment, thereby attenuating that concern. But, as I argued in Chapters 3 and 4, Cohen became increasingly discontented about the generality and plausibility of

112

The Political Philosophy of G. A. Cohen

luck egalitarianism. To add to the confusion, Cohen on occasion intimated rejecting both (1) and (2). In connection with (1), he wrote: Under its most abstract description, the motivation [required by communal reciprocity – NV] might be consistent with hierarchy: Prince Charles’ motto is ‘Ich dien,’ and serfs and lords alike who buy feudal ideology wholesale can describe themselves as being motivated thus. If community motivation is indeed consistent with hierarchy, then the principle of equality informs the principle of community, in its socialist form. (Cohen 2011, p. 218)30

The truth of the antecedent in the last sentence implies that (1) is false. Communal reciprocity must therefore be ‘informed’ by a ‘principle of equality’. We are therefore back to the indeterminacy faced by Cohen’s original notion of justificatory community, which I discussed in the previous section. The ‘principle of equality’ that Cohen envisages in this connection is: ‘from each according to his ability, to each according to his needs’: [F]or present purposes, the unambiguous message of the slogan is that what you get is not a function of what you give, that contribution and benefit are separate matters. Here the relationship between people is not the instrumental one in which I give because I get, but the wholly noninstrumental one in which I give because you need … Accordingly, the ideal in the primeval socialist slogan constitutes a complete rejection of the logic of the market. (Cohen 2011, p. 220)

Whether or not Cohen’s argument for the conclusion (that the socialist slogan is completely anti-market) is sound, the premiss (that human relationships be non-instrumental) is not very informative for our purposes: saying that treating people as equals consists in institutionalizing non-instrumental relationships sounds very much like a tautology.31 And if the conceptual lacuna left by Cohen’s dissatisfaction with luck egalitarianism is not filled (see the section ‘Cohen’s change of mind’ of Chapter 4), we are left almost completely in the dark as to what it means for a ‘principle of equality’ to ‘inform’ community. We are left, at best, with a more egalitarian version of Rawls’ difference principle. I turn now to (2). One way in which Cohen tries to hoist Rawlsian liberals with their own petard consists in charging them with a fallacy of equivocation: the equality that engenders community and vindicates (2) can

Community

113

only be inspired by a strict reading of the difference principle. But when Rawls affirms equality, he tends to fall back to a lax reading. According to Cohen, therefore, the only notion of equality that meets the demands of Rawlsian community – and does not commit the mooted equivocation – is ‘strictly maximinizing’. ‘Strictly maximinizing’ community is, however, weaker than Cohenite communal reciprocity. The gulf between these two distinct notions of community might explain some of Cohen’s ambivalence on (2). In an unfinished essay published four years after his death, for example, Cohen broaches the following suggestive questions: What do we acknowledge when we nod at a stranger in the street, without smiling? Can we represent that as a case of respecting without any friendliness? Is what’s acknowledged when we acknowledge a stranger the answer to what it is to regard people as equals? (Cohen 2013, p. 200)

An affirmative answer to the second question is plausible and amounts to a repudiation of (2). Perhaps the case can be made that communal reciprocity is not a kind of ‘friendliness’32 and therefore that an affirmative answer to the second question is compatible with the truth of (2). Communal reciprocity could, alternatively, be conceived as a form of institutionalized rejection, writ large, of the quid pro quo nature of market transactions. More work on Cohen’s account of community might shed light on what exactly this means. The distinctive moral significance of socialism, and its capacity to represent itself as distinct from liberal egalitarianism, may yet hinge upon the understanding and prominence socialists attach to this question.

The division of moral labour I conclude this chapter by showing that a certain division of moral labour underlies Cohen’s commitments to justice and community, such that justice is concerned primarily with outcomes, whereas community is concerned primarily with individual motivation. In discussion of the role of the egalitarian ethos for justice, Cohen writes: Under abnormal conditions, justice might be consistent with universal self-interested maximising: if, for example, talents and utility functions

114

The Political Philosophy of G. A. Cohen

are identical, then initial equality of tangible assets might be considered sufficient for justice. (Cohen 2008, p. 73, emphasis added)

It follows that, in the possible world in which self-interested maximizers get universal equality, justice is fully done. It is, of course, very unlikely that this possible world will become actual, but the preceding modal claim shows that Cohen is an externalist about justice. Externalism about justice asserts that justice is primarily concerned with the evaluation of states of the world,33 especially distributions, independently of the motivation or standing of the agents that bring them about (for the connection between externalism and Cohen’s conservatism, see the section “Conservatism” of the Introduction).34 That Cohen asserts externalism about justice can be further corroborated from a passage where he discusses James Meade’s social-democratic egalitarianism. The Meadian ideal is that people should act altruistically at the ballot box, so that they can (permissibly) remain selfish in everyday life: Suppose that all citizens have the same tastes and talents, or at any rate that no differences among their tastes and talents would prevent an initial stateenforced equality of resources from reproducing itself under subsequent market-maximizing behaviour. Here the Meadian prescription would be sound, from a purely egalitarian point of view (as opposed to from the point of view of an egalitarianism enriched by a principle of community). But the required conditions will never obtain. Selfish market behaviour will induce inequality of reward, and state intervention could mitigate but not (consistently with efficiency) reverse that tendency. (Cohen 2008, p. 175, emphasis added)

Since the Meadian prescription is ‘sound’, externalism about justice is true. But the same passage confirms that Cohen rejects externalism about community. For, the italicized fragment implies that ‘the principle of community’ judges more than person-independent states of the world. I try to further uncover the polemical significance of this claim in Chapter 6.

6

Socialism and the Market In Chapters 1 and 3, I argued that Cohen’s secular theodicies, the first from the standpoint of a theory of history and the second from the standpoint of a theory of justice, are both animated by a commitment to the desirability and feasibility of a fully emancipated society. But what exactly would such a society look like? There is a long tradition within Marxism of resistance to socialist ‘blueprintism’, that is, to ‘writing recipes for the cook-shops of the future’ (Marx 1976, p. 99). Despite his original positivist aversion to blueprintism, Cohen was gradually converted to the idea that such resistance is misguided. Indeed, he eventually came to argue that familiarity with at least the basic ingredients of the recipe is not merely optional ornament to emancipatory political theory but an integral part of it (see Cohen 2000b, pp. 75–77 and passim). This chapter seeks to bring together some of Cohen’s arguments to bear on the issue of the desirability and feasibility of socialism. Socialism is perhaps the one topic that unites the disparate intellectual labours and political advocacies that animated Cohen’s philosophy. These labours and advocacies are usually attempts to rethink the basics of socialist commitment and strip it down to its essentials. Any study that attempts to make Cohen’s philosophy whole cannot, therefore, succeed without assigning pride of place to the content of the socialist ideals he advocated. At the same time, the piecemeal and immanent nature of Cohen’s arguments makes it difficult to extract a positive vision with some measure of precision. For this reason, this chapter is more speculative than the foregoing ones. I proceed as follows: in the first section I outline Cohen’s early critique of market socialism. That critique has a justice component and a community component. After offering some brief remarks on the justice component, I develop and discuss Cohen’s anti-market community argument while trying

116

The Political Philosophy of G. A. Cohen

to vindicate his view that ‘every market, even a socialist market, is a system of predation’ (Cohen 2009a, p. 82). I conclude with a discussion of Cohen’s ambivalent stance on the question of market socialism and offer a communitybased argument by way of explanation of Cohen’s withering, but enduring, life-long commitment to socialist planning.

Market socialism and its criticism The development of Cohen’s vision of socialism is perhaps most transparent in an essay he never published, entitled ‘David Miller on Market Socialism and Distributive Justice’ (Cohen 1989b). This essay contains the seed of Cohen’s later critique of market socialism and offers a glimpse of his account of the emancipated society. By ‘market socialism’, Cohen understands a decentralized form of worker management based on profit maximization, such that: (1) the means of production are jointly owned (either by workers, or by the state); (2) the firms are democratically controlled, either by the workers or through the state,1 allowing the former to share in the profit; and (3) production is for a market, in which values are determined either by the cost of production or marginal cost. Under market socialism, production equilibrium is achieved through maximization of profit (per worker). Thus in long-run competitive equilibrium the firms that survive are those, and only those, that make an economic profit greater than, or equal to, zero. It follows that different firms will come to own differential assets, depending on how successful they are at accumulating capital. The whole point of allowing the existence of profit-seeking firms, after all, is to take advantage of the coordinating informational signals putatively furnished by markets open to capital accumulation. It follows that some people will do well and others badly, depending on where and whether they work. Inequalities thus generated will tend to undermine the combination of egalitarianism and community that socialists are after. The enthusiasm for market socialism by some leftists is therefore difficult to explain: [T]here is among socialist intellectuals an intelligent movement, but also alongside it, an unthinking and fashion-driven rush, in the direction of

Socialism and the Market

117

a non-planning or minimally planning market socialist society. Market socialism is socialist because it overcomes the division between capital and labour: there is, in market socialism, no class of capitalists facing workers who own no capital, since workers themselves own the firms. But market socialism is unlike traditionally conceived socialism in that its worker-owned firms confront one another, and consumers, in competitive market-contractual fashion; market socialism is also, and relatedly, unlike traditionally conceived socialism in that it reduces, even though it does not entirely eliminate, the traditional socialist emphasis on economic equality. (Cohen 1995a, pp. 255–256)

According to Cohen, part of the explanation for the unthinking rush in this movement is to be found in adaptive preference formation, that is, the ‘irrational process in which a person comes to prefer A to B just because A is available and B is not’ (Cohen 1995a, p. 253). David Miller is not guilty of adaptive preference formation, for he offers an explicit argument in defence of the market (see Miller 1990). That defence has two premisses, says Cohen: the first is that the market tracks desert and the second that desert tracks justice. Miller infers that the market tracks justice. Cohen questions both premisses, but it is only the first one that matters for our purposes:2 Market earnings are a function of sales minus costs, and that magnitude is in turn a function of how much agents manage to produce. But how much agents manage to produce is not a well-behaved function of, or good proxy for, the amount of care they devote to production, since accidents of native talent and social fortune and market contingencies also affect output. (Cohen 1989b, p. 13)

Cohen is here being uncharacteristically sloppy, since – as he knew very well  – the market registers even less than ‘how much agents manage to produce’. It only registers output that is productive of surplus value, that is, production that leads to the realization of profit. So one of the reasons why the market fails to track desert, and therefore why Miller is mistaken, is that ‘the only thing the market can register is achievement or success, of which effort [for example] is only one cause, not itself registered by the market’ (Cohen 1989b, p. 13).

118

The Political Philosophy of G. A. Cohen

‘Happenstance variation in demand’, inequality in the original distribution and similar deficiencies in background conditions imply that the market does not, indeed cannot, reward equal effort equally. Markets only reward objective achievement, in the sense of realized surplus value or profit. Cohen thus turns Miller’s argument on its head: to the extent that desert mandates reward for effort, markets necessarily fail to reward desert. This anti-market gambit is related to, but does not entail, a broader set of criticisms. Cohen distinguishes between an equality- or justice-based critique of the market and a community-based critique. The equality critique of the market says that ‘the market distributes in unjustly unequal amounts’ (Cohen 1989b, p. 26a). The community critique says that the market ‘motivates contribution not on the basis of commitment to one’s fellow human beings and a desire to serve them while being served by them, but on the basis of impersonal cash reward’ (Cohen 1989b, p. 26a). I shall deal with the equality critique for the rest of this section. I will then take up the community critique. Cohen’s equality critique makes two claims. The first claim is that capitalist markets are unjustly exploitative because they reproduce an unequal distribution in the means of production.3 Cohen approvingly cites John Roemer to this effect: [T]he agendum of our age is to eliminate capitalist exploitation and replace it by socialist exploitation, an exploitation based on talent differences in a society which has abolished differences in access to capital. (Cohen 1989b, pp. 1–2)

More precisely, capitalist exploitation obtains when, for any coalition of agents A and its complement B, the following is true: (i) were B to withdraw from the society, endowed with its per capita share of society’s alienable property (that is, produced and nonproduced goods), and with its own labour and skills, then B would be better off (in terms of income and leisure) than it is at the present allocation; (ii) were A to withdraw under the same conditions, then A would be worse off (in terms of income and leisure) than it is at present; (iii) were B to withdraw from society with its own endowments (not its per capita share), then A would be worse off than at present. (Roemer 1996, p. 40)

Socialism and the Market

119

Socialist exploitation is the residual exploitation once capitalist private property has been eliminated. It stems from inequality due to differences in talents, knowledge, etc., and is identified by replacing ‘alienable’ by ‘inalienable’ in (i) above. A truly emancipated society would tolerate inequality neither due to arbitrary social differences, such as ownership in the means of production, nor due to arbitrary natural differences, such as ownership in talents, organizational knowledge, etc. Thus Cohen’s second claim is that equality proper – which Cohen (2009a) dubs radical equality of opportunity – requires an equal distribution that enfranchises both alienable and inalienable resources. Most of Cohen’s researches on socialism post1990 focused on conceptualizing such a system. He thought he had finally discovered it in the work of Joseph Carens. Carens (1981) defends a system of market-based firms (possibly cooperatives), which produce for profit, and whose profits get redistributed after production, such that the post-tax distribution, in whatever metric, is exactly equal, or responsibility-constrained-equal (see the section ‘The problem of luck’ of Chapter 3) or satisfies some other just and/or socially desirable pattern.4 Carensian socialism enlists a set of moral incentives in production, which, together with redistribution, generate the desired equality of outcomes in the relevant metric. Cohen called the Carensian model the ‘Platonic ideal of market socialism’ (Cohen 1995a, p. 264). Here is how he characterized it: Carens described a society in which what looks like a standard capitalist market organizes economic activity, but the tax system cancels the disequalizing results of that market by redistributing income to complete equality. There are (pretax) profit-seeking capitalists, and workers who own no capital, but people acknowledge an obligation to serve others, and the extent to which they discharge that obligation is measured by how close their pre-tax income is to what would be in the most remunerative (and therefore, on standard assumptions, the most socially contributing) activity available to them, which taxation effects a fully egalitarian post-tax distribution of income. Here, then, producers aim, in an immediate sense, at cash results, but they do not keep (or otherwise benefit from) the money that accrues, and they seek it out of a desire to contribute to society: a market mechanism is used to solve the social technology problem, in the service of equality and community. (Cohen 2009a, pp. 63–64)

120

The Political Philosophy of G. A. Cohen

The Carensian ideal is ‘Platonic’ because, unlike other forms of market socialism, it makes sufficient room for eliminating objectionable inequality due to differences in holdings of alienable and inalienable resources.5 But in order to do this, Carensian socialism must give a very prominent role to the market. For this reason Cohen retains some ambivalence towards it, not least because of a withering, but enduring, commitment to a nonmarket, democratically planned, form of socialism. The next two sections discuss the basis of Cohen’s ambivalence.

Community against the market In the section on communal reciprocity of Chapter 5, I discussed the ‘second mode’ of community advocated by Cohen (2009a), communal reciprocity. In a society where human relationships are regulated by communal reciprocity, people can properly be said to constitute each other’s need, and the emphasis is not just on how they fare but also on how they approach and treat one another. Fear and greed thus cease to be the dominant motives in their everyday lives, as they are in market society, and are replaced by dialogically acceptable mutual concern. Some liberals object to the identification of market motivation with motivation by fear and greed. Jeremy Shearmur, for example, discusses ‘mixed motive’ systems in connection with the market for blood, adding that ‘there seems to be no special reason to insist on austerely altruistic motives here’ (Shearmur 2003, pp. 264–265). Drawing upon Andrew Carnegie’s life, Hillel Steiner also argues that neither fear nor greed are necessary elements in the motivational explanation of market behaviour. (Steiner 2014, pp. 140–150) The liberal objection confuses different senses of the term ‘market’. As I understand that term, there are two kinds of markets. The first kind is innocuous, the second is devilish. An innocuous market system is what may be called small-scale commodity production: a network of markets, where people bring their privately owned wares for trade, to the exclusion of a labour market.6 Injustice or lack of community may supervene under small-scale commodity production, but that social formation is not intrinsically unjust or unfraternal and can be regulated to produce both justice and community.

Socialism and the Market

121

A non-innocuous market system is what Marxists call generalized commodity production: a network of markets, where people bring their privately owned wares for trade and which includes a labour market. Indeed, what distinguishes capitalist from pre-capitalist modes of production is that, under the former, and unlike the latter, capital has invaded the sphere of production and turned labour power into a commodity.7 I maintain that the distinction between different forms of commodification sheds light on Cohen’s remarks about community and therefore on his opposition to the sort of ‘handling’ that markets (regulated or not) generate: The capitalist market does not, of course, require people to handle people roughly, but … the market does require people to handle people, to manage them, in a particular sense. Business is, among other things, people treating people according to a market norm – the norm that says they are to be dispensed with if they cannot produce at a rate which satisfies market demand … Business turns human producers into commodities. (Cohen 2000b, p. 181)

The distinction shows, I think, how wrongheaded the ‘mixed motives’ response is. The capitalist market, qua generalized commodity production, by definition involves treating humans like commodities. And commodification must perforce institutionalize fear and greed, for otherwise it cannot perform its functional role. It institutionalizes fear because the owner of labour power (who does not own means of production) will always be under (the threat of) unemployment, poverty, alienation, misery and so on.8 It institutionalizes greed because no owner of the means of production can maximize her profit without maximally exploiting (in both the ethically neutral and the non-neutral sense) her workers.9 To be sure, not all market relationships constitute institutionalizations of instrumental treatment of others, as smallscale commodity relations demonstrate (e.g. my relation to the local grocer). Furthermore, no moral complications need arise with commodification if I get my wealthy neighbour’s child to help me move my sofa in return for some pocket money. Commodification is objectionable if and when it embodies or implies some sort of domination, that is, the instrumental treatment of others through power over them.10 A rough and ready liberal rejoinder to this argument may grant that capitalists, qua capitalists, must behave with greed, and that workers, qua

122

The Political Philosophy of G. A. Cohen

workers, must behave with fear under the structural constraints capitalism imposes. But one might add that systems of market socialism, such as the Carensian one, very nearly eliminate the suspect forms of motivation or render them morally negligible. There is, therefore, nothing wrong with the market per se. Cohen’s position on this rejoinder is ambivalent. On the one hand, he claims that market exchange in the Carensian economy is ‘only superficially market exchange’ (Cohen 2009a, p. 69). On the other hand, all forms of market socialism, including Carens’, retain some inferiority to ‘planned mutual giving’, ‘primarily on grounds of fellowship’ (Cohen 2009a, p. 6). One explanation for this may be that even Carensian market socialism fails to meet Cohen’s standard of communal reciprocity. To see this, consider what Carens requires of the typical capitalist: in the morning she goes to work and treats others exclusively as sources of enrichment, fires and hires them, and so on. The typical worker, on the other hand, views the capitalist and her fellow workers as constant threats to her own autonomy and selfrespect.11 In the afternoon the capitalist and the worker join hands and return home together to enjoy their high – and equal – levels of material wealth. In the afternoon they ‘storm heaven’, but in the morning they must reluctantly take their chances in a profane world. The scent of bourgeois schizophrenia is in the air. What’s wrong with schizophrenia?12 I say nothing about schizophrenia in general, but the bourgeois schizophrenia defended by market socialists is psychologically unsustainable and morally problematic, or, at least, it must be for someone (like Cohen), who affirms the value of community. It is psychologically unsustainable because few sane persons can treat others as mere sources of enrichment in the morning and as fully equal persons in the afternoon. It is morally problematic because Carensian morningafternoon interactions are surely analogous to someone kidnapping your child in the morning and returning it (with some added benefit) in the afternoon (for Cohen’s use of the kidnapper analogy, see the section ‘Against liberalism’ of Chapter 5). Market socialism, even Carensian market socialism, is a generalized, self-reproducing kidnapping, in the sense that both the victims of kidnapping and the victims of the market (in the noninnocuous sense of ‘market’) are dominated: only a handful of workers

Socialism and the Market

123

are ever given a reasonable exit option from the sort of institutionalized relationship that breeds fear and/or greed, in which their lives are embedded and reproduced.13 To be sure, under Carensian socialism, market domination results in substantial net benefits for the receivers of post-tax redistribution (whereas in the case of kidnapping, the domination typically results in net harm for parents and child). But this does not detract from the inherently objectionable features of the relationship in which market agents are embedded, at least during their working lives, including the tendency of this relationship to constitute yet another ‘nursery for those vices of character’ which include servility and greed.14 Since communal reciprocity by definition rules out these ‘vices of character’ and the structures that generate them, Carensian socialism is pro tanto inconsistent with communal reciprocity. To summarize the argument of this section: Cohen thought communal reciprocity to be a value of ‘supreme’ importance. He also held that value to be inconsistent with standard (capitalist) market motivations, such as fear and greed. Some liberals counter this criticism by arguing that markets are consistent with ‘mixed motives’, which need not involve fear and greed (or, at any rate, their morally suspect forms). To this I have responded, on Cohen’s behalf, that whatever motives people happen to (want to) have, once we properly disambiguate the term ‘market’, we will come to see why the capitalist market has an inherent tendency to generate fear and greed.15 The liberal rejoinder says that elimination or minimization of fear and greed is possible under a system of market socialism (e.g. resembling Carens’ model). To this I have replied that even such a system might fail to meet Cohen’s standards of communal reciprocity, insofar as it retains an extensive labour market. For, the labour-market-inclusive Carensian system would uphold the liberal distinction between ‘earthly’ and ‘heavenly’ aspects of the social world, in a way conducive to structural domination. I conclude that full realization of communal reciprocity requires the abolition of such relationships. It requires, in a nutshell, the decommodification of human labour power: decommodification is necessary for community.16 The next two sections further develop the decommodification strand of Cohen’s thought.

The Political Philosophy of G. A. Cohen

124

Commodification and personal autonomy Cohen’s argument against the commodification of human labour power is a recurring leitmotif throughout his work. In the context of discussion of the market for kidneys and sex, Cohen writes: One might condemn rape solely because it is a species of assault. In that case the condemnation of rape is consistent with indifference toward prostitution. Indeed, if one condemns assault simply because it is a breach of self-ownership, and the principle of self-ownership exhausts one’s ethical armoury, then … the very basis for condemning rape also confers unlimited license on prostitution. But if one condemns rape because it is forcing something that should be given only in love … then that reason for condemning rape is a also a reason for disapproving of prostitution. (Cohen 2008, p. 224)

And he adds: Few think about labour in general in the way that most think about a prostitute’s labour … In the widespread view, it is all right to sell your labour, though not, perhaps, for absolutely whatever you can get. But I belong to the stated few. I have some sympathy with the communist slogan that says ‘From each according to his ability, to each according to his needs.’ … [I] n the communist view, labour, like love, should, if given, be given freely. (Cohen 2008, p. 225)

At this point Cohen approvingly cites Marx’s claim that ‘prostitution is only a specific expression of the universal prostitution of the worker’ (Marx 1963, p. 156). But what makes the sale of one’s labour power relevantly analogous to prostitution? Cohen’s answer is that prostitution is only a special case of a broader phenomenon in which ‘the wanted thing is yielded for the wrong reason’ (Cohen 2008, p. 224). What does that mean? Part of the answer has already been provided in the previous section. The argument there was that Cohen’s ideal of communal reciprocity, which implies the regulative ideal of ‘serve-and-be-served’, is directed against social systems which institutionalize relationships of fear and greed. And since the market just is a generalized institutionalization of such relationships, the market is incompatible with communal reciprocity. Therefore ‘wrong reasons’, or

Socialism and the Market

125

reasons of the wrong kind, are those motives, nurtured by capitalism, which embody some mixture of fear and greed through the institutionalization of buying and selling. This idea, which is ubiquitous in Cohen’s writings (see Cohen 1988, pp. 243–244, 1995a, p. 199, 2008, pp. 223–225), can be further elaborated as follows. Say I have a rational, and, as it happens, true belief that I will be beaten if and only if I φ, and I also have a rational desire not to be beaten. I am then rationally required to form the intention not to φ. This (rational) intention expresses fear and is therefore the wrong kind of intention to have. ‘Wrong kind’ here implies an injunction to eliminate the conditions in which people are constrained to act in such a way that their actions and/or intentions express fear, deference, greed etc. to other people. There are two ways to elaborate this claim further: one is teleological and the other is deontological. On the teleological reading, not acting for the wrong (kinds of) reasons is a necessary condition for human flourishing. Acting systematically out of fear of others, or out of seeing them as possible sources of enrichment, makes the good life, living well, impossible. On the deontological reading, acting for the wrong kinds of reasons fails to satisfy a necessary condition for autonomy. Cohen never explicitly affirms either of these readings. But he does commit himself to an account of personal autonomy in discussion of the distinction between free action and acting freely: I say that it is usually, not always, true that when I am forced to do something I do it unfreely, because I am inclined to accept something like Gerald Dworkin’s claim that ‘A does X freely if and only if A does X for reasons he doesn’t mind acting from’, and on that view some forced action is freely performed … But that is an unusual case. In the standard case forced action is performed unfreely, even though the agent was, because he must have been, free to perform the action he performed unfreely. What a person is free to do is a matter of his situation. Whether or not he does what he does freely is a (complex) matter of his mental state. (Cohen 1988, p. 243)

One finds in this passage rudiments of Cohen’s (2008) discussion of prostitution. For it is natural to think that ‘acting freely’ in that discussion is to be interpreted along the lines of the personal autonomy model just rehearsed.17 A caveat must be registered here, however. On Cohen’s model, all of the following possibilities are conceptually open:

126

(1) (2) (3) (4)

The Political Philosophy of G. A. Cohen

A is forced to φ and A φs freely. A is forced to φ and A φs unfreely. A is not forced to φ and A φs freely. A is not forced to φ and A φs unfreely.

Those who object to rape merely on the grounds that it constitutes forceful exertion or a violation of rights (e.g. to self-ownership) cannot object to prostitution on the ground that sex is given unfreely. They are, therefore, unable to discriminate morally between cases (3) and (4) or, indeed, between (1) and (2). On the mooted interpretation of Cohen, prostitution is ‘deplorable’ (Cohen 2008, p. 224) because it instantiates cases (1), (2) or (4). This is important for the following reason. Say Smith has reasonable alternatives to selling her labour power as a teacher, such that she is not strictly forced to sell her labour. Smith does, however, (have good reason to) fear the loss in social standing among peers if she leaves her job, or the stigma attached to unemployment or the bank’s reaction upon falling back on her mortgage, etc. In this case, Smith’s predicament is best described by (4) and is analogous to that of an unforced prostitute. One can fail to act freely, or indeed be dominated by others, without being forced to be or do anything.18 If we take Cohen’s discussion of prostitution at its word, then a hallmark of the market is that acting freely is institutionally obstructed. One group of people (namely, those on the wrong side of the market) is constrained to act out of fear or deference to others, while another group of people (namely, those on the right side of the market) is constrained to act out of greed or covetousness towards the former. This is how capitalists and workers treat one another under capitalism. But it is also, significantly, how worker-owners treat one another under market socialism. This is a further illustration of how ‘a market, even a socialist market, is a system of predation’ (Cohen 2009a, p. 82).

Planning vs. markets The appeal to autonomy is significant to Cohen’s argument against the market and, by extension, against market socialism. But Cohen stops short of spelling out what the alternative to the market is supposed to be. Indeed, he disavows a commitment to collective autonomy in any argument for socialist

Socialism and the Market

127

planning. He thus argues that the values of equality, freedom, community and democracy are central to socialist theory and practice but criticizes the ‘Hegelian legacy’ which compels Marxists to see planning as good in itself, because it imparts ‘self-knowledge’19 or ‘conscious social purpose’: Unlike collective self-direction, democracy is good in itself. The case for it, if one is a principled democrat, is not exhausted by the claim that it produces better results. One can believe that, if A was a bad decision, and B would have been a better one, it says something for A that it was taken democratically. By contrast, I would not say about bad things caused by planning: well, at least they were planned. (Cohen 1995a, p. 261)

In other words, there is the undemocratic (because not democratically planned) market allocation; there is the democratic (because democratically planned) nonmarket allocation; and there is nothing worthwhile standing midway between the two. It follows that, if planning is desirable, then that is just because it is democratic. Call this the exhaustiveness claim. This claim raises a formidable challenge, one whose force is rarely appreciated – let alone met – by socialists who defend planning. The most common response among those socialists is to bite the bullet, by saying that planning is a necessary condition for ‘genuine’ or ‘complete’ democracy, and this is all they have been saying all along. This response grants too much to Cohen’s challenge. For it implies that Marx, Engels and the numerous scholars behind the Marxist critique of political economy were fools who either forgot to append the term ‘democracy’ to their most important writings (the term does not appear once in Marx’s Capital, for example) or did not know what they were talking about.20 But Cohen’s challenge can be met. Indeed, it can probably be met on Cohen’s own terms. To do this, I must be allowed a digression through the intricacies of sexual desire. According to Thomas Nagel, sexual relations are successful if and only if they involve ‘some version of reciprocal interpersonal sexual awareness’ (Nagel 1991, p. 49) and ‘perverted’ otherwise. What is reciprocal awareness? Consider a case where A and B occupy seats on adjacent sides of the main train isle. They can see each other’s reflection in the window. A sees B’s reflection and comes to desire B sexually. B also, quite independently, sees A’s reflection and comes to desire A sexually. At this stage neither A nor B is aware of each other’s

128

The Political Philosophy of G. A. Cohen

perceptions. But then A realizes that B is looking at him, indeed that B desires him. Nagel remarks: ‘this is definitely a new development, for it gives him a sense of embodiment not only through his own reactions but through the eyes and reactions of another’ (Nagel 1991, p. 45). Then B also becomes aware that A sexually desires her. A reciprocal awareness of sexual desire is achieved. Nor is this the end: A may then become aware that B is aware that A sexually desires her. This adds to his arousal, ‘for he becomes conscious of his sexuality through his awareness of its effect on her and of her awareness that this effect is due to him’ (Nagel 1991, p. 46). More schematically: (5) (6) (7) (8) (9) (10)

A desires B. B desires A. A is aware, and welcomes, that (6) obtains. B is aware, and welcomes, that (5) obtains. A is aware, and welcomes, that (8) obtains. B is aware, and welcomes, that (7) obtains.

Reciprocal awareness obtains when all of (5) to (8) are true. And when all of (5) to (10) obtain, says Nagel, things are going well.21 Indeed, what we have then seems to entail a local advancement of communal reciprocity. For, in a sense, I serve you while also being served by you, ‘and I get satisfaction from each side of that equation’ (Cohen 2009a, p. 41). Inversely, prostitution is a case where things are not going well. That is, whether or not sex with prostitutes is ‘perverted’, it will typically fail to realize reciprocal awareness, since at least one of (5) to (8) is likely to be false. This is, at a minimum, a blow to community. Indeed, it is sometimes more than that: Prostitutes become victims of male hatred and rage, but we have already said that pure lust can be contaminated by things a lot more impure. The reality principle comes back, and the client realizes that what he really desired – [reciprocal awareness – NV] – cannot be bought and has not been delivered. And the resulting deflation, especially when overlaid by … cultural baggage … that is, with self-hatred, disgust, guilt and shame, may prove dangerous for anyone in the vicinity. (Blackburn 2004, p. 110)

What does all this have to do with socialist planning? One way to refute Cohen’s exhaustiveness claim consists in pointing out a set of valuable

Socialism and the Market

129

democracy-independent features that make planning desirable. I claim, first, that the salient pro-planning feature is the generalized institutionalization of reciprocal awareness of fellowship, and, second, that such institutionalization is deemed valuable by Cohen himself. Before defending the first claim, I shall try to vindicate the second. In Chapter 5 (see especially section ‘Marx and Cohen on community’), I discussed Cohen’s jazz-band model, ‘one way of picturing life under communism as Marx conceived it … [as] a concert of mutually supporting selffulfilments’ (Cohen 1995a, pp. 122–123). In seeking to satisfy his own desire for music-playing, which has the same object as the saxophonist’s, the pianist comes to fulfil herself, the saxophonist, and the audience around her. The same is true, mutatis mutandis, of the saxophonist. Successful music-making and successful love-making are therefore structurally similar in presupposing reciprocal awareness. Conversely, when reciprocal awareness of fellowship obtains, the joint action of agents is motivated not by fear, deference, shame, greed or covetousness but by reasons of the right kind. Cohen introduces his own notion of community along just these lines: The requirement of community that is central here is that people care about, and, where necessary and possible, care for, one another, and, too, care that they care about one another. (Cohen 2009a, pp. 34–35)

The ‘care that they care’ part can be analysed as follows: A cares that [(A cares about B) and (B cares about A)] and B cares that [(A cares about B) and (B cares about A)]. The basic idea can be put very succinctly: I can’t be your friend without both you and I caring that I care about you and that you care about me. This is just a more general version of (7) and (8) above. It follows that Cohen is committed to what may be called generalized reciprocal awareness of fellowship (see also Cohen 2008, p. 43).22 An ardent reader of Cohen might object that Cohen does not commit himself to the jazz-band model of community and therefore does not commit himself to reciprocal awareness. He does not do so because he thinks that model presupposes an overoptimistic ‘technological fix’ on Marx’s part: it is ‘lofty material endowment’ that makes the concert of mutually supporting self-fulfilments generalizable (I discuss this objection in Chapter 5). Whatever one thinks about the mooted presupposition,23 it does not follow that Cohen repudiates the values embodied in the jazz band model.

130

The Political Philosophy of G. A. Cohen

Indeed, the second-order form of care posited in the ‘central requirement of community’ above offers evidence that Cohen is committed to some sort of reciprocal awareness of fellowship. Such awareness seems to be a necessary condition for the sort of communal reciprocity realized in Cohen’s successful camping trip (Cohen 2009a, pp. 44–45). None of this establishes a strong presumption in favour of socialist planning, and none of it strictly refutes Cohen’s claim that the case for planning is exhausted in the case for democracy. A fuller repudiation requires a disquisition whole to itself. Instead of that, I shall outline some reasons, grounded on the rudiments just provided, to think the exhaustiveness claim false. We have seen how Cohen argued that all markets, even socialist markets, are morally tainted because they embody the ‘universal prostitution of mankind’. One way to interpret this claim is community-grounded: it says that markets make it impossible for people to revel in knowledge that the activities in which they are jointly engaged are done for their own sake, or, more generally, for the right kinds of features. What is more, a mutual expectation, and awareness, of a general commitment to one another is necessarily absent from market society. What we get, instead, is ‘typically some mixture of greed and fear’ (Cohen 2009a, p. 40), a mixture that taints even the ‘Platonic ideal’ of market socialism. This leads to a generalization of the kind of frustration Simon Blackburn describes for the case of prostitution: deflation, emptiness and a persistent failure to ‘assert one’s true individuality’ (Marx and Engels 1956, p. 176) in community with others.24 In the camping trip model defended by Cohen (2009a), participants relate to one another non-antagonistically – at least insofar as production-related activities are concerned – and are mutually aware of this way of relating. This awareness is expressed in their joint production activities – fishing, cooking, building, etc. – through a collectively drawn-up to-do list (a plan), which is amenable to adjustment depending on exigency of need and circumstance. Were money to be introduced as mediator between the products of these activities (thereby establishing a market), the upshot would be not just inefficiency (see Cohen 2009a, pp. 6–7) but also an obfuscation of that awareness. Markets systematically impugn reciprocal awareness of fellowship (in addition to reintroducing antagonism itself), thereby undermining a key aspect of community that Cohen is committed to.25 True, Cohen never commits

Socialism and the Market

131

himself explicitly to this interpretation. But it seems to follow from his account of community and has the benefit of explaining his life-long commitment to some form of planning: [I]sn’t this, the socialist way, with collective property and planned mutual giving, rather obviously the best way to run a camping trip, whether or not you actually like camping? (Cohen 2009a, p. 10, emphasis added)

Since the egalitarian distribution that Cohen wants is realizable without planning (that is, after all, the crowning achievement of market socialism, if and when realized), and since democracy features nowhere in the vicinity of this argument for ‘the socialist way’, there must be some intimate conceptual link between the ‘mutual giving’ mandated by communal reciprocity and its being planned, that makes planning desirable. The link is provided by the reciprocal awareness of human fellowship, which is a necessary part of communal reciprocity.26 This is, I submit, the only explanation available in Cohen’s conceptual armoury for his commitment to socialist planning (for democracy-independent reasons). One might now object to the camping trip analogy on the grounds that it makes a demanding form of community with others inescapable, and therefore oppressive. What the market allows and planning disallows, the objection goes, is people adjusting their individual life plans on the basis of different preferences for consumption, work and leisure. This objection overextends the scope of Cohen’s camping trip analogy. As I understand that analogy, Cohen is not saying that introverted people should join hands and ‘kumbaya’ with others, whether they like it or not. Rather, for as long as humanity needs to toil in order to reproduce the necessities of life (summed up in the fishing, cooking and building activities of the campers), the division of labour will bind its members together at the point of production. This is what Marx (1981, p. 959) calls the ‘realm of necessity’. Is it not better, within the ambit of this realm, that productive activity be organized along the lines of an egalitarian camping trip than of a capitalist workhouse? The exigencies of necessity met, the campers enter into a ‘realm of freedom’, a realm, that is, of free and independent spontaneous activity. On this restricted interpretation, the camping trip offers a more humanized vision of the realm of necessity, not a totalizing account of life in socialist society.27

132

The Political Philosophy of G. A. Cohen

Coda In the 1982 edition of his Capitalism and Freedom, Milton Friedman writes: This, I believe, is our basic function: to develop alternatives to existing policies, to keep them alive and available until the politically impossible becomes politically inevitable. (Friedman 2002, p. xiv)

‘Our’ of course refers to the scions of neo-liberalism, those pro-capitalist ideologues who draw their social science from neoclassical economics and their moral philosophy from right-libertarianism. Since the 1980s, this postmodern cocktail has, quite literally, proven explosive: the combination of capital market liberalization, aggressive privatization, wage austerity and welfare-state demolition – the process of ‘accumulation through dispossession’, as Harvey (2012) succinctly puts it – has unleashed a series of economic crises culminating in the recent 2008 recession (see Brenner 2006, Krugman 2009, Callinicos 2010 and for sapient discussion). At the same time, income inequality has risen to new heights: in the United States about 10 per cent of the population collects half of all national income, while the 1 per cent collects over a fifth. Income inequality in the United Kingdom, Germany and France is about 25 per cent higher than it was in the mid-1980s (see, for example, OECD 2012). These developments undermine fairness, diminish freedom28 and deny community to all. Given these developments, socialists are today charged with the mission Friedman promulgated – albeit for a different audience – back in 1982. In the midst of a long-standing impasse within the ranks of British socialism, Cohen (1995a) admonishes leftists along exactly these lines. He laments the leftist willingness to make theoretical compromises in the interest of ostensibly leftist practice (and commends Marx and Engels who did not oppose ‘partial transformations of capitalism’ but did oppose ‘restricting political activity to a “war against the effects of the existing system, instead of simultaneously trying to change it” ’ (Cohen 2000b, p. 72)). Echoing Friedman, he argues that a reason for the success of right-wing ideology during the 1980s consisted precisely in the ‘uncompromisingly fundamental’ views put forward by theorists such as Nozick and Hayek:

Socialism and the Market

133

Politicians and activists can press not-so-crazy right-wing proposals with conviction because they have the strength of conviction that depends upon depth of conviction, and depth comes from a theory that is too fundamental to be practicable in a direct sense. (Cohen 2011, pp. 212–213)

This is one way in which Cohen’s defence of (fact-independent) principles might turn out to be less utopian than the thought of those self-styled nonutopians who dilute fundamental principle in light of the demands of so-called non-ideal theory (see the section ‘Metaethics’ of the Introduction). According to Cohen, the left’s ideological appeal can return only when leftists realize that: [F]undamental socialist values which point to a form of society a hundred miles from the horizon of present possibility are needed to defend every half-mile of territory gained and to regain each bit that has been lost. (Cohen 2011, p. 213)

Cohen’s normative political philosophy is an imaginative and honest attempt to depict what stands a hundred miles from the horizon of present possibility – no longer in certainty nurtured by belief in its pre-ordained arrival but in resolute adherence to its desirability, established through a set of scientifically respectable methodological and ethical commitments. In a milieu of historical non-inevitability, it is the task of socialists to defend every half mile of socialist territory gained as well as to fight for every half-mile lost. In this task they can draw concepts, arguments and inspiration from Cohen’s defence of socialism, his political theory and its practice.29 And they can even draw confidence and guarded optimism from his lifelong conviction in the desirability and feasibility of full human emancipation: There’s actually much less inequality now than there was, say, a hundred years ago. Then, only a few radicals proposed that everyone should have the vote. Others thought that was a dangerous idea, and most would have considered it to be an unrealistic one. Yet today we have the vote. We are a political democracy. But we’re not an economic democracy. We don’t share our material resources, and most people in this country would regard that as an unrealistic idea. Yet I think it’s an idea whose time will come. Society won’t always be divided into those who control its resources and those who have only their own labour to sell. But it’ll take a lot of thought to work out

134

The Political Philosophy of G. A. Cohen

the design of a democratic economic order, and it’ll take a lot of struggle, against privilege and power, to bring it about. The obstacles to economic democracy are considerable. But just as no one, now, would defend slavery, I believe that a day will come when no one will be able to defend a form of society in which a minority profit from the possession of the majority. (Cohen 1986)

Notes

Introduction 1

See Cohen (2000b) and Wolff (2014).

2

See, for discussion, Tooley (2002).

3

The notion of ‘species being’ comes from Marx, whose secular humanism looms large in Cohen’s writings throughout his career. I take it up in Chapters 1 and 5.

4

This is the subject of Chapter 5.

5

‘Merely’ in the sense that the latter theory purports to go through in the absence of ethical or axiological commitments, and ‘predicts’ because it purports to prospectively map humanity’s future on the basis of social-historical laws (see the introductory section to Chapter 2).

6

It is no coincidence that there is no single identifiable work standing out as Cohen’s own major contribution to philosophy: none of his monographs have the programmatic ambitions of Kant’s Groundwork, Marx’s Capital, or Rawls’ Theory. To put it in the terms set out by Cohen’s teacher, Isaiah Berlin: Cohen was the archetypal philosophical fox.

7

For illustrations of this strategy, see the section ‘Cohen against Berlin’ of Chapter 2, sections ‘Patterns of justice’ and ‘Where the action is’ of Chapter 3, and section ‘Against liberalism’ of Chapter 5.

8

Cohen thought of himself as a ‘reactive thinker’. See Wolff (2014, p. 342).

9

Cohen’s moral intuitionism is a form of foundationalism. Foundationalism is the view that there exist certain fundamental, or basic, beliefs that are neither justified by, nor inferred from, other beliefs. This theory went out of fashion in philosophy in the 1960s, and has remained out ever since.

10 In fact ‘intuitionism’ and ‘radical pluralism’ are not equivalent. The former is a method for discerning moral truth, whereas the latter is a theory about the nature of values (and the inscrutability of trade-offs). See the section ‘The role and significance of political philosophy’ for further discussion. 11 P1 may also be sensitive to fact F1, in which case we need to know what explains that sensitivity. And this ensures that there is a further principle P2 which does

Notes

136

the explaining, and so on. Cohen claims that this justification-regress cannot go on forever, since ‘an unending sequence of justifications would run against the requirement … that she who affirms P has a clear grasp of what her principles are and why she holds them’ (Cohen 2008, p. 237). He infers that all fact-sensitive principles are grounded on fact-insensitive principles. 12 Moral realism is the thesis that moral statements can be true or false, and they are made true by a class of moral facts. Cohen was indeed a moral realist, but he took his realism to be independent of his Platonism (Cohen 2008, pp. 257–258). 13 A further caveat might be worth registering here: some philosophers take Cohen’s fact-insensitivity thesis to be a denial of moral naturalism, the idea that moral properties are natural properties. This is a non-sequitur: there is no reason to think that fact-insensitive moral properties can’t be natural properties. 14 Indeed, on Cohen’s view it is historically necessary (see Chapter 1). Not all Marxists need affirm this stronger claim. 15 There is a tension here between what would-be revolutionists can do and what their principled commitments imply that they ought to do. I take this up briefly in Vrousalis (2011b). 16 See the section ‘The structure of justice’ of Chapter 3 for Cohen’s attempt to remedy this confusion. 17 See, for example, Feltham (2009) and Miller (2013). 18 To the extent that we are concerned with texts and their meanings, Frankfurt’s argument amounts to an intentional fallacy (in the sense of Wimsatt and Beardsley 1946): when engaged in textual interpretation, we seek to understand the text itself, not the extra-textual state of mind of its author. 19 He says that Cohen-bullshit is ‘just one flower in the lush garden of bullshit’ (Cohen 2013, p. 96), thereby disputing Frankfurt’s ascription of an ‘essence’ to bullshit. 20 This definition sits well with Cohen’s lifelong habit of taking arguments apart through precisification and analysis. 21 The negation test furnishes no necessary condition either: the statement ‘John is very tall’ is unclear, but its negation changes the force of the statement in salient contexts. 22 For defences of unclarity de re from different sides of the epistemicism/ supervaluationism divide, see Williamson (1994) and Tye (1990). 23 In the work of Alain Badiou, for example, ‘nothing’ and ‘the void’ are systematically identified with the empty set (see, for example, Badiou 2005). But the empty set is not nothing, and it does not represent nothingness.

Notes

137

Badiou’s argument presents itself as interesting truth. By dint of that selfrepresentation, and in virtue of obvious incoherence, it is merely bullshit. I do not know if Cohen would endorse my characterization of Badiou’s argument (but see his take on what may be called the epidemiology of bullshit, in connection with the French academy, in Cohen 2013, pp. 108–114). 24 Cohen claims that such things include future existents. This gives rise to a timeinconsistency problem. Let Vx stand for the value of x, and Vy for the value of y. From the point of view of the present, where only x is in existence, the replacement of x by y is acceptable if and only if Vx – Vy < ε, ε > 0, where ε is the ‘particularity mark-up’ on x’s value in virtue of its existence in the present. But from the point of view of the future, in which y is also in existence, the replacement of x by y is acceptable if and only if Vx < Vy, since y carries its own mark-up in particular value. These acceptability conditions are incompatible. See Bader (2013) for discussion. 25 There is, I think, an intimate connection between Cohen’s understanding of subject–object unity and his acceptance of ‘agent-centred prerogatives’ to depart from the demands of equality, in that the former presupposes the latter (see the section ‘Where the action is’ of Chapter 3). The connection between the humanization of value through personal and particular valuings, on the one hand, and said prerogatives, on the other, needs more attention that I can give it here.

Chapter 1 1

One of Cohen’s earliest published discussions of materialism offers a critique of Marcuse’s One Dimensional Man. Here Cohen attributes to Marx the vision of a ‘conquest of and reunion with nature, which was once so hostile that men had to separate themselves from it and wage a long technological war that reproduced itself in battles between and inside men’ (Cohen 1969, p. 35).

2

The Second International was the world’s largest and most powerful federation of socialist parties before the First World War. It was founded in the 1880s with the active support of Friedrich Engels. Its chief theoreticians (Karl Kautsky, Eduard Bernstein, August Bebel and Franz Mehring) were all associated with the Social Democratic Party of Germany.

3

It is remarkable that this set of distinctions scarcely reappears in Marx’s later work. Excessive reliance on the Preface undermines Cohen’s belief that Marx

Notes

138

actually held the theory of history ascribed to him, or indeed any theory of history. I take this up briefly in the section ‘Is historical materialism dead?’ below. 4

‘Non-normative’ does not here entail ‘necessarily not based on social norms’. It only entails ‘not necessarily based on norms’ (see Cohen 1988, pp. 35–36).

5

Historical materialism cannot get off the ground if its categories are not developed with sufficient clarity and rigour, says Cohen, and people tend to ‘reject the theory because they apply these standards not too severely, but not severely enough’ (Cohen 1970, p. 121). For Cohen’s commitment to the attainment of clarity through analysis, see the section on bullshit in the Introduction.

6

Note that the individuation of each epoch rests entirely on the level of the material surplus (that is, the mass of use-values over and above what is necessary for universal subsistence). The nature of control over the surplus matters only indirectly, that is, insofar as it impinges on the level of development of the productive forces.

7

The Marxian distinction between ‘abstract’ and ‘concrete’ labour plays no role in Cohen’s discussion, in part because he wants his argument to remain independent of the labour theory of value, which this distinction seems to presuppose. For discussion of Cohen’s critique of the labour theory of value, see the section ‘The labour theory of value’ of Chapter 4.

8

Cohen eschewed the term ‘functional’, referring instead to ‘consequence explanations’. One reason for this may have been his insistence that one cannot infer from: ‘event A is functional for B’ that: ‘A functionally explains B’. Marxists often observe a social phenomenon functional for capitalist development (for example, clever young people recruited into ‘think tanks’ doing apologetics for capitalism), and infer that the relevant explanans is capitalist economic structure. This is too quick, says Cohen (see his prophet example in Cohen 1978, p. 281).

9

It is worth noting that Althusser’s (2011) lamentation of ‘cet hégélianisme’ extant in Marx’s (1977) distinction between form and content implicitly vindicates Cohen’s interpretation of Marx.

10 For the methodological significance of the non-bullshit (not: no-bullshit) commitment, see section on bullshit in the Introduction. 11 For a famous argument that it very likely would not, see Nozick (1974, pp. 274–275). 12 In this connection, David Harvey (1986, p. 686) claims that Cohen’s book defends ‘Stalin’s interpretation of Marx’s theory of history’. This is a cheap shot.

Notes

139

For anything remotely recognizable as ‘Stalin’s interpretation’ can be found in the works of Lenin (1974, pp. 43–91) and Bukharin (1925), who preserved – almost unadulterated – what they learned in that domain from Kautsky, Plekhanov and the tradition of the Second International (see, for example, Lih 2011). Harvey inadvertedly dismisses thirty years of Marxist scholarship in one frivolous jibe. 13 Charles Mills (1989) offers near-exhaustive evidence that Marx does not draw the distinction the way Cohen does (see also Rosdolsky 1977). This undermines only the exegetical ambitions of Cohen’s argument. 14 Which is not to deny that there are such properties. I borrow the golf ball analogy from Williams (1995, p. 84), who puts it to different use. 15 ‘Hand-in-glove’ because the social assumes the form material content gives it. 16 This may be where it differs from Althusser’s (1969, Althusser and Balibar 2009) structuralism of ‘overdetermination’ and ‘structures in dominance’. For although there are important similarities between the two views, Althusser and his followers assigned explanatory primacy ‘in the last instance’ to the economy (whatever is meant by ‘instance’ and ‘economy’). 17 Finlay (1959), for example, remarks that slavery in Ancient Athens gets its significance not from the proportion of slaves in the Athenian population, but from the concentration of slave-ownership in the hands of a few landowners. Such base-heavy elaborations make it easier to explain the persistence of slavery and rid Marxists of the difficult task of tracking that persistence down to productive growth. 18 See Hobsbawm (1999, pp. 15–16). 19 Cohen’s work up to the mid-1970s was heavily informed by logical positivism and positivist philosophy of science. It is no coincidence that Cohen’s (1978) explanatory model takes its cue explicitly from Hempel’s (1966) deductivenomological model. 20 As C. J. Arthur (2004) has tirelessly pointed out, moreover, Marx’s categories in Capital are not historical categories. 21 One of Marx’s innovations in Capital vol. I is the idea of abstract labour. Cohen, and other analytical Marxists, associated this idea with the labour theory of value (which they took to be a theory of equilibrium prices asserted by Marx) and dismissed it as a useless relic of Ricardian economics (see Chapter 4). But the abstract labour gambit is, pace Cohen, an ethically loaded gambit, indeed one that is central to the critique of political economy. It follows that ethics is central to the critique of political economy. 22 Cohen published one of his celebrated papers on Nozick in 1977 (Cohen 1977).

Notes

140

Chapter 2 1

Section ‘Problems with the argument for primacy’ of Chapter 1 called these the agency and scarcity problems.

2

Berlin is aware of the thesis that Cohen defends (namely, that lack of money reduces freedom), but it is not clear that he rejects it. See, for example, Berlin (2002, p. 53).

3

More precisely: under market capitalism, the private production of goods or services for sale (commodities) and the social ascription of freedom become existentially interdependent through the mediation of money. Some might object that the 1 per cent deserves its share. But this consideration is extrinsic to the argument from freedom. For discussion of the moot desert claim, see the section ‘Market socialism and its criticism’ of Chapter 6. For extensive empirical refutation, see Piketty (2014, chapter 9).

4

For the exact meaning of ‘dispersal’ and ‘inequality’, see Chapter 3. In Friedman’s discussion, the object of the ‘independent foci’ is both ‘political power’ and ‘civil freedom’. In both the power and the freedom case, the pro-dispersal premiss tells in favour of (market) socialism, and against capitalism.

5

Cohen points out that any objection to the effect that pavement users are not really free because their users necessarily lack control over them ‘treats freedom fetishistically … [F]reedom, in the central sense of the term with which we have been occupied, is freedom to act, and if there is a concept of full freedom in that central sense, then it is inappropriate, if we want to identify it, to focus, from the start, on control over things’ (Cohen 2011, p. 156).

6

Unlike pavement-commuting, pre-privatization train travel cost commuters money. But, unlike public trains, private train fares must include a profit markup. At comparable levels of efficiency, therefore, private trains are likely to be more expensive than public trains, and therefore more restrictive of freedom for the poor. Of course, freedom is not the only casualty of privatization. For discussion of the disastrous short-term consequences of rail privatization in Britain, and its long-term legacy, see Bowman et al. (2013).

7

Roughly: relations of effective power arising from the distribution of control over productive forces and labour power – see the section ‘Definitions’ of Chapter 1.

8

For a discussion of Cohen’s distinction between being free to act and acting freely, see the section ‘Commodification and personal autonomy’ of Chapter 6.

9

Collective unfreedom eo ipso generates individual unfreedom: as long as there is only one key, A is free to exit if, and only if, B does not.

Notes

141

10 Cohen produced these examples in the 1970s. Making them realistic would require multiplication by a factor of at least a hundred. I have kept the sums intact for consistency. 11 I emphasize ‘his’ because Locke’s proviso seems more demanding than Nozick’s (interpretation of the former): Locke offers an analogy with a river, where one’s drinking does not leave any less, or less good, for other potential drinkers. Land was abundant at Locke’s time, but that is no longer the case (see Cohen 1995a, p. 77). 12 For non-proprietarian readings of Locke, see Waldron (1990) and Fried (1998). 13 The sweater example is due to Otsuka (2005, pp. 17–19). Cohen notes the mooted discrepancy on a number of occasions. It is brought out most starkly in his example of a world in which people ‘float in space, and all necessary and luxury services take the form of other people touching [each other] in various ways … It is far more difficult to object to inequality in this world than it is to object to [inequality traceable back to – NV] unequal division of non-human resources’ (Cohen 1995a, p. 170). 14 For Cohen’s contribution to egalitarianism, see Chapter 3. 15 Note that the kind of joint ownership contemplated by socialists is significantly weaker than the strict joint ownership conjured by Cohen in the midst of polemic with Nozick (see the section ‘ … on self-ownership’ above). 16 I have argued for the conceptual integrity of this notion in Vrousalis (2011a).

Chapter 3 1

Cohen maintained that Marxists are committed to moral condemnation of capitalism as unjust because that is the only way (i) to make sense of central claims within Marxian and Marxist political economy (such as the ‘theft’ of labour time under conditions of ‘fair’ capitalist exchange), (ii) to make sense of the conviction and passion Marxists invest in questions of equality, (iii) to explain Marxist participation in political struggles when each can harmlessly stay at home (since the contribution of each makes little, if any, difference to the outcome of these struggles). See Cohen (1983b, pp. 442–445, 1995a, pp. 139, 195).

2

There are numerous excellent introductions to this debate. See, for example, Kymlicka (2001), Swift (2006) and White (2006).

Notes

142 3

This is sometimes called a ‘second-price’, or ‘Vickrey’ auction (although ‘Sotheby’s model of distribution’ might sound more apt in this case). Dworkin never questions some underlying assumptions of signal importance, such as the existence and stability of the putative market equilibrium, or indeed why such equilibrium will necessarily reflect opportunity costs.

4

In the first few chapters of Sovereign Virtue, Dworkin argues that some form of market mechanism is somehow constitutive of justice. It does not follow that justice requires markets. But as we read on, Dworkin (2000, pp. 173, 317, 492 and passim) relies more and more on the market for the institutionalization (or ‘practice’) of equality, as opposed to its merely theoretical characterization.

5

There are two possible specifications of luck egalitarianism that have a bearing on the question of inequality between the prudent and the imprudent. On a weak specification, such inequalities are deemed acceptable, but not required, at the bar of justice. On the strong interpretation they are required. Much of the recent debate on luck egalitarianism consists in finding an acceptable middleway formulation between these two. I discuss the ‘undeserving poor’ problem more extensively below.

6

I assume that actions are individuated with reference to volitions, not intentions, and that it therefore makes sense to ascribe actions when they are voluntary but unintentional.

7

See, for example, Griffin (1986).

8

Note that the problem of cheap preferences cannot be solved by simply stipulating that informed preferences cannot be cheap (in the relevant sense). For that would be tantamount to (arbitrarily) assuming that one can never informedly prefer, say, potatoes to salmon.

9

See, for an independent but related argument, Barry (1999).

10 Resourcism comes in two forms: commodified and decommodified. The former is exemplified by Dworkin’s (2000) egalitarianism, and the latter by the ecological footprint idea (see, for example, Rees and Wackernagel 1994). Cohen never takes up the latter, so I will not discuss it. 11 Insofar as the auction involves actual preferences, Dworkin’s account is also subject to the objections of the two preceding paragraphs. 12 These criticisms do not afflict resourcism alone, but also the Rawlsian focus on primary goods. 13 See Dworkin (2000) for a response to Sen, and Williams (2002) for a convincing Senian rejoinder.

Notes

143

14 Indeed, Cohen notes that Sen’s ‘equality of what?’ question loads the dice in this direction. For it makes it look as if there exists some unique and simple desideratum that will allay all our axiological worries about the metric of justice. See also Wiggins (1998, p. 136). 15 ‘Mysteriously’ because Cohenite advantage has striking similarities with the mixed prudential bag that is commonly taken to constitute human flourishing. One reason for resisting the latter term may have been its strongly teleological ring. 16 Sen could, of course, restructure capability propositions so as to include references to welfare, functionings and goods (as in ‘A should have C’, where C is ‘capability to function without pain’). But in doing so he would be making a concession to Cohen, not just offering a trivial reformulation of his view. For the relevant metric would then be: capability set C-cum-functioning set F-cumwelfare set W-cum- … This is just what Cohenite advantage consists in. 17 Although perhaps more acute for Cohen, they equally confront Sen and sophisticated welfarists, such as Richard Arneson (see Arneson 1999). 18 The war experience itself might further vindicate Cohen’s argument for the feasibility of equality. During the Second World War, the UK systematically devoted more than half its national income to the war effort and less than half to personal consumption (see Harrison 2000). This sufficed to satisfy the basic needs of the entire population for the whole duration of the war. And all that at levels of scientific and technical knowledge that pale in comparison with present capabilities. 19 And so does his acceptance of agent-centred prerogatives to depart from equality. See Cohen (2008, pp. 70–72). 20 A distribution X is a Pareto improvement over Y if and only if X makes at least someone better off and no one worse off compared to Y. The Pareto principle always mandates Pareto improvements, that is, movements from X to Y when X is Pareto superior to Y (and vice-versa). 21 On some readings of Rawls, ‘is compatible with’ must be replaced with ‘requires’. For the purposes of this exposition the former reading suffices for, indeed strengthens, Cohen’s argument. 22 Temkin’s reply might strike some as horrendous. Cohen thinks this particular reaction issues, in part, from insufficient appreciation of the parallels between ‘the luck of the genetic draw’ and the luck entailed by state-backed inequality in the distribution of eyes. See his discussion of the ocular tree in Cohen (1995a, pp. 243–244).

Notes

144

23 For this recurring pluralist motif, see Cohen (1989a, p. 908, 1995a, pp. 25, 257, 2008, pp. 274–279, 315–320). It is an interesting question to what extent Cohen is influenced in this pluralism by his teacher – and arch-pluralist – Isaiah Berlin. 24 Say A works for 5 hours, producing 5 widgets, and B works 10 hours, producing 10. It is plausible to assume that A non-comparatively deserves 5 and B noncomparatively deserves 10. Suppose the wind destroys 3 of A’s widgets and 6 of B’s. Then A gets 2 and B gets 4, which upsets the non-comparative, while preserving the comparative, aspect of desert (see Cohen 1989b). 25 The passage implies that distributive justice includes desert. This conclusion is contradicted by other things that Cohen says (perhaps with non-comparative desert in mind). Nothing of significance hinges on inclusion, or non-inclusion, of desert under distributive justice. 26 For Cohen’s distinction between material and spiritual properties, see the section ‘Is historical materialism dead?᾿ of Chapter 1. 27 This dilemma is not, strictly, exhaustive. For Cohen could keep his Marxism and his commitment to the ideal of equality of advantage, but drop the claim that such equality is feasible. This gambit is worthwhile only if one thinks that understanding certain (true) ideals matters independently of whether they can be realized. And Cohen does think this: see the section ‘The role and significance of political philosophy’ of the Introduction. But pace Cohen (2008), it is not clear that Cohen lost confidence in the feasibility of equality (see the concluding section of Chapter 6 and the last sentence of Cohen 2009a).

Chapter 4 1

By ‘etiology’ of some entity or property, I understand its causes or the causal structure that gives rise to it. By ‘individuation’ I mean the giving of necessary and sufficient conditions for an entity or property being what it is.

2

Capitalism is the first economic formation in human history in which the dominant mode of exploitation is not direct coercion. It does not follow that there is no coercion under capitalism: ‘force figures considerably among the immediate causes of economic power, just as rights do. Indeed, effective rights are effective partly because of the force behind them’ (Cohen 1978, p. 234). See also pp. 294–295 for the functional explanation of the state’s ‘relative autonomy’.

3

For a discussion of effective rights and their relation to effective powers, see the section ‘Definitions’ of Chapter 1.

Notes 4

145

Cohen also points out the obvious, but often overlooked, truth that capital, by definition, does not create what has value. For ‘if that claim were true, then capital and labour would not be distinct factors of production: qua capitalists, capitalists supply capital, which is not a kind of labour’ (Cohen 1988, p. 237). As Joan Robinson (1966, p. 18) succinctly put it: ‘owning capital is not a productive activity’.

5

This has numerous important implications which Cohen addresses throughout his engagement with normative political philosophy but all of which are treated more or less in abstraction from the critique of political economy. For further elaboration on the notion of ‘success’, see the section ‘Market socialism and its criticism’ of Chapter 6.

6

Cohen claims that exploitation is a form of ‘forced and unreciprocated flow’ in Cohen (1995a, p. 200). This claim directly contradicts the quotation just provided in the main text. More substantively, it is inconsistent with Cohen’s view, to the effect that proletarians are individually free to escape their proletarian predicament but not collectively free (as a class) to do so ( see the discussion of proletarian unfreedom in Chapter 2). It follows from that structural argument that individual proletarians, although exploited, are not forced into their proletarian predicament. Hence, exploitation of A does not presuppose A is forced. Cohen (1983b) explicitly repudiates force as a necessary condition for exploitation.

7

Resources, in the relevant sense, can be of two kinds: alienable (cars, factories, etc.) and inalienable (talents, capabilities, etc.). Cohen (1988) restricts his attention only to the former, whereas Cohen (1995a) explicitly enfranchises the latter (see below).

8

Cohen (1988, p. 235) explicitly claims to prefer the distribution-based argumentative strategy to the argument just rehearsed.

9

In his debate with Roemer, Cohen affirms (i): A flow F is exploitative only if it issues from a distribution D that is unjust. James Hall points out (in private communication) that (i) is consistent with (ii): some distribution D is unjust solely because of D’s tendency to produce flow F that is exploitative. Hall attributes (ii) to Cohen (1995a) and infers that Cohen rejects the distributive paradigm (since (ii) seems to contradict (7)). I disagree with the attribution of (ii) to Cohen (1995a). For (ii) contradicts the following thesis, which Cohen affirms (iii): if flow F is exploitative then D, the distribution that generates F, is unjust and D is unjust independently of (its tendency to produce) F. The first conjunct in (iii) follows from (i). The second conjunct follows from

Notes

146

Cohenite claims like the following: ‘a maldistribution of means of production is intrinsically unjust (unjust, that is, apart from its consequences) … although still derivatively unjust’ (that I, of morally secondary interest, although still of F-independent interest: Cohen (1995a, pp. 201–202, emphasis mine)). (ii) is therefore false. 10 This formulation of luck egalitarianism makes it look as if exploitation is a fairness-independent concern, such that luck egalitarianism covers only a part of justice. I shall come back to this at the end of this chapter. 11 John Roemer has struggled with the set of sufficient conditions in most of his work on exploitation (see the first five chapters in Roemer (1996)). It is interesting that, despite his original plea in favour of the distributive paradigm – explicitly contrasted with the Marxian unequal-exchange-of-labour definition – labour takes the centre-stage once more in Roemer (1996, pp. 104–114). 12 This section borrows heavily from Vrousalis (2013). 13 I am here nailing my colours in the vicinity of the accounts put forward by Wood (1995) and Goodin (1987): to exploit is to treat another’s vulnerability as an opportunity for self-enrichment. 14 This does not mean that capitalists will not ‘take care’ of their labourers: the efficient utilization of machinery mandates that sufficient care be taken. Even the most humane of capitalists, however, can indulge only in as much humanity as market competition (literally) affords her (Cohen affirms this conclusion in his remarkably succinct and touching Envoi in Cohen (2000b).). 15 See Vrousalis (2013), which defends the domination view. 16 Some will resist this characterization, for it contains no gambles. The objection retires if we assume a low, but positive, probability that Charlotte will give in to the courtship (but in fact she never does). The choice/gamble distinction can have no purchase here. 17 Notice that standard luck egalitarian responses, to the effect that ‘who you fall in love with is a matter of brute luck’ are moot in two-person worlds. Yet the exploitative element survives even in such worlds. 18 Cohen explicitly rejects the converse discriminating position, that of dubbing the unanimity cases fair and the gamble cases unfair in Cohen (2011, p. 140). Furthermore, he himself has argued that a fair (50–50) slavery gamble is unfair (in its outcomes), even when freely consented to by all parties (see Cohen 1995a, p. 49). 19 Cohen claims that we must go ‘back to the drawing board’ in this connection (Cohen 2011, p. 141).

Notes

147

20 The following passage comes from a one-page response to Otsuka (2010), written by Cohen in mid-2009 but not intended for publication. I thank Michael Otsuka for allowing me to include it here.

Chapter 5 1

For excellent recent discussion of Marx’s early work, see Leopold (2007) and Brudney (1998). By contrasting Marxian humanism and materialism I mean to assert no opposition between these two conceptual schemes nor anything like an ‘epistemological break’ in Marx’s work. On these questions, I remain agnostic for the purposes of this work.

2

Since Cohen treats ‘community’ and ‘fraternity’ as synonymous, I shall follow him here and use them interchangeably.

3

For Marx’s characterization of community under communism, see, for example, Marx and Engels (1975, vol. 3, pp. 293–306), Marx (1986, p. 32) and Marx (1973, ‘Chapter on Money’).

4

See, for example, Marx and Engels (1975, vol. 12, p. 126). For an influential argument defending Marx’s ‘dialectic of labour’ along these lines, see Cohen (1974).

5

I am here grateful to Keat (1981).

6

It lends itself to an ‘Aristotelian’, rather than ‘Kantian’, reading of Marxian communism, where the former places emphasis on individual self-realization and the latter on self-realization-independent moral obligation. To the extent that the Aristotelian reading of communism is too optimistic about material possibility, and therefore about the feasibility of the form of society it envisages, it must either be significantly revised, or wholly abandoned, in favour of the Kantian one. Cohen does think the Aristotelian reading is overoptimistic and takes the Kantian route in his normative political philosophy. See the section ‘Planning vs. markets’ of Chapter 6.

7

Cohen argues that a commitment to this sort of Star Trek superabundance also explains the ‘life’s prime want’ passage in Marx (1986, pp. 162–168). See Cohen (1995a, pp. 124–127) and Cohen (1978, pp. 204–207).

8

He deems it necessary to note that, for Marx, community is ‘a means to the independently specified goal of the development of each person’s powers’ (Cohen 1995a, p. 123, emphasis in original). To the extent that Cohen here ascribes to Marx an instrumental view of (the value of) community, Cohen’s own account

Notes

148

of community is at odds with that ascription. For, as we shall see, Cohen thought community to have intrinsic value. 9

See (Marx and Engels 1975, pp. 3, 154) for Marx’s distinction between the ‘communal’ being of political community and the ‘profane’ being of civil society.

10 I discuss the term ‘market’, and Cohen’s critique of markets, more generally, in the Chapter 6. 11 For a telling passage explaining his enamourment to Rawls’ moral philosophy, see Cohen (2008, p. 11). 12 See Pogge (2007) for an introduction to Rawls’ work. 13 Cohen’s last major work (Cohen 2008) is devoted to this dual task. See also the first three sections of the Introduction. 14 The first horn of this dilemma (between a narrow and an inclusive basic structure) has been grasped by Joshua Cohen (2001) and the second by Estlund (1998). Cohen responds to both in Cohen (2008, pp. 374–394). Andrew Williams (1998) is the only author I know who has argued that the dilemma is not exhaustive. See Cohen (2008, chapter 8) for a response to Williams. 15 By ‘the talented’, Cohen means all those ‘fortunate people’ who are ‘so positioned that, happily, for them, they do command a high salary and they can vary their productivity according to exactly how high it is’ (Cohen 2008, p. 120). 16 Reprinted as the first chapter of Cohen (2008). 17 Cohen accepts an ‘agent-centered prerogative’ of individuals to ‘pursue selfinterest to a reasonable extent’ (Cohen 2008, p. 61). 18 A note on metrics: if A gets 5 and B gets 10 units of, say, preference satisfaction, and this difference is solely due to the fact that B worked two times harder than A, then the said distribution involves no inequality in Cohen’s sense. Cohen (2008, pp. 106–107) enfranchises labour burden into his metric of distribution. 19 There are at least two possible interpretations of this test. On the metaethical interpretation, it expresses an intrinsic property of all moral claims as secondpersonal (see Anderson 2010). On the ethical interpretation, it expresses a substantive moral requirement. According to the ethical reading, if P is patently motivated by, say, vileness, and/or has committed vile acts, then P cannot (always) justifiably or excusably condemn others of vileness, even if what P says is substantively true. See Cohen (2013, pp. 115–133) for development of the substantive reading. 20 Though not sufficient: see the ‘Martian’ passage in Cohen (2008, p. 44). 21 ‘Poignant’, because Rawls invokes the image of the family to explain how community is related to the difference principle.

Notes

149

22 He grants, of course, that the behaviour of the kidnapper and the behaviour of the talented egoist are disanalogous in many respects (Cohen 2008, p. 41). But there is one respect, he thinks, in which they are analogous, namely in encroaching upon justice and community. 23 Of all the discussions of Cohen’s argument I know, only Richard Miller (2010) has noticed that the justice grounds and the community grounds for intentioninclusion are distinct. The reason why many commentators fail to distinguish between them may be that Cohen himself is vague as to the exact relationship between justice and community. 24 Cohen did not advocate maximin as the fundamental principle of (distributive) justice (see Chapter 3). Anderson (2010, pp. 5–6) is, however, mistaken in her judgement that Cohen’s fraternity critique of incentives requires a luck egalitarian premiss, that is, a premiss embodying Cohen’s own theory of fairness. For, that critique is immanent – that is, it goes through on the assumption that justice is compatible with, or requires, some form of maximin. Cohen’s method of immanent critique is here in full swing. 25 Cohen refers explicitly to a ‘trade-off between fraternity and fairness’ in Cohen (2011, p. 120). 26 There is some textual evidence that Cohen views the lack of justificatory togetherness in the bus case and in the incentives case as relevantly similar (cp. Cohen 2008, p. 45, 2009a, p. 36). But even if I am mistaken about this, Cohen thinks that the bus case violates some norm of community. 27 In the light of all these difficulties, perhaps, Cohen is better off giving up luck egalitarianism – his own theory of justice. Cohen was, in fact, flirting with rejecting, or at least significantly revising, its scope towards the end of his life. See the section ‘Cohen’s change of mind’ of Chapter 4. 28 In addition, Cohen does not say what the relevant class of actions falling under justificatory community is. His argument against the basic structure restriction entails that this class is much broader than some liberals are inclined to think, but we are left in the dark as to how much broader. 29 He does not say how these two modes are related or, indeed, whether they are consistent. For the polemical significance of communal reciprocity in the debate on socialism, see Chapter 6. 30 By ‘community motivation’, Cohen must mean the fraternal motivational structure that excludes fear and greed, that is, something in the conceptual vicinity of communal reciprocity. 31 I further discuss this conclusion, and its institutional ramifications, in Chapter 6.

Notes

150

32 For an argument that it is, see the section ‘Planning vs. markets’ of Chapter 6. 33 I eschew the term ‘consequentialism’ because consequences can include states of mind. 34 Externalism seems to be at odds with Cohen’s claim that ‘justice in citizens was put, above, as a necessary condition of a just society’ (Cohen 2008, p. 129). Much emphasis needs to be placed on ‘was put’, since Cohen’s argument, in this context, is immanent to the Rawlsian system.

Chapter 6 1

Whether the relevant form of organization is one-person, one-vote, for example, is an open question.

2

Cohen’s rebuttal of the second premiss charges Miller with equivocating between comparative and non-comparative aspects of desert (see the section ‘The structure of justice’ of Chapter 3).

3

For a sketch of the development of Cohen’s views on exploitation, see Chapter 4.

4

For discussion of the distinction between the predicates ‘just’ and ‘socially desirable’, see Cohen (2011, pp. 225–236).

5

Elaborate models of market socialism have recently been drawn up by Roemer (1994) and Schweickart (1996, 2002). Both Roemerian socialism, which equalizes individual shares to the means of production, and Schweickartian socialism, which gives all members of workers’ cooperatives an equal say over the management of the means of production, imply (more or less) equalization of alienable resources. But what happens if, say, the more talented members of a cooperative down the road systematically make more money than others? Schweickart, unlike Cohen (and unlike Roemer), takes the inequality to be not unjust. But the resulting discrepancy in resources and power is hardly congenial to Schweickartian economic democracy. Carensian market socialism addresses that form of inequality and is, in that respect, more attractive.

6

This is, roughly, what some Marxists (not Marx) call ‘simple commodity production’.

7

What distinguishes capitalism from pre-capitalism is therefore neither private property, nor the profit motive nor the existence of capital: all of the above predate capitalism and are necessary but not sufficient for capitalist production. It is the generalization of commodity production that completes the set of sufficient conditions.

Notes

151

8

For a more dramatic description of the evils of commodification, see Marx (1976,

9

As Cohen puts it, ‘to the extent that people take their eyes off money-grabbing,

p. 799) they disturb the market’s allocative function’ (Cohen 1989b, p. 26a). 10 I discuss this in Chapter 4 above and more extensively in Vrousalis (2013). 11 She proceeds cautiously, ‘timid and holding back, like one who is bringing his own hide to market and has nothing to expect but – a tanning’ (Marx 1976, p. 280). Threats to one’s self-respect survive even if she is guaranteed to earn as much as others do or indeed to fare as well as they do. Any focus on mere distribution is bound to miss this relational dimension of concern. 12 This question was put to me by Hillel Steiner in the course of my presentation of this objection. 13 ‘The market, one might say, is a casino from which it is difficult to escape…’ (Cohen 2009a, p. 33). The casino, one might add, is situated on a large boat and enforces a strict policy of fear and greed. Those who fail to wear fear and greed on their face during their gambling endeavours are summarily thrown out to sea. 14 I am paraphrasing Mill (1869, p. 66), who is here describing the relation of ‘superiors to dependents’ in the context of the patriarchal family. His description nevertheless applies as much to ancient slavery as to the contemporary wage relation. 15 A system that has this tendency may be bad because the tendency is itself bad or because what grounds it is independently bad. I take Cohen to be making both claims: the tendency is itself bad, and its presence is evidence of something bad underlying it. Under capitalism, the ground of the tendency is objective relations of dependence, themselves constituted by the commodification of labour power. 16 Cohen denied that decommodification is sufficient for communal reciprocity. Section IV of Cohen (2009a) is devoted to the question whether it would be possible, using a form of social technology morally superior to that of the market, to produce a reasonably efficient and egalitarian economy on the basis of communal reciprocity. His answer is that we do not know whether such a technology will ever be forthcoming, which has the – only slightly encouraging – consequence that it is false that we now know that such a technology will never be forthcoming. And, because ‘every market, even a socialist market, is a system of predation’, it follows that we should not give up on ‘our attempt to go beyond predation’ (Cohen 2009a, p. 82).

Notes

152

17 The contrast between free action and acting freely is, in a sense, like the contrast between true belief and knowledge. You know that p only when you believe that p for the right reasons. Likewise, you φ freely only when you φ for the right reasons. 18 Cohen (1988) does not argue for the conceptual integrity of (4). All he is concerned to show is that being free to φ is compatible with, indeed required by, (1) and (2). But the possibility that (4) is true seems to follow from the conditions of acting freely (see the last sentence of the Cohen quote provided two paragraphs back). 19 For discussion of the self-knowledge motif in Marx, see Cohen’s lectures on Marx, reprinted as Chapters 3 and 5 of Cohen (2000b). 20 Some have even claimed that the omission makes these theories, or their implications, undemocratic. The claim is thoroughly unjustified, a reactionary fairy tale as old as the institutions that nurtured it. See, for extensive refutation, Balibar (1977), Draper (1987) and Gilbert (1981). 21 Nagel argues that ‘sexual perversion’ obtains when the ascent from (6) upwards fails to materialize. So, for example, narcissism and voyeurism obtain when A blocks, or is indifferent to, the ascent to (6) and beyond: the voyeur does not want to be the object of another’s desire, and a fortiori does not want to be the embodiment of that desire. Sadism and masochism obtain when either (7) or (8) fail to obtain, and so on. 22 As opposed to the merely local reciprocal awareness of successful sexual or musical relations. 23 Which is probably false. Flourishing sexual relations do not presuppose an abundance of sexual instruments (even if one is, as Freudians put it, ‘polymorphously perverse’). Something similar holds for social relations, quite generally. 24 It is difficult to authentically enjoy any joint activity in the knowledge that it takes place, as Cohen puts it, ‘for the wrong reason’. To return to the prostitution example, the prostitute’s acting unfreely normally impinges on the client’s satisfaction, indeed on his own freedom to have sex (without concealing a very basic fact about the prostitute from himself). In a sense, the prostitute’s unfreedom renders the client unfree. Of course some people have sex with prostitutes, and some of them no doubt enjoy it. But such enjoyment is typically forthcoming only by mobilizing large amounts of self-deception. If Cohen is right, then capitalist institutions universalize both the general form of prostitution and the self-deception required to reproduce it (see, for example, Jeffries 2009).

Notes

153

25 This suggests a set of concerns distinct from democracy that makes planning desirable. It is, of course, possible that democratic planning is a necessary condition for the achievement of reciprocal awareness. Indeed, democracy seems to be required by any defensible community-based argument against the market, for that is the only way of making bonds of fellowship transparent. But the goal is not democracy per se. It is, rather, the consummation of individual flourishing in community with others. 26 It is worth noting that reciprocal awareness has strong family resemblances with one interpretation of Rawls’ (1971) requirement of ‘well-orderedness’. 27 Cohen (1978) argues that economic planning alone makes possible a significant expansion of the realm of freedom for the vast majority. It does this by eliminating ‘capitalism’s distinctive contradiction’, the bias in favour of toil and against leisure. See the section ‘The argument for primacy’ of Chapter 1. 28 Indeed, they seem to vindicate the view that the title of Friedman’s book is an oxymoron. For Cohen’s argument that it is an oxymoron, see the section ‘Unfreedom and private property’ of Chapter 2. 29 Cohen was not active in party politics. He did, however, pen several pamphlets, including one for Socialist Renewal in 1995. There he argued against the removal of the common ownership proviso, clause IV, from the Constitution of the Labour Party (Cohen 1995b). He left the party after Blair revised clause IV.

Works Cited

Althusser, L. (1969), For Marx. Harmondsworth: Penguin. Althusser, L. (2011), Sur la Reproduction. Paris: Presses Universitaires de France. Althusser, L. and Balibar, E. (2009), Reading Capital. London: Verso. Anderson, E. (2010), ‘Cohen, Justice and Interpersonal Justification’, mimeo (http://ssrn.com/abstract=1642011). Arneson, R. (1999), ‘Human Flourishing vs. Desire Satisfaction’, Social Philosophy and Policy, 16, 113–142. Arthur, C. J. (2004), The New Dialectic and Marx’s Capital. Leiden: Brill Academic. Bader, R. (2013), ‘Review of Finding Oneself in the Other’, Notre Dame Philosophical Reviews. (http://ndpr.nd.edu/news/38715-finding-oneself-in-the-other/). Badiou, A. (2005), Being and Event. New York: Continuum. Balibar, E. (1977), On the Dictatorship of the Proletariat. London: Verso. Barry, B. (1999), ‘Sustainability and Intergenerational Justice’, in A. Dobson (ed.), Fairness and Futurity. Oxford: Oxford University Press. Berlin, I. (2002), Liberty. Oxford: Oxford University Press. Blackburn, S. (2004), Lust. Oxford: Oxford University Press. Bowman, A., Williams, K., Folkman, P., Froud, J., Sukhdev, J., Law, J., Leaver, A. and Moran, M. (2013), The Great Train Robbery. Manchester: Centre for Research on Socio-Cultural Change. Brenner, R. (2006), The Economics of Global Turbulence. London: Verso. Brudney, D. (1998), Marx’s Attempt to Leave Philosophy. Cambridge, MA: Harvard University Press. Bukharin, N. (1925), Historical Materialism. Moscow: International Publishers. Callinicos, A. (2010), Bonfire of Illusions. London: Polity. Carens, J. (1981), Equality, Moral Incentives and the Market. Chicago: University of Chicago Press. Cohen, G. A. (1969), ‘Critical Theory: The Philosophy of Marcuse’, New Left Review, I/57, 35–51. Cohen, G. A. (1970), ‘On Some Criticisms of Historical Materialism’, Proceedings of the Aristotelian Society, 44(supplement), 143–156. Cohen, G. A. (1974), ‘Marx’s Dialectic of Labour’, Philosophy and Public Affairs, 3, 235–261.

156

Works Cited

Cohen, G. A. (1977), ‘Robert Nozick and Wilt Chamberlain: How Patterns Preserve Liberty’, Erkentnnis, 11, 5–23. Cohen, G. A. (1978), Karl Marx’s Theory of History: A Defence. Princeton, NJ: Princeton University Press. Cohen, G. A. (1980), ‘Reply to Elster’. Political Studies, 28, 129–135. Cohen, G. A. (1982), ‘Reply to Elster on Marxism, Functionalism and Game Theory’, Theory and Society, 11, 483–495. Cohen, G. A. (1983a), ‘Reply to Four Critics’, Analyse und Kritik, S, 195–222. Cohen, G. A. (1983b), ‘Review of Allen Wood’s Karl Marx’, Mind, 92, 440–445. Cohen, G. A. (1986), ‘Against Capitalism’, The Listener, 4 September 1986. Cohen, G. A. (1988), History, Labour, and Freedom. Oxford: Oxford University Press. Cohen, G. A. (1989a), ‘On the Currency of Egalitarian Justice’, Ethics, 99, 906–944. Cohen, G. A. (1989b), ‘David Miller on Distributive Justice and Market Socialism’, mimeo. Cohen, G. A. (1995a), Self-Ownership, Freedom and Equality. Cambridge: Cambridge University Press. Cohen, G. A. (1995b), ‘Is Socialism Inseparable from Common Ownership?’, Socialist Renewal, 1, 1–8. Cohen, G. A. (2000a), Karl Marx’s Theory of History: A Defence (Expanded edition). Princeton, NJ and Oxford: Princeton University Press and Oxford University Press. Cohen, G. A. (2000b), If You’re an Egalitarian, How Come You’re So Rich? Cambridge, MA: Harvard University Press. Cohen, G. A. (2004a), ‘Rescuing conservatism from the Conservatives’, mimeo. Cohen, G. A. (2004b), ‘Expensive Taste Rides Again’, in J. Burley (ed.), Dworkin and His Critics. Oxford: Blackwell. Cohen, G. A. (2006), ‘Luck and Equality: A Reply to Hurley’, Philosophy and Phenomenological Research, 72, 439–446. Cohen, G. A. (2008), Rescuing Justice and Equality. Cambridge, MA: Harvard University Press. Cohen, G. A. (2009a), Why Not Socialism? Princeton, NJ: Princeton University Press. Cohen, G. A. (2009b), ‘Reply to Michael Otsuka’s Justice as Fairness: Luck Egalitarian, not Rawlsian’, mimeo. Cohen, G. A. (2011), On the Currency of Egalitarian Justice, M. Otsuka (ed.). Princeton, NJ: Princeton University Press. Cohen, G. A. (2013), Finding Oneself in the Other, M. Otsuka (ed.). Princeton, NJ: Princeton University Press. Cohen, G. A. (2014), Lectures in the History of Moral and Political Philosophy, J. Wolff (ed.). Princeton, NJ: Princeton University Press.

Works Cited

157

Cohen, J. (1982), ‘Review of Karl Marx’s Theory of History’, Journal of Philosophy, 44, 253–273. Cohen, J. (2001), ‘Taking People as They Are?’, Philosophy and Public Affairs, 30, 363–386. Coram, B. T. (1989), ‘Social Relations and Forces of Production’, Social Theory and Practice, 15, 213–229. Davidson, D. (1970), ‘How Is Weakness of the Will Possible?’, in D. Davidson (1980), Essays on Actions and Events. Oxford: Clarendon Press, 21–42. Draper, H. (1987), Karl Marx’s Theory of Revolution: The Dictatorship of the Proletariat. New York: Monthly Review Press. Dworkin, R. (2000), Sovereign Virtue. Cambridge, MA: Harvard University Press. Elster, J. (1980), ‘Cohen on Marx’s Theory of History’, Political Studies, 28, 121–128. Elster, J. (1982), ‘Marxism, Functionalism and Game Theory: The Case for Methodological Individualism’, Theory and Society, 11, 453–482. Estlund, D. (1998), ‘Liberalism, Equality and Fraternity in Cohen’s Critique of Rawls’, Journal of Political Philosophy, 6, 99–112. Feltham, D. (ed.) (2009), Justice, Equality, and Constructivism. Oxford: Wiley Blackwell. Finlay, M. I. (1959), ‘Was Greek Civilization Based on Slave Labour?’, Historia, 8, 145–164. Frankfurt, H. (2005), On Bullshit. Princeton, NJ: Princeton University Press. Fried, B. (1998), The Progressive Assault on Laissez-Faire. Cambridge, MA: Harvard University Press. Friedman, M. (2002), Capitalism and Freedom. Chicago: University of Chicago Press. Gilbert, A. (1981), Marx’s Politics. Rutgers: Rutgers University Press. Goodin, R. (1987), ‘Exploiting a Person and Exploiting a Situation’, in A. Reeve (ed.), Modern Theories of Exploitation. London: Sage Publishers. Griffin, J. (1986), Well-Being. Oxford: Clarendon Press. Harrison, M. (ed.) (2000), The Economics of World War II. Cambridge: Cambridge University Press. Harvey, D. (1986), ‘Review of Making Sense of Marx’, Political Theory, 14, 686–690. Harvey, D. (2005), A Brief History of Neoliberalism. Oxford: Oxford University Press. Harvey, D. (2012), The New Imperialism. Oxford: Oxford University Press. Hempel, C. (1966), Philosophy of Natural Science. New York: Pearson Books. Hobsbawm, E. (1984), Worlds of Labour. London: Weidenfeld and Nicholson. Hobsbawm, E. (1999), Industry and Empire. Harmondsworth: Penguin. Hurley, S. (2005), Justice, Luck and Knowledge. Cambridge, MA: Harvard University Press. Jeffries, S. (2009), The Industrial Vagina: The Political Economy of the Global Sex Trade. London: Routledge.

158

Works Cited

Kautsky, K. (1996) [1909], The Road to Power. New York: Humanity Books. Keat, R. (1981), ‘Individualism and Community in Socialist Thought’, in J. Mepham, and D.-H. Ruben (eds), Issues in Marxist Philosophy, Volume 4. Hassocks: Harvester Press. Krugman, P. (2009), The Return of Depression Economics and the Crisis of 2008. New York: W. W. Norton. Kymlicka, W. (2001), Contemporary Political Philosophy. Oxford: Oxford University Press. Lenin, V. I. (1974), ‘Karl Marx: A Brief Biographical Sketch, with an Exposition of Marxism’, in Institute for Marxism-Leninism (ed.), Collected Works, Volume 21. Moscow: Progress Publishers. Leopold, D. (2007), The Young Karl Marx. Cambridge: Cambridge University Press. Levine, A. and Wright, E. O. (1980), ‘Rationality and Class Struggle’, New Left Review, 123, 47–68. Lih, L. (2011), Lenin. Montreal: Reaktion Books. Locke, J. (2003) [1689], Two Treatises of Government. New Haven, CT: Yale University Press. Lukács, G. (2009), Lenin: A Study on the Unity of His Thought. London: Verso. Marx, K. (1963), Early Writings, T. Bottomore (ed.). London: C. A. Watts. Marx, K. (1973), Grundrisse, M. Nicolaus (trans.). Harmondsworth: Penguin. Marx, K. (1976), Capital, Volume I. Harmondsworth: Penguin. Marx, K. (1977) [1859], Contribution to a Critique of Political Economy. Moscow: Progress Publishers. Marx, K. (1981), Capital, Volume III. Harmondsworth: Penguin. Marx, K. (1986), Karl Marx: A Reader, J. Elster (ed.). Cambridge: Cambridge University Press. Marx, K. and Engels, F. (1956), The Holy Family. Moscow: Progress Publishers. Marx, K. and Engels, F. (1975), Collected Works (MECW), vol. 1–50. New York: International Publishers. Mill, J. S. (1869), The Subjection of Women. London: Longmans Press. Miller, D. (1990), Market, State, and Community. Oxford: Oxford University Press. Miller, D. (2013), Justice for Earthlings. Cambridge: Cambridge University Press. Miller, R. (1984a), Analyzing Marx. Princeton, NJ: Princeton University Press. Miller, R. (1984b), ‘Producing Change: Work, Technology and Power in Marx’s Theory of History’, in T. Ball and J. Farr (eds), After Marx. Cambridge: Cambridge University Press. Miller, R. (2010), ‘Relationships of Equality: A Camping Trip Revisited’, Journal of Ethics, 14, 231–253.

Works Cited

159

Mills, C. (1989), ‘Is It Immaterial That There’s a “Material” in “Historical Materialism”?’, Inquiry, 32, 323–342. Munoz-Darde, V. (1998), ‘John Rawls, Justice in and Justice of the Family’, Philosophical Quarterly, 48, 335–352. Nagel, T. (1991), Mortal Questions. Cambridge: Cambridge University Press. Nozick, R. (1974), Anarchy, State and Utopia. New York: Basic Books. OECD (2012), OECD Factbook 2012. Paris: Organization for Economic Cooperation and Development. Okin, S. M. (1991), Justice, Gender, and the Family. New York: Basic Books. Otsuka, M. (2005), Libertarianism Without Inequality. Oxford: Oxford University Press. Otsuka, M. (2010), ‘Justice as Fairness: Luck Egalitarian, Not Rawlsian’, Journal of Ethics, 14, 217–230. Parfit, D. (1997), ‘Equality and Priority’, Ratio, 10, 202–221. Piketty, T. (2014), Capital in the Twenty-First Century. Cambridge, MA: Harvard University Press. Plekhanov, G. (1947), The Monist View of History. London: Lawrence and Wishart. Pogge, T. (2007), John Rawls: His Life and Theory of Justice, M. Kosch (trans.). Oxford: Oxford University Press. Rawls, J. (1971), A Theory of Justice. Cambridge, MA: Harvard University Press. Rawls, J. (1999), A Theory of Justice (revised edn). Cambridge, MA: Harvard University Press. Rawls, J. (2001), Justice as Fairness: A Restatement. Cambridge, MA: Harvard University Press. Rees, W. E. and Wackernagel, M. (1994), ‘Ecological Footprints and Appropriated Carrying Capacity’, in A. Jansson (ed.), Investing in Natural Capital: The Ecological Economics Approach to Sustainability. Washington, DC: Island Press. Robinson, J. (1966), An Essay on Marxian Economics (2nd ed.). London: Macmillan. Roemer, J. (1994), A Future for Socialism. London: Verso. Roemer, J. (1996), Egalitarian Perspectives. Cambridge: Cambridge University Press. Rosdolsky, R. (1977), The Making of Marx’s Capital. London: Pluto. Rosen, M. (2010), ‘Jerry Cohen – an appreciation’, mimeo. (http://scholar.harvard. edu/michaelrosen/publications/jerry-cohen-appreciation). Rubin, I. I. (1972) [1928], Essays on Marx’s Theory of Value. Detroit, MI: Black and Red. Schweickart, D. (1996), Against Capitalism. Boulder, CO: Westview Press. Schweickart, D. (2002), After Capitalism. London: Rowman and Littlefield. Sen, A. K. (1984), Resources, Values and Development. Oxford: Blackwell. Shearmur, J. (2003), ‘Beyond Fear and Greed?’, Social Philosophy and Policy, 20, 247–277. Steiner, H. (2014), ‘Fear and Greed’, Politics, Philosophy and Economics, 13, 140–150.

160

Works Cited

Sweezy, P. (1946), The Theory of Capitalist Development. London: D. Dobson. Swift, A. (2006), Political Philosophy: A Beginner’s Guide. London: Polity. Temkin, L. (2003), ‘Egalitarianism Defended’, Ethics, 113, 764–782. Tooley, M. (2002), ‘The Problem of Evil’, in E. Zalta (ed.), Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford, CA. Tye, M. (1990), ‘Vague Objects’, Mind, 99, 535–557. Vrousalis, N. (2011a), ‘Libertarian Socialism’, Social Theory and Practice, 37, 211–226. Vrousalis, N. (2011b), ‘Lamentation in the Face of Historical Necessity’, in Y. Vanderborght and A. Gosseries (eds), Arguing about Justice. Louvain-la-Neuve: Louvain Press, 367–376. Vrousalis, N. (2013), ‘Exploitation, Vulnerability, and Social Domination’, Philosophy and Public Affairs, 41, 131–157. Waldron, J. (1990), The Right to Private Property. Oxford: Clarendon. Weber, M. (1905), The Protestant Ethic and the Spirit of Capitalism. Harmondsworth: Penguin. White, S. (2006), Equality. London: Polity. Wiggins, D. (1998), Needs, Values, Truth. Oxford: Oxford University Press. Williams, A. (1998), ‘Incentives, Inequality, and Publicity’, Philosophy and Public Affairs, 27, 225–247. Williams, A. (2002), ‘Dworkin on Capability’, Ethics, 113, 23–39. Williams, B. (1995), Making Sense of Humanity. Cambridge: Cambridge University Press. Williamson, T. (1994), Vagueness. London: Routledge. Wimsatt, W. and Beardsley, M. C. (1946), ‘The Intentional Fallacy’, Sewanee Review, 54, 468–488. Wolff, J. (2014), ‘G. A. Cohen: A Memoir’, in G. A. Cohen (ed.), Lectures in the History of Moral and Political Philosophy. Princeton, NJ: Princeton University Press. Wolff, R. P. (1968), The Poverty of Liberalism. Boston, MA: Beacon Press. Wood, A. (1995), ‘Exploitation’, Social Philosophy and Policy, 12, 136–158. Wood, A. (2004), Karl Marx. London: Routledge. Wright, E. O., Levine, A. and Sober, E. (1992), Reconstructing Marxism. London: Verso.

Index Note: The letter ‘n’ following locators refers to notes. 1 per cent 26, 45–6, 132 akrasia 75 Althusser, Louis 2, 28, 38 and overdetermination 139 n.16 Anderson, Elizabeth 148 n.19, 149 n.24 Ant and Grasshopper case 95 Badiou, Alain 136 n.23 Berlin, Isaiah 15, 42–4, 140 n.2, 144 n.23 Blackburn, Simon 128, 130 Brenner, Robert 132 bullshit 9–11, 29, 136 n.23 as pretentious nonsense 11 as unclarifiable unclarity 10 bureaucracy 46–7 Callinicos, Alex 132 capability approach 70–2 and functionings 69–72, 143 n.16 (see also Sen, Amartya) capitalism and community 100, 120–3, 126, 130–1 and exploitation 65–6, 84–91, 94–6 and historical materialism 23–6, 38–9, 83–4 and injustice 64–6, 117–18, 141 n.1 and money 44–7, 140 n.3 and private property 47–55, 132–4 see also freedom; market Carens, Joseph 119–23, 150 n.5 Chamberlain, Wilt 35, 52–4 civil society 102, 148 n.9 see also market coercion 84 Cohen, Joshua 29, 148 n.14 collective action problems in historical materialism 29 and inequality 54

commodification 121, 124–6 and prostitution 124–8 and rape 124–5 communism and community 100–2 and jazz band analogy 101–2, 129 primitive 11 community 7, 59, 79, 99–114 as communal reciprocity 110–13, 120–3, 129–31 and interpersonal test 106–7 justificatory 106–8 and kidnapper case 107–8 and reciprocal awareness 127–31 see also commodification; market consequence explanation, see explanation conservatism 11–14 big-C 13 constructivism 7, 103 Coram, B. T. 33–5 Darwin, Charles 28 democracy 79, 127, 129–31, 133, 150 n.5, 153 n.25 desert 79–80, 106, 117–18 comparative vs. non-comparative 80, 144 n.24 dialectic 21–5, 36–9 difference principle 6, 66, 74–5 and community 103–5 lax vs. strict reading of 104–5, 112–13 and Pareto argument 76–8 see also Rawls, John; incentives; inequality division of labour and historical materialism 19, 33, 131 in household 73–4 domination 94–6, 122–3, 125–6

162

Index

Dworkin, Gerald 39, 125 Dworkin, Ronald 15, 142 and equality of resources 62–6, 69–70, 91, 96–8 egalitarian ethos 13, 15, 72–6, 80–1, 103–5, 108, 113–14 and integrity 75–6 Elster, Jon 28, 37 explanation consequence 27, 138 functional 27–8, 32, 35–8, 84 see also historical materialism exploitation 5, 17, 31–2, 49–51, 83–98 capitalist vs. socialist 118–19 and distributive justice 91–8 and libertarianism 54–5 mode of 84, 144 n.2 facts, see principles fear and greed 110–13, 120–2, 124–6, 149 n.30, 151 n.13 feasibility 24–5, 30, 76–8 see also inequality fettering 22, 24–7 feudalism and community 100–1, 112 and historical materialism 2, 37, 80–1 Feuerbach, Ludwig 1 flourishing 65, 67–70, 125, 143 n.15 as equal access to advantage 72 (see also justice, metric of) force, see coercion Frankfurt, Harry 9–11, 136 n.19 freedom and unfreedom 14, 17–18, 41–59, 70–1 acting freely 124–6, 140 n.8, 152 n.17 proletarian 49–51, 145 n.6; fallacy of composition and 50 realm of 131 Friedman, Milton 46–7, 132, 140 n.4, 153 n.28 functionalism 27 Gramsci, Antonio 75 Harvey, David 43, 132, 138 n.12 Hegel, Georg Willem Friedrich 1, 14, 99, 127, 138 n.9

hegemony 13, 75, 81 historical materialism 1–3, 17–39, 83–4 and agency problem 31 contradiction and 22, 25–7, 37, 153 n.27 and economic structure 19–21, 24–8, 34–7, 47–51, 63–5 and egalitarianism 80–1 inclusive vs. restrictive 37–8 material vs. social properties in 19, 32–5 and material abundance 17–18, 80–1 mode of production interpretation of 35–8 primacy thesis in 18, 21–7, 29–32, 36–8, 139 n.15 productive forces and 18–19, 21–4, 27–8, 30–8 and scarcity problem 31–2 and superstructure 18–21, 27–8, 37–8 Hobsbawm, Eric 43, 49, 139 n.18 ideology 20–1, 37–8, 50–1, 84, 112, 132 immanent critique 4, 38, 41–2, 73–4, 103–4, 110, 149 n.24 incentives justification of 74–8, 104–8, 149 n.26 moral 119–20 see also Rawls, John inequality Pareto argument for 76–8 and political power 47 right-libertarian argument for 51–4 see also justice, distributive intergenerational justice 69 justice in acquisition 51–2, 54–5 circumstances of 101–2 distributive 47–9, 51–9, 62–72, 76–80, 86–94, 96–8, 103–5, 108–10, 117–19; aspects of 79–80; socialist principle of distribution and 65–6, 112, 124–5, 128 externalism about 113–14 metric of 62, 67–72; midfare 71–2 (see also Sen, Amartya); Tiny Tim case 68–70 pattern of 52–4, 61–6

Index site of 61, 72–6 in transfer 52–4 Kant, Immanuel 1, 17, 135 n.6, 147 n.6 Kautsky, Karl 30–1 Krugman, Paul 132 labour embodied vs. counterfactual 85 and historical materialism 25, 131 and primary goods 63, 153 n.27 labour theory of value 84–7, 139 n.21 Lamarck, Jean-Baptiste 28 Lenin, Vladimir Ilyich 30–1, 138 n.12 liberalism 5–6, 73–6, 103–8 see also Rawls, John libertarianism left- 57–9 right- 51–7, 132 (see also Nozick, Robert) Locke, John 54–5 Lockean proviso 55 lottery 96–7, 109 love 12, 96, 100, 124, 129 luck brute vs. option 62–3, 65–7, 91–2, 96–8; and unanimity cases 96–7 luck egalitarianism 62–7, 91–4, 96–8, 149 n.27 Lukács, Gyorgy 3 market 44–7, 52–4, 63–6, 74–5, 103–5, 115–18, 120–4, 126–31 as commodity production 120–1 see also commodification; socialism Marx, Karl 1–2, 5–6, 17–39, 41, 44–7, 49–51, 65, 80–1, 83–6, 90, 99–102, 110–12, 124, 127, 131 and 1859 Preface 17–24, 35–7 and Capital 38–9 see also historical materialism; communism Marxism analytical 9, 29, 38 classical 18, 31, 35, 137 n.2 see also historical materialism, communism

163

Miller, David 116–17 Miller, Richard 35–8, 149 n.23 moral responsibility 64, 66–7 Munoz-Darde, Veronique 73 Nagel, Thomas 127–8 Nozick, Robert 2, 41, 48, 51–7, 132–3 minimal state and 51–3 Okin, Susan 73 Otsuka, Michael 93–4, 96–7 philosophy Harvard vs. Oxford approach to 4–5 Socratic method in 4 Platonism 5–8 Plekhanov, Gyorgy 27 pluralism about value 7–8, 79, 135 n.10 explanatory 37 positivism 38 power effective 19–20, 84, 140 n.7 political 46–7, 102 preferences adaptive 68, 117 satisfaction of 63–4, 67–72 principles fact-insensitive 3–6 (see also Rawls, John) privatization 48–9, 100, 132 profit motive 25, 87–8, 94–6, 107–8, 116–23, 134 see also fear and greed property common 55–9 private 47–9, 54–5, 73, 87–8, 90, 118–19, 150 n.7 Rawls, John 2, 62 and basic structure 72–6 and community 103–8 and democratic equality 76–8 and freedom 43 and reflective equilibrium 4–8 see also incentives; inequality; difference principle

164

Index

Roemer, John 86–96, 118–19, 145–6, 150 n.5 Rosen, Michael 2 Rubin, Isaac Ilyich 86 self-ownership 51–2, 56–9, 124–6 see also libertarianism Sen, Amartya 67–72 serfdom 23, 112 sexual desire 124–30 see also community Shearmur, Jeremy 120 slavery 2, 22–3, 35, 80–1, 83–4, 134 socialism 2–3 desirability of 23–6, 46–51, 59, 65–6, 95–6, 109–13 feasibility of 30, 80–1 market 116–26, 130, 150 and planning 117–31 sociobiology 29 Soviet Union 30–1 species-being 2, 102, 135 n.3 Steiner, Hillel 120–1, 151 n.12 surplus material 22–3, 37, 83–4 value 84–6, 117–18 Sweezy, Paul 86 talents 58, 76–8, 101, 104–10, 118–19 see also incentives; Rawls, John

tastes cheap 68 expensive 63–6 technological materialism, see historical materialism Temkin, Larry 79 theodicy 1–2, 17 Tye, Michael 136 n.22 utilitarianism 67 and preferentialism 68–9 utopia 133 and blueprintism 115 voluntariness 53 vulnerability and exploitation 92–6 wage labour 23, 71–8, 86–91, 116–19, 132 see also commodification welfare 62–6, 68–9 Williamson, Tim 136 n.22 Wolff, Jonathan 135 n.1 Wolff, Robert Paul 99 Wright, Erik 26–7, 37, 39 Young Hegelians 99