The Philosophy of Rawls
A
Collection of Essays
Series Editors
HenryS. Richardson
Georgetown University
Paul J. Weithman
University of Notre Dame
A GARLAND SERIES READINGS IN PHILOSOPHY
RoBERT NozicK, ADvisoR
HARVARD UNIVERSITY
Contents of the Series 1.
Development and Main Outlines of Rawls's Theory of Justice
3.
Opponents and Implications of A
2.
4.
5.
The Two Principles and Their Justification
Theory of Justice
Moral Psychology and Community Reasonable Pluralism
Moral Psychology and Community
Edited with an introduction by
Paul J. Weithman
University of Notre Dame
GARLAND PUBLISHING, INC.
A MEMBER OF THE TAYLOR & FRANCIS GROUP
New York & London 1999
Introductions copyright 0 1999 Paul J. Weithman. All rights reserved.
Ubrary of Congress Cataloging-in-Publication Data Moral psychology and community I edited with an introduction by Paul J. Weithman. p.
em.
-
(The philosophy of Rawls; 4)
·A Garland series, readings in philosophy.* Includes bibliographical references. ISBN 0-8153-2928-8 (alk. paper)
l. Self-esteem. 2. Community. 3. Rawls, John 1921Contributions in political science.
I. Weithman, Paul J., 1959- II. Series.
HM106l.M67 1999 155.2-dc21
Printed on add-free, 250-year-life paper Manufactured in the United States of America
99-048609
Contents
vii ix
Series Introduction Volume Introduction MORAL PSYCHOLOGY
Self-Esteem and Self-Respect
l 22
37
Shame and Self-Esteem: A Critique John Deigh How to Distinguish Self-Respect from Self-Esteem David Sachs Rawlsian Self-Respect and the Black Consciousness Movement Larry L. Thomas
Moral Development and a Sense of Justice
50
67
85
The Claim to Moral Adequacy of a Highest Stage of Moral Judgment La..vrence Kohlberg The Motivation to Be Just Stanley Bates Human Evolution and the Sense of Justice Allan Gibbard
Moral Psychology, Stability and the Support they Provide for "Justice as Fairness"
l 02
1 39
Reason and Agreement in Social Contract Views Samuel Freeman Justice and the Problem of Stability Edward F. McCiennen
Vi
CONTENTS
167
The Stability Problem in Political Liberalism Thomas E. Hill, Jr.
187
Moral Independence and the Original Position Samuel Scheffler COMMUNITY
Group Identity and the Communltarlan Critique
195 217
Rawls on the Individual and the Social Wayne Proudfoot The Procedural Republic and the Unencumbered Self Michael J. Sandel
Defense of Rawlslan Liberalism
234 246 261
Rawls and Individualism C.F. Delaney
Communitarian Critics of Liberalism Amy Gutmann Rawls, Hegel, and Communitarianism Sibyl A. Schwarzenbach
The Uberallsm·Communltarlanlsm Debate and Beyond
295 321 343
Rawls's Communitarianism Roberto Alejandro The Individual, the State, and the Common Good John Haldane Acknowledgments
Series Introduction
John Rawls is the pre-eminent political philosopher of our time. His 197 1 masterpiece, A Theory of Justice, permanently changed the landscape of moral and political theory,
revitalizing the normative study of social issues and taking stands about justice, ethics, rationality, and philosophical method that continue to draw followers and critics today. His Political Liberalism (rev. ed., 1996) squarely faced the fundamental challenges posed by cultural, religious, and philosophical pluralism. It should be no surprise, then, that turn-of-the-century searches of the periodical indices in philosophy, economics, law, the humanities, and related fields turn up almost three thousand articles devoted to a critical discussion of Rawls's theory. In these Volumes we reprint a wide-ranging selection of the most influential and insightful articles on Rawls. While it was impossible, even in a collection of this size, to reprint all of the important material, the selection here should provide the student and scholar with a route into all of the significant controversies that have surrounded Rawls's theories since he first began enunciating them in the nineteen-fifties- issues that the Introductions to each Volume of this series delineate. Eight criteria guided our selection. First, these volumes form part of a series devoted to secondary literature. We reprint no articles by Rawls: most of these have just appeared together for the first time in his Collected Papers.1 Second, we reprint only self-contained articles published in English, rather than selections from books or articles in other languages. Third, the articles reprinted here are all about Rawls's view, as opposed to being original reflections inspired by Rawls's work. Fourth, we aimed for a broad coverage of controversies and of the main features of Rawls's theory that they surround. Since the Volumes are organized in terms of these controversies, we include very few overall assessments or book reviews. Some central elements of Rawls's theory, while relatively novel and well-articulated, have not been controversial enough to draw critical fire in the secondary literature. The Volume Introductions mention many of these features. Fifth, we aimed to include the most influential articles that have appeared. In identifying these, we used a systematic search of the citation indices to supplement our own judgment. Naturally, we also took special notice of pieces cited by Rawls himself. Sixth, we sought to reprint articles by a large number of authors representing the widest possible range of points of view. In some cases, this meant refraining from reprinting a certain article because its author was already well represented in the selections. Seventh, we have sought to exhibit through
Viii
SERIES INTRODUCTION
�
our selections the broadly interdisciplinary influence of Rawls's writings. We ha e included articles by political theorists, economists, lawyers, religious thinkers, and sooal sdentists as well as by philosophers. Eighth, we have favored including articles that are now relatively hard to find. For this reason, with the exception of H. L.A. Hart's exceptionally influential essay, we refrained from including any of the fine articles that were reprinted in Norman Daniels's 197 5 collection, Reading Rawls/ which the reader interested in the early reception of Rawls's views should consult. Utilizing all of these selection criteria did not leave us without painful choices. The secondary literature on Rawls is so deep that another set of five volumes could cover all the main issues with a completely non-overlapping set of fine articles. Some articles unfortunately had to be cut because of their sheer length: dropping one of them allowed us to include two or three others. Others, more arbitrarily, fell victim to the high permissions costs set by their initial publishers. We particularly regret that it proved impossible to find a short enough, self-contained essay by Robert Nozick that would have represented his trenchant libertarian critique of Rawls. While we do include (in Vol. 3) some of the secondary literature that responds to and picks up on Nozick's influential arguments, one should consult Nozick's Anarchy, State, and Utopia ( 197 4) to appreciate their richness, subtlety, and power.' The five volumes are arranged in roughly chronological order. The first volume includes articles on Rawls's early statements of his view and on its central contractarian ideas. Volume 2 covers the two principles of justice as fairness and Rawls's most general ideas about their justification. Volume 3 focuses on the concrete implications of Rawls's view and on the debates between Rawls and his utilitarian, perfectionist, libertarian, conservative, radical, and feminist critics. Volume 4 treats of Rawls's moral psychology and his attempt to accommodate the value of community. Volume 5, on Rawls's most recent work, is entitled "Reasonable Pluralism. The serious student of Rawls's initial impact is greatly assisted by John Rawls and His Critics: An Annotated Bibliography, put together by J.H. Wellbank, Denis Snook, and David T. Mason, which catalogues and provides abstracts for most of the secondary literature in English prior to 1982.4 While this work was of great help with that earlier •
period, completing the onerous task of collecting and sorting through the voluminous secondary literature, which has since continued to balloon, would not have been possible without the able and thorough research assistance of Rachael Yocum. We are
grateful to the Dean of Georgetown College and to the Graduate School of Georgetown University for their generosity in supporting this research assistanc e. Henry S. Richardson Paul J. Weithman Notes
Rawls, CDI/tatd Paptrs
1
John
•
J.H. Wrllbank, Denis Snook and D
�an (Cambn.dge, Mass.: Harvard University Press, 1999).
ed Samu e1 Free
'Norman Daniels' rd., Be-�: ""1"11 Ra. wIs (N .Y.: BaSic Books, • 1 Roben NOZick , AMrdty, Statt, and Utqpia (N.Y.:
�
(New York: Garla d
1975). 1974) 'd T· Mason, Jo 1rn Rawls ' and Hu · Critics: An Annotattd Bibliography
��
198
Basic Books
Volume Introduction
One of the most important pieces of Rawls's theory of justice is the moral psychology it incorporates. Rawls gives an account of the moral and psychological capadties citizens are presumed to have, and of the moral development and motives of citizens in a just society. Elements of this account bear great weight at pivotal steps in the arguments of A Theory of Justice1 and Political Liberalism.2 Rawls developed and connected those elements with an eye to the burdens he intended to place upon them. A good deal of attention has been devoted to determining exactly what burdens he intended them to bear, whether they are able to bear them and whether the arguments they are intended to support establish what Rawls thinks they do. The account that Rawls produced is also of intrinsic philosophical, psychological and political interest. Some of the commentary it has received therefore pays scant attention to the roles it plays in arguments for justice as fairness. Instead it is viewed as a sophisticated contribution to the studies of emotion and moral development, worthy of careful examination in its own right. Rawls's treatment of moral psychology also supplies premises crucial to his argument that a society that conformed to his two prindples would enjoy the goods of community. Challenges to this claim sparked one of the most important debates in political philosophy during the 1980s, the liberal-communitarian debate. The articles in the present volume are among the most probing pieces of critical literature on Rawls's discussions of moral psychology and community. Their content, range and influence demonstrate both how important these topics are to Rawls's work and how far-reaching their implications are taken to be. Since the articles depart from such different questions and range over so wide an area, it will prove useful to map the terrain, plot their connections and show where they venture into territory covered in other volumes. One indication of how central Rawls's moral psychology is to his theory is the connection between it and the distinctive features of the original position, which are designed to model the moral powers Rawls attributes to citizens. While there is some intimation of this in TJ (see p. 18), Rawls could fully articulate the connection between the moral powers and the original position only after having given a richer and more Kantian conception of persons in response to H.L.A. Hart's criticisms (see PL, pp. 299, 304ff.: also the introduction to Volume 2). Because essays in other volumes treat of the original position (Volume 1), Rawls's response to Han (Volume 2) and the increasingly
X
V o LUME INTRODUCTION
�
L:���::
he Dewey n present in Rawls's work from Kantian conception of the perso nted here . A ers are not pursued in articles repn onward (Volume 2), these matt ttach es to th e central role is the importance Raw! indication of moral psychology's s ays, es with self-esteem. This IS, Raw ood of self-respect, which he equat rima les nnap two 's Rawls that fact The (TJ, p. 440). he m t mportant of the primary goods to t itmen comm 's society publicly proclaiming would support citizens' self-respect by e averag over them nt for e in the a gu treating them as ends in themselves is decisiv conceptions . utilitarianism (TJ, pp. l78ff.) and over "mixed . or self-esteem is the most 1 po ant ect self-resp why of Rawls's explanation argument for the two pnnaples, pri!T}'lry good, and hence is suited to play this role in the TJ, part ill. There Rawls argues of ns discussio depends upon the moral and psychological it is shown to be stable until justified that a conception of justice cannot be fully moral support when It own until, that is, it is shown to be capable of generating its th refore regulates basic institutions (d. TJ, p. 6; PL, p. xix). In the third part of TJ Rawls _ ly attempts to show that justice as fairness would be stable. This is an extraordman
� ��
��
: ��
�
�
�
� �
�
�
ambitious undertaking. It requires Rawls to address one of the most difficult problems in moral philosophy: that of showing that it is good to be just or, as he says, that of showing that a sense of justice informed by the two principles is Mcongruent" with a person's good (TJ, p. 513). This, in tum, requires him to develop both a general accou nt of the good, which he calls •goodness as rationality," and an account of the good as it applies to persons. Only with this last account in hand can he explain what it is for persons to have a sense of their own goodness or worth, why it is good for them to have it and how the sense of self-respect would be supported by a just society. Someone has self-respect, Rawls says, when she is secure in the sense that her plan of life is worth executing, and when she is confident that she has the resources, abilities and excellences of character needed to carry out her intentions (TJ, p. 44 0) .
Its possession depends upon affirmation by those with whom one associates, upon public respect for citizens' ends and upon the just adjudication of their claims. All these conditions, Rawls says, would be satisfied in the well-ordered society of justice as fairness
(TJ, p. 442). Conversely, when someone fails to exhibit the virtues necessary to execute his plan of life, he experiences the diminution of self-esteem Rawls associates with shame (TJ, pp. 444-45). Since justice is one of those virtues and knowledge of its possession engenders a sense of self-worth, acts of injustice give rise to shame. This connection between shame, self-eSieem and regulative principles of right, Rawls implies, supports the Kantian interpretation of justice as fairness (TJ, pp. 256, 44 5). John Deigh's article in this volume challenges the connection Rawls asserts between shame and a loss of self-esteem. He ultimately locates the problem i n Rawls's conception of the worth or good of persons, according to which persons have worth depending upon how they live or direct their lives. This conception of worth is, he says, "the wrong one for explaining the sense of worth that makes one liable to shame.• Deigh draws a conclusion about a Rawlsian account of the emotions. He does not draw out the implications of his conclusion for the Kantian interpretation.
If the arguments
in his paper are sound, however, one of the arguments Rawls uses to support that
Interpretation will have to be recast. David Sachs argues that self-respect and self-esteem
are different but related ideas, a point Rawls has since conceded to him (PL, p. 404 ,
VOLUME
INTRODUCTION
Xi
note 39). This difference points to a question Sachs himself does not raise: if self-respect and self-esteem are different, which if either should be included among the primary goods? Larry L. Thomas, like Sachs, seizes on the distinction between self-respect and self-esteem and criticizes Rawls for ignoring it. What part III of TJ relies upon, he says, are the conditions of self-esteem. The conditions required by self-respect, he argues, are quite different. Securing the conditions of African-American self-respect, he continues, was the goal of the Black Consdousness Movement. His discussion suggests that he, at least, would think self-respect and not self-esteem is the most important of the primary goods. The argument that justice as fairness would be stable depends, not just upon an account of the goodness of persons, but also upon an argument that persons growing up under institutions which conformed to the two prindples would develop an effective, corresponding sense of justice. As part of his argument that they would, Rawls sketches an account of moral development according to which persons pass through three stages, which he calls "the morality of authority," "the morality of association," and Nthe morality of principles" (TJ, pp. 462-79). This account of moral development has been disputed by feminist psychologists who allege that it unjustifiably elevates a distinctively male form of moral reasoning.4 In his article, Lawrence Kohlberg, to whose work Rawls says he is indebted (TJ, p. 460, note 6), defends Rawls's claim that the morality of principles is the most advanced state of moral development. Stanley Bates's article is an early appreciation of Rawls's discussion of the development of a sense of justice and its debt to Kant. Reading it gives some indication of how Rawls's moral psychology was received at the time it was published; it remains the best introduction to Part III of TJ.
Progression through the developmental stages Rawls identifies depends upon
the operation of three psychological laws (TJ, pp. 463, 470, 473-4). These laws are laws of redprodty. They assert that human beings will respond in kind to benefits shown them by persons and assodations, and will develop allegiance to institutions which operate for their good. This is not the place to survey and systematize the many ways in which reciprocity surfaces in Rawls's work. What is clear is that for Rawls, the tendency to reciprocate is a "deep psychological fact" about our species which makes it possible for human beings to cooperate on fair terms (TJ, pp. 494-95). It is therefore the psychological basis of the sense of justice. This tendency, Rawls conjectures, has an evolutionary basis: reciprocity has survival value for creatures of our kind (TJ, p. 503). Allan Gibbard, who says he learned from Rawls the centrality of reciprodty to human moral experience,� attempts to substantiate this conjecture in the article reprinted in this volume. The surprising upshot of his argument, Gibbard thinks, is non-cognitivism about justice. Samuel Freeman's very rich article contrasts Rawls's version of contractualism with that of David Gauthier and argues for the superiority of the former. His argument turns on claims about the moral capadties that Rawls attributes to citizens, including the capacity for reciprocity. As noted earlier, Rawls thinks it necessary to show that justice as fairness would be a stable conception of justice, capable of generating its own support. Rawls frames his stability arguments with two audiences in mind: parties in the original position, and readers of his work. Parties in the original position ask whether various conceptions of justice under consideration would be stable and know enough about
V 0 L U M E IN T R 0 D U C T I 0 N
Jdj
the principles of moral learning to make an informed judgment
��J, P· 138). Rawls
�
argues that the likely stability of justice as fairness and the reasons auzens woul come . to support it, are among the reasons the parties adopt it rather than the p naple of . average utility (TJ, p. 177). For their pan, readers of Rawls's work must determme whiCh
�
conception of justice can be put into reflective equilibrium with their own considered judgments. The stability arguments Rawls offers- including his argument that parties in the original position would prefer justice as fairness to average utilitarianism because it can better generate its own support - are intended to convince readers of the superiority of justice as fairness to competing conceptions. This presupposes, of course, that the function of the original position is not merely to represent the moral powers of citizens, but that it also has some justificatory force. Edward Mc C!ennen, however, is sufficiently worried about choice under conditions of extreme uncertainty that he raises the possibility of arguing for justice as fairness without the original position. His article reformulates the stability argument for the superiority of justice as fairness to average utilitarianism without relying on it. Rawls's later work has famously taken a political turn, officially announced in his article Justice as Fairness: Political Not MetaphysicaV6 and cemented in place •
with the publication of PL. Rawls says that he first considered revising justice as fairness in 1977, when Samuel Scheffler sent him a copy of the paper included in this volume (PL, p. xxxiv). The explidtly political turn, he says, was motivated by his own growing
recognition that the treatment of stability in TJ had been inadequate. In the introduction to PL, he says: The fact of a plurality of reasonable but incompatible comprehensive doarines -the fact of reasonable pluralism- shows that, as used in Theory, the idea of a well-ordered society of justice as fairness is unrealistic. This is because it is inconsistent with realizing its own principles under the best of foreseeable conditions. The account of the stability of a well-ordered society in pan III is therefore also unrealistic and must be recast. (PL, p. xix) The stability argument in TJ presupposed, Rawls thinks, that everyone in a well-ordered society of justice as fairness would agree on what he now calls "comprehen si ve philosophical doctrine.• It presupposed, for example, that everyone would accept
Rawls's account of the good in general and the good for persons. Once it is recognized that reaso ble persons will disagree on fundamental questions about the good life, the . p esupposltlon Is seen to be false. Since this disagreement implies that citizens may have fferent grounds for assessing their own worth and since the account of self-respect is VItal to the stabT 1ItY argument of TJ, discarding the presupposition required Rawls to . ._ reVIse the stabliity argument for justice as fairness. He now argues that justice as fairness w�uld be the object of an "overlapping consensus• of reasonable philosophical and . ehglous VIews. Though this argument departs signific antly from the stability arguments m an of TJ, it like them turns on theses in moral psych ology (PL, pp. 141, 163). C !aims m moral psychology therefore remain cruda l to Rawls's defense of J'ustice as f almess.
�
� �
��
� �
The notions Rawls relied upon to effect his political tum, the notions of a
VOLUME INTRODUCTION
Xiii
comprehensive doctrine, an overlapping consensus and reasonable pluralism, are examined in some of the essays included Volume 5. What matters for present purposes is what the new argument for stability is intended to prove. Is it supposed to show that an overlapping consensus on justice as fairness is likely under favorable conditions? Rawls's remark that his earlier idea of well-ordered society of justice as fairness is "unrealistic" might suggest this. In an article reprinted here, however, Thomas Hill maintains that Rawls wanted only to argue that a stable overlapping consensus is possible under reasonably favorable circumstances. That it is possible for a conception of justice to win stable allegiance tells in its favor, Hill thinks, quite apart from whether it is likely to do so. Though Hill does not say so, his imputation of this lesser aspiration to Rawls fits well with other remarks Rawls himself makes. Near the end of the "Introduction to the Paperback Edition" of PL, Rawls writes that the Holocaust and the wars of the twentieth century "raise in an acute way the question whether political relations must be governed by power and coercion alone.· He continues "[i]f a reasonably just sodety that subordinates power to its aims is not possible and people are largely amoral, if not incurably cynical and self-centered, one might ask with Kant whether it is worthwhile for human beings to live on the earth?" (PL, p. lxii). A great deal therefore depends upon showing that a just society is possible. If this possibility cannot be vindicated, it would be reasonable for human beings to acquiesce in cynicism and despair. The task Rawls assigns political philosophy is precisely that of vindicating the possibility. It does so by showing how a just society could win stable allegiance, given reasonable
psychological assumptions like the laws appealed to in part Ill of TJ and in the discussion
of an overlapping consensus (PL, pp. 101, 172).
Rawls argues that a well-ordered society of justice as fairness would realize the goods of community because it would be composed of social unions in which members esteem one another's pursuits and excellences. It would also be a "social union of social unions" in which "the successful carrying out of just institutions is the shared final end of all the members of sodety, and these institutional forms are prized as good
in themselves" (TJ, p. 527). The argument for these claims depends upon a psychological
principle Rawls calls "the Aristotelian principle" and upon his analysis of the conditions
of self-respect. These arguments have not convinced everyone. From the beginning, Rawls was said to have built his theory around the moral psychology of autonomous individuals, a fact clearest, it was said, from the construction of the original position.7 Critics contended that a theory like Rawls's, which seems to be individualist in basis and to privilege autonomy, is more contentious, and less even-handed, than Rawls took his theory to be. Some have also argued that a sodety which privileges autonomy either cannot enjoy the goods of community at all or can enjoy them in only an attenuated way. The essay in this volume by Wayne Proudfoot is an early, but clear and concise, statement of these criticisms. Published in 1974, it anticipated many of the so-called "commmunitarian critiques" of liberalism which were so prominent in the 1980s. The best known communitarian criticism of Rawls's work is Michael Sandel's Liberalism and the Limits ofJustice.8 This volume reprints an essay by Sandel that distills and recapitulates
the more extended critique of Rawlsian liberalism found in Sandel's book.
Responses to communitarian criticisms were not long in coming. Cornelius
xiv
V0 L U M E l N T R 0 D U C T I 0 N
Delaney argues in his paper that justice as fairness is an individualistic conception of
justice only in an innocuous sense. The form of individualism Delaney finds in TJ does
not, he contends, have the damaging implications critics have claimed. Amy Gutmann's article contains relatively little explicit discussion of Rawls. It is reprinted here because it is the locus classicus of the liberal response to communitarian criticisms and is clearly Intended as a defense of Rawlsian liberalism. In a bold and strikingly original piece, Sibyl Schwarzenbach argues that while Hegel is often the inspiration for Rawls's communitarian critics, Rawls's political philosophy bears important similarities with Hegel's. Failure to appreciate these similarities, she concludes, has led to a corresponding failure to appreciate the resources Rawls has to respond to his critics. Roberto Alejandro explicitly takes on Sandel's criticisms of Rawls. He argues that the individualist conception of persons that Sandel finds in Rawls's work rests upon a misreading. Like Schwarzenbach, Alejandro believes Rawls has the resources to reply to his critics, though he cautions that there are costs to drawing on them. The last essay in the volume is by John Haldane. Its relatively recent date of authorship afforded Haldane the benefit of a studied distance from the liberal communitarian debate. Haldane argues, in effect, that that debate depends upon a false dichotomy, at least if the liberalism at issue is the Rawlsian liberalism of PL. The most appropriate form of stability for a modern liberal democracy, Haldane maintains, is provided neither by a robust sense of community nor by an overlapping consensus. Rather, it is provided by what Rawls would call a stable modus vivendi (for this notion, see PL, p. 146£.). The extent to which a society held together by a modus vivendi can be stable, establish justice or realize the goods of community depends upon political culture and so varies from society to society. Haldane writes that political philosophers should take their tasks and problems, as well as their bearings, from the political cultures of the societies they address. This makes of political philosophy a piecemeal enterprise without the synthetic or systematic aspirations of Rawls's liberalism. Paul J. Weithman
Notes 1 John Rawls, A Theory of Justice (Harvard .
u mvemty · · Press, 1971). References to this work will hereafter . . b e given paremheucally tn the body of the text, with the title abbreviated 'TJ· • , J ohn Ra w1s, Poltt!ca/ Liberalism (Columb· · ta u mverstty · Press. 1996). References to this work will hereaf· . ter be gtven par mhetically in the body of the text, with the title abbreviated • PL . � C • • , J oshua �en, h "Democratic Equality; Ethics99 (1989) 2 : 7 7-51. . Carol Gtlhgan , In a Different Voice (Harvard University Press ' 1982) · 'Allan Gtbbard' Wist Choias, A t ret · · p " 1·rnas (Harvard Umver sity Press 1990) 'John Rawls, "Justice as Fairness: Politica l N t Meta hys· cal • Phr./ o ' h an ' p. 1�'Affairs 14 ( 1985): 223-51. p P d p ublic y os . t ' 'Thomas Nagel. "Rawls on Justice. Ph'l o0 h aRMtw8 2(1973):220-34 IOs p�e l ' . . . C c · IS ae ande, l Libtralrsm an d tht Lrmr 'Mih · 15 0J .r1USIIa ( ambridge University Press, 1982).
·
Shame and Self-Esteem: A Critique*
John Deigh Twenty-five years ago the psychoanalyst Gerhan Piers offered what remains the most influential way of distinguishing shame from guilt. Reformulated without terms special to psychoanalytic theory, Piers's distinction is that shame is occasioned when one fails to achieve a goal or an ideal that is integral to one's self-conception whereas guilt is occasioned when one transgresses a boundary or limit on one's conduct set by an authority under whose governance one lives. Succinctly, shame goes to failure, guilt to transgression. Shame is felt over shortcomings, guilt over wrong doings.' More recently, writers who have addressed themselves to the way shame differs from guilt, notably, among philosophers, John Rawls, have characterized shame as an emotion one feels upon loss of self-esteem and have analyzed self-esteem and its loss in a way that bears out Piers's influence.2 Rawls plainly is in Piers's debt. He explains self-esteem in terms of the goals and ideals one incorporates into one's life plans, and he makes this explanation central to his account of our moral personality, in particular, our capacity to feel shame. A characterization of shame like Rawls's, when set in the context of distinguishing shame from guilt, we are likely to find intuitively appealing. And we may feel a further pull in its direction when we think of shame in comparison with other emotions to which it is thought similar-for instance, embarrassment. For we associate both shame and embarrassment with an experience of discomfiture, a sudden shock that short-circuits one's composure and self-possession; yet we would agree, •
1
think, that
1 am indebted to Herbert Morris for helpful comments on an earlier draft of this
article.
1. Gerhart Piers and Milton B. Singer, Shame and Guilt: A Psychoanalytic and a Cultural Study (Springfield, 111.: Charles C. Thomas, 1953) , pp. 1 1 -12. 2. John Rawls, A Theory ofjwtice (Cambridge, Mass.: Harvard University Press, 197 1) , pp. 440-46. For similar views see Helen Merrell Lynd, On Shame and the Search frrr Jtkntiry (New York: Harcourt Brace &: Co., 195 8), pp. 23-24; Raben W. White, "Competence and the Psychosexual Stages of Development," in Nebraska Symposium on Motivationl960, ed. Marshall Jones (lincoln: University of Nebraska Press, 1 960), pp. 125-27; and David A. J. Richards, A Theory of Reasurufrrr Action (Oxford: Oxford University Press, 197 1 ), pp. 250-59.
Etlaics 93 (January 1 983): 225-245 C 1983 by The University of Chicago. All rights reserved. 0014-1704/8319302-0001$01.00
1
226
January 1983
Ethics
��
mfiture that, unlike sha�e. do embarrassment is an experience of disco e sense of worth. An expenenc be not include a diminishment in one's may of worth. Here we shame, by contrast, strikes at one's sense �ell and we were somew at going were s reminded of times when thing lves, when sudde nly, qmte inflated by the good opinion we had of ourse lie to our favorable �if unexpectedly, we did something that gave the in fa� ess fiat�enng assessment, and we were shocked to see ourselves pos1uve sel -1mage light. Such are the circumstances for sham� , and the a negauve one that disappears in these circumstances and IS replaced by spells loss of self-esteem. em arThese contrasts between shame and guilt and shame and W lch, rassment present the bare outlines of a charact�rization ?f sham.e, arucle. when filled out, appears rather attractive. It IS the top1c of this nce, appeara first at attractive though tion, characteriza this My thesis is that . shame of conception dubious a contend, I , represents It ry. is unsatisfacto e In particular, I mean to call into question its central idea that sham signifies loss of self-esteem. The paper is divided into three parts. In the first I lay out what I . shall call the Rawlsian characterization of shame, Rawlsian in that I retam the controlling thesis and overall structure of Rawls's account but do not concern myself with its specifics, an exact rendering of Rawls being unnecessary for my purposes. Though my approach here is largely un critical, my aim is to set up a well-defined target for subsequent criticism. In the second, then, I begin that criticism. I set forth a case of loss of self-esteem and some cases of shame that pose problems for the char acterization. By themselves these cases stand as counterexamples to it, but my hope is that they will have a more illuminating effect, that they will produce a sense or spark an intuition that its central idea is problematic. Accordingly, in the third part I complete the criticism. I draw from the cases two lessons about shame intended to give definition to the intuition I hope will already have been sparked. Each lesson points to a key feature of shame that the characterization leaves out or misrepresents, its central idea being implicated as the source of these failures. Thus, while the criticism of this t ird part is aimed at the target set up in the first, the _ . force ?f the cnUC1sm should lead us to consider rejecting the idea at the target s center.
�
�
�
� �
?
We need at the start to fix our understanding of self-est eem, since the . concept ts at the base of the Rawlsian characterization. To this end I shall p�nt some considerations leading up to a defini tion of self-esteem, f� m wh1ch an ext:>la�ation of its loss will follow direct ly. This will then �eld the charactenzauon of shame we seek. Let us begin with the general 1de t�at self-esteem relates to what one make s of oneself or does with one s life One has self.-teem 'f I one s spmts are h'1gh because one bebeves that one has made or �: w1U make something of oneself, that one has been
�
·
·
•
2
·
·
Deigh
Shame and Self-Esteem
227
or will be successful in one's life pursuits. Conversely, one lacks self esteem if one is downcast because of a judgment that one has failed to make or never will make something of oneself, that one doesn't or won't ever amount to much. Something of this idea is suggested in W illiam James's equation that sets self-esteem equal to the ratio of one's successes to one's pretensions.� The first thing to note in this general idea is that self-esteem connects up with the condition of one's spirits. We speak of vicissitudes of self esteem: highs and lows. One's self-esteem can plummet. It can also be boosted or bolstered. Indolence and languishing in doldrums are signs that one's self-esteem is at a low ebb. Enthusiasm for and vigorous en gagement in activities in which one chooses to participate are signs of an opposite condition. We also describe persons in these conditions as having or lacking self-esteem. And though subtle differences may exist between a person's having self-esteem and his self-esteem's being high and between his lacking self-esteem and his self-esteem's being low, I shall treat the two in each pair as equivalent. A second point of note, which is corollary to the first, is that self esteem goes with activity. But to assert that having self-esteem requires that one be active would be an overstatement. We should allow that the esteem a person has for himself is relative to that period in his life with which he identifies for the purpose of self-assessment. Thus, a person may retain his self-esteem after having retired from active life if he looks back on his endeavors and accomplishments with pride while content to take it easy. He maintains a high opinion of himself while leading a rather leisurely and unproductive life because his self-assessment proceeds from recollections of an earlier period when he was active and successful. Or, to take the viewpoint of a youth looking forward in time, he may have esteem for himself in view of the life he aspires to lead if he believes in the accuracy of the picture he has of his future. He identifies, for the purpose of self-assessment, with the person he believes he will become, his present self having little bearing. Consequently, he may even at the time be leading an altogether easygoing and frivolous life while exuding self-esteem. I mention these possibilities only to set them aside. We simplify our task of explaining self-esteem if we restrict the discussion to self esteem had in view of one's current doings and development. Besides this simplifying restriction, we must also add a qualification to the statement that being active is a condition of having self-esteem. As a third point, then, one's actions, if they are signs of self-esteem, must have direction. They must be channeled into pursuits or projects and reflect one's goals and ideals. A wayward vagabond does not present a picture of someone who has self-esteem. Nor do we ascribe self-esteem to someone who, having no settled conception of himself, tries on this 3. William James, The Principles of Psychology, 2 vols. (1890; reprinted., New York: Dover Publications, 1950), vol. 1, p. 310.
3
228
January 1983
Ethics
ere�t styles, as he would sunglasses of diff and that trait of personality, ' s had b t comf�rtab e lo?k. Self-esteem . to see which gives him the mos fa1rly well defimte d1recuon an some persons whose lives have a fairly have self-esteem reqmres that one defined shape, which is to say that them. values and organize one's life around and ideals, and a settled con aims One's values translate into one's steem. Specifically, we may take stellation of these is necessary for self-e ly, �omeo ne who had this as a precondition of self-esteem. For, argua d1recuon and coher e�ce no aims or ideals in life, whose life lacked the neither an appro pna�e that such aims and ideals would bring, would be understand 1s object of our esteem nor of our disesteem. We would impelle d h1m behavior as the product of primitive urges and desires that his life, he to design or order no given at the time of action. Having We should . purposes d short-live for or would act more or less at random re on literatu ic philosoph recent frequents who recognize in him a figure who urges, and desires of battery a by assailed man the human freedom: of n definitio clear a lacks he because them overpower to is helpless ves mo but once at directions many in impelled is man a Such himself.•
�
?
�
�
�
in no particular one for any great distance. Frustrated and disorie nted by inner turmoil, he lapses into nonaction. He would, were we ever to encounter his like, properly evoke in us pathos indicating abeyance of judgment rather than scorn indicating low esteem for him. By contrast, when a person has aims and ideals that give order and direction to his life, counterpoint between primitive forces that impel him and his wanting to fulfill those aims and ideals becomes possible. Thus, at those times when he acts in conflict with his aims and ideals, he may declare that he was caught in the grip of some emotion or was overpowered by some urge or desire. He would then convey the idea that he had bee� acted upon or compelled to act as opposed to doing
the act or choo�mg to act. Undeniably, the emotion, urge, or desire is . attnbutable to h1m; but by such declaration he disowns it and so disclaims
�
aut orship of th� act it pro�p�ed. Authorship, not ownership, is the key nouon here, that IS, a�thorsh1p m the general sense of being the originator
or creator of somethmg. When one has a settled constellation o f aims and ideals, then one distinguishes between the acts of which one is the au�hor and those in which one serves as an instrument of alien forces.� Without any such constellation, one is never the author of one's actions
d
though many times the instrument of alien forces that act on one, triggere by external events.
�t! joel Ff!inberg, "The ldu of a Frt!e Man," in EdUt;alional Judgmnw: Pap"s in of james Doyle (London: Roudedge & Kegan Paul, 1975), pp . !48-49. n un, Fr dom of the Will and the Concept of a Person," Journal of Pllilosoph, 6 (1971). 5-20; Wnght Neely, "Freed ' om and Desire•" Philo�vr• -nMical RnJUW · 83 (1974): 32-54·'and Cary Wauan, "Free A gency,., Journal of Philosophy 72 (1975): 205-20 . 5 · I h•.ve dnwn here from Harry Fran . Mun, "ldenu.ficauon and Externality " in The Jdntti�Vl of�mons , cd . Amelif! Rony (Berkeley: University of California Press, 1 76), PP· 239-5J .
�
� Ph
i:?
;r
;due�, e.?· �
�
9
4
Shame and Self-Esteem
Deigh
229
It is in view of this contrast that I suggest we take one's having a settled constellation of aims and ideals as a precondition of self-esteem: when one is the author of one's actions, one is an appropriate object for esteem or disesteem; when one is only an instrument of alien forces, one is not. We can then look to this precondition for the defining conditions of self-esteem. So while we would have said, loosely speaking, that self esteem came from one's having a good opinion of oneself, we may now say more strictly that it comes from a good opinion of oneself as the author of one's actions, more generally, one's life. Accordingly, this opinion comprises a favorable regard for one's aims and ideals in life and a favorable assessment of one's suitability for pursuing them. Lacking self esteem, one would either regard one's aims and ideals as shoddy or believe that one hadn't the talent, ability, or other attributes necessary for achieving them. Either would mean that one lacked the good opinion of oneself that makes for self-esteem, and either would explain the dispirited condition that goes with one's lacking self-esteem. These considerations then yield an understanding of self-esteem as requiring that two conditions jointly obtain. This we can formulate as a
definition. Specifically, one has
self-esteem
if, first, one regards one's aims
and ideals as worthy and, second, one believes that one is well suited to pursue them.6 W ith reference to the first we say one has a sense that
one's life has meaning. With reference to the second we speak of a confidence one has in the excellence of one's person. And this combination of a sense that one's life has meaning and a confidence in one's ability to achieve one's ends gives one impetus to go forward. Turning then to loss of self-esteem and, in particular, the sudden loss taken on the Rawlsian characterization to be explicative of shame, we obtain immediately from the foregoing definition an account of this experience. One loses self-esteem if, because of a change in either one's regard for the worthiness of one's aims and ideals or one's belief in one's ability to achieve them, a once favorable self-assessment is overturned and supplanted by an unfavorable one. The loss here is the loss of a certain view of oneself. One had self-esteem and correspondingly a good opinion of oneself: one viewed oneself as having the attributes necessary for successfully pursuing worthy ends around which one had organized one's life. The change in judgment about the worthiness of one's ends or the excellence of one's person destroys that view. One's good opinion of oneself gives way to a poor one. This constitutes loss of self-esteem. The Rawlsian characterization has it that shame is the emotion one feels when such loss occurs. Moreover, shame is to be understood as signifying such loss. Shame on this characterization is the shock to our sense of worth that comes either from realizing that our values are shoddy or from discovering that we are deficient in a way that had added to the confidence we had in our excellence. Either is a discovery of something 6. The: definition
matches Rawls's (sc:e
p. 440).
5
230
Ethics
January 1983
-di scovery of ourse1ves, and suc h self false in the good opinion we had . spells loss of self-esteem. not figure m every expene ce Of course' self-discovery of this sort does ess opinion of himself is non ethe of shame for a person who has a poor poor t that is his re�son for the liable to eel shame when the very defec s y , as a last pomt, we mu . opinion is brought to his notice. Thus whose self-esteem IS a rea Y something about shame felt by someone ham's novel Maug in low . While a schoolboy, Philip Car�y, . clubf�t. Hts feehn� h1s over Bondage, feels shame innumerable Urnes eem 1s low t� begtn self-est his do not involve loss of self-esteem, since But 1t would with, nor, obviously, do they reflect any act of self-d.isc�very. on the grou�d be uncharitable to object to the Rawlsian charactenzauon wlt that it does not cover such cases, for they can be treated on analogy hts on foot cripple his have always cases it does cover. Philip does not e thes 0� . 1t of forgetful is he when mind; there are plenty of times to hable 1s he himself, with comfortable is he occasions, especially when as feel shame when made conscious of his "freakish" condition, when, teem, self-es any lose not does while he Then, it. rediscovers it were, he his being comfortable with himself is certainly lost to him. ·
f
�
t � f!/ Hu�:
�
�
.
II In this section I shall set forth a case of loss of self-esteem and some cases of shame that present real problems for the Rawlsian characterization. 1 begin with the former. The case itself is quite straightforward. We have only to think of someone who suddenly loses self-esteem because he discovers that he lacks the ability to achieve some aim he has set for himself, who is crestfallen, dispirited, and deeply disappointed with himself, but owing to circumstances or a philosophical temperament, does not feel shame. And such a case is not hard to construct.' Imagine, for example, some youth who is indisputably the best tennis player in his community. He defeats all challengers; he wins every local tournament; and he has recently led his high school team to a first-place finish in a league consisting of teams from the high schools of his a nd the neighboring towns. His coach rates him the most promising player to come along in a decade, and he is highly touted by other ten nis enth�sias�� in the ar��· Quite naturally, he comes to have a high opi nion of h1s abthty and V1s1ons of winning tournaments on the profess onal i tour. At some point early in his high school years, he makes profess ional . tenms a career goal and devotes much time to impro ving his game. In trut h though, the gro nds for his high opinion of his ability and for h is � . : deas1�n to �ake tenm a career are shaky. The competition in his a nd � the netghbonng towns 1S rather poor, these being rural and isolated fro m urban centers. And the aging coach's hope s have distorted his judg me nt 7· �pies &imilar to this fint case were suggested to me by Herben Morris Rogers Albntton.
6
and
Deigh
Shame and Self-Esteem
23 1
of his star player's talents. Thus, when this young player enters his first state tournament, he quickly discovers that his skills are below those of the top seeded players. His first defeat need not be humiliating, just convincing. And though he will surely lose some self-esteem, we need not suppose that he feels any shame. One explanation of his losing self-esteem but not feeling shame is this. The first defeat is sufficiently convincing that it alters his view of himself as a tennis player, and given his aims, this means loss of self esteem. But just as others close to him would respond that his defeat is nothing to be ashamed of, so his own attitude toward it may reflect such judgment. Accordingly, he would be deeply disappointed with himself but not ashamed. This possibility becomes even more vivid if we suppose that he has gone to the tournament alone or with friends who, unlike him, have only a passing rather than an abiding interest in tennis. For then he does not find himself having to face someone like his coach before whom feeling some shame would be natural, though even here the presence of the coach does not necessitate the emotion. This case thus broaches the question what distinguishes those cases of loss of self esteem whose subjects feel shame from those whose subjects feel dis appointment but no shame. The inability of the RawI sian characterization
to answer this question implies that the understanding of shame it gives is, at best, incomplete. Let us next take up cases of shame. The first comes from an obser vation, made by several writers, that shame is commonly felt over trivial things. One writer instances experiences of shame had on account of "one's accent, one's ignorance, one's clothes, one's legs or teeth."8 Another, to illustrate the same point, mentions shame felt over "an awkward gesture, a gaucherie in dress or table manners, . . . a mispronounced word."9 To be sure, none of these examples poses a threat to the Rawlsian charac terization, since each of the things mentioned could be for someone a shortcoming the apprehension of which would undercut the confidence he had in the excellence of his person. This would certainly be true of someone who consciously subscribed to ideals the achievement of which required that he not have the shortcoming. For then, though others would disparage these ideals as superficial or vulgar and accordingly think the shortcoming trivial, to him it would still appear as a serious Haw in himself. Naturally, the more interesting case is that in which the subject also thinks the shortcoming trivial and is surprised at having felt shame on its account. This case too can be understood as coming under the Rawlsian characterization. For one need not fully realize the extent to which one places value on certain things, and one may even deceive oneself about one's not being attached to certain ideals. We need, then, 8. Stanley Cavell,
Press, 1 976), p. 2 86.
Mwl We Mean What We Say1
9. Lynd, p. 40.
7
(Cambridge: Cambridge University
232
Ethics
January 1983
�
�
edare wer e one hand, h�t one wou ld . to distinguish between, on the s tf � ne bute ld list as one s Impo rtan t attn wou and ls idea and s aim s one' conc epu o � and , on the other, one's self. were asked to describe oneself e�phcl t any to vior apart from or in additio� as it is reflected by one's beha tdea ls ption I mean th � atms an� self-description. By one's self-conce s one behe life together wtth the around which one has organized one's th t IS tand what we u nder� has about one's ability to pursue them. And . s one ut one's behaviOr witho these aims, ideals, and beliefs can guide n perso a . Consequently . being conscious of having subscribed to them ng of soup, thoug h he slurpi the or teeth d who feels shame over crooke rance and proper would have thought himself unconcerned with appea or good �able form, shows by his emotion that a pleasant-looking face of comeh ness manners are important to him, that he subscribes to ideals signifyi n g or social grace. Hence, we can easily understand his shame as loss o f self-esteem. over At the same time, such examples invite us to look for things made which someone might feel shame though he did not believe they e him ill suited to pursue his ends . Shame one feels over somethi ng on not could not believe affected one's excellence, say, because one could regard it as a fault in oneself, would present a problem for the Rawlsi an
�
�
characterization. Thus, consider shame felt over a humorous surname. The example comes from Gide. He describes to us the experience of a young French girl on her first day of school, who had been sheltered at home for the first ten years of her life, and in whose name, Mlle Peterat, something ridiculous is connoted, which might be rendered in English by calling her Miss Fartwell . "Arnica Peterat-guileless and helpless had never until that moment suspected that there might be anythin g laughable in her name; on her first day at school its ridicule came upon her as a sudden revelation; she bowed her head, like some slu ggish waterweed, to the jeers that flowed over her; she turned red; she turned pale; she wept." 1o With this e�mple we move from attributes that one can regard as . . mor flaws an �nstgmficant defects to things about a person that lea ve � h1m open to ndtcule, though they do not add to or detract from his excellence . The morphemes of one's surname do not make one better or worse suited for pursuing the aims and ideals around which one has organiz � d one's li�e . Hence, shame in this example, because it is felt over somet mg that hes outside its subject's self-conception opposes the ' Rawlstan characterization. e s�cond case of shame is cousin to the first. One finds onese lf . m a sttuauon in which others scorn or n "d"1cu1e one or express som e . d�precatory J�dgment of one, and apprehending this one feels sham e. Gtven only thts general description, such a case presen no real problem
?
�
�
�
10. Andn Gide LojcaditJ's ltdwntures, trans. Dorothy Bussy (New York: Alfred A . . Knopf• 1955) • P· 100 ·• The rendenng of her name in English is suggested by the translator.
8
Deigh
Shame and Self-Esteem
233
for the Rawlsian characterization. I t serves t o remind u s that one's self esteem depends to s ome extent on the esteem others accord one-certain
others, anyway-and the greater that dependency the more readily one will feel shame in response to any deprecatory judgments they express. This can be understood by way of the amount of confidence one has in one's own independent judgments about the worthiness of one's aims and one's ability to fulfill them, for this, we might say, varies inversely with the strength of the dependency of one's self-esteem on the esteem of others. That is, the greater that dependency, the less one's confidence will be in independent judgments one makes about oneself and, con comitantly, the more accepting one will be of the judgments others make about one. Consequently, given a strong enough dependency, if they criticize or ridicule one for some fault, one accepts their criticism and thus makes the same judgment about oneself, where before one did not notice the fault or it did not much matter �o one. This arouses shame inasmuch as the judgment issues in an unfavorable self-assessment that replaces a favorable one, that is, in loss of self-esteem. We have then an account of the case that is fully in line with the Rawlsian characterization. But we must also admit cases of shame felt in response to another's criticism or ridicule in which the subjects do not accept the other person's judgment of them and so do not make the same judgment of themselves. And these cases do present a problem for the Rawlsian characterization. Consider Crito and his great concern for what the good citizens of Athens will think of him for failing to deter Socrates from meeting his demise. " I am ashamed," he says in vainly trying to argue Socrates out of accepting his fate, "both on your account and on ours your friends'; it will look as though we had played something like a coward's part all through this affair of yours." 1 1 And though Crito is in the end convinced that Socrates' course is the right one and knows all along that he has done everything one can expect of a friend, we still have, I think, no trouble picturing this good-hearted but thoroughly conventional man feeling ashamed when before some respectable Athenian, who reproaches him for what
he believes was cowardice on Crito's part. Examples like this one dem onstrate that shame is often more, when it is not exclusively, a response to the evident deprecatory opinion others have of one than an emotion aroused upon judgment that one's aims are shoddy or that one is deficient in talent or ability necessary to achieve them. The third problematic case of shame is this. We commonly ascribe shame to small children. Shaming is a familiar practice in their upbringing; "Shame on you" and "You ought to be as hamed of yourself " are familiar admonishments. And, setting aside the question of the advisability of such responses to a child's misdemeanors, we do not think them nonsensical 1 1 . Plato, Crito 45d-e. Quoted from the Hugh Trednick translation, T� Colkcud Dialogues of Plalo , ed. Edith Hamilton and Huntington Cairns (Princeton, N.J. : Princeton University Press, 1 96 1 ), p. 30.
9
234
January 1983
Ethics
;:' t� �
ore, al capaci�ies. Fur ther m ous in view of the child's emotion or incon e � em to ascnbe sha ers of small children do not hesitate close obs c development, observed t at Erik Erikson, writing about human gan t? deI ve P stage when they acquired a sense of shame at the _ bou harles Darwm, wnung a muscular control and coordination.12 C age n to blush ar?und the blushing, noted that small children bega s ed on occas1ons that shyne "notic had three and later remarked that he g youn ited in the eyes of or shamefacedness and real shame are exhib of blushi ng."!! power the children before they have acquired who at the age of four child ious But it would certainly be a precoc zed his life arou nd or five had a well-defined self-conception, who organi _ d 1deals and the pursuit of certain discrete and relatively stable aims a� them . In measured himself by standards of what is necessary to ach1eve e , d es other words, a child at this age, though capable of feeling sham . � s1gmfy not have self-esteem . Hence, the shame he experiences does not loss of self-esteem. Finally, a fourth problematic case of shame emerges once we juxtapo� the orientation of an aristocratic ethic and that of an achievement ethiC.
�
�
�
�� � �
The Rawlsian characterization with its emphasis on making something of oneself, being successful in one's life pursuits, is tied to the latter. The experiences of shame it describes are at horne in a rneritocratic society one in which social mobility is widespread, or, at any rate, the belief that •
it is constitutes a major anide of faith. On the other hand, some experiences of shame reflect an aristocratic ethic; one feels shame over conduct un becoming a person of one's rank or station. The experiences are better suited to a society with a rigidly stratified social structure like a caste society. And, as we shall see, they stand in marked contrast to experiences the Rawlsian characterization is designed to fit. The contrast is this. With shame reflective of an achievement ethic, the subject is concerned with achieving his life's aims and ideals , and he measures himself against standards of excellence he believes he mu st meet to achieve them. So long as he regards his aims and ideals as worth y ,
they define for him a successful life, and accordingly h e uses the standards to judge whether he has the excellence in ability or of character necessary for success. He is then liable to shame if he realizes that some of his aims
�
and deals are shoddy �r that he has a defect portending failure where prev10 �sly he ad ascn d to himself an excellence indicating success. And th1s fits mcely the 1dea that shame signifies loss of self-esteem. On the othe� ha�d, wi�h s ame reflective of an aristocratic ethic, the su bject's . concern IS wtth mamtauung the deportment of his class and not necessarily
?
�
?e
with achieving aims and ideals that define success in life. He is concerned with conforming to the norms of propriety distinctive of his class, and 12. Erik. Erikson, CAildhood and Socil,, 2d ed. (New York.: W. W. Nonon &: Co., 1963),
pp . 251-54.
Charles ��·
I S .. 1M &fwwion of 1M EIIIOtimu in Man and Animal ( 1 872; reprint ed., Ch101go: Uruvenuy of Chicago Press, 1965), p. 33 1 .
10
Deigh
Shame and Self-Esteem
235
conformity to these is neither a mark of achievement nor an excellence that forecasts achievement. In the usual case one is born into one's class and conforms to its norms as a matter of course. Failure to conform, that is, failure to deport oneself as becomes a member of one's class, invites comparison to persons of lower classes on whom the members of one's class look down. Thus, someone from a social class beneath which there are other classes may be liable to shame over such failure: someone wellborn may be liable to shame if he behaves like the vulgar. And such shame does not fit the Rawlsian characterization. For the subject neither realizes that his aims and ideals are shoddy nor discovers a defect in himself that makes him ill suited to pursue them. In other words, given the analysis we have laid out, he does not lose self-esteem. But, one might ask, can't we say of someone who feels shame over conduct unbecoming a member of his class that he too has ideals that regulate his actions and emotions? After all, with his class we associate a way of life, and this implies an ideal or set of ideals. To feel constrained to act as becomes a member of one's class is to feel pressed to conform to its ideals, and conduct unbecoming a member is, in other words, conduct that falls short of an ideal. Granted, one doesn't so much achieve these ideals as conform to them, which shows perhaps that the conception of self-esteem on which the Rawlsian characterization is built must be modified. But supposing we make whatever modification is needed, isn't it sufficient to bring the experience under the Rawlsian characterization that we can redescribe it as shame felt over one's falling short of an ideal? Something, however, gets lost in this redescription. When we redescribe the experience as shame felt over falling short of ideals around which one's life is organized, our focus shifts from who one is to how one conducts one's life. The subject's identity as a member of a certain class recedes into the background. We see it as the source of his ideals but do not assign it any further part. This, I think, is a mistake. In this experience the subject has a sense of having disgraced himself, which means he has an acute sense of who he is. We do not have an understanding of shame otherwise. It is revealing that on the Rawlsian characterization this shift in focus does not register. For the characterization recognizes no distinction between questions of identity and questions of life pursuits, between who one is and how one conducts one's life. From its viewpoint, a person says who he is by telling what his aims in life are and what ideals guide him through life. 14 This makes it an attractive characterization of the shame felt by persons who are relatively free of constraints on their choice of life
�
pursuits owing to class, race, ethnic origins, and the l ke . For su�h p erso ns . . tend more to regard their aims and ideals as constttutmg thetr tdenttty
�
and their ancestry, race, class, and so forth as extrinsic facts a ut th� m selves. So the characterization explains the shame they feel as mcludmg 1 4. See Rawls, p. 408.
11
236
Ethics
January 1 983
�
ident y because it restricts a. perso.n's an acute sense of who they are. But et e sens to explain as includmg thts to his aims and ideals in life, it fails d s ? n ified society, feels whe � � shame someone, living in a rigidly strat g 10 hvm or the shame some one not act as befits a member of his class ople. p ath the dignity of hts � a multiethnic society, feels when he acts bene deals cond uct falls short of � his that Granted, such a person recognizes s tdeal e follow, but thes members of his class or culture are expected to d coul , ce der for instan do not constitute his identity. Another, a preten y , just as a tombo identity same the have the same ideals as he but not to fail Hence , we has the ideals of a boy but not the identity of one. . over one' s havm g felt being as it describe account for such shame if we fallen short of ideals that regulate one's life. Mas p�e Thus, about the following experience, which Earl Mills, a will mstst Indian, relates, a defender of the Rawlsian characterization ide als that sometime during the episode Mills must have embrac ed the z d, of an Indian way of life or, alternatively, that he must have reali � g um though he nowhere suggests this, that the ideals he was then purs lai were shoddy. But ignoring the Rawlsian characterization , we can exp � s Mills's feeling shame without importing either of these assumptions : ht e c ran having, in the circumstances he describes, to acknowledge his igno •.
�
of Mashpee traditions disgraced him as an Indian, made him betray , as it were, his Indian identity, and this aroused shame. This explanation accepted, his experience directly opposes the Rawlsian characterization, for it suggests that, despite the aims and ideals around which a man organizes his life, circumstances may arise that make him, becau se of an identity he has that is independent of those aims and ideals, liable to experience shame. When I was a kid, I and the young fellows I ran around with cou ldn' t �ave cared less �bout our Indian background. We never participated m any of the tnbal ceremonies, we didn't know how to dance , and we wouldn't have been caught dead in regalia. We thought anyon e . who made a fuss about our hentage was old-fashioned, and we even used to make f�n of the people who did. Well, when I came back from the Army m 1 948, I had a different outlook on such matters. Yo':' �e, there happened to be two other Indians in my basic trammg company at Fort Dix. One of them was an Iroquois from Upper �ew York State, and the other was a Chippe wa from Montana. I w� mnetee� years old, away from Mashpee for the first time in my hfe and, hke most soldiers, I was lonely . Then one night the � �r uo fellow got up and did an Indian dance in f;ont of eve o ne c The hippewa got up and joined him, and when to a mtt �.I d'd 1 n t know how, I felt terribly asha med.'�
� ::a �\�J �
ry
�
1 5. Paul Rrodeur, " A Reponer at Large.. Th e Mashpees," New YorA.tr 54 (November 6, 1978): 62-150, p. 103.
12
Deigh III
Shame and Self-Esteem
237
Before drawing any lessons about shame from the discussion of Part II, I should say something to allay doubts about the import of the cases of shame presented there. Such doubts naturally arise because one might think that some, if not all, of those cases exemplify experiences of the emotion the subjects of which one could criticize for being irrational or unreasonable . That is, while agreeing that many persons are liable to such experiences, one might wonder whether they ought to be so liable and then note that a case's force as a counterexample lessens if it only describes an experience of irrational or unreasonable emotion. The first and last cases of shame are especially in point. To feel shame over one's surname and because of conduct unbecoming a person of one's class seem good examples of shame one ought not to experience. For one is not responsible for one's parentage and thus ought not to judge oneself according to facts wholly determined by it. Inasmuch as shame in these cases reflects such judgment, they exemplify experiences to which one ought not to be liable. These doubts arise under the assumption that, in giving a charac terization of an emotion, one specifies those conditions in which the emotion is experienced reasonably or rationally. Such an approach to characterizing an emotion requires that one regard as its standard cases those in which the subjects are fully rational individuals and not at the time of the experience in any irrational frame of mind. But we ought to question this requirement. Why should we restrict the class of standard cases to these ? While there is, for instance, something absurd in the familiar picture of an elephant terrified at the sight of a mouse, why should this absurdity lead us to regard the elephant's terror as any less important a case to be considered in characterizing that emotion than the terror a lynch mob strikes in the person on whom it takes revenge? To be sure, the elephant is not a creature capable of bringing its emotions under rational control, whereas a human being, if sufficiently mature, is. And for this reason there is a point in criticizing the emotional ex periences of human beings, whereas making similar criticisms of an el ephant's emotional experiences is altogether idle. But this provides no reason to regard the class of rational or reasonable experiences of a given emotion as privileged for the purposes of conceptual inquiry. To have brought one's emotions under rational control means that the range of one's emotional experiences has been modified through development of one's rational capacities : one no longer responds with, say fear, to certain sensory stimuli that before the development provoked fear, and conversely. But far from instructing us to discount the elephant's or the toddler's emotions in our conceptual inquiries, this bids us to examine emotional experiences had in response to sensory stimuli unmediated by rational thought as well as experiences the occurrence of which we explain by reference to rational thought.
13
238
Ethics
January 1 98.3
us pri�arily characterizing shame. To f� _ Similar points then apply to bem g n�au� nal one would not criticize for on cases the subjects of which e tzat � n cing distortion into the charact � or unreasonable is to risk introdu m w lC s to examine closely those case Indeed ' one might be well advised 1 d'tsp ay more grounds that they may . such criticism is forthcoming on the uon. acteristic features of the em� prominently than others certain char l f ine closely the �ham e �yptca � Thus, one might be well advised to exam l sooa iCal , hterarch homo hierarchicus, even though one thought that rigid ht g thou even though one structures lacked rational foundations (i.e., on ment to social class) , nal attach that the emotion indicated an irratio am sh than in the grounds that in such shame one sees more clearly _ of the subject s parts those rian society egalita _ typical of persons living in an . M�reover, self-conception in virtue of which he is liable to the emouon mouon that, though the resultant characterization rendered shame an � one_ ne�er from the perspective of an egalitarian or meritocratic ethtc, enzauon had good reason to feel, this would not in itself show the charact use of to be faulty: no more than that gentlefolk like the Amish, beca r has certain theistic beliefs, regard resentment as an emotion one neve re good reason to feel shows that they harbor misconceptions about al on sentment. Since we are capable of bringing our emotions under rati ble control, we may regard our feeling a specific emotion as incom pati with OUT moral principles and SO try tO make ourselves UO longe r liable
�
h
�
to it. Alternatively , we may regard this emotion as essential to our humanity
and so revise our principles. The conflict makes evident the importance
of having a correct understanding of such emotions; at the same time we should see that altering the understanding one has in order simply to avoid such conflict or the criticism of irrationality would be misguided. Turning then to lessons that come out of our discussion of the problematic cases, I shall draw two. The first is that a satisfactory cha r acterization must include in a central role one's concern for the o pinions of others. 'Ibis is really a lesson in recall. From Aristotle onward, discussions of shame have focused attention on the subject's concern for the opinions
�
�
o ers have �f him 16 Aqu �as, Descartes, and Spinoza each incorporated : thts concer � mto hts defiruuon of shame.'7 And latter-day writers, Darwin _ and Sartre m parucular, took the experience of shame before another as ke� to an understanding of the emotion. 1a Thus, we should not be surpnsed to find that the Rawlsian characterization founders since it ' regards s�ch concern as not internally related to shame . Its fculure, owever,_ is not due to neglect. The characterizatio n, through emphasts on the dependency of one's self-esteem on the esteem of others, can accord the concern an important role in an overall un-
�
16. For Aristode'a view S« Rheloric bk 2 cha , . , . P· 6 . For 1 7 . '! definition · 1 a2ae, 4 1 ,4. For Descanes's see The 17w 1ogtae p4SSiofls 11uAqwnas'a Soul 2 artid S« S • pt· • e _66 . For Spmoza s see The Ethics, pt. S, definition 5 1 . 1 8. See last chap. of Darwin. Sartre's view is . found in . a'lld Nothangmss, trans. 9 Bang Hazel E. Barnes (New York.: Philosophical Lib rary, 1nc., 1 56) , PP· 2 52-502.
�
�
14
Deigh
Shame and Self-Esteem
239
derstanding of shame. 19 B u t this makes the concern part o f a mechanism
that induces shame rather than part of our conception of shame. A mechanism exists which, when put into operation, transforms high self esteem into low; part of that mechanism is the concern one has for the opinion of others; and one way in which the mechanism gets going is when others on whose good opinion one's self-esteem depends deprecate one and one apprehends this. In this way, the characterization gives one's concern for the opinion of others an important role. B ut it is only a supporting role and not the central one I think it deserves. And this is one reason for its failure. Each of the first three problematic cases bears out this last point. It is evident in the second and third cases, where the subjects feel shame but do not lose self-esteem. In the third case shame is felt directly in response to another's scorn or reproach. Thus, an expressed low opinion of the subject induces in him shame without affecting his self-esteem. In other words, the mechanism is not engaged, though the subject's concern for the opinion of another is clearly operative. In the second, Mile Peterat, even apart from the context in which she feels shame, jeering classmates, feels the emotion because of something about herself that is laughable. It invites deprecatory responses. Thus, she may feel ashamed because of it, even though it is not a deficiency. It is not a ground for reassessing her excellence, though, of course, the whole ex perience could cause her to think less of herself. Here, too, there is shame reflecting a concern for the opinion of others without the mechanism's being engaged. We can also mine the first case to bring out the point that the Rawlsian characterization has misconstrued the role one's concern for the opinion of others has in shame. Consider again our young tennis phenom. I n the circumstances described, h e loses self-esteem, i s disappointed with himself, but does not feel shame. On the other hand, as we noted, if the circumstances had been different, if he had had, say, to face his coach after the defeat, then his feeling shame would have well been imaginable. There would then have been someone at courtside whose look he could not meet. He would have averted his eyes, lowered his head, gulped to fight back tears. That the coach's presence could spell the difference between disappointment and shame cannot be explained by reference to the player's losing self-esteem, for the loss occurs in either case. The mechanism would be in operation whether or not the subject felt shame, so it would not account for the role his concern for the coach's opinion would have had in his experiencing shame. We can thus conclude from these three cases that one's concern for the opinion of others has a role in shame apart from the way in which their opinion can support or bring down one's self-esteem. 19. See, e.g., Rawls's discussion of the companion effect to the Aristotelian principle,
pp. 440-41.
15
240
Ethics
January 1983
h . The Ra wlsia n charThe second lesson is about our sense of wort . of worth according , an understand'mg of a person's sense . acterization Vlf'lds tha t he has . conviCtiOn to which it has two sources. One ts t.he person• s h e h as .m his o w n gt·ven meaning to his life. The other l S the confidence . . ding h ts atm s and excellence as a person. The first comes from h.lS r� gar . tha t he is well ideals in life as worthy. The second comes from hts .behef . . . ract en za t 10n ' suited to pursue them. Thus, accordmg to the Rawl stan cha . . . o r 1'd eals shame since it is felt e1ther upon a JUdgment that 0ne's atms . at makes are shoddy or upon a judgment that one ts de fiClent m a w ay th . , s sense one to k h s a 'bed as oc one ill suited to pursue them, 1s aptly descn . . . ' . . one' s s ense of worth smce of worth. One expenences a d1m1mshment m , ence d fi con s ne o or life either one's sense of having given meaning to one's .-
·
•
·
·
� �: �
in one's excellence has been struck down. . . There is difficulty in this, however, because, while the descn t of shame as a shock to one's sense of worth is apt, the accou nt 0 various ways in which the sense gets shocked is, at best, too meag er . T h reason for this is that the characterization omits important sources 0 our sense of worth. The point is directly evident in our last two c� ses. The child of four who feels shame over some misdemeanor has not gtve n
meaning to his life and does not have confidence in his excellenc e as a person. Hence, he has a sense of worth the source of which th� ch ar acterization does not acknowledge. Similarly, we recognize in an anstoc::at who feels shame over behaving like a plebian or in an American In dta n who feels shame over betraying his Indian identity a sense o f worth the source of which is neither a conviction about the worthiness of his ends nor a belief about his suitability to pursue them. A sense of wo rth that comes from knowledge that one is a member of the upper class or a noble people also lies beyond the sight of the Rawlsian character izati�n. To put the point generally, the Rawlsian charac terization fails to reco gmze aspects of our identity that contribute to our sense of worth independe ntly of the aims and ideals around which we organize our lives . We should note here the structural, as well as the substantive, diffe rence between the sense of worth the Rawl sian characterization reco g ni zes and the one it excludes. We can get at this structural difference by looking at the theory of worth that unde rlies the characterization. That theory is based on a conception of us as the authors of our action s. We are authors in the sense discussed in Part l, that is, in virtue of hav ing a constellation of aims and ideals according to which we live our lives . We have worth on this theory in accordance with the value of our lives, such as they are and such as they pro mise to be. An author has worth in view of his wo�k, completed or in progress , and our lives, so to spea are o u r k, work 1s an�logy can be pressed. Our work has value to the degree : . that 1t 1s the kind of thmg that when well made has value and is itse lf well made So we have wo to the degree that we prod : uce such thin gs or have dtrec�ed our energtes toward producing them and pos sess the talents and skills that augur successful production. Ou r live s, con ceiv ed
�
�
16
Deigh
Shame and Self-Esteem
24 1
as our work, thus have value to the degree that the ends that give them order and direction define a kind of life that has value and those ends have been realized. And we have worth as authors of our lives to the degree that we live lives of value or have directed our energies toward living such lives and possess the attributes that promise success. In capsule form, what might be called the auteur theory of worth is that what a person does with his life, how well he directs it, determines his worth. On this theory, then, we attribute different degrees of worth to someone depending on how valuable we deem the kind of life he lives and how successful we think he has been in living it or how suitable we think he is for it. In other words, we attribute to him more or less worth according to how well or badly he conducts his life. Contrast this with attributions of worth made because of one's class or culture. Judging from these attributions, we might say that a person's worth is determined by his status in the context of some social hierarchy. The salient feature here is that one's status, and so one's worth, is fixed independently of one's conduct. To be sure, one can change classes through marriage or cultures through immigration, but short of this the general conduct of one's life, that is, however well or badly one conducts it, does not increase or decrease the worth that is attributed to one because of one's status. And pretty much the same holds of worth that is attributed to human beings because of their s pecies or to persons because of the kind of beings they are conceived to be: rational ones, say, spiritual ones, or autonomous ones. That is, worth attributed to one because of one's essential nature
is, like worth attributed to one because of one's status, fixed independently
of how one conducts one's life .
Consequently, the dynamics o f t h e sense of worth that comes from knowing the worth that goes with one's status or essential nature, that is, the understanding we give to augmentations and diminishments in that sense, are altogether different from those of the sense of worth the auteur theory recognizes. Statically, both kinds of sense correspond to the degree of worth one attributes to oneself. But an augmentation in one's sense of worth, as is experienced in pride, or a diminishment in it, as is experienced in shame, is not, if this sense originates in a recognition of one's status or essential nature, to be understood in terms of a n attribution t o oneself o f greater o r lesser worth than one attributed to oneself before the experience.20 A college boy who wears his fraternity pin with pride does not regard himself as having greater worth for having worn it, and a man who feels ashamed of having eaten like a pig does not regard himself as having less worth than is attributed to human beings as such. This contrasts with the way the auteur theory would have us understand augmentations and diminishments in one's sense of worth. In particular, it would have us understand a diminishment in one's sense 20. Of course there are exceptions to this, e.g., the white supremecist who discovers he has a black ancestor.
17
242
Ethics
January 1 983
�
self-estee shame, as amounting to loss of of worth as is experienced in les� wort ssment of oneself as havi ng · and so c rresponding to an asse on e h reflects conc ern �1th On the auteur theory, a sense of wort ce 0f and appearance as evtd en real worth, and one takes one's conduct h. wort uting to ones elf that or more strongly as the grounds for attrib or s tu t a one •s s · know mg By contrast a sense of worth that comes from one s conduct en betwe ency congru essential nature reflects concern with the rn could say, one's c� nc� or appearance and one's real worth. Here, we htg one's sta�us lS is with the relation between appearance and reality. If behav tor that lS then noble, is relative to that of others or one's nature for prid e , and ' congruent with one s worth and so displays it is occasion ce of lesser behavior that is at variance with it and so gives appearan
�
�
•
'
'
•
�
worth is occasion for shame. sensibility This model better accommodates the idea that to have a verges to shame means that one is prepared to restrain oneself when one the on the shameful and to cover up the shameful when it come s into no open . We speak in this regard of having shame as opposed to hav ing ty shame, and we connect this with modesty, particularly sexual mode �
•
which involves a sensibility to shame in matters of decorum. Havmg shame, that is, having a sensibility to shame, can be understood here as self-control that works to restrain one from giving the appearance of lesser worth and self-respect that works to cover up shameful things that, having come to light, give one such appearance.21 This suggests that we should conceive shame, not as a reaction to a loss, but as a reaction to a threat, specifically, the threat of de m ean ing treatment one would invite in giving the appearance of someo ne of les se r 22 w�rth. Its analogues then are, not grief and sorrow, but fear and shyneSS . d nte Like fear, shame serves to protect one against and save one from unwa exposure. Both are in this way self-protective emotions. Fear is self protect ve in that it moves one to protect oneself against the danger on e senses ts present or approaching. From fear one draws back, shi elds
�
�
one �lf, or ees. Of course, it may also render one immobile, th ereby putung one m greater danger, so the point does not hold witho ut qual2 1 . On th�� points, � Carl Beacon Pr�ss. 1977), pp. 24-27.
D. Schneid�r. Shame, Exposure and Privacy (Boston:
22. W��th�r to patt�rn sham� aft�r grief and sorrow or aft�r f�ar and shyness is an . tssu� a. r�vt�w of th� lit�ratur� r�v�als. On� oft�n finds in the writings of thos� off�rin g . · d�fimuons of sham� use of on� or th · � oth�r of these emotions as analogues, sometimes . . . �v�n as a ��n�nc emo on of whtch shame i& defined as a specific type. For definitions of � f or sorrow se� Ho�s (Leviathan chap. 6) and Desca�es (pt . . 3 . � . ug . ?assag� ts equtvocal smce he also says there that shame IS a speCies . · · of modesty) For defimuon of shame as a type c of .ear �e Aqumas; it is also suggested tn Plato•s Euda ro \ 2 . In connection with this issue see Havelock. Ellis's "The Evolution of Mod�sty. m Stud�LS in 1M p h 1o of S � 0 gJ ex, 2 vols. 3d ed. (New York: Random H ouse , 1942) vol 1 ' P· 36 ' n. 1 . Ell'IS htmself appe ars to hold that shame is a kind of f�ar ( see PP · 36-52 .
:'::; :s� : ro 0� �: � 72) .
�
18
Deigh
Shame and Self-Esteem
243
ification.2� Still, the general idea is clear. Shame, too, is self-protective in
that it moves one to protect one's worth.24 Here the general idea is not so clear, though a trope may be useful. Shame inhibits one from doing things that would tarnish one's worth, and it moves one to cover up that which through continued exposure would tarnish one's worth. Less fig uratively, we might say that the doing or exposure of something that makes one appear to have less worth than one has leaves one open to treatment appropriate only to persons or things tha t lack the worth one has, and shame in inhibiting one from doing such things and in moving
one to cover them up thus protects one from appearing to be an unworthy creature and so from the degrading treatment such appearance would invite.
This idea that shame is a self-protective emotion brings together and explains two important features: first, that a liability to shame regulates conduct in that it inhibits one from doing certain things and, second, that experiences of shame are expressed by acts of concealment. The second is crucial. Covering one's face, covering up what one thinks is shameful, and hiding from others are, along with blushing, the most characteristic expressions of shame. Students of shame commonly note them. A quote from Darwin is representative, " U nder a keen sense of shame there is a strong desire for concealment.''25 Moreover, etymology reinforces the point. According to many etymologists, a pre-Teutonic word meaning 'to cover' is the root of our word shame.26 Now the Rawlsian characterization, since it conceives shame as a reaction to a loss, can explain, on the model of fear of loss, how one's liability to shame regulates one's conduct. Where it has trouble is in explaining shame's moving one to cover up and hide. For it does not have in itself the materials needed to construct such an explanation. Because it conceives shame as a reaction to the loss of something one prizes, it yields an account of the emotion as at first giving way to low spirits and dejection and eventually moving one to attempt to recover what one lost, that is, to regain through self-improvement one's good
opinion of oneself and so one's self-esteemY Acts of concealment, however,
are nowhere implicated in this account. Hence, if one adheres to the characterization, one must make use of supplementary materials to explain them. One must go outside the characterization by, say, citing certain fears associated with shame : fear of ridicule or rejection by those upon 23. I owe this point to John T. MacCurdy, "The Biological Significance of Blushing and Shame," British )ourrwl of Psychology 2 1 ( 1 930): 1 74-82. .. 24. The idea is one of the central themes of Max Scheler's essay "Uber Scham und Schamgefiihl ," in Gesammelte Werke, ed. Maria Scheler and M. S. Frings, I I vols. (Berne: Franke Verlag, 1954), vol. 10, pp. 65--154. 25. Darwin, p. 320. 26. See Oxford English Dictionary, s.v. "shame"; also Ernest Klein, A Comprehensive Etym ological Dictionary of the English Language (Amsterdam: Elsevier, 1967). 27. See Rawls, p. 484; Lynd, pp. 50-5 1 ; and Richards, p. 256.
19
244
Ethics
January 1 983
, zs But such an lf-esteem d epen ds whose good opinion of one one s se al ent fails to explain acts of.conce � explanation would not be adequate, for it f lears o s k these as expres sion as expressions of shame. lnstead , It ta _es ex any for tion would hold associated with shame. And the same objec cter cha � outside the planation one constructed from materials found explam as to e unabl is , words other in ization. The characterization, us that somethmg expressions of shame , these acts. And this should tell has gone wrong. appears to . be The adherent to the Rawlsian characte rization thus that .covenng in an untenable position. We would dismiss any suggestion ressiOn� . �e up and hiding were not really among shame's natural ex p et es, Rection on shame, particularly shame concerning sexual Improp � � auon nz alone suffices to rule this out. And we should reject any charact� se th of an emotion that misrepresents its natural expressions. Faced with � g posm objections, the adherent might retreat to a weaker thesis by pro all, not that the characterization gives an adequate account of some, but the experiences of shame. But this thesis is no more defensible than of ase original. For our adherent, as we saw from the first problematic � . on zau Pan II, has the burden of showing how the emotion the characten describes is distinguishable from disappointment with oneself. Since he ·
.
•
.
admits on this weaker thesis that some experiences of shame elude the characterization, he has, in other words, the burden of showing that the experiences of emotion the characterization captures are classifiable with these as shame. What reason could he give to show this? That they h�v e the same feeling-tone is itself questionable, insistence on the point bem� question begging. That they involve a shock to one's sense of worth IS insufficient. For the characterization identifies this shock with one's suffering loss of self-esteem, and this by itself does not qualify an c;xperience as one of shame. The trouble with this proposal, I think, is that it wo uld, in effect, divide shame into disparate kinds, one kind having fear as its analogue, the other grief. That is, we should suspect of any conceptio n of shame the proposal spawned that it covered a mismatched set of experiences. We can trace the characterization's problems back to the understanding . . It gtves to the sense of wonh that makes one liable to shame and ultimately to the auteur theoiJ' of wonh, which grounds that unders nding . On ta that th�ory one attnbutes �o oneself worth accord ing to how one conducts . one s hfe, and so perceptions of that conduct determine one's sense of worth. Shame then, since it is felt upon discov ery of shortcomin gs in oneself that falsify the �orth one thought one had, includes a sense th at one lacks worth. And thiS proves problematic because it leaves unexplained . how sham� mou�ates acts of concealment. By contrast, when we conceive shame as mcludmg a sense that one has worth, we can readily explain 28. See Pic� and �. P· 16; and Rawls, p. 5 44 . White, however, expresses reservations . t connccun agams g shame to such fears (pp. 12�27).
20
Deigh
Shame and Self-Esteem
245
its motivating such acts : one covers u p because one senses that the worth one has is threatened. This speaks in favor of the understanding of the sense of worth the idea that shame is a self-protective emotion entails, which understanding is grounded on a conception of worth that opposes the one the auteur theory yields. Consequently, we should suspect that the conception of worth the auteur theory yields is the wrong one for explaining the sense of worth that makes one liable to shame, and, because the Rawlsian characterization presu pposes that conception, we should give up the view of the emotion it represents.
DAV I D SACHS
Ho w to Di sti n g u ish Self -Re spe ct
fro m Self-Este e m
I f you tell someone that he should respect a certain person you may or may not be telling him that he should esteem that person . If you tell someone that he should respect everybody you surely will not be
telling him that he should esteem everybody . These facts suffice to show what, in any case, ought to be obvious : that respectin g pers ons and esteeming them are not the same . Perhaps the princip al or e ven y the sole reason for supposing they are the s ame is that freque ntl . "respecting someone" has the sense of "esteeming someon e ." Thu s , if
you say you respect so-and-so greatly, you are saying you hold him in e high esteem, that you regard him as indeed praiseworthy ; that you ar saying anything else is indeed doubtful. Parallel remarks would ofte n apply were you to say you have little respect for so-and-so; often, that is, you would be saying you hold him in low esteem or regard him unfavorably. Moreover your saying this would be compatible with your adding that, of course, you respect him as you respect-or en· deavor to respect-persons generally , including many whom you hold in much higher esteem.
, The situation is similar with self-respect and self-estee m. Th at is although at times "self-respect" and "self-esteem" are used inter·
changeably , self-respect and self-esteem are not identical. Thus if you complain that someone is lacking in self-respect you may not be de· ploring any lack of self-esteem on his part. It is not unheard-of for
persons to esteem themselves highly and for others who are fully t aware of that fact to complain of them that they are notably defici en © 1g81 by Princeton Philosophy & Public
Univenity Press AffaiTs 1 0, no. 4 0048-391 5/SI /040346-15$00.75/ J
22
347
How to Distinguish Self-Respect from Self-Esteem
in self-respect. Respect and esteem , then , are differen t , whether it i s a m atter o f others or of oneself. But how d o they differ? There also can be little doubt that they are importantly related to each other. What are some of the main relations between them ? In the following I try to answer these questions chiefly as regards self-respect and self esteem ; however, I shall incidentally make some observations about differences and connections between respecting others and esteeming them . I
Often it is thought of individuals that they have too much self-esteem , and there is no difficulty in understanding the thought. But it is not often thought-indeed it seems a hard saying-that a person has too much self-respect. 1 That a person esteems himself overmuch is easy to grasp regardless of whether he esteems himself too highly in gen eral or in some particular connection or connections. But that an individual respects himself too highly in general or in some specific way is a puzzling notion . There is, it should be noted, nothing prob lematic about the notions of someone's h aving either too little self esteem or too little self-respect . The perplexing though t , again, is that of an excess of self-respect. Yet counsels of self-preservation and self-advancement , advice re lating to the security and flourishing of those dependent upon one , recommendations as to what would enable a c ause one supports advice on these and other matters m ay include, in certain situation s , one's being urged t o sacrifice a measure o f one's self-respect. Through others' counsel , one's own deliberation, or in other ways, one may come to think th at certain situations warrant the sacrifice or surren der of some of one's self-respect. Just conceivably one migh t proceed from that reflection to another : perhaps one can h ave too much self1 . To say of a man, not that he has too much self-respect, but instead that he h as too much self-respect to do or submit to such-and-such is not to attribute
to him an excess of self-respect. To critici:z:e a man for having too much self respect to do or submit to such-and-such is to complain of his loftiness or stuffiness or of some kindred pretension; such uses of "too much self-respect"
are sarcasms.
23
Philosophy & Public Affairs
respect for one's own good-or the good of one's family or the promo tion of a cause, and so on. And from such thoughts it m ay be no large step to supposing there can be such a thing as too much self-respect. But to what would that supposition amount here? It would amount to supposing that there are .circumstances when one should not be averse to other persons disregarding, without good reason, one. s wishes, cases when one ought not be inclined to take exception to the flouting of one's rights, situations when one should not resent or find . shameful being used or degraded, or m anipulated or explmted.
�
Clearly, there is a confusion involved in this sequence of thoug t.
Judging a measure of self-respect excessive and dispensable, that lS, an
encumbrance, is confused with judging it something that may
more or less painfully have to be sacrificed.
Certainly it is plausible that a variety of situations can occur in
which a person may reluctantly have to sacrifice a measure of his self respect. (To be sure , Kant, the moralist of the pride of self-respect, speaks with conviction against one's ever doing so. ) But if one con cedes that situations occur when it is at least arguable that a person should do so, there is no argument or consideration thereby offered to the effect that he ought not be averse to it, resent it, find it shameful . Or to the effect that he ought not to hope and strive for social and per sonal circumstances in which sacrifices of his self-respect will neither be nor appear necessary. In light of the foregoing, one can see that it is not just that the notion of an excess of self-esteem is easily comprehended , while that of an excess of self-respect is perplexing. What may have appeared a conceivable and even faintly plausible way of understanding the notion of too much self-respect turns out to be nothing of the kind . Instead, what emerges i s the notion of possibly h aving to m ake a more
or less painful sacrifice of some of one's self-respect-a sacrifice of part of what is, in its entirety, v aluable. Clearly, this is not a con
ception of anything excessive. In contrast, consider an excess of self esteem. Whereas, again, there may well be no such thing as unwar ranted self-respect, there is, of course, such a thing as unwarranted self-e steem. If a person is disabused of an unduly high estimate h e has of himself, he gets rid, perhaps at some cost in dismay or cha grin ,
of a misconception of himself ; and this is likely desirable. But p ain-
24
349
How to Distinguish Self-Respect from Self-Esteem
fully surrendering some of one's self-respect is not being disabused of any misconception or falsehood. One is subjecting oneself, with re luctance and resentment, to an abuse or abuses. I h ave not, it should be noted, claimed that the notion of an excess of self-respect is unintelligible . I have repeatedly claimed it is problem atic. In effect I have contended that, within the framework of the notions of self-respect and self-esteem which we understand and ap ply to ourselves and others, what may just seem a plausible way of making sense of the idea of an excess of self-respect does not bear scrutiny. So far as I am able to tell , there is no other even faintly plausible way of making sense of the notion within that framework . Stepping outside of that framework-outside of what proves to be the structure of our pre-analytic comprehension of, and concern for, self respect and self-esteem-may, however, yield a sense for the notion of too much self-respect.2 Th us from a cynical standpoint, or possibly from one or another utilitarian perspective, it could be argued that some such notion is intelligible. To sketch an argument to that effect would not , however, serve any of my purposes here. Here I want merely to observe that, if such a conclusion follows from any doctrine whatever, it appears that this would show that the doctrine involves a deformation of our understanding of and concern for self-respect. II I turn now to a systematic difference-the range of possible diver gences-between self-respect and self-esteem. If one takes pride in something , one is, to some extent or other, proud of it; moreover, one will probably feel proud of it in the sense of experiencing feelings of pride and related feelings in connection with it. Conversely, to be proud of something is to take some pride in it, and so, again, one is likely to experience feelings of pride and re lated feelings in connection with it. These claims seem unquestion able , whether or not what one takes pride in or is proud of is some thing upon which one prides oneself. Imagine, now, a man who is not proud of anything. There is 2.
In speaking of our pre-analytic concern for self-respect I emphatically d o
not mean concern for merely conventional respectability.
25
3 50
airs Philosophy & Pu blic Aff
e hime does no t perceiv nothing whatever in which he takes pn.d e. H are advanta ges-t here self as a person of any accomplishments or as f . . . not to be thou gh t o. none' at any rate , of which he IS proud . H e IS ns I·t no occasiO experiencing feelings of pride ; there are, as h e sees ' d pn or feel for him t o do so; not even perhaps, occasions t o take neither proud 0 vicariously . Although the man takes no pride in-is vably be f lse nor feels pride about-anything whatever, it could concei you With· to claim of him that he has no pride. To the questiOn , "Are he could out pride?" it is possible that he could truly answer no; that htly pro protest as undeserved the reproach, "Have you no pride ? " Rig t th t tha test it on the ground that he has his pride , and sincerely add ThiS is nothing of which to be proud , let alone a matter of being proud. he is to claim that it could be categorically true of a person both that. e. Id takes no pride in anything whatever, and yet that he has his pr .m _a This is a real, however remote or lamentable , possibility because well-understood sense of "pride," what is required for one to h ave Is pride, not to lack it, is to possess one's self-respect. To do so , again . y carries no implication about being proud. It is compati ble with ver s nt cou little self-esteem , compatible indeed with self-disesteem on all but one-one which can be psychologically , and of course mor ally , of
�
•
�
�
�
great importance : the possession of self-respect . To continue with the man I am imaginin g : he disesteems himself generally with the exception-the absence-of one ground for dis esteem, that is , that he has allowed or even h as had to allow in
roads upon his self-respect. But he does not regard retaining one's self-respect as worthy of esteem ; he neither esteems himself nor e x pects others to esteem or admire him for it. Conceiv ably , he m ay vie w as grounds for self-esteem cases of either of the following kinds : fir st,
keeping one's self-respect in the face of sore urgency or severe temp tation to sacrifice a me asure of it; second , after havin unhapp y sur il g rendered it to some extent , making earnest efforts to restore it. But he
knows that neither of these has been the case with him. And he believes that neither of them has been the case with great numbers of persons. The man I am imagining thinks of himself as a norma lly rational and ordinarily effective, self-respecting agent ; one , moreove r , who has n ot been so beset by temptation or driven by exigency that he has sacri· ficed, or even seriously considered sacrificing , his self-respect to an y
26
35 1
How to Distinguish Self-Respect from Self-Esteem
important extent. In this or some equivalent way he ch aracterizes himself, and he extends the same characterization to much of m an kind. But again , by way of accomplishment s and natural and social advantages , he judges himself to be distinctly and generally inferior or, at best, mediocre. There may be no one who, for all or even most of his adult life , satisfies the description of my imaginary man. But I believe th at ap proximations to the type exist. At the least it is likely that some actual persons , in more or less protracted phases of their lives, have seen themselves quite in the way I h ave pictured him as seeing himself. Although the strong contention-that there are people very similar to the man I h ave imagined-is, I think, true, a weaker contention-th at he represents a psychologic ally possible type-will suffice. It will suf fice, that is, for one extreme of the possible divergences between self esteem and self-respect : self-esteem wholly or almost wholly missing , together with self-respect, t o put i t a s it i s sometimes p u t , intact. I shall try now to sketch the other extreme of possible divergence between self-esteem and self-respect. Toward the outset I noted that some persons are thought of as esteeming themselves highly but also as notably lacking in self-respect. Why not imagine, among possible types of persons, a man whose self-esteem knows no limits apart from those set by sanity-a m an , moreover, to whom the description "wholly lacking in self-respect" would be applicable? There are, however, dif ficulties with any such conception. First, the idea of sane , unqu alified self-esteem is troublesomely vague. '1 Next, to imagine a total absence of self-respect is more difficult than may be thought. What limits does sanity impose upon self-esteem ? Certainly, being sane allows for extravagant conceit and inordinate vanity. In allow ing for extravagant conceit, being sane allows for somewhat illusory beliefs about the magnitude of one's advantages or accomplishments, about the magnitude, for example, of one's intelligence, social stand ing, or role in an enterprise. And in allowing for inordinate vanity, 3 · As I trust will be clear, I
am
le aving aside lunatic self-esteem, for ex
ample, delusions of grandeur. Throughout I take no account of deranged or
mentally defective humans. To them too, of course, considerations of respect and esteem and also and importantly, self-respect and self-esteem, regularl y
pertain.
27
airs Philosophy & Pub lic Aff
352
e of ions of the importan� being sane also allows for absurd exaggerat - g s, indeed , for pnzm one's advantages and accomplishments, allow e. h1ch sam· ty co unt . them to an irrational degree. The extent to w conceit or the Irra nances either the illusory beliefs connected with uncertain . Some tional prizing that typifies vanity is, evidently , very . · al"1tY or 1nc on· instances of self-esteem that involve illusion or 1rrat10n sanity and sistency or more than one of these will be compatible with decidable. other instances not, while yet other instances may be un ·
•
Thus, as I said, the matter is troublesomely vague. n wh o I tum now to the second difficulty , that of imaginin g a perso wholly lacks self-respect. In regard to the total absence of self-respect, two somewhat con· nected thoughts may present themselves. First, a point of fact : when persons are said to lack self-respect utterly, those who say it of th�m , if there i s any truth at all in what they say, are likely t o b e m akin g
wild overstatements, indeed, resorting t o abusive hyperbole. Second. however, it may be thought that there could be a person of whom t what they say would literally, without any exaggeration, be true . B u could there be; could anyone wholly lack self-respect ? To im a gine someone could, I am taking it, is to imagine a socialized , rational human, not an imbecile, who is without any self-respect whatever.
Were there to be such a person, he would not find it reason for re
sentment that persons ignore, capriciously or even blankly ignore , what they know to be his wishes. He would not be inclined to object to any impositions; not, that is, because they were impositions. Nor would he be averse to submitting to anything on the ground that it was degrading. Also he would not resent or be indignant about the flouting of any rights he possessed; not, that is, for the reason th at rights of his were bein g flouted . To be sure , he could dislike , be an gered by, or resist treatment of any of those kinds, but not in that they
were of those kinds. (Nor, it should be added , would he find it shame ful or con�emptible that he was quite prone to self-deception or weak ness of will ; of course he could regret bein quite prone to either of
g
them.)
What, it may be asked with some bafflement, is so difficult about imagining such a person? Does not history show that multitude s of
m topeople have existed in circumstances which have reduced the
28
3 53
How to Distinguish Self-Respect from Self-Esteem
or never permitted them to rise from-the condition of total absence of self-respect ? That is, does not history and many a contemporary chronicle show that there h ave been multitudes of socialized, rational humans to whom the characterizations in the foregoing paragraph categorically apply? What history su ggests about this matter can, I believe, be helpfully thought about as follows. First , there very likely h ave been societies whose codes of conduct did not include the notion of a right. ' If it is claimed that, even so, members of those societies had "natural" rights , presumably they could not avail themselves of appeals to them , resent or be indignant about violations of them , and so on. A fundamental social provision for-and one of the bases of our conception of-self-respect would be missing. Members of those societies could be said to have a relatively impoverished c onception of self-respect. They would not suffer losses of self-respect due to their being unable to resist violations of their righ ts ; for of these, again, they would have little or no notion. Second, institutions of slavery and caste, and multifarious social and legal practices that make for inferior degrees of citizenship or commun al membership also involve the absence of certain provisions for self-respect. But usually, if indeed not always, slaves, the lowest castes, and lesser-status citizens are conceived of and conceive of themselves as possessing some rights, however few and pitiable they may be. If such unfortunate persons did not resent violations of those rights, and if they did not regard voluntary submission to abuses of them as more or less shameful, they would thereby show a lack of self res pect. But here it m ay again be urged that the life-circumstances of many such persons have been so m arked by deprivation that they h ave been unable to attain, let alone preserve, any self-respect. And, it m ay be said, when one notes among the historical inhumanities of men to men the perh aps distinctive contributions m ade , for example, in the four centuries of the Inquisition or in our century, how can it be doubted that there are and have been persons wholly without self respe ct? The mistake involved in this line of thought i s that it neglects to con sider the possibility that, no matter how deprived or degraded a 4. See H.L.A. Hart, "Are There An y Natural Rights?"
64, no. 3 ( April 1955 ) : 1 7 5-91, esp. 176-77 ·
29
Philosophical
Review
Philosophy &
354
Public Affairs
as havin g equ als person may be, he yet may have or be imagined s at least the peers-in deprivation and degradation . There is alway of others whose Possibility, and often the actuality , of the presence ve as comwretchedness any particular wretched person will perce1 respec t , he Parable to his own. If, then , he were utterly to lack self· the · . m d auon would relate to any of his real or possible equals m degra onn i ways mentioned earlier : he would not resent their arbitrarily Imposi his wishes; he would not regard any impositions of theirs as compel tions ; he would not find anything whatever they might ask or m he of him degrading. There would be-there could be-no one to who �elf would respond in any of these ways if be literally wh olly lacke ed aliz i respect . It is indeed difficult, I submit, to imagine a rational , soc ·
.
·
�
�
�
human of whom all this would be true .
In trying to set forth the extremes of possible divergen c e between self-respect and self-esteem the following position h as been reach ed. On the one hand , persons who are without, or virtually without, self esteem but whose self-respect is intact are clearly conceiv able . On the other hand , persons who wholly lack self-respect are hard to im a gine , if not in fact unimaginable . It is plausible, therefore, to assume that
everyone has some measure , however pathetically little it m ay b e , of self-respect. The question now is, How much self-esteem-vaguely limited by sanity-is compatible with a minimum of self-respect? In
answering this qestion one ought not focus solely on person s whose deficiency in self-respect is due, at least in part, to social condition s
of coercive and extreme degradation . One ought also think of rel a tively free persons who are yet slavishly, and more or less gener ally , dependent upon others. To suppose a minimum of self-respect is to suppose some sensi tiv ity, however slight, to considerations of self-respect-few and re stricted as they will be . But that, in tum , is to suppose the pos sibility of a perceived occasion or occasions of loss of self-respect. And , should such occasions occur, they will tend to affect self-esteem adversely.
Losses of self-respect, then, may somewh at qualify a person's self esteem even if he has only a minimum of self-respect , a minimum of iniJ 1um of considerations of self-res ct. concern for a But apart pe from t at qualificat1 and, again, within the vague limits imposed by saruty, an otherw1se boundless self-esteem seems compa tible with
�
�
�
�
30
355
How to Distinguish Self-Respect from Self-Esteem
a minimum of self-respect. This, then , will be the other extreme of possible divergence between self-respect and self-esteem. Obviously, there is much room for divergences between the possible extremes, divergences that are less great. III
Let us return to the man who does not view his having maintained his self-respect as deserving of any esteem. He would be a man who
does not pride himself upon having his pride. And again, apart from what he may think of as exceptional cases, he does not accord esteem
or admiration to others for their h aving maintained their self-respect.
Characterizing a man in this way, together with the further charac
terization that he views himself as generally inferior in terms of ad
vantages and attainments , provided one of the possible extremes of divergence between self-respect and self-esteem. In fact, actual per sons who maintain their self-respect will generally account themselves deserving of esteem for doing so; moreover it may be presumed that they would regard it as a ground for self-esteem whether or not they supposed it exceptional. It is, after all , not a requirement for some thing's being a ground or reason for self-esteem, or for esteeming others, that it be unusual or extraordinary. However, my imaginary man thinks it is a requirement, and one or another psychological con dition , such as a pervasive feeling of inferiority, may h ave to be in voked to accommodate his thinking it is one . ( Incidentally, it does seem to be a requirement for admiration ; and there are, unhappily, persons who live by standards such that they do not esteem them selves in any connection in which they judge themselves less than
admirable . ) The point to be stressed, however, is that in the case of
this imaginary man the possession of self-respect still constitutes a ground or reason for self-esteem ; that it fails to serve as one for him does not mean -in his case or in any c ase w�atever-that the retaining of one's self-respect is not a ground for self-esteem. The possession of one's self-respect, I am t aking it, then , is in every case a ground for self-esteem, and, in the usual case, contributes
to it ; in the usual case, that is, a man who has his self-respect will pride himself upon it. The pride taken in it can, of course, vary
31
Philosophy & Public Affairs
�
ealous attitude of fiercel i widely-from some self-contentment to .an . thus . 1t 1s lf -respect 1f independence. Contrariwise, any lessemng of se so even 1' f it is not perceived will occasion a loss of self-esteem; and ( As IS only to be experceived, furnish a ground for some loss of 1t. n s that bear pected, self-deception thrives in the area of matter s, ized , tha t losses of self-re spect. ) If a loss of self-respect is recogn invited, mif a person sees himself as having allowed, perhaps even. . . ate d and m1t1 cursions upon his self-respect , indeed, sees that he h as gs of selfpursued a course that entails some loss of it, then his feelin y selfdisesteem will be experienced as shame and, more paru. cul arl when e contempt. Analogously, a person who enjoys his self-respect , m m finds other persons lacking in it, may well not simply bold th 1f ap disesteem. He is likely to feel some contempt for them and, propriately related to them, to experience some feelings of shame . p I have stressed that, in every case , retaining one's self-respect su ·
'
·
·
� �
•
�
�
plies a ground or reason for self-esteem ; and that any loss or lack
�f
self-respect is a ground for self-disesteem. In doing so I h ave agam
�
relied on our pre-analytic comprehension of and concern for se respect; first, on the fact that, in general, persons who retain tbeu self-respect esteem themselves to some extent on that account , taking it that there is no question but that they are right to do so, and that they likewise disesteem themselves for what they view as their lapses or failures of self-respect; second, on the fact that they e steem and dis esteem others for the corresponding reasons and in corresponding ways. Earlier, when discussing the problematic notion of excessive self-respect, it appeared that the idea of an excess of self-respect in volves a deformation of our common understanding of self-respect. Similarly , it seems that a distortion of th at common understandin g would be effected by the idea that a lack or loss of self-respect provides no ground for self-disesteem. (The man I imagined e arlier would also see it thus ; but he would not count his having fully retained his self respect a ground for self-esteem. ) If this is well taken, it points to a further and fundamental difference between self-esteem and self respect. For neither a lowering of one's self-esteem nor any enhance
ment of it furnishes any ground or reason whatever for one to come to possess less or more self-respect. Thus if a man no longer takes
pride in something upon which he had once esteemed himself-realiz-
32
357
How to Distinguish Self-Respect from Self-Esteem
ing he was not, for example, quite the gymnast or algebraist he once thought he was-his later lower estimate does not in itself constitute a ground for any lessening of self-respect. So far as his lower self estimate goes, h e is warranted in expecting his wishes-barring, as always, reasonable objections-to be honored as much as before,a in
expecting his rights to be acknowledged as fully as ever, and in resist
ing and resenting, as always, any attempts to use or manipulate, ex ploit, or degrade him. Similarly, in whatever ways a person's self esteem may be augmented, none of them affords a reason or ground for greater self-respect. For it is not to be forgotten that, although one certainly can h ave too little self-respect it is surely doubtful that one can have too much ; that there can be an excess of it. If that is true,
it is true regardless of any lessening or heightening of one's self esteem. IV Some additional complexities about self-respect and self-esteem will emerge in what
I shall now, by way of conclusion, try to do : to under
cut misgivings to the effect that I am not doing justice to the interde pendence of self-respect and self-esteem. In this connection it may be worth examining a specious difficulty
about self-respect. Consider the following remark : "For example, if a
man says
'I could never respect myself again if I did X,' and then he
does X, we think he ought to lose his self-respect."0 This remark ends
on a jarring note ; that jarring note may help to point up a complica
tion in our use of "'self-respect." It is, of course, possible for a man to
do something such that he will thereafter not be able to respect him self; and other persons, on learning what he has done, may correctly s. To honor a person's wishes, unless there is good cause not to do so, is basic to respecting him, whereas to honor a person is to express esteem and admiration for him, or to hold him in honor and accord him admiration. Hon-
0_ring undertakings and debts at least strongly suggests respecting such obliga tion s; indeed , I believe it constitutes part of what it is to respect other persons ; in any case, it carries no suggestion of esteem. Honoring one's parents, as Scripture enjoins, seems to involve special concern as regards respecting them. 6. R. S. Downie and E. Telfer, Respect fOT Persons ( London : Allen and Un
Win,
Ig6g),
p. 84.
33
air s Philosophy & Public Aff
�:�
e e his self-respe�t . How conclude that he will, as a result, have lost or to respect h1msel it is also clear that, if a man is no longer able fol· hmg done , not lost his self-respect because of something he has or he will then be lows as to how concerned or relatively unconcerned er ct . Amon g oth ought to be about matters that bear on his self-respe eats further thr possibilities , he may be steadfast in withstanding any involve any to it and staunch in the face of temptations that would is unabl� t� further lessening of it. So the apparent difficulty is : if he , h ow IS It respect himself, if he has irretrievably lost his self-respect ? By now possible for him to be resolute, even zealous, in guardin g it that, to it should be obvious that this difficulty is a verbal one ; and phrases resolve it, one need only recognize that in various contexts the em or "self-respect" and "respect oneself" are used to signify self-este �d esteeming oneself insofar as that is grounded upon self-res pect ; that, accordingly '1oss of self-respect" sometimes signifies self-dis
�
esteem insofar as a ground or grounds for it are losses or failings of self-respect. To be sure , a man may disesteem himself for the re
mainder of his days because of some grievously shameful loss or
failure of self-respect. To suppose it could not be true of him both that he had forever lost his self-respect and that he there af ter un yieldingly maintained it would be to mistake for a contradiction a
play upon uses of "self-respect." Toward the beginning I observed that "self-respect" and "self esteem" are at times used interchangeably; to that observation I h ave here added that sometimes "self-respect" signifies self-e steem insofar as it is predicated upon self-respect ; and a parallel observ ation about '1oss of self-respect" and self-disestem e , namely, that '1oss of self
respect" sometimes signifies loss of self-esteem insofar as th at is predicated upon a loss or failing in self-respect.
To tum now to a more substantial question about the interdepend· ence of self-respect and self-esteem, one relatin to fairly familiar g phenomena : changes of manner and condu ct that seem to evince in· creased self-respect subsequent to an enhancemen t of self-e steem. and changes in at tude and conduct that se em to e xpress diminished self-respect followmg upon a loss of self-estee . These m phenomena may suggest that the psychology of some perso ns is such that to some degree their self-respect-their "inner worth ," to deploy Kant's phrase
�
34
3 59
How to Distinguish Self-Respect from Self-Esteem
-varies for them with their self-esteem. For them, having one's pride appears to depend in some measure on how deserving of esteem they take themselves to be in connections other than those that constitute having one's pride. Further, some of the persons who appear to ex hibit these phenomena, whose psychology seems to be of this kind, tend to think that what warrants changes in self-esteem is to at least some extent good reason for parallel changes in self-respect. Con sistent with this, their respect for other persons may seem to vary somewhat in accord with what they take to be good reason for its varying : that they rightly enough esteem certain persons more than oth ers, and the same persons more at some times than at other times. In sum, within a range of self-respect and similarly as regards respect for others, some individuals seem not only to subordinate respect to esteem, but also to regard that subordination as reasonable. They think it good reason for more or less respecting oneself or another that there is good reason for more or less esteeming oneself or another. The phenomena of the apparent subordination, in some measure, of self-respect to self-esteem, together with the opinion that reasons for self-esteem provide reasons for self-respect, may appear to call into question the claim I made earlier-the claim that neither an increase nor a decrease in self-esteem supplies any ground or reason for one to com e to possess more or less self-respect. The suggestion is that there may be a greater interdependence between self-respect and self-esteem th an th at claim allows. To begin with the phenomena. First, they are far from universal. Second, concerning many instances of them , it is doubtful that self respect is actually increased or diminished, appearances notwith st anding. For if a person's enhanced self-esteem were really to carry in its train an increase of self-respect on his part, then for example, he would c ome to regard being servile as shameful, and as no less sh ameful in the past, when he did not esteem himself so highly. He Wo uld come to regard others' attempts to get him to be servile as at tempts to be resisted and resented, and equally to have been resisted and resented earlier. Otherwise his change in manner and conduct will not signify increased self-res pect. Instead, it will signify the decorum an d conduct that he thinks suits · a person as deserving of esteem as the one he has become . It will be a matter, so to say, of status rather 35
Philosophy & Public Affairs
than self; of the manners and deeds he thinks appropriate to one who has what he now has to be proud of, and not of his coming, in any greater degree, to have his pride. Usually it is this kind of change, a merely specious increase in self-respect, that occurs when it seems that enhanced self-esteem has led to increased self-respect.
Finally , in cases of seeming decline in self-respect subsequent upon
a loss in self-esteem, the change again is usually one in self-esteem
alone. Self-respect is much more likely to be revealed as h aving been
less than it was thought to have been, and not to h ave actually di minished.
Thus the matters canvassed in this last section serve only to cl arify
and confirm the distinctness of self-respect from self-esteem.
I am grateful to Gerald Barnes for helpful comments on earlier drafts of this essay.
RAWLSIAN SELF-RESPECT AND THE BLACK CONSCIOUSNESS MOVEMENT LARRY L. THOMAS
One of the most important effects of what is called the Black Consciousness Movement is thought to be the enhancement of the self-respect of blacks. But surprisingly enough, the analysis of self-respect which we find in John Rawls ' book, A Theory of Justice, cannot account for the fact that this is so. ' And consider the case of the famed black educator Booker T. Washington. Alth ough it is undoubtedly true that few blacks of his time ever attained his prominence, it is not, however, a foregone conclusion that he was not an uncle tom, and so that he failed to have self-respect . 1 But according to the Rawlsian analysis of self-respect, we could have no reason to suppose that Washington lacked self-respect, given his prominence . As these brief remarks would suggest, I think that Rawls' analysis of self respect is defective. And as I hope to show, this is so because Rawls has failed to distinguish between self-respect and self-esteem. Now the distinction between self-respect and self-esteem, while often overlooked by philosophers, is one which psychologists frequently make and which, to some extent, is reflected in ordinary discourse. 3 For instance, the distinguishing characteristic of an uncle tom, or servile persons gener ally , is said to be a lack of self-respect, and not self-esteem. And we do not say "No self-esteeming person would do such-and-such , " but rather "No self-respecting person would do such-and-such. " To be noted, also, is that We can esteem persons only for (what we take to be) their accomplishments or abilities. There is no such thing as esteeming a person just because he or she is a person. But we can respect persons just because they are persons. There seems to be a sense in which we can respect all persons equally - the rich, the poor, the talented, and so on. At any rate, I suspect that one reason why the distinction between 303 37
LARRY L. THOM AS
have to do us is that both conc epts self-respect and self-esteem often escapes sense of self- respe ct has a with a person's sense of worth: a person with of the Wha t is the natur e worth , and so has a person with self-esteem . vide a ? l hope to pro difference between these two senses of worth that what we have satisfactory answer to this question as I attempt to show m t, but self-estee in A Theory of Justice is not an account of self-respec
instead. conviction that Rawls defines self-respect (self-esteem) as having the to Rawls , one one's plan of life is worthwhile (178, 440). And accordin g nal plan of comes to have this conviction in two ways: (l) by having a ratio finding one' s life which satisfies the Aristotelian Principle, and (2) by association person and deeds appreciated and confirmed by people whose life is to one enjoys and who are likewise esteemed (440). A rational plan of ugh t of. be understood in the way that rational planning is traditio nally tho on as For Rawls remarks that it is " . . . the plan that would be decide d up t of the outcome of careful reflection in which the agent reviewed, in the ligh and all the relevant facts, what it would be like to carry out these plans e thereby ascertained the course of action that would best reali ze his mor fundamental desires" (417). The Aristotelian Principle reads as follows: other things equal, human beings enjoy the exercise o f their realized me nt �apacities (their innate or trained abilities) , and this enj oy mcrease � the more the capacity is realized ' or the greater its complexity (426).
�
on And the rist�telian Principle has what Rawls refers to as the compa ni effect which iS that
�
s we witness the exercise of well-trained abilities o f others , t hese . iSplays are enJo ed by us and arouse a desire that we should be able � . to do the same thmgs 0urse1ves. w e want to be hke those persons w ho . can exercise the abilities that we find laten t in nature (428).
. Now the important th s t0 bear m mmd r garding Raw ls ' (so-called) theory of self-respect is t (1) and are not dependent of each o ther · m Rather, (2) holds true of a person only if holds true of that person ' s plan . . of life. Although Raw1s does not say this e�priCitly, it seems clear that we should understand and-rh m th"iS way glVen the various remarks which : he makes throughout A 1 eory of Just1ce .
�!
(1)
·
(l�
(l) .
'
304 38
.
�1)
�
RA W LS I A N S E L F-R E S P ECT A N D T H E B LA C K CONSCIOUSNESS M O V E M E N T
Our self-respect, Rawls maintains, depends upon the respect of those with whom we associate ( 1 78, 44 l f) . If this is so, then the obvious question to ask is " When are those with whom we associate disposed to respect us?" The answer, it appears, is that our associates are disposed to respect us when they see that our plan of life satisfies the Aristotelian Principle, that is, when they witness the well-trained abilities which our plan of life displays. For as the companion effect would suggest, when others witness these abilities, they come to judge that our plan of life is worthwhile. And for others, in particular our associates, to judge that our plan of life is worthwhile is for them to respect us. For Rawls thinks of a person as a human life lived according to a plan (408). These considerations would seem to indicate that , as I have remarked, ( l ) and (2) are not independent of one another. What is more, it should be obvious, given just these con siderations, that the Aristotelian Principle and its companion effect are the backbone of what Rawls presents as a theory of self-respect. Now the Aristotelian Principle is relative to the abilities of persons . Thus for any two persons S and N, it does not follow from the fact that S and N have similar plans of life, that their plans equally satisfy the Aristotelian Principle. For if the abilities and talents of S far exceed those of N, then although they are in pursuit of similar plans of life, S's plan will satisfy the Aristotelian Principle to a lesser extent than N's plan. For S's plan will not realize the abilities of S to the fullest extent. Since the Aristotelian Principle is relative to the abilities of persons, Rawls seems to think that all persons can have the secure conviction that their plan of life is worthwhile. But not so. Consider two plans of life K and J. And suppose that K, on the one hand , is such that all persons have the ability, and know that they have it, to pursue K. On the other hand, suppose that J is a plan of life which most lack, and know that they lack, the ability to pursue. Given the Aristotelian Principle and its companion effect, only persons who successfully pursue the J plan of life are the ones who should have a more secure convicton of the worth of their plan of life. For they are the ones whose plan of life will exhibit the well-trained abilities which others lack , and so have their plan judged to be more worthwhile by others. For by hypothesis, the K plan of life is such that everyone knows that they could success fully pursue it i f only they should try. Thus, in contrast to the J plan of li fe, the K plan will appear less attractive to persons. Now Rawls would like to think that it does not matter that some persons are able to pursue plans of life which are beyond the reach of others, so long as each pers on can find an association of persons which affirms the plan of life that he or she pursue (44 1). But as Rawls' further explanatory remarks 305 39
LAR RY L.
THOMAS
. . . t that som e persons are make clear, what 1s meant 1s th1s: It does not mat er reach of other s, so long able to pursue plans of life which are beyond. the . . ilar to theu ow n. 1 each person associates with those whose ab1 1t1es are s1·m to think •. see the For in so doing, a person does not, so Rawls would like contentlOn here plan which she pursues as lacking in worth. But is Rawls ' sound? I think not. • W� can s e For we are not oblivious to the abilities of people around us. c1ate lac . that others have abilities which we and those with whom we asso , given hls feel And although each person amongst a group of associates may ber �f mem or plan of life, that her abilities equal those of any other • m mber the group, it may be symptomatic of the group itself that each me y ver be contrast to the plans of life of non-members, views her abilities to y el em minimal. Moreover, what about persons whose abilities are extr f o minimal? Is the answer to their having a secure conviction of the worth their plan of life, and so their self-respect, that they associate with onlY ·
�
�
·
those whose abilities are equally minimal? Surely not! Now as my discussion of Rawls' account of self-respect should m ake clear, it is mere wishful thinking to suppose that the self-respect of all persons can be equal, if self-respect consists in having the conviction that one' s plan of life is worthwhile . But is the sense of worth thus grounded that ' which we should want to call self-respect? As I think the case of Boo ker T. Washington and the Black Consciousness Movement will show, the answer to this question is a negative one. As I have already remarked, few blacks ever attained the prominence of Booker T. Washington, who lived during his life time. As a black who founded an educational institution, the Tuskegee Institute, and who w as consulted by virtually everyone concerning monetary appropriations to blacks,' Washington surely had to have the conviction that his plan of life was worthwhil e. For he was universally esteemed for his accomplishm ents . Few blacks, and for that matter, few whites could boast of similar accomplishments for which they also were universally esteemed . And so if ( 1) in finding our person and deeds appreciated by those with whom we associate we come to have the conviction that our plan of life is worthwhile, and (2) in having this conviction we thereby have the sense of worth which constitutes self-respect, then it should be a settled question that Washington was not an un� le tom, and so failed to have self-respect, since on any one' s account Washmgton stands as one of the most esteemed blacks of his time . But it is not a settled question as to whether or not Washington was an uncle tom. For his famed "Atlanta Exposition Address" renders Washington vulnerable to the view that he was an uncle tom.'
306 40
RAWLS I A N S E LF- R E S PECT A N D THE B LA C K CONSCIOUSNESS M O V E M E N T
Of course, I do not mean to say that Washington was an uncle tom. Rather, I claim only that it is not a foregone conclusion that he was not one. And if this is so, then either ( 1 ) or (2) of the preceding paragraph is false. (1) is surely not false, thus i t follows that (2) i s false. Turning now to the Black Consciousness Movement (BCM), one of the most important changes which came about as a result o f this movement was a change in blacks' self-description from "colored" or " Negro" to "black" . Of course, there were other changes as well. • But I shall take the change in blacks' self-description as indicative of the fact that the self respect of blacks was enhanced as a result of the BCM. But is this so because blacks came to pursue more worthwhile plans of life, as Rawls' account of sel f-respect would have us to say? Clearly not. The change from "colored" or "Negro" to "black " has a significance which cannot be fully explained by the fact that blacks came to pursue more worthwhile plans of life. To be sure, many blacks did come to pursue more worthwhile plans as the governmental, educational, and business institu tions of America began to hire blacks for their more status positions. And there is no question but that the i ncreased appreciation of many aspects of the black culture by the American society has made it possible for many blacks to pursue more worthwhile plans. The increased appreciation of music generally characterized as "soul" music proves to be the most striking example o f this . ' But the plans of many blacks remained unchanged with respect to their worth. That is to say, many blacks who pursued a plan of life of little worth proper to the BCM did not come to pursue a plan o f greater worth as a result of the BCM. Yet one thing seems clear: The BCM enhanced the self-respect of not only those blacks who came to pursue a more worthwhile plan o f life , but of those who did not as well. The change in blacks' self-description, then, was not indicative of the fact that they came to pursue more worthwhile plans of life, though many in fact did. Rather it indicated a more fundamental change, namely, a change in the way blac ks came to view themselves as persons qua persons. For the BCM was a rej ectio n of a conception of persons according to which to have a certain pigm entation of the skin was ipso facto to be less worthy of the rights and liberties to which other members of the American society had been so long accustomed. The adoption of the self-description of "black" serves to indicate that blacks came to view themselves in this way. And it is in so com ing to view themselves that, as a result of the BCM, the self-respect of blacks was enhanced. Together, the case of Booker T. Washington and the BCM show that having a plan of life which is worthwhile, and hence one that satisfies the 307 41
LARRY L. THOMAS
�
��
Aristotelian Principle, is neither a necessary nor su fficient con ition for having self-respect. And if this is right, then the sense of worth whlch Ra l calls self-respect is not that at all. As I remarked earlier , I want to sa� t a _ lS not what Rawls calls self-respect is self-esteem instead. And to say th1s simply to make a dogmatic assertion. For as I have noted, we esteem persons for their accomplishments and abilities. And it should be clear from my discussion of Rawls that, in the last analysis, his account o f self-respect turns upon persons being esteemed for just these things . For to have the conviction that one's plan of life is worthwhile is to have a sense o f worth with respect to one's accomplishments and abilities. Given that the � ense of worth thus grounded is that which is properly called self-esteem th1 s mu�h � . _ is evident: Self-esteem is not a sense of worth which we are JUStlfled m having in virtue of the fact that we are persons. 1 0 Turning now to self-respect, it i s perhaps obvious from m y remarks concerning the BCM that I want to say that self-respect is a sense of worth which we are justified in having in virtue of the fact that we are persons. Moreover, though, 1 want to say that it is a sense of worth which should be equal among persons. But what aspect of persons can serve as the basis for this sense of worth'? The answer , I believe, is the moral status o f persons. An obvious, though far from trivial, fact about the world is that not every thing in the world has the same moral status . The moral status of some things, most notably those which are inorganic, is nil. One can do no moral wrong to a book lying on the side of the road. It has no rights, nor are there duties or obligations owed to it. To be sure, a moral wrong is done when, in the absence of permission to do so, a person wilfully destroys a book knowing that it belongs to someone else. But the moral wrong done in this case to the person to whom the book belongs, and not to the book itself. Sentient animals which are non-human have, so it would appear, partial moral status. For at the very minimum, it may be said that persons have a moral duty not to treat such animals cruelly. Some, in fact, would want to say that such animals have a moral right not to be treated cruelly . I shall not enter the dispute as to whether or not animals have moral rights. Suffice it to note, however, that the so-called moral rights of animal£ are conceptually predicated upon the fact that persons exist. For it is all but absurd to speak of non·h �man animals as exe�cising or acting within their rights. Consider also that 1f, as happened dunng the Mesozioc era there were animals b ut not persons, it would certainly be out of place to s ak of animals as hav ing a right not to be treated cruelly. And even though animals are said to violate the territorial space of others or defend their own, we do not countenan ce
�
308 42
RA W L S I A N S E LF- R ES P ECT A ND T H E B L A C K CONSC I O U SN ESS M OV E M ENT
these activities as the violation or defense of rights of any kind. Finally, there are those living things, namely persons, which have full moral status. Their full moral status is evidenced by the following: First, the moral rights of persons and the duties which persons have towards one another are not conceptually predicated upon the existence of any other form of life; and secondly, the moral duties which persons have toward any other living things are a subset of the moral duties which persons have towards one another. And this remark holds mutatis mutandis with respect to non-human animals having moral rights, if indeed this is so. Now it is clear that the moral status of a thing determines, in part at least, the modes of conduct towards it which are proper. It is precisely the fact that a piece of wood has no moral status at all that one can do no moral wrong to it, however one treats it. Similarly, it is because sentient but non-human animals have some moral 'status that it is not morally wrong to kill them for the purposes of food, although it is morally wrong to inflict needless pain upon them . And, finally, it is just the difference between the moral status of non-human animals and that of persons in virtue of which it is morally wrong to treat persons as mere property which can be bought and sold for labor, but not non-human animals. Now on my view the sense of worth which is properly called self-respect is grounded in the fact that persons have full moral status. Self-respect, I want to say, consists in having the belief that one has and is worthy of full moral status. And to believe this is to believe that one is as worthy as any other person of the set of moral rights to the the recognition of which persons are entitled . At this point there are two questions which immediately arise: (1) In virtue of what i s i t that persons are worthy o f full moral status? And (2) What are the moral rights of which persons are worthy? I shall discuss these questions in tur n. To ou r first question, the answer, I believe, is that persons are worthy of full m oral status in virtue of the fact that they are self-conscious beings. I shall not attempt to fully vindicate this reply here; instead, I shall offer a few considerations which indicate that it points in the right direction. We may thin k of a self-conscious being as one which has a concept of itself as a continuing subj ect of experiences and other mental status, along with the Hef that it is such a being. It is one which as the capacity (a) to identify With the past , present, and future stages of itself, and (b) to formulate and execute long-term plans of action. 1 1 No w i t seems that moral systems can address themselves only to beings �ho h ave this capacity, since a fundamental characteristic of moral systems Is what we may refer to as temporal indifference: If at time t, R is a moral
t>:
309 43
LARRY L. THOMAS
things being ��ual) (i) reason for doing act A in circumstances C, then (other time, and (n) R was R is also a moral reason for doing A in C at some future . also a moral reason for doing A in C in times past 1 1 who has �he Accordingly, a being with full moral status must b e one it for domg to s capacity (a) to recognize that a reason which presently applie · (other 1' t 1s s act A in circumstances C can be, as in the case of moral system . C, Am things equal) a reason which will apply to it in the future for doing . cordmgl� . and (b) to formulate and execute long-term plans of action a� mo.re, �t This is something that self-conscious beings can do. But what 1s hty lS seems that it is precisely among self-conscious beings that mora sel.f possible. These considerations do not, of course, show that 10 consciousness is a sufficient ground for having full moral status . But a as � tha indicate offering these considerations my aim has been only to . 10 ts pom ness sufficient ground for having full moral status, self-conscious the right direction. I should like to think that at least this much has been accomplished. . I now take up the issue of the moral rights of which persons are worthy 10 virtue of the fact that they are persons . To begin with, it should be clear that such rights are not, to adopt H .L.A. Hart's terminology , special moral rights. u For these are rights which rise out of special transactions between individuals, as in the case of promising; and so they cannot be the sort of rights which we have in virtue of the fact that we are persons. But needless to say, it would hardly be informative to say simply that the m oral rights of which persons are worthy qua person are non-special rights, in the absence of an enumeration of such rights. Yet, aside from the so-called natu ral rights, it is anything but obvious as to what a complete list of such rights would look like. Fortunately, though, my view of self-respect is in no way affected by this difficulty. True enough, I want to say that self-respect consists in having the belief that one is worthy of one' s moral rights in virtue of the fact that one is a person. But from this it does not follow that a person has self-respect if and only if he knows what all the moral rights of persons are, and therefore his own moral rights. For having self-respect is not so much the ability to give a run down of what one' s moral rights happen to be, as it is a conviction of one's standing in the mo�al community . 1 4 Differently put, then, we may saY that to have self-respect 1s to have the belief that one is a member of good standing in the moral community in virtue of the fact that one is a person. �nd although this belief entails. the belief that for any non-special moral . worthy of M , 1t does not call for knowing what all the nght M , one 1s non-special moral rights happen to be. I want now to bring into sharper focus the connection between my view
3 10 44
RA W L S I A N S E L F- R E S P ECT A N D T H E B L A C K CONSCIO U SNESS M O V E M E N T
of self-respect and the BCM. Although self-respect pertains to our moral status, we must not lose sight of the fact that it is amongst social institutions that we live, that we are members of a society. And as others have noted, the view that we have of ourselves is fundamentally influenced by these institutions. This is no less so in regard to our moral status. The social institutions in which we live can either sustain or undermine the view that we have full moral status, that we are members of good standing in the moral community, in virtue of the fact that we are persons. As one might suspect, I want to say that a society does the former when and only when the grounds for being a member of good standing in that society are co extensive with the grounds for being a member of good standing in the moral community, which is that one is a person. This means, then, that the basic rights and liberties of society are to be distributed equally. Now as one has no doubt noted, the very claim which Rawls makes in A Theory of Justice is that the equal distribution of basic rights and liberties sustain the self-respect of persons (sec. 82). But he thinks that this is so because such a distribution affirms the worth of each person's plan of life. For what he should like to think is that the self-respect of each person is secured in virtue of the fact that each has the conviction that she is free to pursue her plan of life, whatever its end may be. " But the equal distribution of rights and liberties does not affirm the worth of the plans of life which persons pursue. That is to say, the knowledge that one is free to pursue one's plan of life whatever its end may be is not sufficient for one to also have the conviction that one's plan of life is worthwhile. If all that one can do well is sweep floors, the knowledge that one is guaranteed the freedom to do this will not in itself result in one also having the conviction that one' s plan of life - that of floor sweeping - is worthwhile. At least this is so in societies of the Western Culture, where ends such as the arts and sciences and the high forms of culture are more highly valu ed . Moreover, as I have already shown, it is a consequence of the �ristotelian Principle and its companion effect that a person whose plan of hfe exhibits the greater and more subtle number of well-trained abilities is the p erson who will have a more secure conviction of the worth of his plan of life. And between a floor sweeper and a symphony conductor, say, it is clear whose plan of life is more subtle and calls for the greater number of well-t rained abilities. Presumably any (healthy) person could be an excellent floor sweeper, if only they should try; though not so with symphony condtJcting. Wh at is more, if ( 1 ) the equal distribution of rights and liberties were to suffice for persons to have the conviction that their plan of life is
311 45
LARRY L. THOMAS
worthwhile, then why is (2) having a plan of life which satisfies �he Aristotelian Principle a necessary condition for having a plan of life whlch is worthwhile? Either (l) or (2) is true, but not both . It is obvious, I believe, that (2) is true, at least in general. The equal distribution of basic rights and liberties does indeed affirm t�e self-respect of persons. For, as I should like to think, it is the only way m which society can affirm the fact that each person is a being with full moral status in virtue of the fact that he or she is a person . The Black Consciousness Movement supports this view. For slavery and racism in America has been a deliberate and systematic denial to blacks the rights and liberties to which others in America have been so long accustomed. As a result of the Black Consciousness Move ment, we did not all come to regard ourselves as brilliant. N or , again, did we all suddenly discover talents which were hitherto latent in us. These things could never have been accomplished by the Black Consciousness Movement as such. Rather, we came to believe that our blackness to the contrary notwithstanding, we were worthy of first-�lass citizenship in the American society. And on my view, to believe this is, in part, what it means to have self-respect. '6 Un iversity of Maryland Harvard University
NOTES un t 1 Belknap Press, 1 97 1 . All parenthetical page references are to this text I give an acco · of the Black Conscio usness Moveme nt later on in th e essay. · Th. 's paper ts an attempt to spell out some o f the practical implications o f the dis. tinction betwe n self-respect and self-estee m as 1 have drawn it in my " M o ral ity an d Ou r S lf-Concep t, . The ourna of Value Inquiry, v . l 2 0 978). For implicatio ns of a � dtffere t sort, see my . M arx1sm versus Capitali sm : A P sycholog ical Discu ssi o n , " Stu dies � rn Sov1et Thought (forthcomin g). ee e · as ngton' s Up From Slavery: An A utobiogr d aphy, (Doub leda y a n _ . s Atlanta address whi 1 t IS as mgton e · to • h le ch m vulnera akes him so b · t I abe1 unc 1e tom. S ee note 7 below. 3 For examp le, see Stanle y Coope rsmith Th e A ntecedents of Self-Esteem, ( w · H · · Freeman and Co · • l 967) • p. 26 ,. K aren Horn ey · (W · W · Ne w ways rn psychoanalys1s, . Norton and Co · • 1 939) , ch . s espec1a . 11 y; ' £·A nalys1s, (W .W . N o r t o n -- , se f" . and Co · • 1 942) • pp · 43 f · R obert W Whit e • "E go and Reality in Psychoa na1 yu· c Theory Psychological Issues• v. 3 · ( I 9 63) Mon o grap h 1 1 ; and Alb ert Ba nd ura • • · "Self-efficacy: Toward a u n I· rymg Theory o r Behav ioral Change," Psyc:h ological Review, v. 84 ( 1 977).
'
�
�
�
�! ��� � � � � '"
.,
' ' -----
312 46
RAW L S I A N SE L F - R E S P ECT A N D T H E B L A C K CONSC I O U SNESS M O V E M E NT
4 This criticism of R awls' account of self-respect is developed more fully in the " M o rality
and Our Self-Concept" paper mentioned in note I above, See Section I .
S Rawls writes: " I t is o u r plan o f l i fe that determines what we feel ashamed o f , a n d so
feelings of shame are relative to our aspirations, to what we try to do and with whom we
wish to associate. Th ose with n o musical ability d o not strive to be musicians and feel no shame for this lac k . I n d eed it is n o lack at all, not at least i f satisfying associations can be formed by doing other things"
(444 ) .
6 O n this point, see John H o pe Franklin, From Slavery t o Freedom, (Vintage Books,
1947), c h . 2 1 , a n d C h arles E. Silberman, Crisis in Black and While, (Vintage Books, 1 964) , pp. 1 2 5 ff.
7 In his " A tlanta E x p osition A d d ress , " Washington emphasized the im portance o f
industrial education for blacks, rather t h a n their obtaining t h e su ffrage. Thus for him,
the political a n d civil status o f blacks were, o n the face o f it at any rate, o f secondary importance. I n deed, in this address, Washington im plies that the su ffrage was something of which blacks should have to beco me worthy. Consider the following passages: No race can prosper till it learns that there is as much dignity in tilling a field as in writing a p o e m . I t is at the bottom of life we must begin, and not at the top . . . . . . . . In all things t hat are purely social we can be as separate as the fingers, yet one as the hand in all t h ings essential to m u t ual progress . . . . I t is important and right that all privileges o f the law be ours, but it is vastly more important that we be prepared for the exercises of these privileges. Of course, as John H o p e Frankli n observes, op. cit. , p . 395, wea k n esses in Washington's view s may seem more obvious today than they were then. All the same, it should be noted that not all o f W ashington's black contem poraries viewed his A tlanta speech favorably. Most notably, W . E . B. DuBois thought Washington's speech to be a call for comprom ise and subm issi o n . See DuBois' The Soul of Black Folks, ( A . C . M cClurgh, 1 903), p p .
41-5 9.
8 As the expressio n s " I ' m black and I ' m proud" and " Black is beautiful" would indicate, from the standpoint o f physical appearance, blaclr� no longer thought o f their not possessing w h ite features as a lack on their part. The standards of w h ite beauty simply no longer applied. And as blacks, w e came to take greater pride in various aspects of our cul tur e. 9 See , for exam ple, M a ureen Orth, "Stevie, The W o nder M a n , " Newsweek, (October 28 ,
1 974 , p p . 59 -65. S h e writes: " I t w a s A m erican b l a c k m usic t h a t provided t h e inspiration ) - and often the m u sic itsel f - for the explosion o f white-d om inated rock in the ' SOs and ' 60s . But with the rare exception . . . , it was white stars w h o got the big money and the Wh it e audi ences. Now the sheer creative power o f black m usic has pushed it into the mai ns tream . . . Today the color line is almost completely erased, and the record charts are st udde d with black names and groups . " 1 0 The co n c ept o f self-esteem i s far more complex than m y remarks would indicate. Asi d e from our abilities and acco m p lishments, another factor is the ends which we value. Th e suc cess ful p u rsuit of a given e n d will be a source of self-esteem to us only if we value th at e nd itsel f. A n d as 1 have remarked in m y " Capitalis m versus Marxism " essay, cited in no te 1 above, t here is also the factor of what I have called derivative self-esteem . Su pp ose that g i s an e n d w hich i s valued b y society a t large, and that as i t happens persons mo st s ucc essful a t p u rsuing g are of kind K. If the association of end II with people o f
313 47
L A R R Y L. T H O M A S
�
kind K becomes strong enough, then being o f kind K c a n itself su ffice t o enhance on 's
self-esteem. The most striking example of derivative self-esteem is the self-esteem wh .ch . often stems from being a member of a high socio-economic class . See S. Coopersmith,
op. cit., pp. 82f.; and M . D . Vernon, Human Motivation, (Cambridge U n iversity Press,
1 969), pp . 1 3 2f. This is the point which 1 believe Thomas Nagel to be making in his "Rawls on Justice," The Philosophical Review, v. 82 ( 1 973), note 6. I I For a fuller account o f self-respect, see my "M orality and Our Self-Concept , " Section II. 1 2 Here I borrow from Thomas Nagel, The Possibility of A ltruism, (Claren don Press, 1 970), p. 58, and Michael Tooley, "Abortion and Infanticide," (Prince ton U niv ersity Press, 1 973), pp. 52-84. Against Tooley, though, 1 want to say that a being's havi ng the
potential for these capacities is a morally relevant consideration. See my " H uman P o ten tiality: Its Moral Relevance," The Personalist, v. 59 ( 1 97 8 ) . 1 3 Henry Sidgwick makes this point i n The Method of Ethics, 7th ed . , pp. 3 1 , 207, 379.
Each in section 3 of their respective books and chapters. 1 4 "Are There Any Natural Rights," The Ph ilosophical Review, v. 64 ( 1 9 5 5 ) . I 5 Here I a m much indebted to t h e essays of Thomas E . Hill J r . , " Se rvilit y and
�elf-Respect," The Monist, v. 57 (1 973) , and Gregory Vlastos, " J ustice a n d Equality,"
ed. Richard B . Brandt, Social Justice, (Prentice-H all, Inc. 1 962) , p p . 3 1 -7 2 . 16 The proviso is that a person' s plan of life conform to the principle s of j u st ice as fairne ss.
m
1 7 I wish to thank I rving Thalberg and Newton Garver for com men ting upon the pe nul umate draft of this essay.
3 14
THE CLAIM TO MORAL ADEQUACY OF A HIGHEST STAGE
N I
OF MORAL JUDGMENT
•
tensive re previous publications I have outlined : (a) the ex of roo al search facts concerning culturally univenal stages wh ich judgment,t (b) the psychological theory of development ts to m atte � best fits those facts,1 and (c) a metaethical view which of es n eo bridge the gap between naturalistic and nonnaturalistic th r pape moral judgment and their grounds of adequacy.• The present a at elaborates a claim made in the previous paper •: the claim th
�
higher or later stage of moral judgment is "objectiv ely" preferable g to to or more adequate than an earlier stage of judgment accordin en certain moral criteria. Since these criteria of adequacy are those c m ble pro tral to judgment at our most advanced stage, "stage 6," the ge sta becomes one of justifying the structure of moral judgment at 6. First, however, we shall brieRy review the position taken in our earlier paper, and introduce the claims of the present one. I. JtEVIEW
OJ' PSYCHOLOGICAL THEORY
Over a period of almost twenty years of empirical research, my col
leagues and 1 have rather firmly established a culturally universal
invariant sequence of stages of moral judgment; these stages are J I treat these matters in greatest detail in my papen "Propositional Verbs and Jlnowledge," thia JOUUI.U.. IJUX, 11 ijune 1, 19'72): !01-512, and "The W ages of Scepticism," American Phi lt»oflhical Q.uarterl,, x. !I Ouly 1975): 177-187. • To be presented in an AP A aymposium on Moral Development and Moral Pbiloeophy, Dea:mber 28, 197! : commentaton will be K.urt 'Baier and Richard G. Hemon: tee thia JOUDI.U.. this isaue, 646-648 and 648-649, respectively. 1 Morolizati"!': The Cognitive-Developmental A�oach (New York: Holt, llinehut lc W101ton, 1974). • "Stage and Sequence: The Copi�ve-Developmental Approach to Socializa tioo," lD D. Gollin, ed., Hanclbool of Socializa&ioJI Theory and Research (New York: lland McNally, 1969). • "From II to Ought: How to Commit the Naturalistic Fallacy and Get Away with It in the Study of Moral Development," in T. Mischel, ed., Cognidve De wlof�met�t ontl Efristemoloty (New York: Academic Pre��, 1971).
50
63 I
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
grossly summarized in Table I :
Table I . Definition of Moral Stages
I. Preconventional level
At this level the child is responsive to cultural rules and labels of
good
and bad, right or wrong. but interprets these labels either in
terms of the physical or the hedonistic consequences of action (punish· ment, reward, exchange of favors) or in terms of the physical power of those who enunciate the rules and labels. The level is divided into the following two stages:
Stage I : The punishment-and-obedience orientation. The physical
consequences of action determine its goodness or badness regardless of the human meaning or value of these consequences. Avoidance of pun ishment and unquestioning deference to power are valued in their own right, not in terms of respect for an underlying moral order sup ported by punishment and authority (the latter being stage
4).
Stage 2: The instrumental-relativist orientation. Right action con
sists of that which instrumentally satisfies one's own needs and occa sionally the needs of others. Human relations are viewed in terms like those of the market place. Elements of fairness, of reciprocity, and of equal sharing are present, but they are always interpreted in a physical pragmatic way. Reciprocity is a matter of "you scratch my back and
I'll scratch yours," not of loyalty, gratitude, or justice.
II. Conventional level
At this level, maintaining the expectations of the individual's family, group, or nation is perceived as valuable in its own right, regardless of immediate and. obvious consequences. The attitude is not only one
of conformity to personal expectations and social order, but of loyalty
to it, of actively maintaining, supporting, and justifying the order, and
of identifying with the persons or group involved in it. At this level, there are the following two stages: Stage 8: The interpersonal concordance or "good boy---nice girl"
orientation. Good behavior is that which pleases or helps othen and is
approved by them. There is much conformity to stereotypical images of what is majority or "natural" behavior. Behavior is frequently
judged by intention-"he means well" becomes important for the first time. One earns approval by being "nice."
Stage 4: The "law and order' orientation. There is orientation toward
au thority, fixed rules, and the maintenance of the social order. Right behavior consists of doing one's duty, showing respect for authority, and maintaining the given social order for its own sake.
II I. Postconventional, autonomous, or principled level
. At this level there is a clear effon to define lhoral values and pnn ciples that ba e validity and application apart from the authority of the gro ups or persons holding these principles and apart from the in-
;
51
THE JOURNAL OF PHILOSOPHY
This level again baa dividual'a own identification with these groups.
generaUY WI· th · terms of gen· utilitarian overtones. Right action tends to be de fined m . . lly ex . tlca era1 individual rights, and standards which have been m aware· dear a is There . society whole the amined and agreed upon by and a corre· ness of the relativism of personal values an d opm1ons . nsus. Aside aponding emphasis upon procedural rules for reaching conse upon the from what is constitutionally and democratically agreed 1s an result · The right is a matter of personal "va1ues and opm1on. . n upo s emphasis upon the "legal point of view," but with an empha of s uon the possibility of changing law in terms of rational considera or· social utility (rather than freezing it in terms of stage 4 "law and. bmd· the is der"). Outside the legal realm, free agreement and COh!ract er· ing element o( obligation. This is the "official" morality o( the .Am
two stages: Stage
5: The social-contract legalistic orientation,
•
·
·
•
..
..
ican government and constitution. Stage 6: The universal-ethical-principle
·
�
..
orien tation.
.
. Right IS defin�d
by the decision of conscience in accord with self-chosen ethical pran· ciples appealing to logical comprehensiveness, universality, and con· ule, sistency. These principles are abstract and ethical (the Golden R
the categorical imperative); they are not concrete moral rules like
Ten Commandments. At heart, these are universal principles of
tice,
of the
recriprocity
and
equality
for the dignity of human beings as Ought," pp. 164j5) .
�e
1us·
of human rights, and of respect individual persons ("From Is to
As Table 1 indicates, the last stage, stage 6, bas a distinctively Kantian ring, centering moral judgment on concepts of obligation as these are defined by principles of respect for persons and of j us
tice. In part, this conesponds to an initial "formalist" or "struc· turalist" bias of both our moral and our psychological theory. Our
psychological theory of morality derives largely from Piaget,' who claims that both logic and morality develop through stages and th at
each stage is a structure which, formally considered, is in better equilibrium than its predecessor. It assumes, that is, that each new (logical or moral) stage is a new structure which includes elements of earlier structures but transforms them in such a way as to repre· sent a more stable and extensive equilibrium . Our theory assumes that new moral structures presuppose new logical structures, i.e., that a new logical stage (or substage) is a necess ary but not sufficient condition for a new moral stage. It assum es, however, that moral judgments (or moral equilibrium) involve s two related processes or conditions absent in the logical domain. First , moral judgments in•
Jean Pia��· Th1 Moral fint edlUOD, 1952).
1�8;
Judgment of th1 Child
52
(Glencoe' Ill.:
Free Press,
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
633
volve role-taking, taking the point of view of others conceived as
subjects
and coordinating those points of view, whereas logic in
volves only coordinating points of view upon objects. Second, equili brated moral j udgments involve principles of justice or fairness. A moral situation in disequilibrium is one in which there are unre solved conflicting claims. A resolution of the situation is one in which each is "given his due" according to some principle of jus tice that can be recognized as fair by all the conflicting parties in volved. These "equilibration" assumptions of our psychological the ory are naturally allied to the formalistic tradition in philosophic ethics from Kant to Rawls. This isomorphism of psychological and normative theory generates the claim that a psychologically more advanced stage of moral judgment is more morally adequate, by moral-philosophic criteria. The isomorphism assumption is a two way street. While moral philosophical criteria of adequacy of moral judgment help define a standard of psychological adequacy or ad vance, the study of psychological advance feeds back and clarifies these criteria. Our psychological theory as to why individuals move from one stage to the next is grounded on a moral-philosophical the ory which specifies that the later stage is morally better or more ade quat e than the earlier stage. Our psychological theory claims that individuals prefer the highest stage of reasoning they comprehend, a claim supported by research.5 This claim of our psychological the ory derives from a philosophical claim that a later stage is "objec tively" preferable or more adequate by certain
moral
criteria. This
philosophic claim, however, would for us be thrown into question if the facts of moral advance were inconsistent with its psychological implications. Our assumption of isomorphism implies first the assumption of continuity between the context of discovery of moral viewpoints (studied by the psychology of moral development) and the context of j ustification of moral viewpoints (studied by formal moral philos
ophy). This implies that the philosopher's justification of a higher stage of moral reasoning maps into the psychologist's explanation of
movement to that stage, and vice versa. The isomorphism assum� tio n is plausible if one believes that the developing human being and the moral philosopher are engaged in fundamentally the same moral task.
II. MORAL THEORIES AND NATURAL STRUCTURES
Our notions of moral philosophic adequacy derive, then, from the notion that moral theories are derivative from the natural struc5 Sec my Moralization, op. cit.
53
634
THE JOURNAL OF PHILOSOP HY
.. not in the nat ural . tures we term "stages." The structures are ults the sequen tt al res sense of being innate but in the sense of being ch· tea r cula · Parti . of processing moral expenence, not denvattve from o nott this sense ings or particular moral ideologies or theories. In al stru ctures natur of natural rights, social contract, and utility are " . ruru't s of cus· emerging in nonphilosophers from reflectiOn upon th e . onal ctrcu tomary morality in very varied cultural and educati ories to thl , stances. To clarify the relation of philosophic moral the moral dl· we have interviewed moral philosophers concerning our _ som ng lemmas and tried to relate their published theory to their rea that a on the dilemmas. We feel that our results justify the notion or· in philosopher's formal moral theory is an elaboration of certa os· hl tions of his "natural" moral-stage structure. Not only do p ew l ophers u se their published moral theories to reason about interv dilemmas, but we see that their theories interlock. with " natural" r unformalized elements of their moral reasoning unsta ted in thel cl· published theories but required for actually malting moral de d iew sions. Further, hardly to our surprise, all philosophers interv X· ml reason at the two highest stages, being stage 5, stage 6, or some ture of the two. We believe our data on philosophers are consistent with our assumptions, that: (1) stages are natural struc tur es gener at· ing families of moral theories, and (2) the two major families of formal normative moral theory tend to be generated by tw o natural structures which we term "stage 5" and "stage 6." Our own approach to moral theory is to view it as a constructive •"
'
.
�
·
� �
�
�
� � �
systematization of these natural structures. Without formal moral theory men naturally attain to a "stage 5" in which they judge law s by the light of a social contract, by rule-utilitarianism, and by some notion of universal or natural rights. Much moral philosophy m ay
be undentood as a systematization of this mode of thought, and most moral philosophers have, in some sense, assumed this to be their task. Other philosophers , however, have attempted to general· ize, o ra to a higher level, these "stage 5" postulates, to de fine _ _ a baslS for md1v 1dually principled moral decis ion. As George Her· bert Mead • put it:
�
m:
Kant ge eralized the position invol ved in the theory of natural right � s and contract, which waa that one could claim for h imself only that wh1ch he 1'N'nen\ ;--o..ized PnUa11Y (or o . -"� then. He made a generaluauon . of this the basu for hlS moral doctrine, the categorical imperative that
�al
•
Movements
1956).
·
of Thoucht in the Nineteen th Century
54
·
(Chicago.· University
Pr ess,
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
every
act should be of such
a
character that it could
for everyone under the same conditions.
be made
63 5
universal
As John R awls 7 states it:
My aim is to present a conception of j ustice which gene ralizes and car
familiar theory of the Rousseau and Kant.
ries to a higher level of abstraction the
con tract as
found, say,
in Locke,
social
In our view, there is a family of theories that have the purpose j ust stated of Kant and Rawls. This family of theories may be looked
at as derivatives of a natural structure I term "stage 6." Rawls's the
ory, when traced back to its natural structural roots, is not merely a "generalization" and "abstraction" of the theory of social contract, but derives from a new way of thought, a new system of assumptions, a new decision-making process. This is true in the same sense that "the familiar theory of social contract" is not merely a "generaliza tion" and "abstraction" of the stage 4 conception of an overriding
need for social order, but is an expression of a natural structure we term ..stage 5." The purpose of -this paper, then, is to describe the stage 6 natural
structure underlying various theories like Rawls's and to j ustify the sense in which it is better than structures of reasoning based on the "more familiar" conception of the social contract and of utili tarianism. In terms of description, this means that we view Rawls's theory as further elaborating a natural decision-making structure only partly specified by Kant ;md that Rawls's theory can be further elaborated and specified as a decision-structure applicable to resolving individ ual moral dilemmas, as we attempt to do in this paper. A second claim of this paper is that the adequacy of a moral struc ture is to be j udged primarily in terms of the adequacy of an inter locked series of assumptions for making moral choice. The criteria of adequacy used by stage 6 are implicit in "stage 5" structures, since these are trying to do the same job as stage 6. Both aim at determin
ing moral decisions and j udgments on which all rational men
involved in sociomoral action could ideally agree. Stage 6 can do
this better than stage 5.
III. DIFFERENTIATION AND INTEGRATION AT STAGE
6
Assumi ng for the moment the existence of a natural structure called "stage 6.'' how can developmental analysis aid in justifying its claim T ..4 Theory of Justice (Cambridge, Mass . : Harvard, 1971).
55
THE
636
JOURNAL OF PHILOSOPHY
stage 5 6 is more adequate than to adequacy? By claiming that stage ade quate . . . . make stage 5 more . by certam formal a1tena wh1ch a1so rauon. teg in and . t' 1on . a erenu d'ff a are 1 . than stage 4. Two such cnten 1 ted the forma m to Ought") we elabora Is ("Fro r pape er In our earli th that apply to both concepts of differentiation and integration n, an moral deliberatio psychologi cal and the normative analysis of s as ment mor al ju g . linked the concepts to formalistic concepts of on u a n and mtegr prescriptive and universal. Increased differentiatio udy power in the st are anchored in the "is" side by their explanatory ' gh f side of the development of directed thinking, and on the " ou oice. To 1 their necessary inclusion in rational justification and ch ider some lustrate how and why this might be the case, we shall cons 5 and 6 • usages of the categories of "rights" and "duties" at stages ed and the sense in which this suggests that moral theories deriv den e s eorie stage 6 structures are more advanced than moral th ue s "du from stage 5 structures. Table 2 shows usage of "righ ts" and
�
�
�i
•
£:0m ��
at each stage:
ns Table 2. Six Stages in the Usage of Categories of Rights and Obligatio
Stage 1. Having
on uol Means having the power or authority to c nce rd something or someone, or it is confused with being right (in acco a with authority). Obligation: Or "should" is what one "has to do" because of the de·
a Right:
mands of external authorities, rules, or the external situation.
a Right: Implies freedom of the self to choose and to con· uol the self and its possessions. One has a right to ignore the p ositiv e claims or welfare of another as long as one does not directly violate his free· dom, or injure him. (Having a right differentiated from being right, and from being given the power to, by a status one holds.) Obligation: O ligation or "should" is a hypothetical imperative . c?n�gent on choice m terms of an end. In this sense, obl igations are bout to o self and one's ends. ("Should" or obligation differen tiated (rom 'has to from external or authoritative compulsion.)
Stage 2. Having
•
�
S� !.
�
��
�
�
avin a Right: Implies an expectation of conuol and freedotn which a good . o� ?atural person would claim. A right is based either on a rule or on a lepumate expectation toward others, e.g., you have a right to have your property respected since you worked hard to acquire the . property. llighu are earned. (Having a right differentiated (rom the free· dom to control and choose.)
Obligation: "Should" or "duty" equals a role-obligation, what it is mcumbent on a member of a social position to do for his role-partners as de6ned by rules, by the expectation of the role-partner, ood or by what a g role-oc;cupant (a husband, a good doctor) would do. (Obligation dif · (erenuated from bemg a means to a deaired end .) •
�
56
63 7
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
Stage 4. Having a Right: Rights are: (a) categorical general freedoms and expectations which all memben of society have, and (b) rights awarded to
particular roles by society. General rights usually take primacy over role rights. (Having a right differentiated from a particular legitimate expecta tion.) O bligation: Obligations are responsibilities, i.e., welfare states of others or of society for which one is accountable. These responsibilities
arise through: (a) being a member of society, and (b) voluntarily entering
into roles which entail these responsibilities. (Obligation or duty as com mitment and responsibility differentiated from what is typically expected of a role-occupant.) Stage 5, Having a Right: Has an awareness of human or natural rights or liberties which are prior to society and which society is to protect. It is usually thought by stage 5 that freedoms should be limited by society and
law only when they are incompatible with the like freedoms of others. (Natural rights differentiated from societally awarded rights.)
, Obligation: Obligations are what one has contracted to fulfill in
order to have one's own rights respected and protected. These obligations are defined in terms of a rational concern for the welfare of others. (Obliga tions as required rational concern for welfare differentiated from fixed responsibilities.) Stage 6. Having a Right: There are universal rights of just treatment which
go beyond liberties and which represent universalizable claims of one in dividual upon another. Obligati()n: Any right or just claim by an individual gives use to a corresponding duty to another individual.
Tabl e 2 suggests that each higher stage's usage of the categories of
rights and duties is more differentiated and integrated than the prior stage. We list, in the parentheses, the differentiation of the concept of "right" and "obligation" made by each stage not made at the prior stages. At stage 2, a right as a freedom is differentiated from a physical or social power; at stage 8 it is differentiated as an ex
pectation to be supported by others from an actual physical or psy chological freedom, etc. The sense in which each stage is
tegrated
better in
is seen in the fact that only at stage 6 are rights and duties
completely correlative. The meaning of correlativity of rights and duties is suggested in the following passage by Raphael.• We have accepted the deontological view that the moral use of "ought"
is a basic concept that cannot be derived from the idea of goodness.
We tum next to the notion of "rights." There are two senses of the word, the first meaning, "I have no duty to refrain from so acting,"
a D. Daiches Raphae l, Moral judgement (London: Allen Be Unwin, 1 955).
57
THE JOURNAL OF PHILOSOPHY
�
�
as I descri e by aying, the &econd in which I describe the same· fact of a nght m lght be kind "Someone else baa a duty to me." The second also of action, right a have I called "a right of recipience." Whenever hts, ng definition of . have a right of recipience. In virtue of the second nght and "B has a the two forms of expresSi. on: ..A has a duty to B" . of analyucally sense the in ive correlat (of recipience) against A" are tautologous 10 plying each other. They may not be connotatively language we are ordinary speech, though they are in the more precise
�
. n�·
recommending (pp. 47-49) .
n�..
· hts · usmg Let us accept, for the moment, Raphael's view that m· e mor or and duties on correlative terms is either more "precise" . . d e am tegrated." We have found that such usage, consistently mamt y etel is found only at stage 6. At stage 5, "rights" categories are compl r reciprocal; i.e., the concept and limits of rights are completel l· tnd and rights ciprocal with the rights of others, but individual •
!' �
vidual duties are not completely correlative.
An example is case 2, age twenty-four, who said:
Morality to me means recognizing the rights of others first to life and
then to do as they please as long as it doesn't interfere with somebody else's rights.
Although case 2 is able to define rights clearly, he is unable to specify clearly the conditions under which awareness of rights gen· erates correlative duties. At stage 5, for every right, society has some
duty to protect that right. Duties to other individuals, however, are not clearly specified in the absence of either individual contract or
social contracL At stage 5 there are obligations to the law and there are obligations to the weUare of others, of a rule- or act-utilitarian
sorL But recognition of individual rights does not directly generate
individual duties; i.e., rights and duties are not directly correlative. Even moral philosophers, like our "natural" stage 5 subjects, need
not accept that rights directly imply duties. This is indicated in philosophers' responses to the following d ilemma: In �urope, a woman was near death from a very bad disease, a spe· . oal kind of cancer. There was one drug that the doctors thought might aave It wu a form of radium (or which a druggi was charging st ten umes what the drug cost him to make. The sick woman's husband , Heinz. went to everyone he knew to borrow the money, but he could o ly t togeth about half of what it coat. He told the druggist that hu W1fe was dymg :'"d him to sell it cheaper or let him pay lata'. But the drugtst wd. 'No, I disco vered the drug and I'm going to make money from it." So Heinz got desperate and broke into the man's store to steal the drug for his wife.
�er.
� �
�
,
��k.�
58
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
639
In general, subjects whom we classify on other grounds as stage 5
recognize the woman's right to live, but do not believe that it di rectly generates an obligation to steal to save her. Or they may
recognize a duty to steal for the wife, based on contract, but recog nize no duty to steal for the friend or stranger who equally has a
right to life. This, too, is the position taken by a number of moral philosophers, two of whom are quoted elsewhere ("From Is to Ought," 206-207).
Examples are as follows :
Philosopher 1 :
What Heinz did was not wrong. The distribution of scarce drugs
such in the circum
should be regulated by principles of fairness. In the absence of regulations, the druggist was within his legal righu, but stances he has no moral complainL He still
was
within his moral righu,
however, unless it was within his society a strongly disapproved thing to do. While what Heinz did was not wrong, it was not his duty to do
it. In this case it is not wrong for Heinz to steal the drug, but it goes
beyond the call of duty; it is a deed of supereroga tion
.
Philosopher 2 :
It is a husband's duty to steal the drug. The principle that hwbands
should look after their wives to the best of their ability
is one whose
general observance does more good than harm. He should also steal it for a friend, if he were a very close friend (close enough for it to be understood that they would do this sort
of thing for each other). The
reasons are similar to those in the case of wives. If the person with
cancer were a less close friend, or even a stranger, Heinz would be
doing a good act if he stole the drug, but he has no duty to.
These philosophers agree that stealing to save a life in the situa tion is right, but disagree on when it is a duty. Using rule-utilitarian criteria of duty, it is difficult to make duties and rights correlative.
In contrast, here is the response of philosopher 3, who responds
quite differently to duty questions about this dilemma:
IF THE HUSBAND DOES NOT FEEL VEllY CLOSE OR. AFFECTIONATE TO HIS WIFE, SHOULD HE STEAL THE DR.UG?
Yes. The value of her life is independent of any personal ties. The
value of human life is based on the fact that it offers the only possible source of a categorical moral "ought" to a rational being acting in the
role of a moral agent.
SUPPOSE. rr WER.E.
A
FRIEND Oil AN ACQUAINTANCE.?
Yes, the value of a human life remains the same.
I n general, philosopher 3 's conceptions of rights and duties were
59
THE JOURNAL OF PHILOSOPHY
640
ars more differentiated correlative. In this sense his thinking appe st philosophers. We m and integrated than that of the other two g makin re a structu note' however' that the greater integration of logd of mcrease rights and duties correlative is not a mere matter that it imposes but ical or analytical tidiness, as Raphael suggests, philos pher � a severe price. The price in question is, baldly, that mta ce, has to be prepared to go to jail to steal for a friend or acqua (op. cd.): and philosophers 1 and 2 do not. According to Raphael
�
·
?
�
er ga· From an objective point of view, the sa
,
son benefi.ted does not have a right
(51).
�
We would question whether duties and rights could be made co lS relative, as Raphael wishes, if the moral point of view adopted " 0r nt age that recommended by Raphael, i.e., "the average moral
"the average impartial spectator." To make duties and righ ts cor relative, we must take not the standpoint of the "average moral agent," but philosopher �·s standpoint of the "rational moral agent." The rational moral agent is not a self-sacrificial saint, since the saint's duties do not imply that he has corresponding rights. The ra tional moral agent is fair, not saintly, he does as a duty only what
his is rationally prepared to demand that others do as a duty, or that to which he has a righL The fact that a dying acquaintance in need of the medicine has the right to life does not define a duty for the average moral agent, but it may for the rational or just moral agent.
In summary, there are cenain basic categories used by every moral
stage-structure or every moral theory, such as the categories of rights and duties. Development through the stages indicates a progressive differentiation of categories from one another, e.g., of righ ts from duties, and a progressive integration of them, expressed in the cor·
relativity of rights and duties. This correlativity is not merely a matter of the analytic tidiness of an abstract norma tive or me ta· ethical theory, but leads to very different judgmetns of obligation, judgments that are in better equilibrium with judgm ents of righ ts. As a result, stage 6 moral judgments are able to differen tiate, " among e amo phous was ebasket of " actions of su pereroga tion, those act1ons whtch are duttes prescribed by the rights of othen from those which are not.
�
�
�
60
64 I
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
As stated, a problem remains for stage 6 moral judgment. If the
rights of every human define duties for an individual moral agent,
this seems to open up the abyss of the existence of infinite and simultaneous duties to support the rights of every human being wherever he is. The problem here en tailed is partially solvable, though complex. The individual moral agent has rights, and these rights are incompatible with having duties to every right of every other. Because a human being has a right to life, other humans have a duty to save that life. The conditions under which one human being has a duty to save the life of another human being require clarification of what it means for a "rational moral agent" to choose between conflicting duties since he cannot be an omnip otent saint.
IV. UNIVERSALIZABIUTY AND REVERSIBILITY AT STAGE
6
Since Kant, formalists have argued that rational moral judgments must be reversible, consistent, and universalizable, and that this im plies the prescriptivity of such judgments. We claim that only the
substantive moral j udgments made at stage 6 fully meet these condi
tions, and that each higher stage meets these conditions better than
each lower stage. In fully meeting these conditions, stage 6 moral structures are ultimately equilibrated.
For developmental theory, meeting these conditions of moral judgme nt is parallel with the equilibration of fully logical thought in the realm of physical or logical facts. According to Piaget and others, the keystone of logic is reversibility. A logical train of thought is one i n which one can move back and forth between premises and
conclusions without distortion. Mathematical thinking is an exam ple; A + B is the same as B + A . Or again, the operation A + B = C is reversible by the operation C B = A . In one sense, the elements of reversible moral thought are the moral categories as these apply -
to the universe of moral actors. To say that rights and duties are correlative is to say that one can move from rights to duties and back with out change or distortion. Universalizability and consist
ency are fully attained by the reversibility of prescriptions of ac tions. Reversibility of moral judgment is what is ultimately meant by the criterion of the fairness of a moral decision. Procedurally,
fairness as impartiality means reversibility in the sense of a decision on which all interested parties could agree insofar as th ey can con sider their own claims impartially, as the just decider would. If we have a reversible solution, we have one that could be reached as
right starting from anyone's perspective in the situation, given each person's intent to put himself in the shoes of the other.
61
64-2
THE
JOURNAL
OF PHILOSOPHY
m: unive rsaliz
of formalis Reversibility meets a second criterion _ e ·slogan, "Put yourself m ability. As reversibility starts with the Wlt s start ersalizabi lity other guy's shoes when you decide," univ if everyone use� thts what it; the slogan, "What if everyone did by ability is imph d principle of choice?'' It is clear that universaliz _ ng the conflict� reversibility. U something is fair or right to do from it is somethmg points of view of all those involved in the situation, eversibility tel s we can wish all men to do in all similar situations. R dilemmas, but lt us more than universalizability, then, in resolving
�
� � �
implies universalizabilily.
The concept of reversibility explains the intuitive plausibility of Rawls's conception of justice
(op. cit.)
as a rational choice in an
original position in which one is under a veil of ignorance as to one's role or identity. Rawls argues that this conception leads to the choice of a justice principle of equality, with inequalities accepted only when it is to the benefit of the least advantaged. This concep tion of choice in the veil of ignorance is a formalization of the con ception of fairness involved in having one person cut the cake and a second person distribute it. This conception leads to a mini-max imization solution in the sense that the division must be such tha t
the least advantaged person is better off, i.e., that the cake is so cut that the person cutting the cake is willing to live with getting the sxnallest piece.
Our conception of stage 6 helps to clarify the intuitive plausibil·
ity of Rawls's notion of the original position. This is because the concept of fairness as reversibility is the ultimate elaboration of the concept that fairness is reciprocity, a conception held at every stage.
At stage I . the conception of reciprocity is mechanical equivalence.
an eye for an eye and a tooth for a tooth. At stage 2. reciprocity is
mediated by awareness that self and others are subjectively evaluat ing acton with different interests and aware of one another's in terests. The result is a notion of fairness as positive (or negative) exchange of gratifications: if you contribute to my needs and in terests, it's fair for me to contribute to yours. At stage �. reversibility becomes the Golden Rule, i.e., putting yourself in the other guy' s shoes regardless of exchange of interests or values. The difficult at tainment of this Golden Rule conception is illustrated by the fol lowing interpretation of the Golden Rule by a ten-year-old:
it'a like your brain hu to �eave your head and go into the other guy • head and chen come bad mto your head but you still see it lik.e it waa in the other guy's head and then you decide that way.
�
We l,
62
643
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
Stage 3 interpretations of Golden Rule reversibility, however, do
not yield fair decisions nor are they completely reversible. As a re sult, they lead to no determinate moral resolution of a situation.
In the "Heinz steals the drug" dilemma. the husband reaches one
solution if he puts himself in his wife's shoes. another in the drug gist's. Or again, in the Talmudic dilemma of a man with a water bottle encountering another man equally in danger of dying of thirst, a stage 3 interpretation of the Golden Rule logically leads to
their passing the water bottle back. and forth like Alphonse and Gaston. At higher stages, e.g., stage
5,
these problems are handled by the
conceptions of prior rights and contractual agreements. Reversibility at stage
5
means reciprocity of rights. In the stage 5 subject's words:
Morali ty means recognizing the rights of other individuals to do they
please as long as it doesn't interfere with
as
somebody else's rights.
At stage 6, reversibility is attained by a second-order conception
of Golden Rule role-taking. I n the Heinz dilemma, Heinz must imagine whether the druggist could put himself in the wife's posi tion and still maintain his claim and whether the wife could put herself in the druggist's position and still maintain her claim. In tuitively we feel the wife could, the druggist could not. As a result, it is fair for the husband to act on the basis of the wife's claim. We call the process by which a reversible moral decision is reached,
"ideal role-taking." Stage 6 moral judgment is based on role-taking
the claim of each actor under the assumption that all other actors' claims are also governed by the Golden Rule and accommodated
accordingly. This is what is meant by calling stage 6 reversibility
the second-order application of the Golden Rule. The steps for an actor in volved in making such a decision based on ideal role-taking are:
I . To imagine onsel£ in each person's position in that situation (including the self) and to consider all the claims he could
make (or which the self could make in his position).
2. Then to imagine that the individual does not know which
person he is in the situation and to ask. whether he would still uphold that claim. 3. Then to act in accordance with these reversible claims in the situation.
This is clearly similar to Rawls's notion of choice under a veil of ignorance as to who in a moral situation one is to be. Differences
63
THE JOURNAL OF PHILOSOPHY
�
. trying to arn e at are we (1) that: facts the from spring ed involv ples governmg a concrete decisions in addition to a choice of princi knowledge of all society' and (2) An ideal moral agent decides with · ranee ·1 o f 1gno the facts of interest and claims, and is under the ve1 . is the op sl· only as to who he is in the situation. Also eliminated baSlS of tion between a mini-maximization and a maximi zation
�
��
making a decision.
ot d For the purpose of solving individual moral dilemmas, we . ues bab1h want to assume that the individual is ignorant of the pro that he only of outcome of a given decision to each person involved, in the person is ignorant of the probability of being any particular that in situation, i.e., that he is likely to be any particular person situation.
Returning to the stealing-the-drug situation, let us imagine some
one making the decision under the veil of ignorance, i.e., not know ing whether he is to be assigned the role of husband, wife, or drug·
gist. Clearly, the rational solution is to steal the drug; i.e., this leads to the least loss (or the most gain) to an individual who cou ld be in any role. This corresponds to our intuition of the primacy of the woman's right to life over the druggist's right to property and makes
it a duty to act in terms of those rights. If the situation is that the dying person is a friend or acquaintaince, the same holds true. Here a solution achieved under the veil of ignorance is equiv alent to one obtained by ideal role-taking, or "moral musical chairs" as described earlier.
The notion of stage 6 as a natural structure implies the followi ng:
1. A decision reached by playing moral musical chairs corre· sponds to a decision as to what is ultimately "just" or "fair." Ideal role-taking is the decision procedure ultimately required by the at· titudes of respect for persons and of justice as equity recognize d at
higher stages. This is suggested by Rawls's derivation of principles of justice as equity from the original position. 2. Accordingly, the decision reached by ideal role-taking defines
duties correlative to rights rather than acts of supere rogation. S. If we engage in ideal role-taking in most situations, we reach a determinate decision. Our stage 6 moral judges do agree on a choice alternative in our dilemmas where facts and proba bilities are speci·
fied. 4. A decision reached in that way is in "equilibriu m" in the sense that it ia "right" from the point of view of ofar as all involved
ins they are concerned to be governed by a moral attitude or a concep-
64
MORAL DEVELOPMENT AND MORAL PHILOSOPHY
645
tion of j ustice, i.e., insofar as they are willing to take the roles of others.
5.
The procedure integrates "absolute rights" or equality notions
and utili tarian conceptions in conflict at stage
5.
To support these claims convincingly involves elaborating a "stage 6" analysis and choice in various moral dilemmas, a pro cedure not possible in a limited space. An example of the integra tion of utility and deontological justice more difficult than that of
stealing the drug is that of an individual drowning in the river. A passerby can save him, but at a 25 % risk of death (and a 75 % chan ce that both will be saved). Stated in terms of ideal role-taking, from the point of view of the drowning person, he should. If the drowning person put himself in the bystander's shoes and returned
to his own position, he still could make the claim. I f the passerby took the drowning person's position, he could not maintain his
claim to inaction. S tated in terms of the "veil of ignorance," if an individual did not know whether he was the bystander or the drowning person, he would j udge the right decision that of jumping in as long as the risk of death for jumping was definitely less than
50 % . Utility maximization (or minimaximizatio�) leads to the choice of j umping in if the actor does not know the probabilities of which party he will end up being, but does know the probabilities of the outcome of each alternative (jumpingjnot-jumping) for both Parti es. In other words, in these situations "stage 5" utilitarian con
siderations are perceived from the point of view of stage 6 notions
of correlative rights and duties based upon the fundamental equal ity of persons. The notion that these claims correspond to a greater psychological equilibration of moral j udgment becomes truly plausible in em pirical study. A . Erdynast
11
gave subjects instructions to assume the
original position. This could not be done or was meaningless to highly intelligent subjects below stage 5. Stage 5 subjects, however, Would ofte n change their choices after assuming the original posi tion and would feel that the new solution was more adequate. A number of Harvard undergraduates engaged in small-group discus sions in which original-position solutions were elaborated by the group leader and were discussed in relation to alternative ultimates of decision. A number of those who were initially stage 5 retained
and elaborated this mode of moral thought, a change we called a
Judgment . • "Relationshi ps between Moral Stage and Reversibility of Moral. In an Oririnal Posi , 197!1. Uoiversaty Harvard thesis, Ph.D. published tion." Un
65
THE JOURNAL OF PHILOSOPHY
movement from stage 5 to stage 6. The fact that only students pre viously at stage 5 adopted this mode of thinking suggests that it is
a stepwise progression of an intuitive natural structure rather than
one more moral theory.
LAWRENCE KOHLBERG
Laboratory of Human Development, Harvard University
The Moti vation to Be Ju st S tanley Bates Middlebury College
No ?ther work of recen t analytical philos ophical h istory has had the im mediate and widespread impact of John Rawls's A Theory ofJustice, not only among philosophers and members of other academic disciplines, but also a�ong a general public, which, for a substantial period of time, has been dJ.sconnected from the contemporary philosophical tradition in those coun tnes dominated by analytical philosophy . 1 Part of the explanation is that the k was eagerly awaited because Rawls had published parts of the theory in a hJgh ly in fluenti al series of articles since 1 9 58. 2 Now that we have the book, we �nd that it conta i ns even more than we could have expected simply on the ?as,.s of the articles . Rawls has actually developed in detail a theory of JUStJce-.one only realizes how staggering this is if he considers in detail some of the major w orks in analytical moral philosophy since World War I I. 3 It is characteri stic of these works that they are chiefly concerned with what has �orne to be called by the barbarous n�ologism "metaethics, " namely, a su b Ject concerned with "higher-order" questions of the meaning, justification, and truth valu e of moral propositions as opposed to "first-order" questions of value, or norm ative ethics. As Stuart Hampsh ire notes, Rawls's work marks a deci sive br eak with this recent practice (a practice which rested, at least pa r:Iy , on a skepticism, engendered by pos itivism, about the possi bility of saymg a ny thing p hilosophically worthwhile on questions of value) and a
�
p
.
I . J.ohn . Ra wls , A Theory ofjustice (Cam bridge , .\1ass . : Harvard � niversity Press, 1 9� l l . .. . 2· )u s t tce a s Fairness, " Philoropbical Review 6 7 ( 1 9 58): 1 64--94; The Sen se of J � su �.e ?
hi�sophteal Rt1.1i ew 72 ( 1 96 3 ): 2 8 1 - 305; "Constitutional Libe"Y and the Concept of Jusuce, m )..usttce, N omos VI, ed. Carl ) . Friedrich and John Chapman (:'�lew York: Ath e"on Pres� , 1 96 3); . Dtstr•butiv e Justice " in Philoropby Politics and Society ed. Peter Laslen and W. G. Run ciman, Jd ser. (O xford: Basil Biackw ell ! 967 · "Distributiv e Ju�tice: Some Addenda," Natural Law Forum, IJ ( 1 968); "The Justiti�ation �f Civil Disobedience," in Civil Disobedimct, ed . Hugo .-\. au C'll ew York: Peg asus, 1 969). . , . J guagt, S. Toulmm s TIN Plac,e . I h ave i.n mind such works as c. L. Stevenson's Ethics and Lan H. P. 0/ Reason m RtaJOn, Ethics, R. M . Hare's The Language of Moralr and Frettiom ,and �:�It-S mith 's Ethics, K . Baier's Tbt Moral Point uf VitW, an d .\1. Smger s Gmtrallutwn m
��
l
67
2
Ethics
return to the great "classical nonskeptical tradition . "4 (Rawls, of course, is not the only contemporary moral philosopher to have attempted such a return.) .\n initial problem is, Why does Rawls think that it is possible to return to a tradition which has been so thoroughly discredited in the eyes of many recent philosophers? This is a problem about which , in its ge neral form, I . shall say relatively little. Despite the fact that his main concern IS to work �ut a substantive theory of justice and that he has "avoided � xtenstve methodological discussions," it seems to me the working out of �s theory . with its "methodological comments and asides" is, at least impltcn.ly, a re sponse to certain of the analytical criticisms of substantive moral ph1losophy of recent years. � Of course, to say this is not to show that Rawls's res�onse to these criticisms is adequate, though I think that it is. What I would hke to do is to take a particular problem associated with a metaethical conundrum and to try to outl ine Rawls's position on this issue in order to see whether he can answer the questions a metaethical theorist might raise. The problem I have in mind concerns the motivation to be just . What is the appropriate role of motivation in a moral theory? There are many different questions about motivation which it might
�
seem appropriate to raise when a general moral theory-or a theory of good . right, or justice-has been presented . ( I ) Do all people (or most people) always (or generally) act as morality (or justice) requires? (2) If they do, why do they act morally? (3) Why ought people to act morally? (4) Why ought I to . act morally? (5) Why ought I , here and now, do what it is c lear to me IS morally required of me? (In order to make these questions appropriate for Rawls's theory , we may substitute "justice" and "justly" for "morality" and "morally ," respectively.) The first two questions seem to be questions for sociolo gy , or a n thropology, or psychology . :\nswers to the last three questions , however, would all seem to be interpretable as "justifications" of morality . It is j u st this enterprise of justifying morality which metaethical theorists have s ubjected to searching criticisms, often suggesting that the whole enterprise is miscon ceived, or based on a mistake. Traditional justifications have often been attempted historically by showing that a moral theory "fits" with, or is based upon, a conce�tion of human nature which usually includes, as a major �omponent, a new of normal or standard (or perhaps, universal) motivatio n . fhese attempts were designed t o make i t easy to understan d w h y a m a n will do (or ought to do) what is just . It is interesting to note what a great reversal has taken place between classical Greek moral theory in Plato and Aristot le,
-1. Stuart Hampsh ire, ".\ :'l:ev.· Philosophy of the Just Society Books, ' " Nrw York Rrvitw of hbruary 2-1 , 1972, pp. H-39. S . The ki�d of cri�ici � � s I have in mind deny the possibilit y of "justificat ion" in moral thc:ury (v.hcn . )UShlicatao� . ts understood in a special way). 1 belie\·e that these criticis ms lie behtnd se\·enl of the ma)Or mora l theories in the twentie · 1 trad't' th-<:entury ana1 yttca k · · · 1 ton, for · . . cump , mtultlomsm, e "'?tmsm, and prescripti\'ism (see G. J. Warnock ' Contemporary Moral Pbllrnoplry [ London: �lacmtllan Co., \967]).
68
3
The Motivation to Be just
and the beginning of modern moral philosophy in the seventeenth and eigh teenth centuries. For the former the crucial motivational question (because of the nature of their moral theories) was, Why will a man voluntarily do evil? For the latter, the really puzzling problem is , Why will a man voluntarily do good ? In both cases, though , what is wanted is a way of relating what is taken to be standard (perhaps universal) human motivation to "the good" or "duty" or "obligation." A paradigm of this problem i s Plato's discussion of 'justice' in the Republic. At the beginning of book 2, after Thrasymachus has been subdued, Glaucon and Adeimantus reraise the question of the 'justification' of justice for Socrates to answer . The c hallenge is put in its starkest form. The value of justice in and of itself is to be shown. What is demanded is an adequate reason to choose justice over injustice. 6 We can leave aside the question of whether the remainder of the Republic does provide an answer, let alone a satisfactory answer, to this question. In her influential essay, "Moral Be lief� ,'' Ph ilippa Foot endorses this question as a perfectly appropriate one for a moral theory to answer. 7 She supposes that some "tough atheistical charac ter" raises the question, Why should I be just? and she presumes that a satisfactory answer must show that justice is a good to the just man. "The crucial q uestion is: 'Can we give anyone, strong or weak, a reason why he should be j ust?' "8 (This means, of course , Can we give each and every person a reason to be j ust? not, Is there some person to whom we can give a reason to be just?) Now, of course, it would be very nice to be able to do this . However, it seems that the main tradition of authors who have tried to do this has fai led. Usually, their fai lure · seems to be traceable to limitations imposed by their motivational assumptions. A recurrent candidate for being the explanatory principle of human action is self-interest. It has always been presumed that you have successfully given a reason for action to �omeone �tho�gh not necessarily a dec isive one) when you have shown the actl.on t.o be In h1s self- interest. The difficulties w ith establishing a reason for bemg J USt, on the basis of self- interest, are obvious. It simply is the case that justice can, and often does, i mpose requirements on our actions which confl�ct wit.h . self-m terest. The traditional examples are too well known to reqUJre their rehearsal he re . It is important to emphasize that considerable portion of, for exampl , . � � SOcl �l contract political theories and utilitarian mo al theones can be p� ausl r bly I nterprete d as having been designed to provide reasons, from the pomt of 6. Plato , The Republi , trans. Frances .\1. Cornford (N ew York: Oxford t!nive�sity Press, c 1 5 o the relatJOns�•p betw en 94 �· PP· 5 1-5 2 . For interesting discussions of Plato's account happmess and ju 10 Pklto. A stice , see Gregory Vlastos, "Justice and Happmess m Tbt RepubliC, Colkc s, ed . Gregory Vlastos, vol. 2 (Garden City, N . . : Dou leday & Co. , 197 0, PP · o[Essay trQn 66-95; an4 Ric hard Rt1MW 82 , Kraut, "Egoism, Love, and Political Office m Pla to , Pb•wsopbical 110 • J 0 97 J ): H().....44 . : 011ford . 1. Phili ppa foot, "Moral Bel iefs , " in Tbtories ofEtbia, ed. Phili ppa Foot (Oxford Umversity Press, 1 96 7), pp. 5� 5 7. 8. Ibid., p . 97 .
y
69
:
?,
�
4
Ethics
view of human interest (or self-interest), to all human beings to be moral. 9 Hobbes's account of the basis o f political obligation attempts t o show that the existence of a political community, which existence depends upon the obse
�
vance of certain laws, is in the interest of each and every individual. H � me s account of justice illustrates very well the difficulty of providing � satts fac . tory account of the motivation to be just. H u me provides a very mge� tous "justication" for justice on utilitarian grounds. Justice is not a natura vtrtue
�
like benevolence, according to Hume. Every single act of the natural \'trt � e of benevolence is of value in itself (gives pleasure from a general point of vtew). However, this is not true of every act of justice. This does not mean merely that justice and self-interest can conflict; it means that an act of justi� e c an . conflict with the general interest. It is the systematic character of ! ust � ce which justifies it on grounds of utility . The "whole plan or scheme" of J UStiCe is a necessary condition for the existence of society, and life in any stat� of society is better for every single individual than any life outside of soctety could be. 1 0
Even if we allow that Hume's very clever argument about the v al � e of a system of justice to society is successful, it does not necessarily provtde an answer based on interest for each individual as to why he should act as j u stice requires. It is true that man is better off in a society than he would be without society, and it may be true that the general accepting of rules of j ustice is a necessary condition for the existence of society. It does not follow that my adherence to the rules of justice (either generally or on this particular occa sion) is necessary for the maintenance of society . Hence, it does not follow that it is in my interest to be just. Of course, H ume allows for other than self-interested motives, but his construction is designed to align j ustice and interest, and it fails to do so . In the case of justice . . . a man, taking things in a certain light, may often seem to be
a loser by his integrity. :\nd though it is allowed that without a regard to property no society could subsist; yet, according to the imperfect way in which human affairs are conducted, a sensible knave, in panicular incidents, may think that an act of iniquity or infidelity will make a considerable addition to his fonune, without causing any considerable breach in the social union and confederacy . That
honesty is the best policy
may be a good general rule, but is liable to many exceptions; and he, it may .perhaps
be thought, conducts himself with most wisdom who observes the general rule ami takes ad\·antage of all the exceptions. I must confess that, if a man think that th is l reasoning much requires an answer, it would be a little difficult t o find any which wil i n per h suc t � him ap� ar � atisfactory and convincing. If his heart rebel not against es s, he h �s � tous maxtms, 1f e feel no reluctance to the thoughts of \'illainy or basen mdeed lost a constderable mottve to \'irtue, and we may expect that this pract ice w tll
�
be an swerable to his speculation.u
.
9. Thr \·rry brief remarks whi�h I make about Hobbrs, Hume, am\ Kant are sometimes con�,·er�tal, and they would ob\'lously have to br argued for at more length than 1 h a , e avatlablc •f they were to br completely convincing. . 10. Henry D. Atken, ed., Halfft's Moral tmd Political Philosophy (:-.lew York: Hafner Publ ish. tog Co., 1948), pp. 277, 6S, 66. I I . Ibid. , pp. 2 � 1 .
·
70
5
The Motivation to Be just
The problem here is that, on Humean grounds, there seems to be no possible reason for calling the man who indulges in this line of reasoning a knave. Hume has not offered this man a reason to be just, but that seemed to be what his construction was meant to do. Kant was one of the first philosophers to raise questions not just about the technical problems involved in providing a nonmoral justification, based on interest, of morality, but also about the very propriety of this sort of justification . Kant h a s one of t h e most complex views about the motivation to be moral of anyone in the history of morality. This complexity arises from two features of h is philosophical theorizing. First, there is his well-known at �empt to separate moral motivation from any form of inclination or self mterest. Second, there is his di fferentiation of the phenomenal and noumenal selves; his assignment of positive freedom to the noumenal self; and his doctrine, from the Critique of Pure Reason, of the impossibility of our having knowledge which goes beyond the bounds of possible experience. If we raise the question, What ought I to do? we receive our answer in terms of acting on the categorical imperative. Kant attempts to connect the categorical impera tive with the conceptions of freedom and reason . But if we raise the question, How can I be moral? or How can there be any motivation to be moral? we are told ,
How pure reason, without any other incent i \'e, wh ere,•er they may be deri,·ed, can by itself be practical, i . e . , how the mere principle of the uni,·ersal \'alidity of all its m� xi ms as laws (wh ich would certainly be the form of a pure practical reason), Itho ut any � material (object) of the will in which we might in ad,·ance take some mterest, can itself furnish an incentive and produce an interest which would be called pure ly mora l ; or, in other words, how pure reason can be practical-to explain this, all hu ma n reaso n is wholly incompetent, and all the pains and working of seeking an expla natio n of it are wasted . 1 2 It is t herefore n o objec tion to our deduction of the supreme principle of morality, but a rep roach wh ic h we m ust make to h u ma n reason generally, that it cannot render co mprehe nsible the absolute necessity of an uncondition al practical law (such as the categoric a l impe rative must be) . . . . . \nd so we do not indeed comprehend the �ractica l unconditional necessity of the moral imperative; yet we do �omprehend its _ �nc�mp rehe nsibility, which is all that can be fairly demanded of a ph 1 losophy which
10 Its p ri ncip les stri v es to reach the l i m it of human reason . 1 3
Not only is there no answer to our question , but theoretical reasons are produced whic h show an answer to be impossible. . . th1s Kanua n . �ric hard , in his " Duty and Interes t," give� a ver� ion . of dlssatlsfac ion with the "justification:· of morality wh 1ch �� ��t bas�d on � , Simply general epiStemological and metaphy sical groun ds, as Kant s IS. He
polis: . 1 2 . I. Ka nt, Foundations of tiN Metaphysics of Morals, trans. Lewis White Beck (Indiana L•brary of Libe \ ral . ns, 1 959), p. 8 1 . 1 3 . Ibid p 8 3 · 1 (Ov'o �•· rd·· Oxford University "· Pnchard, Moral Obligation: anJ Duty and [11 tms .. . H · ·�· Pr JA ess, l968), pp. 204, 2 1 1 , 2 1 31 4.
·
·
.
·
71
6
Ethics
t are i rrel� vant to maintains that all attempts to base moralit y on interes �y are mten morality. He says that they "will do nothing to prove � hat th � are t �ue. hfe y ordmar our of ons to prove, t•iz. that the moral convicti own traditio n , h1s Prichard himself offers, as an alternative to this justifyin g j u dg ment s are moral that ons assumpti the moral intuitionism, which retains matel Y to meaningful and capable of truth value, but demes that they are u u category (or be justified on nonmoral grounds and instead accepts some moral . ption s tha t categories) as primitive . One could instead reject the assu m value of Prichard retains, namely , the meaningfulness and capability of truth . et ICS moral judgment . .\.s is well known to all students of twentieth-centu ry vano us in the analytical tradition this alternative has been well explored by
���
·
1·
�
so-called noncognitivists. Does Prichard's objection to attempts to base morality on interest, even if we belie\·e it to be correct, necessarily rule out all reasoning wh ich attempts to show a connection between what morality is and human interest? My own belief is that it does not, and that though the tradition of justification may ha,·e been misguided, nonetheless there is a task for moral philosophy here. The important thing is to be clear about the nature of the task. We shall return for a moment to Kant's moral philosophy , a return encouraged b�
Rawls's placement of Kant at the center of the tradition in which he IS writing, to see an example of reasoning connecting morality and interest b t not in a justificatory way . (I shall later try to show some features of Rawls s reasoning which do bear a similarity to Kant's work . ) W e are, perhaps, surprised to discover that Schopenhauer accus ed Kant 16 of being an egoist in his statement of the categorical impe rat ive . . Schopenhauer is wrong, but it is instructive to see why he is wron g. He IS . thinking of the argument which Kant presents in section 30 of The Metaphyst cal Principks of Virtue, an argument which is made more rigoro usly in t he Foundations of the Metaphysics of Morals in Kant's17 discussion of the fou rth
�
example (an imperfect duty to others) of an application of the first form ula tion of the categorical imperative, an example in which a man con sid ers whether the maxim of a noncharitable act could be w illed to be a u niv er sal law. Kant readily admits that such a maxim could be a unive rsal la w of nature, but he denies that it could be willed to be such a law , "For a will which resolved this would conflict with itself, since instances can often a rise in which he would need the love and sympathy of others, and in w h ich he w ? uld have robbed hims� lf, by such a law of nature springing from his ow n w1ll, of all hope of the a1d he desires . " 1 8 Schopenh auer says of this kind of arg�m�nt, "It is expressed here in the clearest possible manner that moral obhgauon rests absolutely and entirely on assumed reciprocity . Conse-
1 5 . Ibid . , p. H .. . 1 6 . . \rthur Schopcnhaucr, On r« Basis of Moraliry, tra ns. E. F. J . Payne (Indianapolis: Library of Liberal .\rts, 1965), pp. 89, 90-91 . 1 7 . Kant, P · " I ; a nd TM Mtrapbysica/ Principks of Virrut, trans. James Elli ngton (ln . _ \rts, 1 964), p. 1 1 7 . 18. Ibid.
72
7
The Motivation to Be Just
quently it is utterly egoistic and obtains its interpretation from egoism. Under the condition of reciproci ty, egoism cunningly acquiesces in a comprom ise. "19 It is not too d ifficult to see where Schopenhauer has gone wrong here. He has assumed that Kant's argument is designed to provide a nonmoral motive for being moral-that it is presented as an answer to the question, Why ought I to be moral? But Kant's argument really concerns what morality is, in th is case, not the motivation to be moral. The point is that Kant's argument establishes the morality of an action in terms of the rationality of willing the maxim of that act to be a universal law . This does not, of course, mean that acting morally will be acting in one's own interest because, in fact, the maxim of your act is not going to be a u niversal law. Kant does connect the rationality of willing a maxim to be a universal law with self-inte rest , but this only tells us what morality is . It does not provide us with an argument from self-interest for being moral . (In the Critique of Practical Reason, Kant h as a chapter on the incentives of pure practical reason, but this would take us too far afield . ) Schopenhauer's criticism of Kant is quite relevant to our discussion of Rawls, as we shall see, because this same confu sion about what function an argument is serving can easily occur in reading Raw ls. We might finally note something that has often been said of the ques . tion, Why ought I to be moral? This is that the q uestion is not one that can meaningfully be asked. This question is equivalent to, Why ought I to do what I ought to do? and this latter question is not significant (or important, or someth ing) since it is a tautology that I ought to do what I ought to do. One can attempt to rescue s ignificance for the question by saying that there are two "senses" of 'oug ht' (usually identified as a moral and a nonmoral sense) and then reph rasing the question , Why (nonmorally) ought I to do what I . (mora lly) ought to do? However, this question, on the face of 1.t, IS not a q uestion for moral theory, but perhaps for psychology or some exten �ed theory of rational choice. To answer it, one might have to focus attentton back on presu med motivational factors which affect the person who asks it. One difficulty for the ordinary reader in extracting Rawls's view �n this . t ha t A Theory ofjustice "is a long book , not only in pages, " as Rawls hlm �elf IS says. 20 Indeed, he has taken the remarkable step of providing a reader's gmde to an ab ridged version of h is own work in its preface. Unfortunately the suggested central core of the book , though i ndeed it "comprises �ost of the essentials of the theory," does not include the passages e�senttal to our pro bl�m-the problem of the sou rce of the motivati?n to � JUSt. However, even tf one reads the entire boo k straight through, mcludmg chapter 8 (the essential cha pter for our pu rposes), one may still be unclear a bout what . Raw.ls's view on this topic is. One reason for this is � hat there IS so much detatled mat erial presented in each of the different secttons that one may lose . Sight of the connection s between these sections-and one needs to see chap•
19· �hopenhauer, pp. � 1 . 9 2 0. Rawls, A TbtrJry ofjustice, p . viii.
73
8
Ethics
t d reason is that importa� ter 8 with chapter 2 very much in mind. A secon s Rawls of different parts motivational questions are raised in at least two one at ions ering quest theory, and one might easily think that Rawl s is answ ans';er at anoth�r will he whic and even raised yet point which he has not _ f Rawls s th.eory . as point. (Indeed, I believe that some of the early c �ltlc� � . on th1s error.) was presented in the series of articles, based the1r cnt1c1s ms strand of the Rawls presents his view as a self-conscious revival of one aim is to present a great tradition-namely, social contract theory . "My level of ab conception of justice which generalizes and carries to a higher ke, Rous straction the familiar theory of social contract as found , say, in Loc contr�ct social seau and Kant."2 2 Now it is quite true that Rawls' s theory is a w1�h Rawls theory, but for many readers it will be misleadin g to associate s 10 belong this tradition (unless they have a detailed knowledge of why Kant ; hoi not does it) because they are likely to read into Rawls views which he this ed r m for the most sigr�ificant thing about his theory is how he has transfo . of 1t . If tradition (or, at least, how he has re-presented Kant's transformation . s we rea.d Rousseau through Kant's eyes, he may also have held the v1e� i al s oc onal Rawls does about the function of the social contract). Traditi ove contract theory presented an argument which was meant to justify the m s Rou and from a state of nature (variously interpreted in Hobbes, Locke, l ica polit lar seau) to a civil society. It usually attempted to justify particu be d coul utions institutions, or standards in terms of which political instit judged, as being the outcome of a voluntary agreement made by rational self-interested individuals. Since the acknowledgment of these princ ipl�s , �r the acceptance of these institutions, restricted an individual's behav iOr 1n certain ways, he would only accept them if he expected to get someth ing from the arrangement which was for his own interest . The tradi tion e m phasizes what might be called "economic rationality . " (Herbe rt Simon dis cusses this model of rationality in a number of places . )2 3 Rawls assumes so�e�hing like an extended model of economic rationality in the original position. However, he does not assume, as someone like Hobbes does , that we need to e�plain actual moral actions (for example, adherence to the con tract) solely m terms of economic rationality . Much of this traditional theory made a historical claim that this sort of s contract was actually the st�rt of civil society . Moreover, people like H o and Locke made psychological assumptions about the actual indiv iduals m-
�
�!
�
bl�e
2 1 . I shall n01: attrmpt to substantiatr this charge hrre. Examples of what 1 have in mind _ found 10 h . �erelman, }Krticr ( rw Yor ; Random House, 1 967 - 1 ; and can ), PP · 3 9 5 _ wpl11ca Lansm� Pollock, :\ Di_lrmma for Raw s, Pbiil l Studitr 22 ( 1 97 1 ): 37-4 3 . The inte res ted _ readrr •s u�. to fam1hanz:e hsm ;;elf wsth Rawls's articles and then to read this criticis m . Even so ast ute a philosopher as M au Mandrlbaum, in his review of Rawls's book (Hisrory arui TMry no . 2 l 1973J; 240-50), seems to mr to be confused about Rawls's argument on this sssue. It 1s unclrar from R. M. Harr's "Rawls' A TMry of}IUtict-l," Philosophical Quarterly 2 3, no . 91 ( 1973); 14+-S S , what he has understood of Rawls. 2 2 . Rawls, A Tbtory of}IIStict, p. I I . B. Hrrben ;\. _Simon, "Mathematical Construction in Social Science," in Pbil0st1pbical Prohlnrrs crftht Social Seta�Cts, ed · David Braybrooke (�rw York: Macmillan Co. , 196 5), pp. 8+-8 5 ·
5=
�
1�,
nce
�
�
�
74
9
The Motivation to Be just
volved in contra�tin g into the civil society-among these assumptions was one that self-interest was, at least, a dominant human motive, if not the mainspring of all action. Rawls specifically rejects any historical interpreta tion of h is version of the contract theory. He describes a situation in which principles of justice are to be chosen, principles in terms of which judgments concerning the j ustice of competing claims can be made. I shall not review in detail his description of the original position, which incorporates cenain constraints on the concept of right, or his account of the character and knowledge of the indi viduals who are to choose from among the competing conceptions of j ustice. This is all fascinating material, and much of the discussion of Rawls's work has focused (quite properly) on the details of his construction . I presume a general familiarity with his procedures. The one thing I should mention is that the one imponant motivational assumption Rawls makes is that the panies are mutually d isinterested . He wants to rely on noth ing more than limited altruism (or self-interest extending over, perhaps, a larger 'self ' equivalent to a family), and this is the weakest motiva tional assum ption he can make . Thu s, we can see that the first imponant place in which there is a motivational assumption working in Rawls's theory is here, in the description of the origin al position . We must have some motivational assumptions be cause the panies are being asked to choose from among alternative concep . tions of justice, and even an arbitrary choice must be motivated in some way. So far, as Rawls says, we have a situation familiar in social theory. "Th at is, a simplified situation is described in which rational individuals with cenain ends and related to each other in cenain ways are to choose among variou s cou rses of action in view of their knowledge of the circumstances. What these indi viduals will do is then derived by strictly deductive reasoning from these assu mptions about their beliefs and interests, their situation and the opt ions open to them. Their conduct is, in the phrase of Pareto, the resultant of tastes and obstacles. "Z4 It is within this situation as described that one would like to provide a strictly deductive argument for the principles of j ustice. Ra wls does not have such an argument, but he does have i � a limited way a sketch of such an argument . Some of Rawls's critics have m1staken �he . form of his argument. z s They have criticized his description o� t�e panlcl panrs and the situation in whic h they find themselves as unreahst1c-but of course it is not intended to be realistic. It is intended to provide an argument for the choice of certain principles on the basis of minim� ! as� um.ption� about hu man m otivation, where the choice is made within a Situation m which the constraints of having a morality are enforced.
�
It is tlear, then , that the original position is a purely hypothe� ical situation. othing resem bling it need ever take place, although we can by deh berately foll�wmg the con strai nts i t expresses simulate the reflections of the panies. The concepuon of the 24. Rawls, A 2S · Sec n . 2 1
Tbtory ofJustice, p. 1 19. above for examples. 75
Ethics
10
. . s to d uct excep t ·I nsofar as it .trie . original position is not mtended to explam human con ce . JUSti . f o sense · · g a h avtn 1 account for our moral JUdgments and helps to· exp am our d ere "d s1 con our by ted "fes J ustice as fairness is a theory of our moraI senttmen ts as mant ht thoug our affect ably presum ·J udgments in reflecti\"e equilibr ium. These sentime nts · ·ma1 posl� lon "�s part of and action to some degree. So while the concept ion of the on g . ns that actual sttuatto the theory of conduct, it does not follow at all that there are the play cepted ac be resemble iL What is necessary is that the principles that would
··
26 requisite part in our moral thought and action .
the q�es ln particular, the entire argument has said nothing in response to motiva tion, Why will human beings act as justice requires? Indeed, if the hum�n tional assumptions of the original position were the whole truth about t IS, -tha motivation, it seems clear that men would not be just in "real life" outside the original position. ested For the fact that in the original position the parties are cha racterized not as inter the hold who in one another's concerns does not entail that persons in ord inary life Clea her. �ly anot principles that would be agreed to are similarly disinterested in one
the two principles of justice and the principles of obligation and natural duty require us to consider the rights and claims of others . .\nd the sense of justice is a nor� ally effecti\"e desire to comply with these restrict ions . The motivation of the persons m t e original position must not be confused with the motivation of persons in ev ry d ay g who accept the principles that would be chosen and who have the correspondm a situ sense of justice. In practical affairs an individual does ha\"e a knowled ge of his tion and he can, if he wishes, exploit contingencies to his advantage. S hould his se nse _ of justice move him to act on the principles of right that would be ado pted tn the original position his desires and aims are surely not egoistic. He voluntarily t ake s 00
e
� ��fe
the limitations expressed by this interpretation of the moral point of view . 2 7
How then does Rawls avoid exactly the problem that Hob b�s encounters-why will people who would choose the principles of justice tn the situation described act in the ways required by these principles w hen the limitations of the original position are removed? If the principles of just ice are now clear (assuming for the moment that Rawls's argument for his theory is valid), must we not still ask, Why will men do what justice requires ? 2 8 Rawls's answer to this question is quite complex and involves more of his book �han I s all mention in my rather simplified version of his answer. The po:; 10n of 1s book t? which I shall refer is chapter 8, "The Sense of . . Ju�ttce, wh1ch •s an ampltfied version of the article of the same name . Here is the second important place in which motivational consider ations enter the theory . For Rawls it i� important that a conception of justice should relate to actual human mottvatton m such a way that a well-ordered society will be �table with respect "to the justice of its institutions and the sense of ju stice needed to maintain this condition. " He says that "inevita bly, we shall have to
� �
•.
26. Rawls, A
Tbtory of}IISticr, p. 1 20. 27. Ibid. , pp. 147-48. 2 8. Of cou� , we need not assume that they will act as justice requires; we can qualify the _ q uest1on by askmg what are the reasons for acting justly that those who do act justly have?
76
II
The Motivation to Be Just
take up some rather speculative psychological q uestions; but all along I have assumed that general facts about the world, including basic psychological principles, are known to the persons in the original position and relied upon by them in making their decision ."29 I shall not consider here the general problem of stability, nor shall I go into very great detail about the psycholog ical principles which Rawls believes to be true of h uman beings. I am mainly interested in the relation between the two parts of his theory-the argument for the principles of j ustice and the psychological theory of how human beings acquire a sense of j ustice. Rawls gives an account of the development of the sense of justice in terms of a three-stage process with a psychological principle corresponding to each of the three stages. He draws here on philosophers like Rousseau and psychologists like Piaget and Kohlberg. It will be helpful to state the three principles here in order to comment on certain features of them. First law: given that family institutions are just, and that the parents love the ch ild and mani festly e x press their love by caring fo r his good, then the child, recognizing their evident love of him, comes to love them.
Second law: given that a person's capacity for fellow feeling has been realized by attach ments in accordance with the first law, and given that a social ar rangeme nt is just and publicly known by all to be just, then this person develops ties ?f frie ndly feeli ng and trust toward others in the association as they with evident Intentio n comply with their duties and obligations, and live up to the ideals of their station.
acquiring
Third law: given that a person's capacity for fellow feeling has been realized by his forming attac hments in accordance with the first two laws, and given that a soctety's institutio ns are u st and j are publicly known by all to be just, then this person acquires the corre sponding sense of justice as he recognizes that he and those for whom he cares are the beneficiar ies of these arrangements. 30
First of al l , th �se are not "laws" of psychology, which render ev�ry aspect of the develop ment (or nondevelopment) of moral sentiments exphca ble or predictable. They represent tendencies which are pr�s� me� to oper ate, other thi ngs being equal. They may be taken as descnbmg Important �eatures of the way in which human beings normally develop. A second . Interesti ng feature is that these psychologica� principles t �vol.ve references to the moral co ncept of j u stice itself. I earlier said that Rawls � vieW represents a transfo rmatio n of some of the great traditiona l views of soctal contract theory such as Hob bes's . Here is a major point of difference. It was ofte� assumed th at the argu Why ment made to an individu al who posed the questmn, sho�ld I be nmoral n of terms in ? moral? or Why should 1 be just? must be . tnotivation. This of the long attempt to tdenufy duty point was whole the . and I.nterest. may What Rawls's psychological principles suggest IS that there 29. Rawls, A Tb«Jry ofJust�t, p. ,.56. JO. Ibid . • pp . 490-9 1 .
12
Ethics
"natur al" may in � ite be a "natural" interest in morality itself. The word moral mouva misunderstanding here. It does not involve the claim that the Rath�r, R�wls tion will inevitably be present in each and' every ind ividual. ral" m � md), seems to agree with Aristotle (who has a different sense of "natu the VIrtues who says, "Neither by nature, then, nor contrary to nature do are ma�e arise in us; rather we are adapted by nature to receive them, and in morahty perfect by habit."31 What I above called a "natura l" interest would be this adaptation by nature to receive the moral virtues . . olog1cal At any rate, it is in terms of what is explained by these psych can principles-namely, the development of a sense of justice-that one as such explain the motivation for particular acts of justice in hard cases d a Hume provided-particular cases in which the interest of the individual � w le �ll even the interests of society conflict with the demands of justice . Peop Will s act justly in these cases (when they do-for, of course, they sometime fail to do what justice requires) because they have a sense of j ustice . Now we can see the relationship between the two parts of the t eory . The argument presented about the original position tells us what justtce IS. Justice, for Rawls, is defined by the two principles. Methodologica lly, the�e principles are checked against our considered moral judgments about wh at IS just in particular cases, in order to see whether these j udgmen ts cou ld have been derived from the principles. The psychological construction, if c�rre�t, would show us how men can acquire a sense of justice. The sense of JUS�Ice can be used to explain the motivation to act justly in pa rticular cases. Nottce, all of this does not provide an argument addressed to any rational self interested being, who lacks a sense of justice, which will show him th at he "ought' to be just-in a nonmoral sense of "ought." It will be remembered that th 1s was exactly the argument demanded by Mrs. Foot in " Moral Be liefs," but in reconsidering this demand she said ,
�
:
I was in this difficulty because I had supposed-with my opponents-tha t th e
thought of a good action must be related to the choices of each individual in a ver y special way. It ha not occurred to me to question the often repeated dict um that . moral JUdgm� nts gtve �asons for acting to each and every man. This now seem to � me to be � mtstake. Qmte g nerally the reason why someone choosing an A may , be .; expected . to c oose good A s rather than bad A's is that our criteria of goodness for any class of thmgs are related to cenain interests that someone or other h as in those things. When someone shares these interests he will have reason to choose the good A's: otherwise �ot. Since, in the case of actions, we distinguis good an d bad on h account of the mterest we take in the common good, someone who does not care a amn what happens to anyone but himself may truly say that he has no reason to be JUSt. The rest of us, so. long as we continue as we are, will try to i mpose good conduct � pon such a man , saymg "you ought to be just," and there is this much tru th in the tdea that there are categorical imperatives in morals . n
�
�
�
3 1 . Aristoelc, Nicom«bttm Etbia 1 1 03a24- 2 � . in fntrodw:t ion to Aristotle' c:d. R. (�c:w York: Random House:, 1947), p. H l .
32.
Foot,
p. 9.
78
\tcKeon
.
13
The Motivation to Be Just
Hence, althoug h such an argume nt cannot be provided, this does not tell against the argume nt for the princip les of justice: no such argume nt is needed. To see this more clearly, we might ask what questions about morality a�e left unanswered , after the two parts of Rawls's theory that we have been . d1scus smg have been presented . We know what justice is, and we know why men will· be just (to the extent that they will). But, surely, at this point so?'eone is going to object, the very question which has been the subject of th1s whole paper has been left u nanswered: Why ought I to be just? After all, the principles of justice are checked by our considered moral j udgments about justice, but these judgments are made, in general, by those with a sense of justice, while the sense of justice is explained by reference to a set of psychological principles which involve a reference to justice as defined by the principles of justice. I s not the whole system interdefined, so that some parts are justified i n relation to other parts, but the system as a whole is never justified ? Would not a justification of the whole system consist of an argu ment whic h could be presented to any individual on any occasion, which would not appeal to any moral sentiment, but would show him that from his own point of view justice would be good to him? Is not the absence of such an argument simply an admission that this conception of justice is not justified? Is the question, Why ought I to be just? really left u nanswered? In a sense (the so-called moral sense of 'ought'), it is clear that acting justly is among the th ings you 'ought' to do . In the so-called nonmoral sense of 'ought,' this question can be u nderstood as equivalent to, Why ought I to be the kind of creature subject to these psychological laws? But this is clearly not a matter of choice for which action-guiding reasons can be given. Rawls attempts to connect theoretically the moral sentiments with the natural sen timents. It seems to me that , if someone genuinely lacked a sense of justice, th is lack would be connected with other ways i n which such a person would be abnormally related to other people-"if men did not do what justice requires, not only would they not regard themselves as bound by the rin i � � ples of j ust ice , but they would be incapable of feeling resentment and md•g nation , and th ey woul d be without ties of (riendship and mutual trust. Ther woul d lack certain essential elements of humanity."33 If this were true, Jt would indicate that there is, indeed, a "price" to be paid, and a terrible price, from the point of view of one who is capable of friendship and � utu�l affection, for the lack of a sense of justice. Notice, however, that th1� w1ll . ·�· Ex proba bly not appear to be a 'price' at all to those who �ave pa1d hypothesi, they are i ncapable of friendshi p and mutual affection; h �nce mca pable of know ing what they are missing. Rawls himself emphas1zes ·� t �e section "The Good of the Sense of Justice" that his argument there, wh1ch IS designed to show, on certain supposit ions, that it is rational f�r someone to affi rm his sense of justice, is n.ot the attempt to justify being a JUSt man to an •
.
. J J . John Rawls, "The Sense of Justice," in Morlll Con«pts, ed. J. Feinberg (Oxford: Oxford Untversiry Pres s, 1 970), p. 1 20.
79
14
Ethics
egoist. There is no such argument possible, nor is one needed. But then, one
�
��
might say, there is n� explan� tion of morality at all Althoug the of�e : . enigmatic words of Wtttgenstem do not always 1 lu� mnate a top1 � , I le _ that what he said about the temptation to say th1s IS worth cons1denng.
�
��
Here we come up against a remarkable and characteristic phenomenon in ph il� sophi
cal in, estigation: the difficulty-! might say-is not that of finding the solunon bur rather that of recognizing as the solution something that looks as if it were only a preliminary to it. 'We have already said everything.-:\Jot anything that follows from this, no, this itself is the solution ! '
This is connected, I believe, with o u r wrongly expecting an explanation , where � s the
solution of the difficulty is a description, if we give it the right place i n our considera tions. If we dwell upon it, and do not try to get beyond it . The difficulty here is: to stop. 34
e We might briefly return to Kant's moral theory and a passage I hav by aged already discussed, in order to clarify Rawls, a return which is encour _ Rawls himself. Remember that Schopenhauer criticized Kant for bem g l egoistic. When Kant considers whether one could will to make it a univ ersa t tha int po law of nature that there should be no charity, he first makes the
such a state of nature is perfectly conceivable and is indeed, perhaps, prefer able to a state where men claim to be charitable, but are not-but he goes on: "�ow although it is possible that a universal law of nature according to that maxim could exist, it is nevertheless impossible to will that such a principl e _ should hold everywhere as a law of nature. For a will which resolved th1s would c.onftict with itself, since instances can often arise in which he would need the love and sympathy of others, and in which he would have rob bed himself, by such a law of nature springing from his own will, of all hope of the aid he desires."35 This, of course, as I have already pointed out, is not a consequentialist argument, since it says nothing at all of one's actual expecta tion of receiving charity, if it is needed, in the future. Hence, the reasoning i s not egoistic, though i t does connect t h e concept' o f morality with ratio nal self-interest. It corresponds to the kind of reasoning which would occur in the original position as described by Rawls. Suppose we now try to raise th e question, Why should I let all of this reasoning about the categorical impera tive influence my action? Rawls, in discussing Kant, attempts to provi de him with a � answer to this question-that is, that acting morally is the best express1on of one. s character as a free and rational noumenal self-but, al
though Kant may have wanted to say something of this sort, this answer s�ems unjustifiabl� in Kantian terms. (I cannot defend my interpretation of Kant here; I can s1mply refer to its basis-a dissatisfaction with what Stra w son calls the "metaphysics of transcendental idealis m . ")3& I think the best thing we can do is say that, for Kant, we are rational free beings and that it is 34. Ludwig Wittgenstcin, Zmtl (Oxford: Basil Blackwell, 1967), p. S Se. 3 5 . Kant. Fo�
80
15
The Motivation to Be Just
because of those features of us that the categorical imperative applies to us. There is no further argument that we ought to act on it. Why ought I to be moral? for Kant would then be equi valent to, Why ought I to be a rational, free being? and again, as for Rawls, there seems to be no possibility of answering it other than to say-you are one. lncidenta.lly, Rawls appears to have an advantage over Kant here . It is a plausible interpretat ion of Kant that he associates rationality and morality so strongly that any i mmoral action is, ipso facto, irrational. Though Rawls sketches an argument to show that the choice of the principles of justice he proposes in the original position would be rational, and though he also argues that it is rational (on certain suppositions) for an individual to affirm his sense of justice, he is not committed to as strong an association of rationality and morality as Kant is. What is, in fact, rational for a particular individual will depend, not just o n general truths about human desires, but on h is own desires, aims, sentiments, and so on. Thus, it would seem to be possible for an unjust man to be rational (given that his desires, aims, and sentiments are of a certain sort and that he embod ies other standard features of rationality such as the abil ity to calculate probable consequences of his acts, etc.) on �awls's view. According to my intu itions about rationality and morality, this 1s an advantage for Rawls. With the structure of Rawls's theory i n regard to motivational questions in mind, we might briefly consider a particular criticism which one might make of his argument for the second of the two principles of justice . This principle has u ndergone a considerable revision between its first appearance m "Jus tice as Fairness" and its reappearance in A Theory ofJustice. In the former, it was stated as, "Inequ alities are arbit rary unless it is reas�nable to expect that they will work out for everyone 's advantage, and prov1ded the positions and offices to which they attach , or from which they may be �ained, are open to all . "3 7 In A Theory ofjustice, though the principle is first Introduced in roughly the same form, 38 in the course of Rawls's argument he determines the sense of this principle more exactly, and in so doing reformu lates it as follows: "Social and economic inequalities are to be arranged so that they are both (a) to the greatest benefit of the least advantaged and � b) . attached to offices and positions open to all under conditions of fa1r equality f as the op nown � po is ) in rtu (a stated nity. "39 The part of this principle � . . , _ diffe rence principle," and it is fundamental to Rawls s concepu?n of JUStlc� . . Now it is absolutely clear why the persons described as occu pymg the ongl . . na) positio n would choose the second principle in the form m wh1ch It was g.iVen in "Just ice as Fairness . " It wou ld surely be irrational to refuse to accept differe nces w h ich are guaranteed to work out for your own �nefit. Hence, . . even on the slim motivational assumption s made about the partiCipants m the 3 7 . John Rawls, "Justice as Fairn ss," in justice and Social Pot;ry, ed. F. A . 1 91'\ 1 ), p. 8 1 .
e glewood Cliffs, !'J .j.: Prentice-Hall, Inc., 38. Raw ls , A Thf()ry ofjustict, p. 60.
39. Ibid., p . 8 3 .
81
Olafson (En
16
Ethics
original position, the argument is clear. The problem is that the principle in this fonn seems too restrictive on allowable inequalities. The reformulated version has clear advantages, not least that it conforms more closely �o our intuitions about justice. However, the argument that participant� m � he original position will choose it depends upon the claim that t etr bemg _ rational in the situation described involves their acting on a maxtmm rule for choice under uncertainty. Rawls admits that "clearly the maximin rule is not, in general, a suitable guide for choices under uncertainty. "40 Ho'": ever, he thinks that it is the rational approach in the original position. (Davtd Lyons has criticized this argument of Rawls. )41 I will not try to judge the co�en�y of Rawls's argument here . 1 only want to point out that his weak motiVational assumptions about the original position require a very strong notion _o ra tionality (or of the restraints on rationality imposed by the original position). Of course, as I said, the difference principle seems to be just, but, obvious y ' . Rawls does not want to rely on any moral sentiments to influence part1c1. pants in the original posit ion in choosing the principles of justice; this would be inconsistent with the original position as described. Before I close, I would like to mention one other line of criticism that might be employed against Rawls. It might be said that his theory of justice only captures the intuitions about justice of a particular society at a particular historical stage, but that this stage, like all stages of history, is going to be transcended . Therefore, one can make no "universal" claims about just ice on the basis of Rawls's theory . A variant of this line of thought might point out that the psychological principles to which Rawls appeals in his explanation of the sense of justice, even if they are true, are not "necessarily" true. They may not have been true of human beings in the relatively distant past , and they may not continue to be true of human beings into the relatively distant future. Hence, the principles of justice which he derives might, at some point, prove to be irrelevant to the ways in which human beings can live together in society. Rawls himself suggests that the capacity for a sense of j ustice and the moral feelings on the part of humanity has a h istory and that it m ight receive an evolutionary explanation .42 The question is, Is there any reason to sup �se t at the conception of justice he presents occupies some favored place hJStoncally? I cannot pretend to be in a position to answer this question. However, I can conjecture that a kind of Hegelian account of the history of ju �tice_ and t e sense ?f justice might be developed which would claim that th1s history IS the htstory of the conception of justice which Rawls has _ developed commg to human self-consciousness.'n (Rawls himself is far too
�
�
�
�
�
40. Ibid . , p. l B . 4 1 . David Lyons, "Rawls versus Utilitarianism "journal 01'Philo...., s. J•. , October S ' 1972 ' PP· -rw , '
S J S-4 S . 42 . Rawls, A
Theory of]ust"t, p. SOl. 4J. I rulizr that this n:ff'l'l!llCC to He�l is far too brief to be illuminating for anyone who does not aln:ady have so� knowled� of H�l's view. My own view of Hegel's philosph y of history has lxen great� mfl�nced by my colleague Dennis O'Brien, who, however, is not _ n:spons1blc for my poss1bly m1sleading analogy.
82
17
The Motivation to Be just
modest and tentative to make such a claim directly. ) Whether we are sym pathetic toward such an account, or toward the objection to which it is addressed, will depend upon whether, like the right Hegelians, we em phasize the system at the expense of the dialectic, or like the left Hegelians, emphasize the d ialectic at the expense of the system. 44 Rawls's own emphasis seems to me to be systematic. Indeed, my entire account of Rawls has focused on how various parts of his account relate to each other. It seems to me that even more fundamentally important than his detailed account of justice is his revival and mastery of the architectonic of moral theory. Kant said that "by an arch itectonic I understand the act of construct ing systems. As systematic unity is what first raises ordinary knowledge to the rank of science, that is, makes a system out of a mere aggregate of knowledge, architecton ic is the doctrine of the scientific in our knowledge."4s Surely, no one can read Rawls without sensing the power of the kind of organization he has introduced into his material. He himself says, "I must disclaim originality for the views I put forward. The leading ones are classi cal and well-known. My i ntention has been to organize them into a general framework by using certain simplifying devices so that their full force can be appreciated. My ambitions for the book will be completely realized if it enables one to see more clearly the chief structural features of the alternative conception of j ustice that is implicit in the contract tradition and roints t�e _ way to its further elaboration. "46 Interestingly, Rawls, by prov1dmg this framework for the consideration of problems of morality, has laid the ground for those who wish to criticize his own theory of justice. John Searle has written of Noam Chomsky's linguistic theory, "Not the least of its merit� is _ _ provides an extremely powerfu l tool even for those who d1s�gree that It w_Ith m��� features of Chomsky's approach to language."47 The �ost mterestmg �ntJCJsm of Rawls will come from those who accept many of h1s met�odolog Jcal assumptions and argue from w ithin the framework he has provided.
Will ia m J . Brazill, T� Y()Ung Htgtlians (:-.lew Haven, Conn., and London: Y a le ss, 1 970), esp. chap. I . · • tacma'IIa n Co · • 45 K ant, CntiiJUt · (London·• n · · of Purt Reason ' trans . :"lorman Kemp sffil rh 1 96 1 ), p. 6 5 3 .
44· . ee UOlvers uy Pre .
S
· .r;ustrct, 46. Rawls • A T<·· ocvry 0, p. 47 John
... vm.
.r&olu June 29, '
R . Searle , "Chomsky 's Revolution in Lmguasucs, Nt'W Yort t'W'W 01 l97 Z · pp . 1 6-24 . • .
83
. .
"
Human Evolution and the S ense of Justice ALLAN GIBBARD
I
I
�
begin wi s me speculation . First, psychological speculation: why are we con� cern ed wtth J USti_ ce ? A traditional answer is that we are b enevolent: we are con cerned for the well-being of others. Perhaps this is a fundamental fact about our co ncerns, or perhaps it holds rrue b ecause each of us has a capacity for sympathy: · the h app mess of others makes a person happy, the unhappiness of others makes a erson unhappy , and a person wantS to be happy rather than unhappy. Whatever Its psy chologi cal b asis mav be, it is b en evolen ce that underlies our concern with j ustice .
?
·
Raw ls give s a differen t answer: our concern with justice, he says, is grounded . tn a capacit y for "re cipro city" ( 1 9 7 1 , sees. 75-76). It is a psychological law, ac co rdt_ ng to Rawls, th at under certain conditions favorable to moral development, if � person rec ognizes that he and those for whom he cares benefit from his society's Institu tion s, and if, moreover those institutions are just in Rawls's sense and are pu b licly know n to be so, the the person will acquire a sense of j ustice. Roughly, t en, he will supp ort just institutions that benefit him, and that is what constitutes hiS conc ern with j u sti ce. The issue , as Rawls apparently sees it, is whether a person with a normal sen se f · St · JU i ce w ants to benefit o ther people irrespectively of whether he expects thern to be ne f'It h 1m , and whether the person regards th e benentS an mstltutlon con ers o n oth ers as con siderations in favor of that institution, regardless of what e Insti tutio n does fClr him . The utilitarian, according to Rawls. "stresses the capac tty for S}'mp ath v" and relies on unconditional altruistic impulses to make us support the in St! tut 1 0ns he favors. Such altruistic impulses, he concedes, do ex1st, but they are far . . ' weak er than the . disp ositions that underlie his own system : d 1sposmons to ec lpro c ate benefitS. "By appealing straightw ay to the capacity for sympathy as a oundau. o n o f just co nduct in the absence of reciprocity, the principle of utility not
�
�
�
�
0
·
·
·
-
·
'
·
;
85
·
·
32
ALLAN GIBBARD
n depends on weaker and less commo only requires more than justice as fairness, but inclinations." (p.
501)
reciproc ity that u nd erIt is, then, according to Rawls, a sense of justice as fair great burden on the nolies our strongest moral concerns . This, of course, places a . · erms of constitute fa1r cion of "fair reciprocity." When we trade benefits, what fair apporno nment trade? When an institution benefits everyone, what constitutes a that everyon does of those benefits? An inStitution may be reciprocal in the sense · · · · 11 nd yet 1t may better with the institution than he would w1th no msmuuon at a . a . is reciprocal m which well be that for each person, there is an alternative institution · · · n s and ar that sense and which would benefit him more. Our support for msutut!O ng o ur getti are we rangements that benefit us seems to depend on a conviction that _ t said, Tha ry fair share of those benefits - as it does on Rawls's psychological theo · · · . ·11 e about we are left with the problem of what will be regarded as fau. What WI It b tho se some of the many possible reciprocal arrangements that will make us regard of as displaying fair reciprocity? That, in psychological guise, is the central prob lem
�
A
Tbeo")' of justice.
�
.
Game theorists call a situation of the kind I have been discussing a "b argam ing problem." More specifically . a bargaining problem has these features . There are alternative systems of cooperation or mutual restraint, each of which is mutually
�
beneficial in the sense that its implementation would make everyone invol ved b t ter off than he would be with no cooperation or restraint. There is a conflict of In
terest, though, over which of these alternative mutually beneficial systems to im·
0� ·
piement: for each such synem, there is another which makes someone better Price haggling, for instance , involves a bargaining problem on a small scal e : there IS
a range of prices at which a sale would benefit both panics as compared to no sale , but for any price, the seller prefers a higher one and the buyer a low er one . Th e
problem of so cial organization is a bargaining problem on a large scale : many alt er· native forms of social organiz.ation give everyone a bener life prospect than d o es a State of nature, but different forms allo t the prospects differently.
t-;ow an imponant feature of bargaining problems is this: if everyon e exp ects a particular outcome, then it pays everyone to insist on what that outcome gives him and on no more. More precisely, suppose there is a mutually benefic ial system S
with the following feature: Each person thinks that if he agreed to s, eve ryone else would , and that every alternative system by which he would benefit more is una t· uinable, in the sense that if he held out for it, there would be no agreement. Then
if everyone aets rationally and self-interestedly, everyone will agree to s . Thus given suitable expectations, bargaining problems will be resolved by rational panics who are purely interested in how much they get.1 In fact, though, people confronted with bargainin g problems do not seem to care only about how much they get. They care pass i onately about the fairness of the outcome. To get less than expects to be share will
one's fair share is to be "cheated," and a p erson wh o cheated will resent it. In contrast, a person who expects to get his fair
not normaly l resent the outcome, even though he knows that under some arrangements, be would get more. We extend our concern with fairness
alternative
86
H UMAN EVOLUT ION AND THE SENSE OF J USTICE
33
to bargaining situations i n which we are not involved. In the small , this concern is with such things as a fair price or a fair wage; in the large, perhaps, it is at least a part of our sense of justice. The psychological questions with which we are left are these: What is it about a system that makes us regard it as fair, and why do we so much care when we regard a system as unfair? To ask these questions is to inquire into the nature of our sense of justice or fairness.
II
I tum now to another kind of speculation. How might humans have come to have a sense of justice? I am asking here for an evolutionary explanation: What kinds of psychological propensities are involved , and how might evolutionary theory explain humans' having those prope nsitie s ?
One e xplanation is that the human sense of justice is grounded in a propen sity to be benevolent, and that this propensity evolved through kin-seiecrion.2 Let me sketch the theory in as plausible a version as I can. First, some terminology: I shall call a behavioral propensity biologic:Jliy altru istic if it reduces the individual fitness of one who has it but enhances the individual fitness of others . Here individual fitness is roughly one's expected number of de
scendants after a few generations. What kin-selection explains is how there could be
a biologically altruistic propensity to be benevolent. The claim is not t!u.t any pro pensity to be benevolent must be biologically altruistic. Benevolent propensities may well enhance one's individual fitness, for others may be disposed to reciprocate good with good and ill with ill . No special explanation is needed for benevolence in such an environme nt of reciprocity, so long as the propensity is shaped to yield the best net return to the benevolent individual. The explanation of the propensity in th at case is the standard Darwinian one: the fittest reproduce. What kin-selection exp lains is unreciprocated benevolence : why a person might have a �·sternatic pro pensity to be bene volent in ways that go beyond what yields him the best expected net return fro m reciprocity . Might there be biologically altruistic aspectS or our pro pensities toward ben evolence ?
The answer is that there might b e , and that i s because o f kin-selection . which Works as follo ws. One shares genes with close relatives, and so a gene that disposes one to com mit actS that enhance a close relative's individual fitness, even at some expense to one's own, may on balance be more likely to be represented in later gen erations th an an alternative gene which does not. Such genes will tend to multiply . Now th e pro pensity a gene for which would most tend to multiply is a complex on e : a pr op ensity costlessly to identify the expected degree of generic overlap for each of o ne 's asso ciates, and match one's maximu m sacrifice-to-benefit ratio to it. s u ch a pro pensity . though , is too complex to be at all likely, and simpler propensi. ties Wo u 1d · · · wh 1"ch humans m have done much the same job. Under the condltlons evol ved, close asso ciates tended to be close relatives. A propensity intentionally to �o fer benefits on close asso ciates, then, would tend on the whole to increase the l nd tvi du al fitness of close relatives. Such a propensity is simple, in that it requires
�
87
ALLAN GIBBARD
34
no mechanism for determining degree of relationship ; close associates are easy to identify . Indeed it may be quite unclear what propensities toward benevolence would
�
be most likely to evolve through kin-selection. A propensity to benefit close ass
�
ciates is one, but there may be others. Take, for instance, a propensity to ben flt
�
other people, whether or not they be close associates. If, in a hunting-gat enng society , the people one could most easily benefit tended to be close relanves, a gene for such a propensity would tend to multiply. Take also a propensity to bene· _ fit anyone of whose situation one is vividly aware. If in a hunting-gathering soc1ety, the people of whose situation one is most vividly aware tend to be close relatives, a gene for that propensity would tend to multiply . Do evolutionary considerations allow us to choose among these hypotheses? If not, it is unclear what kind of sense of justice might have resulted from kin-selection. What a hunter-gatherer does af· fects chiefly those close to him, whereas in a complex society a person knows that
what he does will affect strangers and mere acquaintances. It may be unclear what moral impulses
�ill
govern these relationships, if those moral impulses are to be ex·
plained by kin-selection. Nature selected genes for their manifestations in a hunting· gathering society , and did not care how they would manifeSt tbernsdves in a com· plex society. The alternative propensities to benevolence that l have mentioned would manifest themselves disparately
in a complex society.
A general propensity
to benefit other people might constitute a utilitarian sense of justice : combined with enough reflection, it could cause people to moderate their s elf-seeking for the
sake of the general welfare . A propensity to benefit anyone of whose situation one
was vividly aware would make us susceptible to appeals to help strangers based on
vivid depiction of their plight. Finally, a propensity to benefit close associates would give us the kind of benevolence Rawls thinks we might have : strong for those to whom we are close, but weak at best for those more distant. This last kind of benevolence would be a poor foundation for a sense of justice in a complex society. Which of
these kinds of benevolent dispositions we might have is not,
as far as I can
see , predicted by the theory of kin-selection; it seems to me that to some degree, we have all three.
What I want to do now is to present
Ill
an alternative speculative evolutionary account of justice? Consider bargaining situations: situations, as 1 have said , in which there are benefits to be gained from cooperation or mutual restrairlt , but there is conflict of interest over how these benefits are to be divided. his talk of T
of our
sense
"benefits" and " confliCts of interest" is normally to be understo od in terms of hu· man goals and preferences - in terms of "preference-satisfaction," let me say . Now, though, I_ wan to bstitute "evolutionary fitness" for "prefere �c.c satisfactio n" and cons1der Sltuatlo that have the same fonna.l StructUre as human bargaining problems. An evol11tJonary bargaining situation is a recurren t kind of interaction among organisms in which the following conditions hold. First, there are mutu al
�
� �
88
H UMAN EVOLUTION AND THE SENSE OF J USTICE
35
benefits to b e gained , with benefits now understood in terms of fitness. That is co say ,
as
compared co some combination of behaviors of the org_anisms involved,
called their threatened combination, some o ther combinations result in greater fit ness for
all . Call any such comb ination a mutual fitness enhancing combination.
Second, there is a conflict of interest over how these benefits are to be divided.
That is, for any mutual fitness enhancing combination, there is an individual whose fitness is more enhanced by some other mutual fitness enhancing combin'ation. Evolutionary bargaining situations are recurrent in the annals of natural selec tion. From the point of view of fimess, having exclusive access to a territory with
out a figh t beats ceding all access without a fight, but that in tum beats getting into
a deadly fight. Various combinations of behavior, chen , are mutually fimess enhanc
ing compared with the threatened comb ination, fighting- namely that A enjoys the
territory and B keeps out, that A and B share the territory, and that B enjoys the territory and A keeps out. Between any two of these combinations, though , the one
that is more fimess enhancing for A is less so for B.
( 1960, 2). It may be that one possible outcome- call it P - has a special prominence, because of its symmetry or because it is like the outcomes of similar bargaining problems in the past. Because of its prominence, there may be a common expecta Now for human bargaining situations, a plausible theory is Schelling's
Chap.
tion that P will indeed be the acrual outcome. Each person, then, expects that if he insists on P, he will get it, whereas if he insists on an outcome more favorable to him than P , th ere will be no agreement. In virtue of that common expectation , each person indeed will insist on his share under P and insist on no more, knowing that h� can gee tha t much and can get no more. This combination of bargaining strate gies - every one's insisting o n p and not insisting on more - is a .Vasb equilibrium: a co mbination of mategies such that each person, given the strategies of the others, can do no bet ter.
No w consider evolutionary bargaining situations. Let P be a mutual fitness enhancing combin ation of behaviors, and suppose each animal is genetically dis p ose d to th e following b ehavioral strategy: play its part in p if the others do, and otherw ise carry out its threatened behavior. Then each animal's strategy will be �h at May nard Smith calls an evolutionarily stable strategy: a strategy such that gIVen the or ganism's environment, which consists in part of the b ehavioral disposi . tio ns of the oth er organisms , its strategy is at least as fitness enhancin g as any other strateg y easily accessibl e by mutation.4 No w what could we expect in the case of human beings evolving in hunting ga h ering so � cieties? First, we could expect an abundance of human bargaining situation5 • In · vol · VI ng mutual aid, personal property, maces, temtory , use of h o �si·ng, . a�d the lik e. Now human bargaining situations tend to be evolutionary bargaimng . . Situ atlo th e ' . ns. The point is not, of co urse, chat a person's sole goal Is to maXJrntze number o f his descendants; few if any people have that as any goal at all. Rather, th e p o · · · that conferred mc co ncerns propensities co develop goals: those propensltles greacesc ' fItness were selected . Hence in a huncmg-gathenng soctety, we rnay conelude , w h en a person enhancing wanted something, his wanting it was usually fitness ·
·
89
·
·
36
ALLAN GIBBARD
most part, fitness enhan cing. For the - presumably because what he wanted was . H e� ce likely to enhan ce their fitn � s� then , people wanted things that were in fact mg slt� most part , evolutionary bargam human bargaining situations were, for the _ m ts benefi s glo we if ons ning situati � ations ; that is to say . what will count as bargai . fitness of benefits in terms terms of human preferences will count if we gloss bar onarily stable strategies for evoluti ed have develop then, s, Might human y w a such equippe d in � gaining situation$� That is, might humans be geneti cally Will ms? g proble that people raised in a hunting-gathering band will solve bargainin will be an out there , face they it be true that for at least some bargaining problems and on no more? come such that e\•ervone is disposed to insist on that outcome eople no oubt When the question is .put this way , the answer is undoubtedly yes - p neti c me ntal-ge did often solve bargaining problems - but the kinds of environme er a few· consid chanism that migh• nave done the job are extremely d iverse. Let me rrent recu to F irst, there might have been the kinds of stereotypical responses b een ve ha situations that are observed in animals. Human genetic make-up might nse defe adjusted by natural selection to produce specific adaptive behaviors in the pro ses, of territory , settling terms of cooperation, gaining exclusive access to spou hap tecting personal property , and the like. Now to some degree, that may have
�
erse pened , but human responses to particular kinds of bargaining situation s are div enough from culture to culture that "wired in" responses to particular bargaining
situations must be far from the whole storv of how humms solve bargaining prob
lems, if it is much of the story at all . More�ver, even among bunting-gathering soci eties, environments md bargaining situations were probably too diverse for "wired
in'' responses to be adequate, and so we should ask whether other me chanisms were
likely to be available.
A second ob\ious mechanism is simply a general ability to learn from exper ience. Adherence to certain kinds of solutions to bargaining problems might be passed
on by parental rev•ard and the more direct rewards of insisting on what in fact one
can get. I raise this possibility not to refute it; the question of how a social me chan ism based on very general learning abilities might work and whether such a mechan ism
can explain our sense of justice is too difficult for me to tackle here. I simply . mention this p ossibility because it is an alternative to the possibility 1 shall e xplore Third, bargaining problems might have been solved bv the kind of high ra
tionality in the pursuit of self-interest that Schelling describ � : because of pas t ex perience, it is common knowledge that everyone expects a
certain
outcome , and so
everyone insists on that outcome because he expects to benefit by so doill£ and ex pects to lose by insisting on more. I think, though, that that too cannot be the whole story . Purely self-interested bargaining often breaks down : there is too much ad\'arltage to be gained from altering expeetations in one's favor, and once expecta·
insists on more than he can self-interest would not account for the human emotions involved in bargaining situations. In dealing with technical problems , we less see how well we an do in pursuit of our goals, md arc disappointed if we do well than we had hoped. More likely emotional responses to a bargaining situation
tions are confused , it
may
often happen that each party
get. Moreover, ra.tional pursuit of
90
37
HUMAN EVOLUTION AND THE SENSE OF J USTICE
in which we do not get what we expected are anger and resentment. In most bar
gaining situations, we simply do not take the attitude of crying to get what we and being pleased or disappointed with the result.
can
I propose another evolutionary story. The problem with rationally self-inter
ested bargaining is that it is unstable, because the mutual expectations that make
solution possible are
a
so easily disturbed. Expectations could be stabilized, though,
by attach ing a moral sentiment to a particular outcome. Instead of everyone's
wanting as much as he can get in a bargaining situation, suppose there is some out
come such that everyone cares very much about getting his share under that out come, but cares very little, not at all , or even negatively about getting more. Sup
pose also that each person prefers carrying out his threat ro settling for less than his
share under that outcome. If these facts are known, the situation will be very Stable: th ere will be no advantage to altering expectations.
Consider a two-person case, with people A and B. Suppose both outcomes P
and Q are mutually advantageous, that Q is the mutually expected outcome, and that P is more advantageous to A and less advantageous to B than is Q. Suppose B is known to A to be rationally self-in terested. Then A knows that if he can convince
� that he will settle for p and
nothing less, then B will se:L!e for P . A, then, has an
mcentive to try to upset B 's expe cration that
Q will be the outcome. If he thinks he
can do so but cannot, b argaining will b reak down. Suppose. on the other hand , B is known to A to ar-...ac h a moral sentiment to Q, so that B v.iil refuse to settle for P ven if he is convin ced that A will refuse to settle for less cll an P . Then A has no mcentive to alt er B's e xpectation that Q will be the outcome: A's best strategy is ro
�
settle for Q.
A moral sentiment Stabilizes the bargaining. Wh at do I mean here bv a "moral sentiment " attached to an outcome? I have in mind a combination of a umber of things. F irst, some behavio ral dispositions : a d is posi tio n to accep t that outcome if others do and to threaten otherwise . Second. emo tio nal disp ositions that und erly these behavio ral dispositions : a disposition w be r ntful or angry if one gets less than one's share under tbat outcome , and to be . san. stled if th at outcome does obtain. Third, a role for language and conceptualiza tio n , w hi ch I shall discuss. We kn ow, of course, . that humans evolved language and powers of con ceptual IZat ion . I kn ow of no compelling evolutionary explana tion for this developmen t, but presu mab ly as ianguage developed it was available for a number of funct iO - If C rtam ou tc o mes ro bargaining problems had the kinds of b ehavio ral and emononal · · on d 1sposm · s I have described attached to them, we m1ght expect 1anguage and conPtualization to ct a word and a co organi ze these dispositions. We might e Pt to b e atta ched to those outco mes to which the beha\1oral and emononal diS Positio ns were attached. (I use the word 'concept' here as a psychological term for Wh atever plays a certain kind of role in our thinking and use of langu age; 1 do not Want a t j udgments in this po int to make the kinds of normative o r epistemological Vo lved i concepts. n d''stmgu1sh want do I epts. o-con mg, say, concepts from pseud . . . as I . am usmg both With beliefs, and g the term, to play a central role in reasonm these l . at ter terms also construed psy chologically .)
�
�
�
�
�
::
·
�
.
91
��
38
ALLAN GIBBARD
or· and emoti onal dispo sit ions A moral sentiment, then, involves behavioral indeed we that t I want to sugges ganized by a concept expressed by a word . Now · ons, and that · · · mg struau of bargam es outcom certain to nt sentime do attach a moral "as fair," then , simp ly the words involved are 'just' and 'fair'. To regard something words that express is to attach this moral sentiment to it. I do not mean that the ons of 'just' or this moral sentiment in other languages have to be good translati ns involved at· 'fair'. In a society in which the emotional and behavioral dispositio confli ct h ave tach to what is lawfuL either invariably or so regularly that case s of as our term job not affected usage, a word meaning "lawful" might do the same st that t e sugge 'just'. A word with theological meaning might also do the job. I do on as tf now from words 'just" and 'fair' do the job in our language, and 1 shall talk ies. societ g it were translations of these terms that did the job in hunting-gatherin reso· the in So far I have not ass igned language and conceptUalization a role of a ena lution of bargaining problems, but simply talked as if they were epiph enom st adju to set of behavioral and emotional dispositions. Language , though, is neede d t en m these dispositions. If, in a bargaining situation, everyone attaches a moral senti e l p peo to the same mutually ad\-antageous outcome, then all goes well. If different break attach their moral sentiments to different outcomes, though , barg aining can ng. down in ways much more intractible than in rationally self-intere sted bargaini d Suppose in a two-person case that outcome P is more advantageous to person A an P to ent sentim outcome Q is more advantageous to person B . If A attaches a moral rs fe pre B and and B attaches a moral sentiment to Q, so that A prefers conflict to Q , conflict to P , then neither outcome is such that both prefer it to conflict. Conflict
�
then, will result.
What is needed, then, is some way of ensuring that the moral sent iments of different parties will attach to the same out come- that , 1 shall say, they are com·
parible. Now one way that compatibility might be achieved is genetically to prepro
gram moral sentiments which are rigidly directed so as to be compatible : say, toward leaving possessors of objects in possession and the like. That wav of achieving co m·
th;
patibility, though, might well be too inflexible to cope with variety of bargain· ing situations hunter�atherers faced . What was needed w as a mechanism for adj u st ing the obje Cts of moral sentiments to make them compatible. That mechanism
hi ch could be provided by language and the workings of small group conse nsus w language makes possible. Things would work most smoothly if it was diffi cult to browbeat someone out of claiming what he regarded as his due , but it was some times possible to argue him out of it.
A propensity toward a sense of justice of the kind 1 have described would an evolutionarily stable strategy in Maynard Smith's sense. Given that
constitute
others have such a sense of fairness , having it prompts an individual to avoid con flict and yet insist on what he can get without conflict. E ither all w ill regard the same outcome as fair and settl e on it, or language will enable to reach a consen sus on which outcome
all is f&ir. When others have a sense of fairness, it is individuallY
advantageous to have one too. That, then,
is the account I propose of our 92
sense of justice. It is a manifesta-
HUMAN EVOLUTIO N AND THE SENSE OF j USTICE
39
tion o f an evolutionarily stable strategy for dealing with a variety of bargaining situ ations. A concept - or at least what is psychologically speaking a concept-is at tached to certain outcomes of bargaining situations; what it is attached to is subject to the workings o f consensus through language. A person is disposed to insist on h is
share of any outcome to which the person attaches that concept, on pain of resent ment, anger, withdrawal of cooperation, or carrying out threats. A person is at least somewh at disposed not to try for more than his share under an outcome to which
he attaches the con cept.
What can be said of this proposed account? First, of course, it is highly specu lative, and needs to be taken with a grain of salt. Our problem at this point is not wheth er we can establish its truth, but whether we can say anything for or against
its plausib ility . Are there features of our sense of justice for which it accounts and features for which it fails to accoun ? t
There is one clear failure in the account as I have given it so far. I have pre sented a person's sense of justice solely as a concern with justice for himself and toleration of justi ce for others e ven at the expense of his own interestS. A person is
normally concerned , though , with justice among third parties , and is often con
cerned with justice among strangers. I have no really convincing explan.arion of this property of our sen se of j ustice, but I think I can suggest some possible kinds of ex
planations. First, the mechanism that produces cooperative solutions to bargaining
problems invo! ..ing a person may involve his strong impersonal concern that out o � mes he regards as just obtain ; if the concern is truly impersonal, it �ill manifest Itself when he cont emplates bargaining situations in which he is not involved. Sec o d , there are advantages to participating in the formation of cons ensus : one � n � eflts from p e ace among one's associates; there may be kin-selection for propensmes that foster pe ace am ong one's asso ciates, and the exercise may help one to achieve s ooth o utco mes to future bargaining situations in which one �ill be involved . � Fmally concern with justice for others may itself be a form of cooperation, for one may expect it to be rec ipro cated .5 . An oth er defe ct of the account I have given is that it speaks of "the workmgs o f c onsens us thro ugh language" but offers no account of the process of reaching c nsensus. o That defect will have to remain , at least for no w 6 · The strength of the account, I think, is its fit with our experience of what the sense of jus tice is like. The account predicts a strong concern for justice and su bsta ntia l sm all group consensus about what arrangements are just. On the oth er ha d , it offers no reason for expecting that the same standards will direct the se� se � of J Ustice in all oeople. On the basis of the account, we would expect a strong dnve to develo p but stan ds of justice and latch on to standards that others propose, there m standards igh t be . for all the those what account tells us, little uniformity in are . That seems to me to be what we find. In a nu mber is like he of ways, the sense of justice predicted by this accou� t � s e o a capacitY f j ust ice Rawls ens says we have. In the first place , it is grounded m · . · · for recip · · In t he rocity: it governs the terms of cooperanon m bargammg problems o . . n c d 1ac , It fitness' se P e IS strong. A sense of JUStice would stron g1Y enh ance I-ndividual
�
•
dar
· .
·
-
93
40
ALLAN GIBB ARD
· . Th us "·e could expect a . 1ecnon than km-se and individual selection is much stronger . or l sentiit to be stronger than a � sense of J·ustice explained as I have explained . · n IS t h 1s · km-s electhen, the SJtuano ment developed through kin-selection. Perhaps, toward except when direct d tion explains benevolence, but benevolence is weak, stron a m but , benev olence close associates. The sense of justice is grounded not in g throu gy stable strate capacity for reciprocity which evolved as an evolutionarily ·
�
·
�
�
individual selection.1
IV
. then be said Suppose the evolutionary story 1 have told were correct. What could . convictions of r, about the cognitive status of ethical convictions, and in particula like noneth concerning justice? First, of course, ethical convictions are very much cities. We ical beliefs in that they draw on the same linguistic and conceptual capa rea as we think and can think and reason about ethics in very much the same way an m son about anything else-as, of course, we know from experience . I d o no : � iOn ificat here that the same kind of philosophical reconstruction of the ultimate just no neth d an l ..hica e-of the two kinds of reasoning could be given, but that ordinary pac ical thought are similar in many ways, and draw on many of the same mental c a .
ities. In \irtue of this, I shall speak of us as having ''beliefs" about fairness. \\�nat I want to ask now is whether on the evolutionan· account I have p ro posed, beliefs about fairness are represen
�tional: whether the;. represent a way the
world might be. To make sense of this question, consider prime cases of representa·
tion . Marks in a teacher's roll book represent pupils' attendance in class . Here the . teacher deliberately puts the fact of whether a particular place in the ro ll book IS
marked in correspondence with the fact of whether a particular pupil is p resent. The teacher, then, maintains a somewhat arbitrary system of correspond en ce b e . tween artendance md marks in the book. With computerization, the system m1g t be maintained automatically ; it would then be a function of the comp uter to mam
�
tain the system of correspondence. The computer is an artifact, and so we may �ll the marks it makes artificial representa tions of the pupils' attendan ce. The sense 10 which the computer constitutes a system of artificial representation, then , s eems to be this: it is a function of tbe computer to adjust one feature 0f the world (whether certain spaces are marked in the roll book) to coTTespond, in a somewhat arb itrary
way, to another feature of the world (which pupils are present on which days )- W e may call the adjusted feature the representatio n , md the feature to which the repre sentation is supposed to correspo nd, the subject matter of the system . In what sense is this adjusunent of the representing feature to the subject marter a "function" of the system of representation? e In the case of artific ial repr sentation , function is a matter of designer's intention ; a t jus ad system is designed to representation to correspond to subject matter, because rre co there being such a
spondence is expected to contribute to a purpose of the designer. This straightfor ward kind of function we can call artificial functio n or function by design . N ow Darwin's achie�ment was to explain how nature could mimic design where in fact
94
H UMAN EVOLUTIO N AND THE SENSE OF JUSTICE
41
there was no intentional design. Here I want to draw on Darwinian theory, and speak of natural function , the Darwinian analogue of artificial function. Natural function , on Darwinian theory, is to be explained in terms of natural selection. It is
a natural function of a system to do F in an organism, I shall say, if and only if the system has some of the fearures it has in that organism because of natural selection
to do F: because , that is, a genetic propensity toward the system's having those fea rures was selected for in the organism's ancestors because systems that had those fearures did F _s
We may now ch aracterize natural representation by analogy with artificial representat ion. A system of natural representation for a feature S of the world (its subject matter) is a system one of whose narural functions is to adjust some feature
R of the world to corr�spond to S , by a scheme of correspondence that is some wh at arbitrary. R. then , can be called a natural representation of subject matter S .9
Ordin ary , prosaic beliefs about our surroundings are prime cases of natural representa tions . Clearly we do have a system that adjusts features of our descriptive
beliefs to correspond to features of the worl d ; the system includes our observation
al capacities and many of our cognitive capacities. The correspondence is somewhat arbitrary - choose whatever explication of arbitrariness is needed to make this so.
�ow it seems dear enough what the evolutionary explanation of our capacity to ad
J USt des cript ive beliefs to correspond with the world must be. A genetic propensity toward having that capacity must have been selected for in our ancestors because it adj ust ed features of their d escriptive beliefs co correspond to features of the world . For such a pro pensity clearly must have contributed co our ancestors' survi val by mak ing possib le thought and communication about their surroundings . If all this is so , the system that adjusts those beliefs is a natural svstem of representa tion by the definiti on I ha ve propos ed , and our prosaic beliefs bout our surrou ndings are nat ural rep rese ntations. Here, then , is a sense in which we can ask whether our beliefs about justice or fair ess are cognitive : Are they natural representations of features of the world? We are Interp re ting that as a question about the hu man sense of justice - the system that yield s comicrions as to the justice or injustice of systems of cooperatio or urual re stra int. Is it a natural function of that system to adjust judgments of JUS· tl ce or inj ustice to corre spond, by a so mewhat arbitrary scheme, to certain features of th e w orl d ? That in turn means, does our sense of justice have some of the fea· tures it has because of natural selection to produce a particular kind of correspond ence)· . . · Was a gen er1c propensity · e w1·th certain features toward havmg a sense o f JUStlC elecr d roduce for in o ur ancest ors because systems with those feat r tende p amcula r kin d of correspondence between judgme nts of J UStice or InJUStice a d features of the world ? What would the answer to this questio n be if the speculanve evolu tion · en IS rrue.' ary ac cou nt of the sense of justice that I hav e g1v The q u est ion e given, the hu is far fro m straightforward. On the account I h av man sens e of ju stice has the features it has because its having them, long ago, �as . . . Wh at led to the toward behefs resolu atton of bargammg problems . 0 ur propensities abo ut · what the aiven J USti· ce evolv 1:0· ed, I have suggested, because for c:ach person,
;
�
�
�
:
�
·
�� ·
95
� �0 �
�
ALLAN GIBBARD
42
�;
bargain him to get as much out of a others were like, those propensities enabled uld a sities evolved as well, I sh � situation as he could. (Aspects of those propen ' ) Sow 1f stereo· Wlth JUStt ce . for whatever reasons account for third-party concern ad ns were evoiut ionarily typed responses to a limited range of bargaining situatio . that case, m yes: well be v:mtageous, the answer to the question I have posed might . t certam eatures represen naturally well might injustice and justice about our beliefs maintain ing 1rst � os· of the world. Suppose, for instance, a disposition to insist on ing sttu auons session and to defer to others' first possession settled almost all bargain di sposition t� �d in a mutually fimess-enh:mcing way , and there were a genetic pre mainummg just beliefs that an outcome was fair to correspond to the outcome 's fair � atur y first possession. Then we could say that the belief that an outcom e is pauble wtth represents the fact that it maintains first possession. That would be com •
·
·
·
·
� �
�
m\' account of the natural function of beliefs about fairness. •
1 have suggested, though. that stereotyped responses to bargaining situations
would not have enhanced fimess as much as a fle xible sense of justice subj ect to the
formation of consensus-and on the account I am now assuming as the basis of dis·
cussion , the latter is what we have . If, in that case , beliefs about justice naturally
represent any class of faets about the world - if, that is, it is a narural fun ction of
our sense of justice to adjust our ascriptions of justice and injustice to correspond to any such
c.!.a.ss of factS - the facts in question are the facts of what will d raw con·
sensus . Beliefs about justice naturally represent, if anything, the fact of what people will believe to be just if consensus is reached.
this formulation is problematic . It has a troubling elem ent of cir· cularity : beliefs about jus--..ice are beliefs about what the consensus "ill be on wh at As it stands,
n is just. Moreover, even ii this circularity can be rendered benign , the formulatio It it. misses an important element of the human sense of justice as I have pictured misses the degree of stubbornness that beliefs about justice must have in th e fa ce of evidence that one can achieve consensus only by yielding up what one regards as
one's fair share. Beliefs about justice, on the account I have given , \\ill n ot easily b e . changed to match what appears to be a coming consensus; and it is in virtu e of thlS stUbbornness that sentiments of justice stabilize bargaining situations .
Perhaps the way t o put the matter is this. There must be laws of gro up d ynam· ics according to which consensus, when it emerges, does so. These laws m ight te ll in vinue of what features of a bargaining situation and its cultural milieu so me parti e·
ular outcome will be the result of consensus if any is. Belief as to whether a given outcome is fair might then be said to be adjusted to correspond with the fa ct of whether or no� it has those consensus-making features. Perhaps, then , w here th e consensus-making features of outcomes are J , the b elief that an outcome is fair nat· urally represents the fact of that outcome's having those features J . The dynamics of consensus here must be those of the societ" of the person whose beliefs are in question. Consider a bargairl ing situatio n in o n cultu ral mili e u and an observer from another cultural milieu. It may happen that cons ensus is
;
96
43
H UMAN EVOLUTION Al'lD THE SENSE OF J USTICE
achieved by the parties to the situation, but the observer considers the outcome manifesdy unfair. The observer is subject to the workings of consensus, but within his own cultural milieu , not that of the situation he is judging. The fact to which the observer's belief as to which outcome is fair is adjusted to correspond, if any, will be that an outcome has cenain properties J , and the properties J will be the ones that would draw consensus among people of his own cultural milieu.
That may indeed be a satisfactory account of what beliefs about justice nat urally represen t, on the pan of a person who inhabits a single, homogeneous cul tural milieu . In a complex society, however, few of us inhabit such a homogeneous
cultural milieu . Hen ce so long as an o bserver belongs to a complex society, we can not sensibly speak of an outcome's having properties that would make it the con sensus outcome in his cultural milieu. Indeed I am at a loss to suggest any account of what the beliefs about justice of a person in a heterogeneous cultural milieu
migh t naturally represent, on the speculative, evolutionary account of the human sense of justice that I have offered; on that account, as far as I can see, those beliefs naturally represent noth ing whatsoever.
Supp ose that is so . Then we are left with a kind of noncognitivism with re·
gard to justice. It may be that for people in a homogeneous cultural milieu , beliefs about justice are representational
in the sense
I have been exploring, but for those
of us in complex, heterogeneous societies, our beliefs about justice are nonrepresen· tational in that sense. The beliefs can be explained, but not in terms of what they
represent. They must rather be understood in terms of their natural function and th e moral sentim ent that is attached to them.
v
We are con ce rned with justice as something to practice and defend, no � m�rely as something to understand - that is part of what it is to have a sense of JUSt1ce . On th e ac cou nt I have given, we each have a strong sense of justice, but that sense has n o clear obj ect to w hich to a,ttach itself How are we to proceed , then, when we . seek justice but are unce rtain what is just ? · even \�.•nat J1'ttle I can say will H ere I w1sh I had · 1s, more to sav than I do; as 1t be · · standards e . 1 t w1ll not do vagu s1mply to g1ve up the concept of JUS0 ce Shared . of J USti ce, if they exis t, can stabilize the outcomes of bargaining S1tuattons m co � plex so cie ties , as they can in small, homogeneo us societies - to the mutual benef�t of all. C realize that, 1t oncern with justice is \'ital to all we hold dear. Once we . seems to ties and d1srtain me , the reasonable course 1s to try to reso 1ve ou r ,,0ce '" agreem ems as to w .. ' apart from a con· - •nns hat JUSt1ce demands by lookmg to our r-..... cern w1t . e Our sense of h JUSt ice, for wanting to main tam shared standards f J· ustic J. Ust1. ce by it self, if t a poor oracle. he account I have offered is right, will make Knowing w may hat ben efi ts shared standards of justice can produ ce, however, we enlist th s. Our e hu man sen · o f our benevolent concern se of justice in the sernce ·
·
.
·
.
·
.
.
.
.
·
·
·
.
.
97
0
·
.
44
ALLAN GIBBARD
�
goal, I think, should be to foster those shared standards of justice th sharing of . be which will be most beneficial to all . If we have a flexible sense of JUStice, let It 10 guided by considerations of human benefit, that we may flourish as a result. •.
Notes
chap . 2, "An Essay on 1 . Th is . in brief, is the theory developed by Thomas Schelling ( 1 960. . see Luce and Raiffa Bargaining"). For a discussion of other treatments of bargaining problems,
( 1 9 5 7 , chap. 6).
. See also Dawkins 2. The: theory of kiD1dec:tion was developed by W. D. Hamilton ( 1 964). ( 1 9 79, PP· 4 3 (1976) , Wilson (197 5 ; 1978, chap. 7), Maynard Smith (1978), and Alexander rizing about theo evolutionary at attempts of a range wide indeed and 48, 1 03·21 ) . The theory, s in Caplan human nature, have been highly controversial ; see, for instance, some of the piece too lon g far discussion a without ( 1 978). I could not reasonably enter the controversy directly ed spec to be appropriate here. As I indicate here. 1 consider the subject speculative, but inform ed publish the of ulation can be of value so long as we recognize it for what it is. I regard some ded. work applying evolutionary theory to human beings u promising and some of it as misgui
ary The contribution that 1 ancnpt in this paper is to illustrate ways in which I think evol ution
ex hypotheses concerning human propensities can be coherently formulated and fruitfully plored.
3 . The theory I develop here is similar,
502-4).
as
noted later, to one broached by Rawls ( 1 9 7 1 , PP ·
4. The theory of evolutionarily suble strategies was developed by Maynard Smith ( 1 9 74 ; ,>.\aynard Smith an d Price. 197 3 ; se e also Maynard Smith 1 9 7 6 , 1 9 7 8 for more popul ar treat
ments).
Trivers ( 1 971 ) in effect
treats
cases that are like evolutionary bargaining situations in that
there exisu a mutual fimess enhancing combination of behaviors, but which are unlike evolu
tionary barpining situations in that there is only one mutual fimess enhancing combination.
Such cases share the formal structure of a repeated Prisoner's Dilemma. (See Luce and Raiffa, 1 957. pp. 94-102. The Prisoner's Dilemma is apparently due to A. W. Tucker.) I shall therefore call such cases biological prisoners ' dilem1PUls. Trivers discusses the circumstances under which there would be likely to be evolution of what he calls reciprocal altTUism , that is, dispositions to behave in accordance v.;th mutually fimess enhancing combinations if others do, and not othetwise. I shall call this biological reciprocity . Presumably much that can be said of the evo·
lution of biological reciprocity applies to the evolution of evolutionarily stable strategies in evolutionary bargaining situations, since evolutionary bargaining situations are in effect evolu tionary Prisoners' Dilemmas v.·ith added complications. Trivers concludes that the evolu tion of biological reciprocity would be favored by a long individual lifespan, a low dispersal rate, a b..ig:b degree of mutual depeodence, a long period of parental care, a low degree of hierarchical
dominance , and a need for aid in combat. Rawls refers to this anicle in his discussion of the evolution of a sense of justice ( 1 9 7 1 , pp. 502-4). F or a fucinating e further investigation of th evolution of biological reciprocity, see Axelrod and Hamilto n ( 1 9 8 1 ). My discussion here assumes that group selection carmot account for important features of h� sociability· That viev.· is widespread, to say the least, among evolutionary theoristS (see Dawkins, 1 976; and Alexander, 1979, pp. 36-43). It has been challenged, however, by Wade
(1978).
5. I have been treating me terms 'just' and 'fai as r' synonymous, although there are con to be drawn between them. One difference between what the two terms suggest may be "Fairness � seenu primarily a maner of the way a person treats conflicts between his ow n mt�. a.od those of others. ..Justice," on the other hand, be a virtue of
�:
trasts
seems primarily to authonoes, who "dispense" justice as they act to settle conflicts among the interestS of others. Thus me have offered might be said to deal only with our sense of "fairness, " where as me: nature of third-pany concern is genuinely a question of our concern with "justice."
account �
98
H UMA..."i EVO LUTION AND THE SENSE OF J USTICE
4S
6. An especially serious problem for the account is why we should genuinely have a sense
as opposed to an ability success fully to pose as having one when it is advantageous to do so. The problem is not, alas, how to explain a complete absence of hypocrisy in human rela tions. The question is how genuine concern with justice can exist at all - as it presumably must
of justice,
if an attemp t to gain by feigning it is to stand any chance of being effective. Part of the answer may lie in economy of means: When feigning a sense of justice can gain one little over what can be gained by genuinely having a sense of justice, the means to perfect feigning may be too com plex to be worth developing. (Plato, in The Republic, spoke of the cost of such feigning as the
loss of a kind of "harmony " of the soul, and that may be suggestive.) The subject, thou gh, needs furth er stUdy ; here recent work of economists on "iDcentive compatible" mechanisms of
preference revelation (for instance, the symposium Public Choice 29-2, Special Supplement to the bibliography there on pp. 1 G- 1 2 ) and the discussion of "bluff mg by Maynard Smith and Parker ( 1 976) may provide clues as to how to proceed.
�p�g 1977, and articles in
1. Although in some respects, what I am suggesting here is similar to a suggestion by Rawls <1971 , pp . 502-4), the tw o suggestions differ in an important respect, if I understand Rawls correctly. Rawls appears to suggest that human beings evolved a sense of justice with a pan:icu l� conten t, giv en by the construction of the original position and Rawls's cwo priDciples of jus· nee . What I am suggesting here, on the other hand, is a sense of justice with no fixed content. . In hts ecent �ey Lectu res ( 1 980, p p . 5 1 7-19), Rawis appears to drop any claim to the psy � c:hologtcal universality of his cwo principles of justice.
8 . This accoun t owes much to Wrigh t ( 1 9 7 3 ). 9• For an account of reference that is somewhat similar to this account of "narural repre· !entation ," see Stamp e ( 1 9 7 7 ) . 10· Work on this d pap er was supported by a grant from the Collegiate Institute on . Scien e of the t:niversity of .\ii chigan. versions of this paper were presented at the C ruverSity � llmois, Prince ton Universitv, Southern Methodist University, Ohio State University, and the . g e of �tichigan. I have benefited gready from discussion at all these places. Amon th Cn!Vtrsity people to whom I owe special thanks are Richard Brandt, Daniel Farrell, Arthur Kuflik, P �ter Railton, Alexan der Ro senberg• Richard Schacht ' �icholas Whit\!:, Stephen Wbire, and Wilham Wimsa They a dr · · • hicb I have awn to my attention a rich variety of coosJderatJoDS, few or w h ve bee tt. n able to ad dress in this paper.
V��es �
0! :
References
Alexander, R. D., Darwinism .md Human Affairs (Seattle . 1 9i9J . .-\xelrod, R., ) : 1 39oand W. D. Ham ilton. "The Evolution of Cooperation," Science 21 1 0 9 8 1 96.
�pi�,
A. L. , The Sociobiology Debate (New York, 1 97 8 ). awkms, R., The Selfis
h Gene (�ew Y ork, 1 9 76) . . . . Ham ilton W gy 7 rnal ofT heo retrcal Bzow · . E vo I u non . • · D ., ''Th e G eneoc o f S oc1'al Beh aVl·or• · jou ( 19 64) : 1-52
�luce,
�
R . D an ., H. Rai ffa' G.Jm es a nd D e cisions (New Yor k , 1 9 5 7 ). .·• av . nard Sm JCts, " jo urna1 of ith J . . "Th e Th eory o f G ames an d •he Evolution of Animal Con fl' , · Th eoret rcal . Bioiogy 4 7 ( 1 97 4): 209 -2 1 . - . " Ev olu tio n and the Theory of Games," .4 mencan Scientist 64 ( 1 9 76):4 145 . 2· . -. "Th e Ev olut ion of Behavior," Scientific .�merican 239 ( 1 9 78 > = 1 76'9 - \. 1 976) .· O Ur .,4 BehiWI -. and G A. P l Anima . arker, "The L ogic of Asymmetric Contests," 1 59-75 . -' and G . ) : 1 5- 1 8 · R. Price ' "The Logic of Animal Conflicrs ," Sarure 246 0 973 R. awls' 1 :·. A Theory ofjustice (Cambridge, Mass., 1 9 7 1 J . . osophy 77 ( 1 9 80 ) -· 5 1 5-72 . Rati on al Cons ofPhrl tructivism in Moral Theory, " journal Jng, T., The Strategy of Conflict (Cambridge · Mass . . 1960) . Sfllltlp 0 · Widwest Sru J'res m " · Represen tation, " ' · W . , Toward a Causal Theory of li ngu istic Pb · 1 osoph y 2 0 9 77 ) : 8 1 1 02.
�·
·
:
99
46
ALLAN GIBBARD
Triven, R. L., "The Evolution of Reciprocal AltrUism," Quarterly Review of Biology 46 (1971 ) : 35·57.
Wade. M.
J., "A Critical Review of the Models of Group Selection:·
Quarterly Review of Bi·
olol.)· S3 ( 1 9 78): 101-14.
Wilson. E. 0., Sociobiology: Tbe New Synthesis --
· On Human Nature
Wrigh:. L.,
(Cambridge. Mass .. 1 9 7 5 ). (Cambridge, Mass . , 1 9 7 8 ).
"Functions," Philosophical R eview 82 ( 1 9 7 3 ): 1 3 9-68.
SAMUEL FREEMAN
Reason and Agreement i n Social Contract Views
Social contract views work from the intuitive idea of agreement. The ap peal of this notion lies in the liberal idea that cooperation ought to be based in the individuals' consent and ought to be for their mutual bene fit. Social contract views differ according to how the idea of agreement is specified: Who are the parties to the agreement? How are they situated with respect to one another (status quo, state of n ature, or equality)? What are the intentions, capacities, and interests of these individuals, and what rights and powers do they have? What is the purpose of the agreement? Is the agreement conceived of in historical or nonhistorical terms? Since these and other parameters have been set in different ways, it is difficult to generalize and speak of the social contract tradition. Hobbes's idea of agreement differs fundamentally from Rousseau's, just as Gauthier and Buchanan conceive of agreement very differently than Rawls and Scanlon. Rather than being a particular kind of ethical view, the general notion of agreement functions as a framework for justification in ethics. This framework is based on the liberal idea that the legitimacy of social rules and institutions depends on their being freely and publicly acceptable to all individuals bound by them. If rational individuals in appropriately de fined circumstances could or would agree to certain rules or institutions, then insofar as we identify with these individuals and their interests, what they accept should also be acceptable to us now as a basis for our cooperation. Seen in this way, the justificatory force of social contract views depends only in part on the idea of agreement; even more essential
grateful to John Rawls for his helpful discussion and comments on a draft of this of am also indebted to Joshua Cohen, John Carriero, jay Wallace, and the Editon s. on s i e t r l P c Aff airs for many vi l & ub aluable suggestion i subsequen hilosophy v s that ed to P I
am
essay.
I
1 02
1 23
Reason and Agreement in Social Contract Views
is the conception of the person and the conception of practical reason that are built into partic ular views. In this paper I discu ss the concept of practical reason and its relation to the idea of social agreement in two different kinds of social contract views. My ultimate concern is to address a criticism, often made of Rawls, Rousseau, and Kant, that because of the moral assumptions made, the principles or institutions sought to be j ustified are not the
product of a collective choice or agreement at all ; the appeal to a social
contract is an unnecessary shuffl e that masks the true character of these views. This criticism has been formulated in more than one way. I shall focus on a version given by David Gauthier, directed against Rawls. 1 I
will show (in Sections III through V) that Rawls's idea of agreement is
not spurious, but is closely tied to his conceptions of practical reason, justification, and autonomy. To do this, I need first to examine the dif ferent ways reasons are conceived of in social contract views (Section I ) and to consider the structure o f Gauthier's own version of the social con tract (Section I I ).
I. Two KINo s
OF So c i A L CoNTRACT VIEWS
Let us begin with a rough distinction between Hobbesian , or purely in terest-based, and right-based contract views. Both take the idea of rec iprocity-th e idea that social cooperation should be for mutual advan tage-as fund amental. They differ, however, in their charac terization of this basic idea. In Hobbesian views, cooperation for mutual advantage involves no irreducible moral elements. Hobbesian views aim to show that morality is a subordinate notion, grounded in individu als' anteced ent desires and in tere sts. Each person's basic desires or interests are seen as defi nable withou t reference to any moral notions, and normally in terms of c ertain states of the person. The objective is to demonstrate that C 1 ) moral to principles are among the rational precepts necessa Promote one's mig�t we nts sentime prior and independen t ends ; ( 2) any have for suc h principles are conditioned by these ends; and (3) compbI. David Gaut no. hier, "The Ideolo g ," Ph ilosophy & Public Affairs 6•
'?'
Social Contract as
y
2 (Wlllter 1 977): 1 3 gn; see also Gau thier's "Bargaining and Justice," in Ethzcs and Eco II01nics lackwell, z g8s). pp. , ed. Ellen Paul ' J effrey Pau l' and Fred D. M il ler, Jr. (Oxford: B 4 . 0-4 s . J ean Hampton develop s a similar but more sympathe tic, argument in greater detail Ill Contracts an of Ph'·1o sCon tract Theory ? " Journal
�
op y 77
d Choices: Does Rawls
( I g8o): 3 1 5-38.
Have a Social 1 03
Philosophy & Public Affairs
1 24
ance with precepts that promote everyone's antecedent purposes is the most rational course of action for each individual to take in order to re alize his interests, whether in himself or in other particular persons or objects.2 So construed, the conception of cooperation Hobbesian views employ is one of efficiently coordinated activity for each person's benefit. The task is to show that, from among several modes of cooperation that might appear to be mutually advantageous when compared to the status quo or a noncooperative baseline, there is a unique set of institutions which will ensure cooperation on stable terms and which is acceptable to everyone. The primary modern proponents of this view are David Gau thier and j ames Buchanan, both of whom argue for a form of laissez-faire capital ism. Major representatives of right-based social contract views are Locke, Rousseau, Kant, and, among contemporaries, Rawls and Scanlon. The common feature of these accounts is not that they base the agreement on an assumption of prior individual rights. (Locke, Kant, and Rousseau make this assumption; but Gauthier does as well, while Rawls and Scan lon do not.) It is, rather, that principles of right and justice cannot be accounted for without appeal to certain irreducible moral notions.3 This assumption affects the conception of social cooperation employed. It has a dual aspect: in addition to a conception of each individual's rational good, the idea of social cooperation has an independent moral compo nent (characterized in Rawls by the notion of fair terms and what is rea sonable, and in traditional views by an assumption of innate moral rights). Moreover, right-based views ascribe to persons a basic interes t defined in moral or social terms. Consequently, in contrast to Hobbesian
�. lbis description represents a particular kind of moral conception, often attributed to Hobbes, and captures the central elements of Gauthier's view. Whether Hobbes himself actually held such a view is open to debate. See Keith Thomas, "Social Origins of Hob bes' Political Thought." in Hobbes St udies, ed. K. C. Brown (Oxford: Blackwell, 1g65), for a different interpretation. refers, then, to principles of right, and is to be distinguished from 3· "rights-based." Among the principles of right relied on by right-based conceptions there as in Locke's and Kant's versions of the social contract. Rawls, may be certain moral his view is a rights-based conception. See John Rawls. "j ustice as however, denies
"Right-based"
rights,
that
Fairness: Political not Metaphysical," Philosophy & Public Affairs 1 4, no. 3 (S ummer 1 g85): �36n (hereafter referred to as JF): "Justice as fairness is a conception-based, or . · · an Ide al-based view," since It works from certain fundamental intuitive ideas that refl ect ideals implicit In the public culture of a democratic society. Certain principles of are, Rawls aims to show, lmpUdt In these ideals.
1 04
right
1 25
Reason and Agreement in Social Contract Views
views, social relations are not defined as a rational compromise among conflicting interests. This affects fundamentally the way right-based conceptions interpret the social contract. These contrasts have the followin g consequence: If we see the role of a unanimous collective agreement as an account of what we have reason to do in our social and political relations, then, according to right-based views, these reasons are not sufficiently accounted for in terms of what it is rational to do to promote our prior and independent ends. There are reasons that apply to u s without reference to our antecedent desires and interests. What is the nature of these independent reasons, and where do they ori ginate? Consider the skeptical thesis, advanced by H ume, that moral consid erations do not give each individual a reason for acting, whatever his ends or situation. 4 Philippa Foot once argued for a similar position. 5 She contended that moral judgments do not have an automatic reason-giving force, but that they "give reasons for action in ordinary ways."6 What we have reason to do depends upon our ultimate purposes, as given by our desires, interests, and affections. Whether one has reason to act on moral considerations is continge nt upon whether it is in her prudential inter �sts to do so, or upon her having a benevolent disposition , a love of jus tic e, or some other moral motivation. Thus, it is not always irrational to be amoral and act against moral requirements. One who rejects morality may be villain ous but is not necessarily acting contrary to reason. 7 We might look upon the H umean position as presenting a challenge Which is taken up by both kinds of social contract views, b ut which they 4· 1 u se "skeptical" h ere to refer not to moral skepticism (which Hume did not hold) but a skept icism about the reason-giving force of moral judgmen ts independent of desires nonmoral ed. L. A. Selbyi nterests. See D avid Hume, A Treatise of Human Nature, and P. H. N idditch , 2d ed . (Oxford: Clarendon Press , 1 978), bk. II, pt . 3, sec. 3, and ' I, I. Ph lippa Foot , "M orality as a System of Hypothetical Imperatives," The Philosophica R 5·. i Essays (Berkeley · I 972 ): 305- I 6; see also Foot' s Vi rtues and Vices and Other C U ru ersuy of C fo t book expand on tha of 2 1 and ali rnia Press, 1 9 78), chap. I I. Chapters I o F � OOt s po81• a·on. . . . . earli_er or su bseher of ·c u · ens t arac h c Foot ,s posiUon no IS essays m these QUent work. Foot V g condition of ha , irt u es and V ices , p. 1 54 . Foot d oes not hold that desire is a · a a o She g a for ns rea c i so departs from Hume in contending that there are prudent al hl s n. states that there are wno �h are independent of an agent's existing desires (p. 1 48). But sheothers' . interests mto moral reason s of this kind that require us to take or act for their good (pp . 1 5jff. ). . Id. , pP . I 52, I6I-62. to
�d 1{fge
�
�leW �I
t
! ind ac;o':�tependent
1 05
:
Philosophy & Public Affairs
respond to in different ways. Hobbesian ·contract views attempt to meet this challenge on its own ground. They share with the H umean view a conception of reasons that I shall refer to as "agent-centered."8 Agent centered conceptions approach the notion of reasons by focu sing on the deliberations of single agents with given desires and interests in fixed circumstances in which they face a range of options from which they must choose. Reasons are then interpreted in instrumental terms, by ref erence to the agent's desires and interests as an individual. Given this conception of reasons, Hobbesian contract views seek to defeat the skep tical argument on the basis of certain empirical assumptions. They posit a noncooperative situation in which persons are described as fundamen tally self-centered and individualistic, and interpret morality as the co operative nonns all can rationally accept in this situation. Other-regard ing sentiments and our sense of duty are then explained as, at best, secondary motivations that effectively promote our basic interests in our selves (self-preservation and the means for "commodious living" in Hobbes, or utility maximization in Gauthier and B uchanan). A leading problem in moral philosophy then becomes how to demonstrate that the amoral or noncooperative person fares worse in cooperative contexts, in terms of the satisfaction of his self-regarding interests, than he would have fared had he steadfastly observed moral requirements and culti vated social preferences and dispositions. Right-based contract views accept the Humean premise that we have primary desires not focused on the self, and reject the Hobbesian ap proach to moral inquiry from the point of view of isolated individu als ab stracted from social relationships. Their response to the skep tical argu ment is directed at the contention that all reasons must refer to th e antecedent ends of particular individuals. The ultimate aim is to show
that moral principles and our sense of duty, while not derivable from given desires and interests conjoined with principles of rational choice, still have a basis in reason. Where does this conception of reason s come from? If we focus on reasons solely from the perspective of single agents un-
8: I use �e term "agent-<:entered.. because on this conception all reasons center on the des1res and mterests of parucular �gents. The term is not meant to imply egoism ; the con· tent of one.s de�s and mterests 1s left open. 1 aim to encompass a wide range of views. The rough idea •s to represent what Kant had in mind by Imperative. Hypothetical the _ _ _ Bes1des Foot. Williams. Hannan, Gauthier, and mar�y others contend that reasons are ad equately characterized in this way. 106
1 27
Reason and Agreement in Social Contract Views
der conditions of choice, and interpret this notion purely by reference to their desires and interests as individuals, then the skeptical question "Why should I be moral?" is a natural one to ask. And from that perspec tive it would appear that the only kind of considerations that can supply
an answer to particular persons are instrumental ones about what pro motes their antecedent ends. But these considerations are too narrow. They leave no place for the intuitive sense that practical reasons are not
ju s t normative considerations that must motivate an individual but also have a justificatory aspect extending beyond the individual's particular concerns. To account for this intuition, suppose we approach reasons differently, from the standpoint of our membership in a social group. When we ask for people's reasons in social contexts, we are not con cerned simply with their intentions and motives, and we may not be in
terested at all in their having adopted effective means to realize their
ends. Instead, our primary concern is whether their ends are legitimate
and
their means j u s tified, as measured by the system of norms generally accepted with in the group or by society as a whole.
Every social group has norms of cooperation, certain practice s and pro cedures that regulate interaction and are necessary to sustain the life of the grou p. The norms do not simply characterize accepted constraints condu ct. They also serve a social role in providing a public basis for JUstifi cation. Members of the group assess one another's activities and pursuit s in t erms of its system of norms. When someon e's conduct de parts from standard practices, he is subject to criticism according to these standard s, and is expected to justify his actions by referenc e to them. The system of norms has a central place in the pu blic life of the grou p: cer tain rules and institutio ns are seen as providing reasons for and again st people's actions and ends, whatever their desire s and inter e sts may be . 9 Seeing re as count ons in a social context as those considerations that in pu blic argum , is ent and structure u blic justification and criticism
�n
�
· s as "The Supreme · sueh c1 aun m which I use "reasons " here comes ou t m . rt eas ns " rt�d "That slavery mvolve� weak, or f bo rather curtailing were � a � 5 rights rtion t. h �om mauon of humans or condemn and holding them as proper ty is sufficient reason f � IIlg It. " The •a ct that an act is deceptive coercive , or mvo1ves the breach of a promise or oth er co mm ument functio ssment of a ns as a reason in the context of argumen1 or asse . uals' c�nsotu action s. In political individ olate v would i fact that a law the debate "0nal righ ts, concre ate unemployment increase poverty, or undernun · e nation al secunty , SlitUtes a r c reasons . eason against that law. "publi call I what of These are examples C
9- The sense
�
·
,
�rson's
.
'
·
·
1 07
·
Philosophy & Public Affairs
l considerations very different from seeing them as purely instrumenta advance their par taken into account by single individuals concerned to social sense to ticular ends. For it is just the function of reasons in this ls' ends and provide a commonly accepted basis for assessing individ ua realize them. desires, and the courses of conduct ·they adopt in order to ns that mo atio A separate dimension is added to the normative consider tivate particular individuals. reasons. This implies a certain ambiguity in the notion of practical delib This ambiguity is often reflected in the structure of our individual views tered erations. Practical reasoning normally involves (as agent-cen de and correctly point out) clarifying our ends, making them consiste nt, ate ciding on the most effective means to achieve them. But in our priv ch whi s, norm deliberations we take for granted a background of social e th to manifests itself in the following way. When some doubt arises as y legitimacy of our ends or proposed actions, the question we normall confront is
not whether abiding by these norms will effectively promote
our purposes. It is, rather, whether our ends and proposed actions can be publicly justified to others according to the system of norms generally accepted within the group. We appeal to certain social norms to appraise our claims and expectations, and to assess the instrumental means that
we have already determined effectively promote our purposes. The prac
tice of public justification is in effect reflected in our private delibera
tions. In this way, certain social rules and institutions occupy a privi leged position in the course of practical reasoning; they provide special reasons that subordinate the reasons that are instrumental to realizin g our particular ends and concerns.
Being an adult member of a social group requires that one have devel
oped the capacities to understand, apply, and act on these "public rea sons" (as I shall call them). These capacities are, on the face of it, differ ent from the abilities of individual agents to deliberate about their particular ends and the most effective means for realizing them. For what is involved is a social capacity, an ability to assess critically and justify the pursuit of one's ends according to the requirements of a dif ferent kind of norm. Hobbesian views need not deny that we have such a capacity, nor must they deny the social role of reasons in public justification. They
contend, however, that since social principles are but an extension of principles of rational individual choice, whatever justificatory force pub -
108
1 29
Reason and Agreement in Social Contract Views
lie reasons have they must have by virtue of their instrumental relation
to each agent's more particular concerns. So the capacities to understand
and apply social norms, and j ustify our actions by reference to them, are still subordinate to each individual's abilities to deliberate about the ef
fective pursui t of his own particular ends. One of the primary points of Rousseau's Second Discourse is to show the shortcomings of Hobbes's
conflation of these independent capacities and the two kinds of reasons they support . Rousseau maintains , contrary to Hobbes, that as an isolated being,
man is a "stupid and shortsighted animal,"10 tranquil by nature, and driven only by sensation and instinct. Being asocial, he is without lan guage, and so also without reason and the realized capacities for rational choice. H e is not moved by a concern for satisfying his future appetites (H obbes's desire for "power after power"), for without reason he has no conception of himself or his future. He has no need for reason, language, or prudential concern s ; in his isolated condition, his needs are wholly
satisfied by natural instinct. His capacities for reasoning are not acti vated until he enters into cooperative circum stan ces. Reason is the in strument of adaptati on man acquires to deal with social environments, as instinct is his mode of adaptation to the state of nature . And as a so cially adapti ve capacity, its primary role is to enable him to unders tand, apply, act on, an d , if necessary, devise the norms of cooperation neces
sary for social life. It is in conj unction with the development and exercise of this social capacity , and not prior to it, that man is able to apply his rational cap acities to the task of adopting and adjustin g his individual ends and deliberating on effective means for realizing them. Rousseau' s state show of nature is an analytic al device, desi gned to what we ow ities capa our e to society: the development and exercise of for reasoni a emg ng according to both prudential and moral norm s. 1 1 mem ber of a social group, recognizing and accepting that group s coo� erative n orms, and understanding how these nonns functio n as ublic for reasons Within the group are c onditions of our realizing our capacity
�
�
�
I O. Jean-J acq u es Rousseau On Social Contract or Principles of Political Right, in Rous :eau's Polit ical Writings, ed Alan Ritter and J ulia C. Bondanella (New York: Norton, nequ alz ty, PP· I Sbk. I, chap. 8, p. 95- See, i n t he s ame vol ume , his Discours e on I
3�),
:
1 1 · R u se . state of nature in o s au, 0zscours e on Inequality ' pp. g-Io. On the ro1e 0f the
lradi .
and � (Cambnd ge:contract views, see Jean Hampton, Hobbes chap 9·
dit
onal social
Cambridge University ,Press, 1 g86),
1 09
.
the Social Contract Tra-
Philosophy & Public Affairs
the moral reasoning in agent-centered terms. This says nothing about perverse. It content of a group's cooperative norms ; they may be quite artificiality of simply brings out their separate function, and shows the reasoning Hobbesian attempts to use the instrumental aspect of practical capac as a sufficient basis for accounting for social cooperation and our nce refere by ities for reasoned justification. By defining reasons purely ignor by to isolated agents with given desires, Hobbesian views start out one ing the public role of reasons in enabling us to justify our conduct to a another, and explain our choice of ends accordin g to norms that are condition for the existence of the social group. n One way to look at right-based contract views is as view s that be gi of with this other conception of reasons by focusing on the social role this in norms in public justification. If we think of morality and ju stice way, and if our aim is to formulate principles that serve this social role, then the question is not whether members of society can be given rea sons, as individuals, to comply with the norms of the group on the basis of their given ends. It is, rather, whether they can freely accept and abide by the normative system that provides the primary basis for public rea sons, or whether they are entitled to complain. U nlike Hobbesian views, right-based views do not seek a set of rules that it is rational for every individual to comply with whatever that person's prior purposes. Basic norms of justice need not connect instrumentally with everyone's given desires and interests. •o. This fact is reflected in the startin g assu mption of right-based views; they do not take individuals' desires as given, or as having value on their own terms. Instead, they seek to provide a notion of our legitimate interests, defined by reference to what can serve as a comm?n basis f�r pu lic agreement, and supply standards for critic Y assessmg people s desrres and expectations. Given the necessity of social cooperation over a lifetime, the requirement imposed on social norms is that no one can reasonably object when social norms are enforced against them. d That ce�tain c�nsiderations count as reasons in public discu ssion an ot argu�ent Is a claun about our justificator y practice s. By itself, it does n t establish anything about the . . th content of moral pnnc1ples, or even a
�
�
I . Cf. John Rawls, A Theory of Justice (Cambrid e, Ma .: ss Harvard University Pres_s· g I �) i:.��5?� (��er: �erted to as TJ), and Thomas Scanlon, "Contractualism and Util Beyond, ed. Amartya Sen and B rnard Williams (Cambridge·. C��ridge mverstty and e Press, Iglh), pp. l OS, u g.
���msm
1 10
131
Reason and Agreement in Social Contract Views
there are any. What is established is an additional basis upon which to build such principles. The idea is to give the notion of public reasons some kind of representation in theory. Accordingly, right-based views as sume that the ability to understand and apply such reasons involves an independent capacity of practical reason; '3 further, that agents have an interest in cooperating with others on terms that are publicly justifiable. The aim of a conception of justice is to specify principles that respond to this capacity and interest, that is, principles that provide a public basis for justific ation all can accept, in order to sustain willing social coopera tion. The problem is to show that there are principles that meet these spec ifications. This is where right-based views diverge in several ways: in their accou nt of our moral capacities for practical reasoning, their con ception of justification, and the role assigned in them to the social con tract. 14 I shall later discuss these aspects of Rawls's view. First, however, notice how the focus of the skeptical conclusion must change once we conceive of reasons as having an independent role in public justification. The question whether there are reasons for us to accept and act on re quirements of justice is no longer simply the question whether justice is compatible with our preexisting ends. That issue may remain, but it is not the primary concern of right-based views. Instead, a different ques tion arises, the question whether there exists a set of principles that is capable of serving as a public basis for justification and that all individ uals can reasonably affirm and accept as providing the basis for the social norm s that regulate individuals' pursuits of their particular ends. It is an 1 3. See TJ where Rawls defines moral theory as an account of our moral capacities (pp. 4 6-47), and justice as fairness as "a theory of the moral sentiments . . . setting out the Principles governing our moral powers, or more specifically, our sense of j ustice " (pp . so5 I ). Also in Thomas Scanlon, ''The Significance of Choice," in The Tanner Lectures on �u man Values (Salt Lake City: University of Utah Press, 1 g88), 8: 1 73ff.. one of the capacIties for agency is the ability to reason about what could be justified to others. . . . m 1 4· For example, Locke sought the ultimate basis for public justification m relig�on, the self-evidence of God's natural laws. The basic moral capacity was thus represented epistemologically, as a capacity for rationally intuiting these laws and drawing. inferences from them. In Kant, moral pdnclples were seen as part of the structure of practical reason. w hich enabled us to conceive of ourselves as free. For Locke and Kant the social contract had a red uced role; it was pdmarily a device for testing the legitimacy of existing poli �ical constitu tions. Their agreements assumed, and were not designed to prove, a nat�ral nght . of eq ual free agre dom, which was seen as justified on sep�ate groun s. awls s . social ms ment has a more basic des1gmng for pnnc1ples significant seeks he role, since tuu.ons, not just for testing constitutions.
� �
111
SOC!�
��
Philosophy & Public Affairs
1 32
it is an open open question whether there are such principles ; therefore, erations consid in basis a question whether there are moral reasons with ed conjoin independent of individuals' antecedent desires and interests we cannot with principles of rational choice. The importa nt point is that . The re answer this question a priori or on purely metaethical grounds substan of lationship between morality and reason becomes a question
tive moral theory. Now let us return to Foot's concerns . Foot is perhaps right in saying ons, that there is nothing rationally inescapable, in her sense of reas cap about morality whatever one's purposes or situation. What is ines the able, however, is one's being a member of a social group, recognizing as n ctio fun group's system of norms, and understanding how the nonns ve vol reasons in public argument. Though that requirement does not in s endorsing all the norms of the group, whatever they may be, it doe mean that one who is unwilling to cooperate with others on any terms except those most conducive to achieving his own particular ends is being unreasonable. The amoral man however rational he may be (in agent-centered terms), is unreasonabl just because he is not convinced or moved by any considerations except those that best promote his an tecedent ends. He refuses to accept as authoritative any system of public reasons. This amounts to a refusal to acknowled ge others as havin g claims of their own which warrant his recognition on some grounds in dependent of his particular purposes . And right- based contract views proceed on the assumption that moral reasons of j ustice involve the in dependent claims of other persons. It is a f�ct about the vast majority of people that we ac knowled ge eac other as mdependent sources of claims . (Indeed it is hard to take sen part ?�sly anyone who says he does not see others in this way. ) I n large . t It lS because of this recognition that we desire to justify our conduc and cooperate with others on terms that can be publicly accepted . ' 5 Thex:e may be exception&--<:ertain people without such sen timents. In the!! case, Foot is again right: there is little to o cares be said to the man wh _ ce of nothing for justice to convince him to act ju stly. 's But the absen _ this moral motivation cannot he is not taken to show that his conduct one's contrary to reason. There is so mething more fundamental than
�
�
1 5- See Scanlon "Contrac . the desire to be ab
tu sm and Utilitarianism," 1 1 6-17, one's actions and institutions. u 1 6. Foot, Virttu::J v�tlfy 166. ICeS,
ali
pp.
p.
112
on
·�lfl ce of the Sl�Y ''..can
133
Reason and Agreement in Social Contract Views
given purposes underlying this assessment. And we need to discover it in order to comprehend why those of us who have moral motives and a desire to justify our actions have these motivations. 17 II. AGENT-CENTERED REASONS AND THE SOCIAL CoNTRACT
Let us now consider the relationship of the idea of a unanimous collec tive agreement to the different conceptions of practical reason in the two kinds of social contract views. I shall focus on Rawls's and Gauthier's views, beginning with Gauthier's Hobbesian account, as set forth in Morals by Agreement. 1 s Gauthier's purpose is twofold. First, he seeks "to defend Western mar ket society by representing its ideal nature in relation to reason" (MA, p. 353). Second, there is a more traditional Hobbesian project, to give a "rational reconstruction" (MA, p. 339; cf. p. 1 93) not only of morality but of human sociability and all social sentiments (including love, friendship, and the sense of j ustice). To carry out these tasks, Gauthier assumes a "natural man" (MA, pp. 3 1 off. ) who is fundamentally asocial, individu alistic, and self-interested. Given these assumptions and the nature of his project, practical reasons are not going to be represented as having an independent social role of public justification. ' 9 Reason is interpreted purely in agent-centered terms, as individual utility maximization/a where the objects of a person's desires are specified "non-tuistically," or purely by reference to oneself (MA, p. 3 1 1 ). Gauthier places these per sons in a noncooperative state of nature, which is Lockean to the extent that private property and rights in one's person are recognized, albeit on self-interested rather than (as in Locke) moral grounds. Each sees that 17· In Rawls it is our conception of ourselves and our relations as democratic citizens th at account for the moral sentiments of justice. See Section V below. I 8. DaVid Gauthier, Mo als by Ag eement (Oxford: Oxford University Press, 1 986). r r Hereafter referred to as MA. 1 9· This is not to say that Gauthier would have to deny what I call public reasons (cf. note 2o below). But he w ould explain and justify the system of public reasons in agent centered terms: some thing counts as a public reason in a society only if it ultimately ad vances the desires of agent s as nonmembers. �0· "Prac tical rationality is the maximization of utility and so the maximization of the satisfaction of present preferences" (MA, p. 343). Of the maximizing View, Gauthier says he agrees " with economists and others that there is simply nothing else for practic al ration all ty to be" ("M orality, Rational Choice, and Semantic Repre sentation," Social Philosophy and Policy [ i g88]: 1 74).
1 13
Philosophy & Public Affairs
1 34
the quiet enjoymen t of his best response to others is to leave them to they respond in kind. their powers and possessions, on the condition that recognize each By a kind of Humean convention, individuals come to other's personal and property rights. of in this way, If persons and their initial endowments are conceived hostility and only two possible modes of interaction are open to them: ( 1 ) 1 9). Ac (2) cooperation on the basis of contractual bargains (MA, p. 3 s with cordingly , rational agents would enter into contractual relation ment, and those with whom cooperation promises a Pareto improve umably, pres would dominate all the rest; the weak and handicapped, ral con would be weeded out. Social cooperation is the result of a gene ts on tractual extension of this baseline. Justice is the set of constrain onal market behavior and the direct pursuit of individual utility that rati im o Paret agents would collectively agree to because it promises them a ese provement. Gauthier sees no need for a coercive state to enforce th st to constraints or preexisting rights among rational beings ; i n con tra Hobbes, that is not the purpose of his social contract.
a Given Gauthier's assumptions, we might ask, what need is there for ns? social contract, a uniform public rule applying equally to all perso a Why could society (and, if need be, a coercive state) not arise out of ls ea p series of private contracts, as in Robert N ozick's view? Gauthier ap to a collective agreement to deal with problems of market failure and externalities (MA, pp. 84, 1 28ft'., 223).�1 His contract is described as an idealized economic bargain in which property-ownin g utility maximizers agree on a principle that corrects the inefficiencies of markets. zz In the absence of this principle , Gauthier argues, rational agents would not all
er di� ::u. "Morality arises from market failure" (MA, p. 84). The market fail ures Gauthi public f cusses are stand� in economics : nuisances o such as pollution, the provision tions to good.s,. and econonu rent. Gauthier does not discuss tu ti in s al politic the need for � administer the requirements of justice, or to decide on public goods. His account IS ex· tremely apolitical. co0 22' � au.thler's �rinciple of relative concession" (MRC) holds that "in any e co . perative mteracnon, the rauonal th ng o am JOint strategy is determined by a bargain n �o operators In w�ch each advances his maximal claim and then offers a concessio th• s greater in relattve magnitude than t ffec e In the minimax concession" (MA' p. 1 45). the pnnc1ple holds that individuals . are to enJoy the benefits of cooperation , and share in . b�s" rde . proporuo e v ra n to contributions to what Gauthier calls the "coope ti �u� p p. :124). MRC IS an attempt go_' to carry through the basic idea of the Mar Prod d Th� of Just Distributions that underlies Gauthier's conception of jusuce to ooperano : �suming private property in the mean s of production, to each ac· l prod· l& con� Ution to final output , as measured by the value of his margina uct. Cf. MA pp 91 ., 97. l lo-1 2, 140, 152, 154, 254.
"�ax
7�: :'::ty :romg �
�eir
'
>
:
•
114
135
Reason and Agreement in Social Contract Views
accept market outcomes and distributions. These are not problems for Nozick, since he rejects economic efficiency as the basis for the social order. Regarding his conceptions of reasons and persons, Gauthier has ar gued that they are the only appropriate basis for justifying appeal to a social contract. For this reason he claims that Rousseau's and Rawls's views are not genuinely contractarian. Rawls's appeal to a social agree ment is spurious, he says; the moral assumptions made are the real le gitimating principles behind Rawls's view. "The real character of the the ory emerges when one asks for the grounds of the legitimating principles. "23 Gauthier's objection should be distinguished from Ronald Dworkin's. Dworkin says that, because Rawls's agreement is not an actual agree ment, it would seem to be nonbinding and of no significance. This is not an objection to Rawls's idea of agreement, but to its hypothetical nature. As such, it applies to the use of any hypothetical decision model in eth ics-not j ust Rawls's, but also Gauthier's, Kant's, and Ideal Observer theories. A short answer to this objection (which I discuss below) is that agreement in the Original Position is heuristic; it is a thought experi ment designed for purposes of self- and political clarification. 24 It helps us understand what the combined force of certain generally accepted intuitive ideas and finnly held moral convictions that shape public rea soning com mits us to. Gauthier's hypothetical agreement has a different role. It operates from conceptions of practical reason and of the person that Gauthier contends are part of the best explanatory theory in the so cial sciences (MA, p. 8) and seeks to develop a stable conception of jus tice consistent with these theoretical constructs. To lay the background for a discussion of the role of agreement in Rawls' s view, we might focus more closely on certain features of Gau thier's acc ount. First, there is nothing about agent-centered reasons that requires that they be self-interested. But once we suppose (as all social
23. Gau thier. "Social Contract as Ideology," p. 1 3gn. See also his "Bargaining and Jus tice," pp. 4G-4S· 24. See JF. pp. lJ6ff. , for Rawls's response to Dworkin's objection. Cf. Rawls's remarks on the "Socratic" nature of reflective equilibrium: "It is a notion characteristic of the study of Principles which govern actions shaped by self-examination" (TJ , PP· 48-49). See also Joshua Cohen, "Democratic Equality," Ethics 99 (1 g8g): 75o-5 1. For Dworkin's objecti�n. . see Ronald Dworkin, Taking Rights Seriously (Cambridge, Mass. : H arvard uruversJty _ Press, 1 976), chap. 6, sec . I . Dworkin's objection seems to be in part rhetoncal, for he �cognizes that the OP is a "powerful mechanic" for clarifying the requirements of equal Ity.
115
Philosophy & Public Affairs
h ave a basis in contract theorists do) that justice and social cooperation assumption reason, then on an agent-centered view it is h ard to resist an s of de of what cooperation-free interaction would look like for purpose benefits termining terms of social cooperation and the division of social mpare co to and burdens. For that picture provides a baseline from which from t the advantages of cooperating with persons with ends differen tion one's own, and allows one to assess how various modes of coopera yed portra y naturall promote his given purposes. Social relations are then . ests inter as a contractual compromise among essentially conflictin g of c This marked tendency toward some form of Hobbism is characteristi agent-centered views when applied to social questions. 2s
of For Rawls there is no place for comparing the benefits and burdens of te cooperation with cooperation-free interaction ; nothing like the sta nature plays a significant role in his view. One reason for this is that (as
Rawls says) noncooperation is not a viable option for us. So the "take it or leave it" attitude implicit in bargains is inappropriate as a model for social relations. But that does not explain the more p articular features of Rawls's agreement. Here it helps to focu s on the structure of Gauthier's contract, and recognize just how limited it actually is. Gauthier con ceives of the social contract as an idealized economic bargain. Given his noncooperative baseline, some noncontractual basis must be provided for the preexisting claims that are needed to carry that idea throu gh . So there is no justification in terms of justice and the social contract of ab solute property, contractual bargains, the price system and unregulated market exchange , or prohibition s on coercion and deception. None of these rules and institution s rests on mutual consent. Gauthier does not see these crucial mechanism s for distribution and establishing claims as forms of cooperation; rather, they define or are an extension of the nat ural state of man . 26 They are necessary conditions for, not the product
gh an 25· See, for example, Gilbert Harman's version of conventionalism which, thou explanatory project inspired by Hume, is also a kind of Hobbesian co�tract view as 1 �se n· �at term. Unlike Gauthier's hypothetical agreement Hannan argues that the moral pnc CJples that apply to a person are the result of actual pli tm the of sense a�eement, in the r� o �ntions people have accepted in dealing with each other. See Gilbert H arman. "M o X· . �er;:tlVIsm O Defe�ded," Philosophical Review 4 ( 975): 3-22 The Nature ofMoralitY ( ; ford: Oxford Uruverslty Press, 1977), chap 9;8 1 men· . and "Rationality in Agreement: A Com r � ) rals � Agre �t," Social Philosophy and Policy 5 ( 1 987 : �-�� � � ��-:�� � 6 � , om [lsj concetved as an extension of the natural freedom enJoY by a Robinson Crusoe (MA, p. 276; see also MA, p. go). .
1 16
1 37
Reason and Agreement in Social Contract Views
of, his social contract (cf. MA, pp.
222, 295).
Agreement on principles of
justice is called for only to deal with problems that arise within this nat ural scheme of distribution. Suppose, however, that Gauthier's parties had come together and agreed on principles, not after they had staked their property claims, but much earlier, shortly after their creation as intelligent beings. Then they would h ave to agree on a lot more than the limited principle Gauthier defines. They would have to settle on standards regarding how property rights are to be defined; how extensive they can be; how claims are to
be initially acquired and then transferred; conditions on exchange, gifts, and inheritance; and many other things. Had they made such an agree ment, would this make the resulting forms of property, markets, and so on cooperative institutions in Gauthier's view? In any event, it seems that his parties would acknowledge principles defining property and dis tribution quite different from those Gauthier argues for on noncontrac tual grounds ; at least it is an open q uestion. What the possibility of an earlier agreement shows is the peculiar way in which the purpose and results of Gauthier's contract are dependent on historical conditions, that is, on
when the agreement is made in the state of nature. Moreover, it
shows how dependent his notion of social cooperation is on these contin gen cies. I do not mean to su ggest here that Gauthier's parties
should come to
an earlier bargain (based, say, on knowledge of their natural endow
ments), but rather that it is arbitrary when they do. For cooperation will always be mutually advantageous whatever temporal point we choose in the noncooperative baseline, and the time at which the agreement oc curs significantly affects the terms of cooperation agreed to. Fortuitous circ umstances and decisions in the state of nature determine the content of Gauthi er's account of j ustice. As it is, given the way Gauthier sets up the state of nature and con ceives of the social contract as an idealized bargain between persons With initial property endowments, the q uestion whether these institu tions need, or even can have, a j ustification in terms of a social agree ment cannot arise for him. But this is just the issue that Rawls aims to address with his version of the social contract. Property, contracts, mar kets, and other forms of transfer (gift, bequest, inheritance, taxation, and
so on) he sees as social institu tions that can be designed in many differ ent ways. How they are individually designed and combined into one
1 17
Philosophy & Public Affairs
social scheme deeply affects individuais' characters, the development of their abilities, and their desires, plans, and future prospects. Given the deep-seated effects of these institutions on the kinds of persons we are and can come to be, the first question of social justice for Rawls is, what are the appropriate principles for designing these basic social institu tions? Rawls aims to show that there are noninstrumental principles im plicit in what we accept as reasons in public life that have an important bearing on this question. If we understand the problem of justice as a question about the proper design of the social institutions Gauthier takes for granted, and see the social contract as a response to that question, then the structure and significance of that agreement must differ markedly from those assigned to it in Gauthier's view. To begin with, the idea that the appropriate point of view from which to assess institutions and decide on principles is a noncooperative state of nature in which a certain form of private property already exists is wholly out of place. Second, the idea that an idealized economic bargain is the proper model for the social agreement must also drop out of view, since it is the legitimate scope of these kind s of agree ments and the justice of the desires, claims, and expec tation s they take for granted that we are concerned with. Third, the kinds of persons who take an interest in the justice of absolute property, unregulated ex change, and so on are going to be very different from Gau thier's rational appropriators, for whom these institutions are unproblematic . Rawls's parties conceive of themselves as free, not in the sense that they m ay act on any desire they happen to have, but in the sense that they are able to control, revise, and take responsibility for their final ends and desires by acting on and from reasonable and rational principle s. Reco gnizing the deep-seated effects of basic social institutions on these cap acitie s and on their interests, they have a basic concern for how such institutions are designed. Not satisfied with the idea that these institu tions answer to their d�sires for the accumulation of objects, Rawls's parties have a able deeper mterest in whether the institutions are structured so as to en iples them to realize their reasoning capacities and whether t he princ ati c on supporting these institutions can serve as basis for public jus tifi among persons like themselves. 'J.?
�
�7��1a�,
er
at least Implicitly, that the set of institutions argu ed for by Gauthi who are Condition (see TJ, sec. 1 2). Given the publicity condition, those
118
1 39
Reason and Agreement in Social Contract Views
Both Rawls's and Gauthier's agreements are hypothetical. But unlike Gauthier's, Rawls's is also nonhistorical; it does not arise out of the status quo or a state of nature but requires that individuals abstract from
all
knowledge of their situation, abilities, particular ends, existing rights, and everything else about their own history and point of view. The gist of Gauthier's objection to Rawls is that there can be no nonhistorical contracts, or, what comes to the same thing, no genuine social agree ments regarding the social institutions that constitute the basic struc ture. This argument is the natural consequence of Gauthier's conceiving of the social contract as an idealized bargain between persons with es sentially conflicting interests and preexisting property claims. This is a very restricted (and perhaps overly legalized) interpretation of the idea of agreement. Not only does it severely limit the range of interpretations of the values of freedom and equality that underlie
all social contract views,
but it also views justice and cooperation as exceedingly narrow notions, as if they had to do only with resolving certain problems that arise within private-propertied, market economies. The consequence of such a narrow interpretation is that there is no place in Gauthier's conception of the social contract for a collective de cision regardin g social worlds. It is as if the kind of society in which Gauthier's agents live and set their ends were imposed upon them by forces beyond their collective control. za It is a separate question to what
extent this follows from his agent-centered conception of reasons (as op-
worse off in society could not reasonably accept and give their support to a system of prop erty and distrtbution that defines entitlements and contributions according to the principle of Pareto efficiency, especially since it is compatible with virtually any distribution. My concern, however, is not Rawls's rejection of economic bargains and Pareto efficiency as the major method of distrtbution, but his rejection of them as the model for his social agreement. 28. Another reason for Rawls's focus on the basic structure, which I shall only allude to here, is raised by Gauthier's mistaken contention that Rawls denies individuals any return on their contributions to society or social productivity (MA, pp. 248--50). Gauthier's argu ment is guided by the underlying premise given in note 22: Assuming private property in the means of production, to each according to his contribution, as measured by his mar ginal product. But we have no well-defined notion of one's contributions to society, or what he is entitled to in return, independent of the social institutions within which contributions are made. That owners of capital produce or contribute something as owners is simply an institutional fact (some would say a legal fiction) that follows from the way certain institu tions assign rights and powers over things. Rawls's point in focusing on the basic structure is just that we need a public standard by which to assess and decide between different institutional systems defining entitlements and contrtbutions (cf. TJ, sees. 47-48).
1 19
Philosophy & Public Affairs
posed to his assumptions regarding the prerequisites of the contract and the self-centered content of people's final ends). A gent-centered views ultim ately must take certain desires and interests as given and beyond rational assessment. Simultaneously, then, they must take as beyond ra tional assessment the background conditions that underlie and support people's primary ends.
To sum up, it is true that Rawls's social contract does not involve any
kind of bargain in Gauthier's sense. This is because the question that underlies his view is not:
given individuals' ends (actual or presocial), is
there a single set of norms that each has agent-centered reasons to com ply with? It is, rather:
given our conception of ourselves as free and
equal, how can the principles that specify the basic social institutions
determining the ends and agent-centered reasons people have be defined so as to serve as a basis for a system of public reasons within which individuals can justify their pursuits to one another consistent with their self-conception? The focus of Rawls's agreement on the basic structure follows from his conception of public reasons and their independence what we accept as public reasons, and what we can accept as jus tify ing a system of public reasons, cannot be reduced to what rationally pro motes each individual's antecedent concems."9 III.
PUBLIC REASONS AND SOCIAL AGREEMENT
But does Rawls's response to his question still warrant talk of a social agreement? I sh all argue that there are at least three ways in which an idea of social agreement functions in Rawls's view. The first follows from )· the conception of the person that he employs (disc ussed in this sectio n o The second is implicit in his practical conception of justification (Secu IV). And the third is informed by Rawls's K antian account of au tonomY
�
(Section V). tion of Rawls's theory arrives at the soc ial contract through its specifica with the emocratic conception of persons as free and equal. Beginnin g ge, the Idea that society is a ta system of cooperation for m utual advan
�
29· This accords With .Rawls's d to ra stipulation that one of the reasonable (as oppose . tional, In his sense) constraints represented ties ar in the Original Position is that the p
� �;
agree on principles for the basic in Mor structure (John Rawls "Kantian Contruc tivism � r e Dewey Lectures I g8o," Journal of Philoso hy 7 7 ( I g8o): 529; he���e fer:ryt� t i w a . pl of . KC) R ls assumes that a concern for "backgrou nd justice" is im ci 10 0 sense JUstice.
�
�
1 20
Reason and Agreement in Social Contract Views
how should terms of cooperation be specified among free and equal per sons with different conceptions of their good? Given their self-concep tion, it would be inappropriate to see these terms as given by a higher moral authority distinct from the persons cooperating (for example, God), or even by a prior and independent moral order of values (such as natural law). For in both instances the ultimate source of public reasons would involve elements upon which free individuals with different ends and worldviews cannot agree. The most appropriate way to determine such principles, Rawls contends, is by an agreement among the persons cooperating, in light of their self-conception and each person's concep tion of his own rational good. This is the first function of the idea of social agreement in Rawls's v:iew. It is a way of working out terms of cooperation that all can accept,
and that can serve as a basis for public justification among persons with different ends who conceive of themselves as free and equal. There would seem to be little problem here, so far as Gauthier's objection is concerned. Gauthier is bothered, however, by Rawls's next move, estab lishing the nature and scope of the moral conditions placed on the agree ment. Because free and equal persons are specified as persons having an interest in the j ustice of the basic structure of their society, a suitably independent perspective is needed that abstracts from existing condi
tions. This perspective cannot be defined by a state of nature, as we have seen, since its accidental contingencies would also distort the purpose of the agreement. So, Rawls specifies a now familiar point of view, the Orig inal Position, to extend the idea of fair agreement to the basic structure. The veil of ignorance is a representation of strict equality.Jo Since parties behind the veil do not know their particular ends, abilities, or anything abou t their endowments or situation, there is no room for a bargain in Gauthier's sense (TJ, p. 1 3 g). So there is no assurance-indeed it is highly unlikely-that the principles agreed to will improve each person's starting position in society. This is the basis for Gauthier's objection: Rawls's agreement is not a contract because it is not modeled on an eco nomic bargain. The stringency of Rawls's equality condition precludes this. The qu estion raised by Gauthier's objection is whether there is room for a different kind of contract or agreement once we conceive of reasons 3o.
Ibid., p. sso. 121
Philosophy & Public Affairs
in a noninstrumental way. Contracts are legal devices, and the condi tions of valid contracts have been defined differently in different legal systems. Gauthier's conception of a contract recalls nineteenth-century Anglo-American law, in which the same conditions were specified to ac commodate an expanding market economy. Still, at no time have con tracts ever been legally defined so as to constitute an equivalent of Gau thier's idealized bargain. All sorts of promises and commitments have been deemed enforceable as contracts that do not involve anything re sembling an economic bargain.J• Furthermore, background moral con straints are a condition of any legal contract; contracts are deemed "un conscionable" for a number of reasons, and no court would enforce a contract by which one party consented to alienate his constitutional rights. And, too, the condition of finality that is part of social contract views has never been a part of any valid legal agreementY Finally, even if we concede that the idea of a contract connotes a fundamental conflict of interests and lack of mutual concern, that does not mean that there is not a place for some kind of social agreement in Rawls's view. Not all agreements are like economic bargains. To begin with, there is a perfectly good sense in which a grou p of per sons can agree to some joint activity (for example, to play basketball to morrow) that all want to take part in for its own sake. Each commits himself by his word, thereby decreasing the likelihood that someone will (con 3 1 · In fact, few contracts today are bargains in Gauthier's sense ; the great majority
sumer purchases) involve acceptance of detenninate offers not subject to revision by ne , gotiation. The basic elements of a legal contract are a promise (or offer and acceptance) w. la ivil c in and what is called "consideration" in the common law and "sufficient cause" ' ed for " onsideration" is often defin by legal scholars as something that is b argain . gxven m exchange for a pronuse, but there have always been promises enforced as co al tracts that do not involve either, e.g., promises to charities, and promises made under se t. en rec y vel pnor to the nineteenth century. The legal definition of "consideration" is relati Its source was not the courts, but the American Law Institu te, which noted that manY other factors count as consideration in law to make promises en orceable in the absence of f b�galns or exchange-e.g., reliance on a promise or commitment by the promisee to a third Party ; de�ent incurred by a promisee r. I n ac�u ; benefit received by the promiso . practice, the notion of consideration is a way of expressing the idea that not all pronu�. the civil are legally enforceable; they will be enf orced as contracts only if there is (to u se law term) suffi cient cau se. ee Arth L. C West ur S , I QS I ). orbin, On Contract s (Minneapolis : chap. s. s of the J:l. The condition of finality, as described by Rawls is the condition that the term sidera· contract are ultimate and conclusiv n co er oth e in practical re oning' overriding all . tract, tions (TJ, p. IJS). Since it refers to con the finality of reasons' it implies that the social · · unlike any legal contract, ts WIthout excusing conditions.
�
�:
�
�
�
1 22
1 43
Reason and Agreement in Social Contract Views
later change his mind. There are also other means we use to bind our selves by such friendly agreements, such as setting into motion certain practices or institutions that limit the range of future options open to those who participate (as in marriage agreements, or joint commitments by members of the same religious association). Here again the agree ment is not a compromise among essentially conflicting interests. In stead, it represents a joint commitment to certain shared ends or ideal modes of interaction which each desires as regulative of his own pursuit of his particular purposes. The parties rely upon one another in order to achieve these shared norms or ends. The point of the agreement is not to resolve disagreement or conflict, for there may be none. It is, rather, to tie down the future, to keep the parties from later changing their minds and deviating from the shared norms or purposes of the associa tion. These agreements are akin to the idea of rational precommitment,33 except that they involve a joint precommitment among several persons, since the shared norms or ends of the association's members cannot be achieved in the absence of widespread compliance. A general precommitment best expresses the kind of social agreement at work in the natural rights theory of the social contract tradition. Locke, and at times Rousseau, might be read as expressing this idea when they call their agreements a "social compact" (rather than a "con tract"). And the idea of general precommitment indicates a significant way in which collective agreement plays a role in Rawls's view. Free and equal moral persons all have, in addition to their diverse ends and world views, a fundamental social interest (their sense of justice) in cooperat ing with one another on publicly justifiable terms that express their con ception of them selves as free and equal (cf. TJ, p. 561 ).34 This interest is social since it cannot be achieved by single individuals, but requires co ordination of activities. It is shared in that each individual desires the same objec t, a background of just institutions. To realize this shared social interest, free and equal citizens mutually precommit themselves (through their representatives in the Original Po33 · See J on Elster, Ulysses and the Sirens, rev. ed. (Cambridge: Cambridge University Press, 1 g8 ) , chap. 2, esp . pp. 37-47 . 34 · The claim that practical principles "express" our self-conception is common among Kantians. See, for example, Thomas N agel, The Possibility of Altruism (Oxford : Oxford U niversity Press, 1 g o), p. 1 g, where it is part of his definition of "interpretation." Rawls 7 . V, esp. note gives sense to the claim via Kantian constructivism . discussed below, Sectton
4
49-
1 23
Philosophy & Public Affairs
1 44
for the design sition) to principles, appropriate to their self-conception, ts. They com of institutions and the regulation of their individual pursui status as equals mit themselves to, and rely on each other to maintain, a that will be in the free pursuit of their ends and in the considerations the con recognized as public reasons. The principles create and sustain inter mental ditions needed to realize their equal status and their funda an ples est in cooperation based on mutual respect. Since these princi views) a swer to a shared social interest, they are not (unlike Hobbesian on. The compromise, but rather everyone's best response to this situati p l usual compliance problems that bedevil Hobbesian views do not a p Y-35 ens In committing themselves to these principles, free and equal citiz sion willingly impose upon themselves certain constrai nts on future deci mit om prec making at all levels of choice, individual and collective. This is It . ment is general, because it is made by and applies to everyone ion shared, since each depends upon the others to hold him to his decis in ual, mut is by maintaining an environment of just institutions. And it er oth ; that each gives his consent only on the condition that others do wise the coordination of activities necessary to each citizen's realizing his basic interest in justice could not be achieved. These features war rant talk of a social agreement. Members of a democratic society make this general precommitment, r not the parties in the Original Position. The parties choose from inte ested motivations and have no basic concern for justice and equality. Here Rawls's argument from the "strains of commitment" plays an es sential role (TJ , pp. 1 45, 1 76ff. ). Because the parties conceive of them selves as free and able to control their future each sees himself as re sponsible to the future claims of the self (TJ , p. 422_23 ). Each know s that he will be held to his decision by the others so each aims to commit elf himself once and for all to principles that he not reproach h ims
� �
�
3?· Rawls descri this social interest as a highest-order interest In realizin g one's ca equal p��lty f?r � effective sense of justice (KC, p. 52 5). We might look upon free and ce ren Citizens desue to realize this interest as a metapreferen refe is, a p ce for justice' that . . ment that all one's future prefierences meet the con nons e e agr By of reasonable principles. di . . on the two pnnclp les, citizens in effect decide to make themselves into the kinds of persons not make trade-oft's between justice and their particular ends. hey will that they T persons whose preference structu re is (in tenns of rational choice analysis) lexl tility Since lexicographic preference structures u a by cannot be represented a nc�on (see Elster, Ulysses and the Sirens, pp. 1 24-27) we mi ght see their choice as deciSlon not to be rational utility maximize rs in Gauthier'; sense.
:� �graphic.
1 24
1 45
Reason and Agreement in Social Contract Views
for choosing even should the worst transpire. Commitment by the parties to principles that secure equal citizenship and enable each freely to pur sue his own good corresponds to the precommitment of free and equal citizens in society. Still, does Rawls really need to talk of an agreement in the Original Position ? Given the veil of ignorance, it might seem, as Jean Hampton says, that "there is no theoretical reason to posit more than one party in the original position."36 One reason Rawls insists on an agreement is to carry through the contrast with utilitarianism (cf. T], pp. 1 84ff.). The Original Position is not set up like Ideal Observer theories, where a single judge with full knowledge of all but his own identity impartially repre sents and sympathetically identifies with the interests of everyone.37 As against the conftation of interests involved from this point of view, an agreement among many persons concerned to advance only their own interests emphasizes the distinctness of persons and the plurality and incommensurability of their conceptions of the good. More important, the contract condition is essential to Rawls's strains of commitment ar gument. Were there but one party choosing principles, there would be no one for him to commit himself to. Agreement, unlike individual choice, implies a joint undertaking where each is held by the others to his decision, thereby ensuring the perpetuity and irrevocability of the principles agreed to.JB 36. Hampton, "Contracts and Choices," p. 334· This objection voices the view of many who say that Rawls's Original Position merely depicts the rationally self-interested choice of one individual. See Gauthier. "Bargaining and Justice," p. 44, and Scanlon, "Contrac tualism and U tilitarianism," p. I 24. 37· Gauthier's noncontractualist argument for his position in "The Archimedean Point" (MA, chap. 8), although it involves interested choice like Rawls's Original Position, still resembles Ideal Observer theories in that it is choice by a single agent with knowledge of everyone's preferences, but not of his own identity. This resemblance is due to the fact that both utilitarianism and Gauthier's contract view take individuals' desires for granted. and seek to maximize utility. The difference is that Gauthier refuses to aggregate utility, which explains why his ideal actor's choice is interested.
38. Hampton ("Con tracts and Choices," p. 330) argues that the finality condition (see note 32 above) already implies irrevocability of the principles selected, and that this renders Rawls's strains of commitment argument and the contract constraint on which it relies unnecessary. But although finality may imply that excusing arguments will not be enter tained (perhaps ensuring a kind of irrevocability), it still does not bring out the idea that the parties are making a good faith commitment to one another, which they can rely on as citizens and cite to each other to justify holding one another to the terms of the agreement. Nor does finality (or irrevocability) imply the perpetuity of the undertaking, which, as Rawls indicates (TJ, pp. 146-47), has its basis in the "Agreement Condition."
1 25
Philosophy & Public Affairs
Perhaps the best answer to Hampton's objection is that the hypothet ical choice of the Original Position makes sense only in the context of free and equal persons trying to come to a social agreement on terms of cooperation and the presence of standards for public justification that all
can accept, whatever their particular situations.J9 The veil of ignorance
is imposed just as an extension of that basic contractualist idea. It cor responds to the fundament-al equality of moral persons in agreement on
the basic structure. Even if the parties are symmetrically situated by that
condition,4o Rawls needs to maintain the idea of distinct individuals com ing to an agreement to make the main idea of his theory go through. So the most that one can say (although Rawls denies even this)41 is that agreement among parties is the same choice as would be made by a sin gle individual
in the Original Position. But there is no reason to describe
the choice in this way. Not only does this description misrepresent the conditions of choice in the Original Position, but also there is no- way to arrive at it given the way Rawls sets up the situation. 4,.
s 39· Cf. John Rawls, "Reply to Alexander and Musgrave," Quar terly journal of Economi� 1� ct r cont a � 88 ( 1 974): 651 , where he says that "the reason for invoking the concept of
the original position lies in its correspondence with the features of a well-ordered SOCiety. A well-ordered society is partially defined as a society in which everyone accepts, and knows that others accept, the same principles of justice. 40. Hampton ("Contracts and Choices," p. 334) says that the parties are complet�ly identical in all their characteristics; hence by Quine's principle of the identification of m t discemibles, we are forced to "construe the indistinguishable parties as one person ." Bu the parties are not described as having aU the same properties (e.g., they have different concep�ons of their good); instead they share the property of not knowing any gene:al properties about themselves that might give them evidence of their chances under social schemes structured by various principles of justice. 4 1 · On grounds that the class of things that can be agreed to is included in and is smaller d than the of things that can be rationally chosen (Rawls, "Reply to Ale xander an
��ass
6SI-S2).
Musgrave, pp. Hampton also argues not only that a contract is uncalled for. but also th at the con s tract device is too weak to yield the irrevocable commilment Rawls needs. For contract lity ibi poss are always r:vocable if all the parties agree. The parties might then count on t he of a re�ocanon of the agreement once they enter society . So they might take risk s, thereby changmg way they reason and endangering Rawls's argument for the two prtnciples (Hampton, Contracts and Choices," pp. 33 1-32). But Rawls's argument assumes that
4 2·
� �
��
m�m rs of society will enforce the agreement not out of self-interest but out of their sense of J �Stice (TJ , p. 145). This moral motivation makes it highly unlikely that they woul be . g t revoke and renegotiate the agreement. But even if they did, a revocation, or .re ?, SCIS&l�n , of a contract is a separate contract. See Corbin, On Contracts, see s. ss And smce they affect the basic structure, any rescission or renegotiation amon g Rawl any u o les ru parties would als.o have to take place behind the veil of ignorance. But this r grounds the parties might have in the original agreement for taking risks in hopes of late rescission and renegotiation.
�
1 236, 1 23?· t
1 26
1 47
Reason and Agreement in Social Contract Views
To sum up, if we conceive of reasons as having an independent social
role of providing a p ublic basis for agreement, then a social agreement is
the most appropriate way to ascertain principles that serve this role among free and equal democratic citizens with different conceptions of their good. And this procedure, while not a contractual bargain, is still an agreement in that it embodies their shared precommitment to main tain their status as free and equal in the structure of institutions and system of public reasons. Interested agreement from the Original Posi tion models this situation.
IV. SociAL AGREEMENT AND PRACTICAL J usTIFICATION We have considered how the social contract plays a role
within Rawls's
conception of justice, from the point of view of free and equal persons and from the perspec tive of the parties in the Original Position. Let us now consider a second way in which agreement plays a role, implicit in a third perspective in Rawls's view, that of ourselves as members of a democratic society. 43 Rawls identifies the aim of political philosophy as a practical one: to define a conception of j ustice that can "provide a shared public basis for the justification of political and social institutions."44 Its task is to locate
a basis for agreement in a culture that all can affirm and accept and that
can serve as a basis for public reasoning and stable social cooperation.
The practical aim of a political conception is to be contrasted with what we might call the "theoretical aim" of a moral conception, which is truth. "J us tice as fairness . . . presents itself not as a conception of justice that
is true, but one that can serve as a basis of informed and willing political agreement between citizens" (JF, p. 23 0). This does not mean that Rawls is not interested in objectivity or truth (clearly Rawls thinks that the general fac ts assumed by his theory and the parties are true [TJ, pp.
54 7-48!). Rather, there is a difference between the primary objects of a practical v ersus a theoretical inquiry. Whether Rawls's principles are or can be true, in the sense that they satisfy a metaphysical account of truth, is a separate issue which Rawls does not address. H e thinks it important to "avoid the problem of truth and the controversy between 4 3 · On the three different perspectives and their significance, see KC, PP· 533-3 · 4 44 · John R awls, "The Idea of Overlapping onsensus," Oxford journal of Legal Studies C 7 ( 1 987): 1 (hereafter referred to as OC). See also JF, p. 230; KC, PP· 54 1 , 543; and TJ, pp. 44. 58).
1 27
Philosophy & Public Affairs
realism and subjectivism" if justice as fairness is to achieve its practical aim in a democracy (JF, p. 230). This point is essential to Rawls's version of liberalism, as well as to understanding the sense in which his is a contractualist view. The practical aim of a political conception does not by itself seem to imply any form of contractualism. For we might imagine a society in which appeals to religious authority, or to self-evident truths about good reasons, provided the basis for public justification and agreement ( cf. KC, p. 557). Rawls's point is that such appeals cannot work in a democ racy. For given that democratic citizens have different and competing philosophical conceptions of the nature and bases of truth, objectivity, and so on, a basis for public reasoning and agreement cannot be achieved by a conception of justice that relies on such premises. Here the idea of social agreement comes in; such an idea is implicit in what I will call the "practical conception of justification" that Rawls sees as ap propriate for a democratic society. For a justification to be practical (as opposed to theoretic al), it must satisfy the following conditions:
( I ) It must have a practical aim: Addressed to those who disagree, it
is designed to yield principles that all can accept and affirm as a basis for public justification and agreement (Rawls calls this the "social role" [KC, p. 5 1 7] or "public role" [OC, pp. 5-6] of princi ples). (2) It must have a practical basis: It proceeds from considerations that count as public reasons in the culture, and so relies on premises and methods of reasoning that all can agree on and endorse (TJ. sec. 87; OC, p. 8). Our "considered moral convictions" of justice are primary among such premises. (3) It must meet a motivational requirement: Practical justification assumes that members of a culture have a desire to give reasons that justify their institutions to one another ( a common "desire for free and uncoerced agreement" [JF, p. 23 1 ; KC, p. 51 8 ] )-more generally, a willingness to cooperate with others on terms they can freely accept (a sense of justice).4s
45· Rawls says that the sense ofjustice and "implies the desire on the part of individuals �u1 � advance their good in ons reas by ways which can be explained and justified , . can_� � accept as free and equal moral persons" ("Social Unity and PrilllafY s, m Utdttananism and Beyo nd, ed. Sen and Wil ' sliam , p. 1 84 ).
�
128
1 49
Reason and Agreement in Social Contract Views
(4) It must show that the principles yielded by it accord with our con sidered moral convictions at all levels of generality, on due reflec tion and after considering alternative views, so that these princi ples and our convictions best fit together into one coherent scheme (what Rawls calls "general and wide reflective equilib rium"). These conditions identify the sense in which Rawls's view is contrac tualist from the point of view of those who are considering this concep tion of justice. None of these conditions, by itself, warrants talk of a con tractualist justification. The mere fact that condition (2) is satisfied that we can agree on certain precepts and methods of reasoning-does not by itself justify anything; contractualism is not conventionalism. Hence the importance of satisfying condition (4)-reaching reflective equilibrium. Then again, we might imagine a single individual trying to achieve equilibrium of all the moral convictions and ideas that stem from his conception of the good and his philosophical and religious views. (So we might interpret traditional philosophical arguments for comprehen sive moral theories.) Others who do not share this individual's views would not be able to accept all of his conclusions, although he might see them as authoritative for everyone. Rawls uses reflective equilibrium in a more public way, as a method of justification appropriate for political agreement on principles all can endorse; hence the significance of con dition ( I ). It relies only on the considered convictions on which we all can agree (condition [21) and so excludes considered beliefs peculiar to our conceptions of the good or our metaphysical views. Reflective equi librium, then, is "general": everyone should be able to accept the same conception. Rawls's conception is contractualist in that it seeks principles with which to provide a public basis for justification and agreement that re flec tively cohere with the considered convictions and beliefs that we hold in common and that serve for us as public reasons. It is an agreement in that it is a mutual accommodation of different philosophical, religious, and moral views, reached by "reconciliation through public reason" (JF, p. 2 30). This reconciliation depends on public understanding that cer tain considerations--those that stem from individuals' particular world Views and on which we cannot all agree-will not be offered as argu ments or a basis for public claims. The Original Position and other elements of Rawls's view model this situation : they model the bases for 1 29
Philosophy & Public Affairs
1 50
justification from which we actually proceed in trying to justify our social institutions to one another.46 And the conception of j ustice that results incorporates these bases for justification. As such, it attempts to give standards for public deliberation where only the reasons specified by that conception-those that can be justified practically-<:ount as reasons in arguments on constitutional essentials. The agreement or reconciliation that Rawls's conception of j ustice rep resents is not a bargain, in the Hobbesian sense, between essentially conflicting views. It does not proceed from people's given ends and con flicting worldviews and seek to strike a compromise. I nstead, it works from shared political convictions, and happens to meet individuals' dif ferent moral views at the point at which they converge (Rawls's "overlap ping consensus"). It does not, then, require one to compromise his basic convictions; or, if it does, it is only because they are incompatible with
what that person accepts and relies on as public reasons. Here condition (3 ), the motivational requirement, is important. The purpose of this re quirement is not to define what counts as public reasons , bu t rath er to e ensure that citizens will act on principles so that the practic al aim of th on political conception can be carried through. Given Rawls's assumpti that we have a shared sense of justice which is effective independen t of our particular conceptions of the good, there is not the same concern, as in Hobbesian theories, that citizens are always prepared to pursue th eir goals at the expense of others and will depart from j ustic e whenever cir cumstances allow.
Since he explicitly avoids claims to truth and objectivity according to epistemological and metaphysical critieria, Rawls's practical j ustification is open to the objection that it is not really a justification, but rather a kind of complicated accounting method that unifies the underlying con cepts and principles of a political culture.47 But there is a stronger sense
46· For this reason Rawls describes the conditions defining the Original Position as a ti representa on of reasonable restrictions on arguments for principles of justice (TJ, PP· 120• 138• SIS; J F, pp. 237-38). They are general convictions that "we do in fact accept (TJ. �· 21 insofar as they incorporate our considered convictions regarding the kinds of �onsiderations that are and are not acceptable as reasons in arguments for principles of JUstice. ti of � 7· This ��s the q uestion, which Rawls mentions but avoids , whether h is concep on �ral JUS cation 18 sufficient (OC, p. M f e o enc epend 1 5) or at least neces sary ('The Ind n tiO Theory, . APA Presidential Address P ia soc , roceedings of the American Philosophical As 48 [1974): 9. 21) for justification in the theoretical sense.
1�:
�
�
1 30
151
Reason and Agreement in Social Contract Views
in which Rawls seems to characterize his principles as practically justi fied, based on the Kantian features of his view. Let u s now return to the perspectives internal to Rawls's conception, and see the role that social agreement plays in defining autonomy and a practical conception of ob jectivity.
V. REASON, AUTONOMY , AND AGREEMENT The purpose of the state of nature according to a Hobbesian view is to provide a perspective from which to rationally assess existing social ar rangements. (Gauthier's assessment, however, is quite limited, since he incorporates into the state of nature the very institutions many feel to be most in need of appraisal. ) But this does not free the view from its anchor in antecedent desire ; it only relocates the basis for the agent-centered reasons people have in certain natural tendencies and inclinations. S up pose, however, that we seek a basis for reasoned assessment in some thing other than existing social arrangements and natural inclination. Following Kant, Rawls appeals to our capacities for practical reasoning. The moral powers of free and equal moral persons are the two separate capacities for practical reasoning as applied to matters of justice. The moral powers are (a) "the capacity for an effective sense of j ustice, i. e. the capacity to understand, to apply and to act from (and not merely in accordance with) the principles of justice," and (b) "the capacity to form, to revise, and rationally to pursue a conception of the good" ( KC, p. 5 25). These are, Rawls says, "executive" capacities (KC, p. 533) (one might add "judicial"), since they enable individuals both to regulate and to jus tify the pursuit of their ends by means of reasonable and rational princi ples. Our concern is the source of reasonable principles and their con nection with a social agreement. Here Kantian constructivism and full autonomy come in; they develop the idea that the moral powers are not only executive but also
collectively legislative.
The idea of constructivism works on two levels in Rawls's theory. First, it applies at the most general level of justification, reflective equilibrium. From certain considered convictions on which we agree, Rawls "con structs" the conception of j ustice that can best serve as a basis for public justification in a democracy. Here the notion of construction gains sense by contrast with appeals to linguistic or self-evident moral intuitions. Rather than raising certain considered convictions to the level of first
131
Philosophy & Public Affairs
principles because they are thought to be implicit in language or accu
rate reports of a prior moral order, Rawls seeks principles that best co
here with all of our shared convictions of justice (what we rely on as
public reasons), revising and, if necessary, discarding recalcitrant ones
along the way. No specific level of generality is assumed to carry the
burden of justification. Other moral conceptions c an be construed as
constructivist in this sense, so long as they rely on this method of justi fication.48
s "Kantian constructivism " is a more specialized notion that work within this constructive model of justification; it explains why we may find some of our considered convictions recalcitrant. It aims to show that
the principles that best cohere with our convictions are not questionable
accidents of culture, but are objective in that they have a basis in our
capacities for practical reasoning. These capacities, Rawls presumes, un derlie many of our considered convictions, includin g our political con s ception of ourselves as free and equal (JF, p . 233 ; TJ, sec. 77) . Rawl l ode seeks to capture this self-conception with an ideali zation, the "m conception" of free and equal moral persons. To construct, in our firs t sense, principles of justice appropriate for us, Rawls construc ts a social world modeled on the self-conception and practical c apacities of free and equal moral persons. The purpose of this second kind of construction is
to give content to Kant's enigmatic definition of autonomy as "reason giving principles to itself." In the absence of some kind of procedure th at shows the relation ship between our capacities for reasoning and moral principles, such phrases are difficult to m ake sense of. Agreement from the Original Position serves this role; it is a "procedural interpretation" (KC, p. 559 : cf. TJ, pp. 256, 2 64) of practical reason in matters of justice, or, more exactly, of a conception of persons as both reasonable and ra tional. Since this procedure is designed to "model" the moral powers,49 48. For a discussion of this sense of construction, see Dw kin , Taking Righ ts Serious ly, or pp. 1 59-68. the 49· Rawls's claims that the Original Position "models " "mirrors " and "represents" es ti moral powers are somewhat obscure. He ar p following: The seems to have mind g to the Qrtmn rdin !&"' 'al pOSIUon represent co ac ge jud the capacity for rationality in that they . ate to the pnnciples of rationality set fonh u eq in TJ chap. 7' and desire to obtain an ad · share of the pnrnary · g�s. It is rational to want the primary goods , since they en able SOCial free and eq al penons to realize their fundament velop � al interests in the exercise and de t �eu moral powers, and their all is This particular conception of the good. the " Re non al (KC, p. 528). The moral conditions of the parties ' rational choice ( or . asonable ') represent our capacity we y, it ac ap for justice in that, in exercising that c ·
in
·
th�
'
P� of
�� ;!
1 32
1 53
Reason and Agreement in Social Contract Views
the content of the principles chosen from that point of view will be de termined by these reasoning capacities and the conception of the person to which they give rise (KC, pp. 5 1 6, 5 18).so In this sense moral princi ples are "constructed" on the basis of reason. The objectivity that Rawls ascribes to his principles rests on his claim that they would be willed and agreed to from a shared point of view, which is objective in that everyone abstracts from their particular (sub jective) aims, beliefs, and perspectives to view society on an equal foot ing (TJ, pp. 5 1 6- 1 9). This conception of objectivity is practical, as op posed to theoretical, in the following sense: Rawls's claim is not that, being impartially situated, we all have a clear, undistorted view that al lows us to make true judgments about a prior and independent moral order. In constructivism, Rawls says, there is no order of moral facts, prior to human reasoning, for our moral judgments to be true of (KC, p. 568). This does not mean that Rawls must deny that a prior order of moral facts or principles can exist, for that metaphysical commitment would conflict with his practical aim. Rather, it means that if there is such an order, it is not because certain principles are true of it that we are bound, as democratic citizens, to follow them. What commits us, as would ju dge these conditions to be fair and reasonable restrictions on arguments for prin ciples of justice for the basic structure (TJ. pp. I B-Ig; JF. pp. 237-38). More particularly, the veil of ignorance embodies our conception of ourselves as equals in matters of justice. The publicity condition follows from our concern to know the underlying bases for laws and social institutions, as well as from Rawls's aim to arrive at principles that will provide a basis for public reasoning and justification in a democratic society. The condition of fi nality conveys our conviction that considerations of justice are conclusive in public affairs; they override all other public concerns. For Rawls's account of how Kant's Categorical Imperative involves a procedure of con struction, see John Rawls, "Themes in Kant's Moral PhUosophy," in Kant's Transcrndental Deductions , ed. Eckart Forster (Stanford : Stanford University Press, I g8g). PP· 8 1 - 1 1 3· so. To show how the content of Rawls's principles relates to his conception of the person would req uire a separate discussion. For Rawls's discussion of the first principle, see John Rawls, "The Basic Liberties and their Priority," in The Tanner Lectures on Human Values (Salt Lake City: University of Utah Press, I glh) . 3 : 1 8-3 9· Briefly, the basic liberties are seen as "fully adequate" to the development and exercise of the moral powers, and to each person 's pursuit of his rational good. Moreover, they provide the external freedoms that enable moral persons to realize their conception of themselves as free- that is, as self Originatin g sources of claims with the rational capacities to have a conception of the good and who take responsibility for their ends (KC, pp. 543ff. ; JF, pp. 24off.). Finally, an equal right to the basic liberties accords with the equality of moral persons, based in their all having the moral powers needed to engage in social cooperation (TJ. sec. 7 7). The �er _ an egalitar ence Principle ensures the fair value of the basic liberties for all and embodies ian conception of reciprocity (TJ. pp. 1 02-3).
1 33
1 54
Philosophy & Public Affairs
citizens, to these principles is that they are "most reasonable for us" (KC, p. 554), in that they best accord with our capacities for practical reason ing in the circumstances of a democratic society. The conception of ob jectivity that informs this claim is practical, since the shared point of view from which we would agree to these principles is designed not to give us privileged access to a prior moral order but to represent our pow ers of practical reasoning in a way appropriate to our democratic concep tion of ourselves as free and equal.s' But what relation is there between the moral powers, autonomy, and a social agreement? Why is the procedural representation of "reason giv ing principles to itself' conceived as a collective decision, instead of, as in Kant, the ideal choice of a single individual? One might say that agreement emphasizes "autonomy" in the sense of devising principles of justice for ourselves cooperatively and not as individuals. But moral prin ciples, unlike positive laws, cannot actually be legislated or chosen, either by groups or by individuals. N or is "autonomy" meant to imply that they are. Autonomy in both Kant and Rawls involves the j ustifica tion of moral principles by reference to a conception of persons as both reasonable and rational, where the content of principles is determined by these persons' reasoning capacities, and not their particular ends, their innate psychological tendencies, or an antecedent moral order. We judge and act autonomously, not by choosing our moral principles, but by pro ceeding from and regulating our activities according to principles that reasonable and rational persons would freely accept and agree to from a shared point of view modeled according to these powers. But although we do not choose, individually or collectively, principles of justice, we do devise basic social institutions. We cooperatively decide, through laws and willing acceptance of social and legal conventions, how the constitution, the economy, property, and so on are designed and 51 · In this connection, it is important to note that Rawls's use of the term "reasonable" is ambiguous (KC, pp. 554, 55!r-7o). In its narrower use it contrasts with "rational" (in the sense of each person's rational good), and expresses the idea of fair terms of cooperation (KC, pp. 5:18-:J.g). In its broader sense, used here in connection with his conception of prin objectivity, it is a stand-in for truth in Rawls's practical account of j ustification. The " nable ciples �ost reas_onable for us" are those that best fit with our capacities for both reaso . and rational de�be�tion, where "best fit" is ultimately Rawls 5 decided by the two stages of d (l) argu�ent for JUStice as fairness: ( 1 ) the argument from the Original Position, an Ra�ls s argu�ent for stability and the congru th our ence of the principles of justice wi rational good m a well-ordered society pt. J).
(TJ,
1 34
I SS
Reason and Agreement in Social Contract Views
fit together into one social scheme. Here the significance of social agree ment comes in. These basic institutions fundamentally affect our final ends, our basic affections. our character, and the direction of our moral sentiments. So in cooperatively making and supporting laws and volun tarily complying with the existing rules of these institutions, we are de ciding what kinds of persons we are and can come to be. In doing that, we deeply affect the kinds of considerations that count for us as goocl reasons. We are, in cooperatively devising and maintaining basic social institutions , indirectly shaping ourselves and defining practical reason.s� The question is whether we are going to design basic institutions that enable us to realize our legislative and executive capacities for practical reasoning, and thereby take responsibility for our characters and final ends; or whether we are going to sustain institutions that create in us desires and states of character that keep us in a state of heteronomy, or subju gation to forces that, although we (often unwittingly) create them, are beyond our direct control. The significance here of a social agreement for the basic structure is threefold. First, it emphasizes the social nature of practical reasoning, how what we accept as reasons in both public and private life is largely influenced by a shared public culture, primarily, the basic institutions we create and sustain. This accounts for the primary importance of our coming to a public understanding regarding the basic structure. That problem affects our personal and associational relations as well as our political relations; its resolution is a precondition of our resolving other deep- seated moral differences. Second, agreement implies the social conditions necessary for auton omy . Rawls does not hold (as Kant sometimes seems to suggest) that we
can realize moral autonomy under any circumstances. We need a back ground of social institutions of a certain kind (cf. KC, p. S SJ ) A cooper ative effort is required to create these institutions, j ust as a cooperative effort is needed to sustain those we now have. Full autonomy presup .
poses a joint effort and a collective decision. I t requires our cooperatively altering existing institutions and devising others so that they conform to
our moral powers and our conception of ourselves as free and equal. These two points u nderscore the importance of Rawls's full publicity condition (KC, pp. SJ 7ff. ), which suggests the final significance of social
5:1. Cf. KC, p. s6o. where Rawls says that the moral powers are not fixed but are shaped and developed by a shared public culture.
135
Philosophy & Public Affairs
agreement. Again, unlike Kant, Rawls does not hold that first moral prin ciples or the conception of ourselves as autonomous is implicit in individ ual moral consciousness. We must be educated about this conception of persons and their relations. Full publicity of principles and their justifi cation makes possible an awareness and understanding of ourselves as autonomous beings. And social agreement on principles, u nlike impartial individual decision, implies the publicity condition. Full publicity is im portant because it ensures that the process of public j us tification can be carried through completely. Everything that counts as a reason and a basis for claims in the public culture can be fully j ustified; there is no need to keep anything hidden from public view. Let us return now to Gauthier. He specifies a conception of the auton omy of utility maximizers that involves critical reflection on means and clarification of ends, but that denies that "reason affords substantive grounds for choice or action" (MA, p. 344). The significance of Gau thier's account of rationally autonomous beings is that it, too, seeks to capture a conception of persons and their social relations that is latent in
a part of our culture: the pure capitalist ideal of fundamentally self-cen
tered and acquisitive individuals whose relations are all based in contrac tual bargains (MA, pp. 3 1 8- 1 g). Gauthier's autonomous beings are not
only moved ultimately by self-interest; they are also u nder no illusions about it. They publicly affirm self-interest as fundamental to their social relations and the system of public reasons. It is a separate question whether Gauthier's principles and conception of the person can serve as a public basis for justification, or provide for stable social relations even among his rational egoists. Still, Gauthier has latched onto an under standing of persons and a source of public reasons in our culture that stem from the institutions we willin gly sustain, and that provide a suffi cient basis for public justification for many who are especially favored by those institutions. Both conceptions of the person-Rawls's notion of the fully au tono mous democratic individual and Gauthier's idealized conception of "eco s. nomic man"-represent types engendered by our basic institution
re, These conceptions reflect a fundamental tension at work in our c ultu er and per aps in ourselves. The ultimate justificatio h ot n of one or the of moral VIew cannot be (as Gauthier would have it) which conceptions l reasons and persons best approximate the model relied on in the s ocia sciences (MA, pp. 8, 3 1 6), for that account is designed to deal with en-
�
1 36
I 57
Reason and Agreement in Social Contract Views
tirely different problems. It is, rather, a practical question : What kind of persons can we most closely identify with and do we finally want to come to be-individuals whose institutions and basic relations are ultimately imposed upon them by their socially created desire for unlimited acqui sition ; or persons who can take responsibility for their desires, charac ters, and social relations by patterning institutions to help them realize their capacities for practical reasoning?
EDWA R D F. McC L E N N E N
J ustice and the Problem of Stab i l ity
I . THE PROBL EM O F STABIL ITY John Rawls suggests that the issue of distributive justice characteristi cally arises between persons who cooperate with one another for the sake
of their own personal ends. The principles of justice, on this account, are to be understood as regulating the distribution of benefits that this type of social cooperation makes possible. Since in general greater benefits
for so me will mean lesser benefits for others, there will be a potential conflic t of interest over what principles to adopt. Rawls argues, however, tha t given an appropriately defined hypothetical setting in which delib
eration takes place-and in particular the assumption that choice is mad e behind a "thick" veil of ignorance-a consensus on principles will nonetheless emerge. 1
S uch a consensus, of course, amounts to a momentary convergence of pru dential judgment under conditions of extreme uncertainty. Even supposin g that individuals can agree under such conditions on a concep tion of justice that will govern their cooperative venture, why should they
be disposed in their actual dealings with one another to do what these Principles req uire? This poses what Rawls terms the problem of stability:
I am indeb ted for helpful comments to John Rawls, Gerald Postema, Geoffrey Sayre McC ord, Laurence Thomas, and participants of philosophy colloquia held at Rutgers Uni versity, the University of North Carolina at Greensboro, Duke University, and Bowling Green State University, where drafts of this article were read. 1 · These are, of course, the concerns of Part One of Rawls's Theory of justice (Cam bridge: Harvard University Press, 1 97 1 ) (hereafter TJ), but for a full appreciation of the tssues involved, the following are also important: "Justice as Fairness," Philosophical Re . vzew 57 ( 1 958); "Constitutional Liberty and the Concept of Justice," Nomos VI: justice, ed. J. Friedrich and John Chapman (New York: Atherton Press, 1g63); ''The Sense of Jus ce," Philosophical Review 52 ( I g6J); and "Justice as Fairness: Political not Metaphysical," hilosophy & Public Affa i 1 4, no. 3 (Summer 198 . 5) rs
�
�·
1 39
Philosophy & Public Affairs
4
cooperation the problem of stability arises because a just scheme of from the may not be in equilibrium, much less stable. To be sure, tively collec standpoint of the original position, principles of j u stice are comply rational; everyone may expect to improve his situation if all ects with these principles, at least in comparison with what his prosp an life would be in the absence of any agreement. . . . Yet in everyday e ben individual, if he is so inclined, can sometimes win even greater ers . fits for himself by taking advantage of the cooperative efforts of oth acting e becaus . . . just arrangements may not be in equilibrium then is fairly is not in general each man's best reply to the j u st cond uct of h associates. 2 Stability of the requisite sort, Rawls su ggests, is not guaranteed by the conditions of the original position (or what he sometimes calls the "ana lytic construction"). This point is implicit in the whole of the argument of the third part of A Theory ofjustice, but it is stated in an even stronger form in an earlier article, "The Sense of ju stice": The aim of the analytic construction is to derive the principles of ju s tice which apply to institutions. How persons will act in the p articular circumstances when, as the rules specify, it is their tum to do their part is a different question altogether. Those engaged in an institution will indeed normally do their part if they feel bound to act on the prin ciples which they would acknowledge under the conditions of the an alytic construction. But their feelin g bound in this way is not itself accounted for by this construction, and it cannot be accounted for as 3 long as the parties are described solely by the concept of rationalit YStability requires that persons possess a " sense of j ustice," an effec tive desire to act as the principles of j ustice require. This sense of j u stice, Rawls argues (paraphrasing a remark of Rousseau's is "no mere moral ) conception formed from the understandin g alone, but a true sentiment ve of the heart, enlightened by reason, the n atural outcome of our prim iti oW affections."4 Chapter 8 of A Theory ofJustice is devoted to explorin g h
2.. Rawls, TJ, pp. _49&-97· As this quotation suggests, Rawls tends to think about the . m pro?lem of stability terms of the economist's model of the free-rider problem. or alter· d., nauvely, the game theorist's model of the prisoners' dilemma. See also, in particular, ibi pp. 2.67-70 and 336. 3· Rawls, "The Sense of Justice," p. 285. 4· Ibid., p. 2.81 . 1 40
5
Justice and the Problem of Stability
and under what conditions such a sense of j ustice arises from more "primitive" affections. The analysis draws heavily on studies of psychol ogists concerning stages in a child's development of moral attitudes.s On Rawls's formulation, three stages can be identified, each governed by its own psychological law:6
( I ) The Morality of Authority,
in which the child takes the parent as
authority, and which is governed by the "first law: given that fam
ily institutions are j u st, and that the parents love the child and manifestly express their love by caring for his good, then the child, recognizing their evident love of him, comes to love them."
(2) The Morality of Association,
the content of which is given by the
moral standards appropriate to the individual's role in the various associations to which he belongs and which is governed by the "second law : given that a person's capacity for fellow feeling has been realized by acquiring attachments in accordance with the
first law, and given that a social arrangement is just and publicly known by all to be j ust, then this person develops ties of friendly feeling and trust toward others in the association as they with ev ident intention comply with their duties and obligations, and live up to the ideals of their station."
(3 ) The Morality of Principles,
in which the person becomes attached to principle s themselves in accordance with the "third law: given that a person's capacity for fellow feeling has been realized by his
forming attachments in accordance with the first two laws, and s. This is the concern of TJ . chap . 7 . as well as the earlier article "The Sense of Justice." Rawls refers in particular to the work of three psychologists, William McDougall, Jean
Piaget, and Lawrence Kohlberg (see TJ. pp. 4 60 n. 6 and 461 n. 8) . R awls's argument, in this respect, strikes out on a path relatively untraveled by mainstream moral philosophers
in this century. For the most part, philosophers have shown little interest in the problem of psychological stability. Most have been preoccupied with the issues of meaning and
Justification. The tacit assumption seems to be that if one can show certain principles to be j ustified, this will suffice to ensure that reasonable persons will accept these principles as re gulative of their affairs. On this view, once the q uestion of acceptable principles is resolved, one has , as a bonus, a solution to the problem of stability. Such a view relies on a dubiou s psy cholog y, however. It supposes that to grasp abstractly the validity or "reason ableness of a princ " iple is ipso facto to be disposed to accept it in one's dealings with others. M oreover, it can not be said that recent treatments of the justificatory issue leave one very san gui ne abou t the prospects for an informed intellectual consensus on a principle of j us tice.
6. See in particular Rawls, quoted in the tex t.
TJ,
sees. 7rr-72, and pp. 4go-91 for the summary statements
141
Philosophy & Public Affairs
6
given that a society's institutions are j ust and are pu blicly known
by all to be just, then this person acquires the corresponding sense
of justice as he recognizes that he and those for whom he cares
are the beneficiaries of these arrangements." Taken together, these laws assert that the active sentiments of love and friendship, and the sense of justice itself, arise from the manifest intention of other persons to act for our good. As a result of recognizing that others wish us well, we come to care for their well-being in return. Thus we acquire attachments to other people and institutions according to how we perceive our own good to be affected by them.' But these are direct attachments: our own goals and interests have been transformed. In this sense, these laws govern changes in the affective ties which be long to our final ends, and not just our recognition of appropriate means. 8 Rawls argues that these laws are expressive of a disposition to recip rocate-to answer in kind-and that this disposition is a deep psycholog ical fact: Without it our nature would be very different and fruitful social coop eration fragile if not impossible. For surely a rational person is not in different to things that significantly affect his good ; and s upposing that he develops some attitude toward them, he acquires either a new attachment or a new aversion. If we answered love with hate, or c ame to dislike those who acted fairly toward us, or were adverse to activities that furthered our good, a community would soon dissolve. Beings with a different psychology either have never existed or mu s t soon have disappeared in the course of evolution. A c apacity for a sense of justice built up by responses in kind would appear to be a condi tion of human sociability. The most stable conceptions of j ustice are presum ably those for which the corresponding sense of j u stice is most firmly based on these tendencies.9
2.
STABILITY AND RATIONALITY
A careful reading of A Theory of Just ice makes it clear that it would be a on mistake to treat the rational choice argumen t of the analytic constrUcti and the psychological theory pertaining om to the sense of j u stice as c 7. Ibid., p. 494.
8. Ibid. , pp. 493-94·
1 42
5 g. Ibid., pp . 4 94-9 ·
7
Justice and the Problem of Stability
pletely disjoint. 1° First of all , the problem of stability is said to be highly
relevant to reasoning in the original position. Distinct conceptions of jus tice can be ranked with respect to their capacity to engender their own affective support, that is, to foster the developmental process that cul minates in the possession of a full sense of j ustice. 1 1 This is something that those in the original position must take into account: " I t is evident that stability is a desirable feature of moral conceptions. Other things being equal, the persons in the original position will adopt the more sta ble scheme of principles. However attractive a conception of justice might be on other grounds, it is seriously defective if the principles of moral psychology are such that it fails to engender in human beings the requisite desire to act upon it."12 But in view of what Rawls has to say about why stability is a desirable feature of a conception of j ustice, it is clear that the connection between the sense of j u stice and rationality goes even deeper. Society being a cooperative venture for the sake of mutual advantage, and hence contin u ally subject to the strains created by a plurality of final ends, a stable consensus on principles of j ustice is essential to the enterprise itself: while men may p u t forth excessive demands on one another, they nevertheless acknowledge a common point of view from which their claims may be adjudicated. If men's inclination to self-interest makes their vigilance against one another necessary, their public sense of j ustice makes their secure association together possible. Among indi viduals with disparate aims and purposes a shared conception of j u s tice establishes the bonds of civic friendship; the general desire for justice limits the pursuit of other ends. One may think of a public con ception of j ustice as constitu ting the fundamental charter of a well ordered human association. J3 I o. They appear somewhat more disjoint in the earlier article "The Sense of justice," since there the discussion abstracts from the question of the relevance of the comparative stability of altern ative principles of justice for the reasoning that takes place behind the veil of ignoran ce, and there is no discussion of what in TJ Rawls comes to refer to as the prob lem of the con gruence of the sense of justice with the good of the agent. I I . See in particular Rawls. TJ . pp. 53-56, and sec . 76, for a discussion of the compar 4 ative stability of alternative conceptions. 12. Ibid. , p. 4 55 . 13. Ibid., p. 5· See also pp. 9 1-92 for a strong statement of the importance of a sense of 4 the j usti ce or injustice of arrangements in detennining our allegiance to Institutions.
1 43
Philosophy & Public Affairs
8
The outline of this part of the argument is clear enou gh. A well-ordered society is an indispensable common means to personal ends. A well-or dered society presupposes a stable consensus on principles of justice. There can be no stable consensus on principles of justice unless the principles agreed upon have the capacity to engender their own psycho logical support, that is, to engender a sense of j ustice.
ation The idea that the acquisition of a sense of j ustice implies an alter way in one's ends represents something of a departure from the ordinary to of thinking about the choice of principles or policies as instrumen tal of p grou a , e the personal goals of the participants . Imagine, for exampl persons whose social structures do not satisfy principles which could be the subject of a stable consensus and thus who have not yet achieved a well-ordered society. Since their society is not well ordered, we m ay sup pose that the members have yet to develop a full-blown sense of justice .
What Rawls's argument recommends is not that they put into place some institution that will of itself stabilize their relations with one an other. The invitation is to adopt an arrangement whose stabilizing effect comes through its impact on their attitudes. To achieve a well-ordered society they are required to take steps that will result in a reconstruction of themselves as purposive beings-a transformation of their affective ties to one another and hence of their final ends. 1 4 To be sure, i t might b e argued that what would be most u seful to any given participant would be for all others to acquire a sense of justice while he (secretly) remains disposed to defect when this works to his advantage. Rawls implicitly acknowled ges this problem and seeks to counter it by arguing that the acquisition of a sense of j ustice is con gruent with a person' s own good. The turn his argument take s here is reminiscent of a line of reasoning that has repeatedly surfaced in philo sophical talk about morality since the time of Plato and Aristotle . The otion is that our own sense of justice is not a means to, but con stitutive some sense) of, our own well-being. Thus Rawls argues, amon g other well thmgs, that we cannot share fully in the good to be achieved in a om ordered society unless we have this sense ourselves, and ac tin g fr far such a sense of justice is something that we directly desire to d o, inso of as we think of ourselves as free t firs and equal rational bein gs. Th e these points is Aristotelian ; the secon d is clearly Kantian. I S
� (l�
1 4 . 1 return to lhis matter in Section 6 belo w. 1 5. These brief remarks of course, . . cannot do J Ustice to ment m . TJ, chap. g. I shall have more to say about
e argu the complex nature of th this in Section 6 below.
'
1 44
9
justice and the Problem of Stability
There is, then, a double connection between rationality and the sense of j ustice. The capacity of a particular conception of j ustice to engender a sense of j us tice-and thu s its own psychological support-is essential to its capacity to serve as a fundamental charter for a well-ordered soci ety, and hence to be the object of a rational (if abstract) consensus in the original position. B u t, in addition, a sense of j ustice is something that it is rational for the representative person to want to affirm in his own life.
3· RETHINKING THE ROLE OF THE STABILITY ARGUMENT
In the seventeen years since the p ublication of A Theory of Justice the original position argument has been subjected to intense criticism. 1 6 Vir tually none of the criticism, however, has been directed at Rawls's views
concerning stability and the sense of justice. Typically, the question has been what it would be rational to choose in the original position, on the
assumption that the options are all more or less acceptable in terms of
their capacity to en gender a sense of justice.
Given the problematic nature of other aspects of the original position argument, one is bound to wonder whether considerations of stability might not provide an alternative and more promising rational choice ap proach to a theory of justice. The most ambitious hypothesis would be that stability brings with it benefits great enough to be decisive from the point of view of rational choice, and that considerations of stability pick out a uniqu e conception of j ustice. If this could be shown, there would be no need to appeal to the original position argument with its problem atic notion of what it would be rational to choose under conditions of radic al uncertainty. 1 7
I 6. In particular there remains the vexing question of whether, under the conditions characterized in the original position argument, persons would agree to something like Rawls's two principles or, alternatively, to some version of the utilitarian principle. I myself have expre ssed doubts as to whether Rawls has made the case in favor of his two princi ples. See "Constitutional Choice: Rawls vs. Harsanyi," in Philosophy in Economics, ed. J. C. Pitt (Dordrecht: Reidel, 1981 ), and the works cited there. This issue is. I take it, distinc t from the one concerning the significance (or implications) of reasoning from be hind a veil of ignorance for a nonnative theory of justice. On the latter issue, Rawls's re marks in "Justice as Fairness: Political not Metaphysical" are very helpful, but I think they do not touch the issue of whether the "derivation" he sketched in "Justice as Fairness" goes throu gh. 1 7. Some years ago Brian Barry, in The Liberal Theory of justice: A Critical Examina . tion of the Principal Doctrines in A Theory of jus tice by John Rawls (Oxford: Clarendon, 1 973), noted that Rawls was prepared to incorporate stability considerations into the rea1 45
Philosophy & Public Affairs
10
What follows is a sketch of an argument in s upport of this hypothesis. Both the argument and its conclusion are deeply indebted to Rawls. I want to suggest that the general conception of j ustice presented in
A
Theory of justice is especially suited to engender its own psychological support and the resultant stability yields significant benefits. M oreover, the defense that I shall offer extends and adapts what Rawls has to say both about the sense of justice and about its congruence with the good of the agent.
I shall, however, depart from Rawls's analysis in a number of respects. First, as I have indicated, the approach taken obviates any reference to the original position : the concern is solely with the s tability of actual so
cial contracts. I shall, of course, still suppose that persons are limited with respect to the projections they can make about the future, so that, at best, they have only probable knowledge of future events that can sub stantially affect their well-being; but this sort of uncertainty is much less radical than that presupposed by Rawls. IB Second, I want to place even more emphasis than Rawls does on the notion that a commitment to
a
sense of justice can be grounded in a theory of rational choice. Third, I want to accord to a counterpart sentiment, which I shall call the sense of injustice, a somewhat more substantial role. And fourth , the approach taken leads naturally to a broader notion of the problem of s tability. On the view to be developed, a consensus on a principle of j us tice will be unstable not only if those who come to agree on it are subsequ ently prone to unilaterally defect from that agreement, but also if they are dis posed to press for the rejection of that principle and the adop tion of some other, that is, to press for a renegotiation of the social contract.
4 · MOV ING AWAY FROM HYPOTHETICAL CONT RACT S
If one gives up talk of what persons would agree to in the original posi tion argument, what is left? Presumably what persons who perfonn the mental experiment of putting themselves in the original position con-
�
l son g of the original position, and remarked that the stability argument "is so powerfu that t seems to be in imminent ginal [ori danger of short-circuiting the whole elaborate pos!Uon! argument In favor of the 'two principles' " (p. 1 4). f 1 8 In effect, h t I propose · to do here is return to something more like the de gree o un ertainty that 15 Incorporated into the version of the argument to be fou nd in "J u stice as
�
�
��
Fauness. "
1 46
II
justice and the Problem of Stability
front when they return to the real world and take up for consideration the social arrangements that define their actual relations with one an other. Whatever the history of the group, their present political and social structures can become the object of debate and deliberation. Since each will know his present place, and be in a position to form at least some estimate of how he expects to fare, the problem of conflict of interest becomes very real. In the language of game theory, any such group of persons will face an n-person strategic interaction problem in trying to reach agreement on the prtnciples which are to govern the basic social and political structure. This will be true even if there exist well-defined procedures for effecting changes in that structure, for such procedures simply serve to constrain the options available to vartous coalitions of par ticipants or the relative attractiveness of those options. Insofar as the in teraction problem is mutually perceived as having a bargaining dimen sion, the controlling expectations of each will be shaped at least in part by his estimate of the relative bargaining power of the various partici pants. And relative bargaining power in turn will be shaped not only by de facto procedural rules, but by the de facto distrtbution of resources, knowled ge, talents, and so forth, which is largely a matter of historical accident. Under such conditions, participants are likely to perceive their rela tions with one another as defined, not by reference to principles to which they would be willing to give their assent under certain hypothetical con ditions, but by reference to principles that simply express the best that they have been able to secure at that time. There is no reason to suppose, then, that the society in question will satisfy Rawls's condition of being "well ordered." B ut then there is no reason to suppose, with respect to the fu ndamental charter of that society, that participants will have a de veloped sense of justice. Thus one can expect that they will be disposed to "free-rtde." Stability, to the extent that it exists, will result not from volun tary compliance, but from the institution of vartous enforcement devices. ' 9 1 9· Among the devices that qu alify in this regard, of course, will be coercive indoctrina . tio n-ideologic al "education." As will perhaps be evident, the conclusions of the argument to be explored here will hold, if they hold at all, only on the assumption that the participants have not bee n successfully indoctrinated. Even the most "unjust" of arrangements can be stable if people have been manip.ulated into believing them to be legitimate. This raises a host of iss ues that 1 cannot deal with in this article. I shall have to content myself with observi ng th at, given the wide range of alternative arrangements that may work to the
1 47
Philosophy & Public Affairs
12
In this setting, however, the problem of instability can be seen to man ifest itself in other ways. One must expect not only that persons will be disposed to unilaterally defect from their arrangements when they per ceive this to be to their advantage, but also that, as conditions change, they will be disposed to press for renegotiation of existing arrangements.
A "contract" whose terms have been fixed by the relative bargaining
power of various competing groups offers no immunity against renego
tiation pressures, and this applies in particular to the distributive aspects of the social contract. "To each according to his bargaining advantage" cannot serve as the basis for a stable agreement on a principle of distrib u tive j ustice.20 Moreover, the two forms of instability just identified-arising from the tendency to defect and the tendency to press for renegotiation-interact. Widespread defection from the terms of an agreement invites reconsid eration of those terms, but continual renegotiation will also tend to call into question the legitimacy of the social contract, and this in tum is likely to encourage free-riding. Moreover, selective defections will lead to changes in the distribution of power and bargaining advantage, and this in turn will generate pressures for changes in the contractY
will
les to mutual advantage of persons, it always remain for us to ask whether the prin cip a which persons appeal in defense of some particular arrangement are really expressive of the at e fair arrangement, or whether they simply serve to rationalize the privile ges of som e expense of others. In this regard, I find the last part of E. Ulmann-M argalit, The Emergenc of Nonns (Oxford: Oxford U niversity Press, 1977), most helpful. Much of the talk abou t t. the value of strictly competitive markets, for example seems to me to be of the second sor , _ g 2.o. R�wls himself has repeatedly remarked that this poses a problem for any b argai m or negotiation approach t the social contract (see , for example , "J ustice as Fairness, � 1 76-77 n. 1 2.). In suggesting that we must widen our conception of the problem of stabili Y arn to encompass tendencies toward renegotiation as well as tendenc ies toward defection, 1 ss, indebted to James Buchanan, The Limits of Liberty (Chicago: University of Chic a go Pre
�
-�i·
1975), in particular chap. 5·
, 2 1 · The relation between the "fairness" of the social contract and its stability is , of cou rse complex. Suppose a scheme is regarded by some participants as cle arly u nfair. Still it might y be that if those who are disadvantaged unilaterally defect all that will happen is that the se e th er subjected to closer control and observation-polic d more effectively. U nd conditions, they may expect to do appreciably better by conscientiou sly ac ting on he agreement, despite what they take to be its unfairness, ou tweigh and this may serve to _ ed �ot to their sense of the mjustice vat moti of the scheme. Such persons may well be ts IS not defect. Thus a sense of the unfairness of a scheme on the part of some participan a su ficient condition for instability. e of the ns se And clearly we must acknowledge th at a R aw s unfa�mes of a scheme is not a necessary � condition for unilateral defec tion : what ven charac nzes as the assurance problem e se ari (TJ , pp. 2�70 and 336-37) can conne �on with schemes that are perceived problem � to be fair. Similar remarks apply to th e . of stability m the face of renegotiation possibilities.
will � �
�
�
!
m
�
1 48
13
Justice and the Problem of Stability
5 · THE SENSE OF ) USTICE AND THE SENSE OF INJU STICE
Consider now Rawls's suggestion that a sense of justice can serve to counterbalance the disposition to unilaterally defect. Suppose that a per son has a well-developed sense of justice but finds that others have acted as free-riders. On Rawls's account, the sense of justice does not demand that we unconditionally continue to do our part. Those who have a sense of justice will be disposed to do their part only if they expect others, or sufficiently many of them, to do theirs.'2 May we suppose, then, that when those who have a sense of justice anticipate that others will not do their part, rational self-interest takes over and determines what they shall do? The characterization of the sense of justice as a disposition to respond in kind invites one to consider the sense of justice as more complex than this-as a disposition not only to do one's part when one expects that others will also, but to respond negatively and even aggressively to those who free-ride at one's expense. 23 On this interpretation, anticipated de fections by others can be expected to engender not only distrust (which has direct bearing on rational self-interest) but also attitudes of resent ment and hostility. The point is important. Rational self-interest does not invariably dictate that we resist or take aggressive action against those who take advantage of us. Reciprocating may simply make a bad situa tion "worse"-may, for example, tip a delicate balance in the "wrong" direction. And valuable resources will have to be expended to identify free-riders and bring them into line. 22. See ibid., pp. 267 and 336. 23. Ibid .. p. 495. Reciprocity within this context is admittedly complex. In certain pas· sages R awls speaks of reciprocity as a disposition to respond in kind. But care must be taken here. I suppose that not only the person who responds to defections on the part of others by defecting himself but also the person who takes action to force defectors back into com pliance responds "in kind." Insofar as being a free-rider implies serving your own interests in such a way as to impose an expense on others, the response in kind that others undertake may simply be to impose expenses on you. The operation of a disposition to reciprocity in this sort of setting can be compared and contrasted with its operation in other se ttings-in situations in which the nonnal state is not mutual forbearance for the sake of mutu al advantage but mutual disengagement or noninteraction ("''ll leave you alone if you'll leave me alone"), or potential mutual hostility ("I'll refrain from attacking you if you'll �frain from attacking me"). It seems clear that what differentiates the sense of justice and llljustice from certain other moral dispositions is not the nature of the perceived intention on the part of others that serves to foster it, but rather the setting in which it is fostered and expressed. What we characterize as a sense of justice and injustice is a matter of feeling favorably or ill disposed towards others in a particular kind of setting, in which one has, for example. come to count on one's expectations concerning how others will behave.
1 49
Philosophy & Public Affairs
A sense that shapes one's response to free-riding as well as one's re sponse to principled behavior on the part of others is a sense of injustice as well as of justice. If the sense of j ustice restrains a person from doing what he may have a "direct" self-interest in doing (defecting), a sense of injustice can encourage him to do what he may have a "direct" self-in terest in not doing (for example, investing costly resources in bringing free-riders into line). In both cases the moral disposition serves to coun terbalance rational self-interest as the latter is usually unders tood. 24 is Finally, recall here the conditions under which the sense of justice s's awl R On fostered and what this su ggests about the sense of inj ustice. account, it is a perception that others are intending to act for our good (as exemplified, for example, in the loving care of our parents) that sets in motion the developmental process culminating in our havin g a sense of justice. But that primal capacity for a loving response is tran sfonned into a sense of justice via the mediating effect of experiencin g succes sively more abstract forms of association-in which the perception of others as directly caring for us is replaced by a sense of others doing their part, and of institutions arranged, in ways that work to our own benefit. That is, our instinctual response to loving parents is transmuted into a commitment to do our part when we perceive that others are doing theirs. Correspondingly, we may conjecture that prim al feelin gs of hatred, engendered by a perception that others are intending ou r hann or otherwise thre atening our well-being, will be transformed, within the context of more abstract associations designed for mutual adv antage, into a sense of anger and resentment towards those who fail to do their •s p art and thereby secure advantages for themselves at our own ex pense. The account just given of what counterbalances the "rational " dispo sition to defect can be adapted to apply as well to the problem of rene gotiation. As I have noted, it is plausible to suppose that persons will be rationally motivated under certain conditions to press for ren e gotiation.
1�
follow, �0 24·. The qualifier is necessary here, since on Rawls's account, which 1 shall � ge acqUJ.re a sense of justice _(and hence also a �h o a sense of injustice) is to underg uce 0 �ecnve ?es, and hence m final ends. In this sense, to act on one's sense of Jus _ IDJusuce Is fully compatible with rational self-interest, where the latter is defined in terms . of choosmg that act which most effectively promotes our ends. I shall return to this pomt below. an 2?· To so conjectu re, of course, is not yet to provide an adequ ate accoun t of how . rans· t is nn ha us msttnc tual reaction to a perceived intention on the part of others to do _ posed mto a sophisticated sense ofinjustice.
1 50
IS
justice and the Problem of Stability
What can act to counterbalance this? One possibility, of course, is that those who have the power to alter the arrangement in ways favorable to themselves anticipate that others will subsequently be in a position to press successfully for alterations in their own favor. If those who pres ently hold the advantage think that others will be more disposed to press
in the future if they themselves now show no restraint, they may decide
to show such restraint. B u t this sort of stability is fragile.
There is an alternative account, one that appeals again to a tendency to reciprocate, to respond in kind. The hypothesis is that a sense of jus tice and inj ustice can function here in a manner similar to the way it functions in free-rider situations, as a disposition not to press the advan tage that one may enjoy at a given moment with regard to renegotiation
if others are similarly willing to refrain. Moreover, just as in free-riding situations, the capacity for restraint with regard to renegotiation is typi
cally coupled with a disposition to resist renegotiation moves by others or to press for new terms m ore favorable to oneself when others have already done so or can be expected to do so. Finally, it seems plausible to suppose that such a sense will in this context, no less than in the context of the defection problem, be engendered by signals of intention. That is, one may suppose that the manifest intention on the part of others to act for or against our good will play a key role in the develop ment of this sense. 26
6. RATIONALITY AND THE SENSE OF j U STICE As we have seen, it is central to Rawls's argument concerning the sense of j ustice in free-riding situations that this sense of justice serves a use
ful purpose. Each agent stands to gain if members of the cooperative
venture have the capacity to restrain their individual dispositions to free ride. A developed sense of j ustice on the part of members of a group
makes it possible for them to realize in their cooperative arrangements efficiencies that are not available to a group of persons all of whom are 26 . Again, of course, much more needs to be said here about how the primitive reaction is transla ted into a full-blown moral response. Moreover, a capacity for reciprocity, partic ularly in this context, is not an unmixed blessing. Nothing I have said rules out the possi bility that well-intentioned participants will end up in an escalating cycle of renegotiation moves-<m the model of escalating. acts of revenge. Thus Hobbes is led in Leviathan to posit as the Seventh Law of Nature that in revenge men respect only the future good.
151
Philosophy & Public Affairs
will defeat the very disposed to be free-riders. Since widespread defection sense of justice purpose of any cooperative venture, in the absence of a tend to be, enforcement devices will have to be employ ed, and these
comparatively speaking, very costly. 27 l in A sense of justice and injustice would seem to be no less u sefu con renegotiation situations.2s To the extent that the fundame nt al social e ther , change tract of a society is subject to perpetual renegotiation and will this will be no stable consensus on basic principles of j ustice, and ct as a pose a serious problem for the effectiveness of the social contra citly cooperative venture for mutual advantage. Rawls himself impli
Justice speaks to this point when he argues at the outset of A Theory of just that "in the absence of a certain measure of agreement on what is their and unj ust, it is clearly more difficult for individuals to coordinate ents plans efficiently in order to insure that mutually beneficial arran gem and ity, are maintained. Distrust and resentment corrode the ties of civil e wis suspicion and hostility tempt men to act in ways they would other avoid."29
y This idea is familiar enough when applied to interactions in relativel m ble small groups. But by appeal to a game-theoretic perspec tive the pro ts can be shown to extend to all forms of interaction in which parti cipan are disposed to think strategically about personal interes ts. Whether one
2.7.
not simply in I ndoctrination can be no less costly than more overt forms of coercion, of virtue of the expenditure of resources it requires. but in terms of the relative rigidity attitudes it instills in persons. What today serves to rationalize the advant ages of some can. under changing technological and economic conditions, work to the disadvantage of all. m· 2.8. Of course, persons who restrain themselves with regard to renegotiation put the e th in s a selves at a disadvantage if others do not also restrain themselves . That is. j u st case of the problem of free-riding, the capacity to restrain oneself with regard to person allY advantageous renegotiation is subject to exploitation by others. Thus it would appe ar t�at what proves useful is the development of a more complex disposition to reciprocate : a will y mgness to refrain from pressing for terms more favorable to oneself if othe rs will similarl �frain, but also a willingness to press when others are similarly disposed. I am deeply mdebted to James Buchanan for forcing me to think about these matters. Some years ago I was invited to comment on a paper of his that dealt with one aspect of this iss ue-spec� IC�y wtth a class of situations in which those who are capable of "other-regarding" mou vanons can be subject to exploitation. The paper, "The Samaritan's Dilemma," was pu b mund � lished together with my reply in Altruism, Morali ty and Economic Theory, ed. E d nan s �helps (New York: Russell Sage Foundation, c 1 975) . At the time, 1 objec ted to B u ha g t�e in line of argument, _urging that those who accepted ern c con his recommendations pns e . th Samantan s Sltuauon would simply find themselves back in another version of oners' dilemma. I now think that I failed to appreciat ysis. e the force of his an al
29. Rawls, TJ. p.
6.
1 52
17
Justice and the Problem of Stability
thinks in terms of noncooperative or bargaining models of interaction, the conclusion is the same : even if an equilibrium emerges it will almost certainly be suboptimal-that is, given the resources, technology, and knowledge available to participants, significant opportunities for mutual gain will fail to be realized. Some might hope to find in the paradigm of perfectly competitive mar kets a form of interaction in which the ou tcome will be not simply in equilibrium but optimal as well. But political, as distinct from market, processes fail to satisfy the conditions of perfect competition, a point con vincingly demonstrated in Buchanan and Tullock's Calculus ofConsent, and in the large body of public choice literature that this work has spawned.Jo Moreover, in addition to the problems cau sed by distrust, resentment, and simple lack of effective coordination, those who are perpetually re negotiating the distributive aspects of their basic agreement with one another face the prospect of a substantial expenditure of resources to effect purely redistributive shifts, resources that have to be diverted from projects that promise mutual gain. In everyday terms, such persons will have to acknowledge that the price they pay for being preoccupied with what percentage of the pie they get is the failure to realize significant opportunities to make the pie grow larger.JI Gran ting that a sense of justice and inj ustice is socially useful and that each of us has an interest in others possessing such a disposition, there still remains the question of whether it is rational for the individual to encourage the development of such a disposition in himself. As I men tioned above, Rawls takes this to be a real problem, and seeks to resolve it by arguing that a person's sense of justice is congruent with, and to some extent constitutive of, his own good. If one adopts the expanded conception of the sense of justice, does this problem become even more severe? 30. For a review of this literature, see my "R ational Choice and Public Policy: A Critical Survey," Social Theory and Practice g ( 1 g83). 3 1 . For a study that sheds some interesting light on this problem as it arises in the con· text of bargaining games, see H. Raiffa, The Art and Science of Negot iation (Cambridge: Harvard U niversity Press, 1 ga2). Efforts expended to effect purely redistributive shifts, of course, must be carefully distinguished from those devoted to mutually advantageous changes. It can be plausibly argued that a vigorous and continuing debate over change� or adjustments that would work to the mutual advantage of those involved is itself somethmg that c an be expected to be mutually advantageous.
1 53
Philosophy & Public Affairs
On first consideration, it might seem that adopting the expanded con ception of a sense of ju stice and inju stice creates no additional difficulty. The problem of free-riding, it might be argued, arises because while agreement on a scheme of cooperation is public, private defection is pos sible. If the defection is observable, others will typically have a self-inter ested motive to force us to do our share, and defection will less frequently be the rational policy. In the case of renegotiation, to press for a new social contract is to do something unavoidably public. The gains that one may be able to achieve by forcing renegotiation on one occasion must, then, always
be balanced against the additional losses that can be ex
pected to arise from others responding in kind. Thu s one may find it prudent not to press whatever bargaining advantage one has at a given moment. When the matter is put in this way, a capacity for ;restraint with regard to
renegotiation may seem to be nothing but ordinary rational
self-interest informed by a more sophisticated sense of the problems of strategic interaction. And a similar line of reasoning would seem to be possible with regard to what I have described as the sense of injustice: once again, it might seem that the agent has a direct interest in devel oping a capacity to resist being exploited by others, both with respect to defection and with respect to renegotiation. Despite these considerations, I would argue that a full-blown capacity for restraint and retaliation clearly runs well beyond anything that could be sanctioned directly by the
usual account of strategic rational choice.
As the above account is designed to suggest, the sense of j ustice in cases where the problem of renegotiation arises, no less than in cases where the problem of defection arises, involves a transformation of affective ties to others, a transformation of final ends. But if questions can be raised concerning why it would be rational for a person to develop a sense of justice with regard to renegotiation as well as defection, still it seems that the problems will be similar. Once again,
that is, the task will be to show why it is rational for the person, in tenn s
of his own good, to develop this sense. The argument that Rawls con structs here is complicated, and has hardly received the attention it de serves. His intention is clear enough : he wants to appeal to as uncon
�
trov rsial a conception of the good of the agent as possible-One spelled
out m terms of the notion of a "rational life plan"-which thereby allows
for pluralism with respect to final ends.3� 32. See in particular Rawls, TJ. chap . 7.
1 54
19
justice and the Problem of Stability
Now, it is not clear to me that the conception of the good of the person to which Rawls appeals is as uncontroversial as he hopes. On the other hand, I find myself strongly drawn to his account of goodness in terms of life plans that satisfy various formal conditions. I have been working for some time on what appears to be a related problem in the foundations of decision theory, namely, that the effective pursuit over time of certain continuing objectives appears to require that the agent develop a capac ity to discipline his individual choices to settled plans. The notion here is that those very objectives may provide the agent with a rational motive not to approach each individual decision as if the choice to be made there must be one that maximizes with respect to those objectives. That is, one may have a rational motive for thinking holistically rather than incre mentally about the pursuit of those objectives. My intuition is that this sort of model might provide an alternative ground for changes in one's affective ties, and hence one's final ends-an alternative that links the sense of j ustice and injustice with a less controversial concept of ration ality.JJ 7 · Is THERE A STABLE CONCEPTION OF J u STICE?
Here, then, if only in outline, is a brief for the benefits of a stable social contract, and an argument as to what is required for stability. The model sketched above, which represents both an elaboration and an extension of Rawls's psychological model, implies that for persons who perceive their s ocial order as a cooperative venture for personal ends, the only thing that could effectively stabilize the social contract would be an ar rangement capable of engendering its own support-of generating in participants a willingness to do what it requires, and a willingness to refrain from renegotiating its distributive terms in ways that work to the advantage of some at the expense of others. Moreover, by pursuing som ething at least parallel to the line of reasoning that Rawls employs in chapter g of A Theory of justice, it may be possible to show that such a 33 · An account of the value of such a holistic perspective in the conte t of standard x Prisone rs' dilemma problems may be found in my "Prisoner's Dilemma and Resolute Choice," in Paradoxes of Rationality and Cooperation: Prisoner's Dilemma and New
comb's Problem, ed. Richmond Campbell and Lanning Sowden (Vancouver: University of
British Col umbia Press, 1 g 8 s). For a discussion of the idea as it bears on the foundations of decision theory, see my Rationality and Dynamic Choice: Foundational Explorations (Cambridge: Cambridge University Press, forthcoming).
155
Philosophy & Public Affairs
20
sense of justice and inj ustice can withstand rational reflection in that the representative rational agent will not merely want to encourage others to have such a sense, but will find it rational to develop such a sense in himself. Nothing in what I have said so far, however, implies that any particular conception of justice qualifies to serve as the fundamental charter of a well-ordered society. In particular, I have not shown that Rawls's prin ciple of justice as fairness is uniquely, or even especially, qualified to engender its own support. Can such an argument be constructed? Here I understand Rawls's principle of fairness as an egalitarian-effi
ciency principle: prima facie, all are to share equ ally in the benefits that social cooperation makes possible, but this presumption in favor of equal
ity can be set aside just to the extent that some inequality can be shown
to work to the mutual advantage of all. This is what Rawls takes to be
the general conception of justice as fairness, of which the now familiar formulation in tenns of the principle of equal liberty and the difference principle is a special case.34 Now, in its prima facie commitment to equality, this principle might be said to "fit" in some sense with the tendency to reciprocity. Insofar as it mandates equality, the principle could be described as requ irin g that no one be accorded an advantage at the expense of others. Recall, how ever, that the very issue that confronts participants in a cooperative ven ture for mutual advantage is the choice of a principle governing the dis tribution of benefits. That is, the debate takes place against the background assumption that no principle even qualifies for consideration if it fails to work to the mutual advantage of all. Thu s the issu e is one of the comparative merits of one distributive principle over another. How ever, if the participants opt for an egalitarian principle rather than a prin ciple that favors some at the expense of others, they end up favorin g some (namely, those who would have received a smaller share u nder the other alternative) at the expense of others (namely, those w ho would have received a larger share under the other alternative). Again, while the qualifier permits only inequalities that w ork to every ad one's advantage-thu s ensuring compensation for those who are less en vantaged, in the sense that they are better off than be they would have nif the inequality had not been allowed-it does this in a partic ularly s tri 34 · See Rawls, TJ, p. 62.
1 56
21
Justice and the Problem of Stability
gent fashion: the least advantaged must be
maximally benefited. Of
course, in virtue of this feature, the principle of fairness would seem well suited to reinforce the sense of j ustice on the part of the less advantaged. That is, the principle does give expression to a maximal commitment to the interests of the less advantaged, and within the framework of Rawls's psychological model this could be expected to engender in them the de sired ties of friendship and fellow feeling, and eventually a sense of j us tice. But what about those who are more advantaged by such a scheme of things (for example, those who are more favored in the natural lot tery)? They end up doing less well under this scheme than they would have if the scheme had conformed to what Rawls terms the principle of natural liberty, as exemplified in the workings, say, of an unconstrained, perfectly competitive market. What serves to engender in them the req uisite sense of j ustice? Why should they not view a distributive order
defined by j u stice as fairness as one that, while mutually advantageous,
is still arranged to maximally benefit others than themselves, and una voidably at their own expense?Js There is an even more troublesome consideration. The version of the
model developed above seems to imply that one's willingness not to de fect from, and to refrain from pressing for personally advantageous redis tributive shifts in, a cooperative venture is even more decisively a func tion of one's perception of the
intention of those with whom one
interacts than it is in the version employed by Rawls. That is, the original position argument, which forms a backdrop for Rawls's own psychologi cal m odel, seems to provide an important exogenous mediating influence on the dynamics of interaction. Those who publicly share a conception of justi ce meeting the test of the original position, with all the implica
tions of impartiality that this carries with it, may still wonder, of course, abou t the motives of those with whom they interact: would they, for ex ample, be prepared to honor this principle were they in a position to im pose their will in such a way as to secure special advantages for them selves? N evertheless, those who share such a conception have the sense
that in virtue of this public consensus they constitute a community. Moreover, and this is crucial, it is not simply that they happen to publicly share some conception of j ustice. If a shared conception of justice is not 35. In this regard, Rawls's remarks in ibid., pp. I OJ-4• concerning what can be said to the more favored seem to me less than fully convincing.
1 57
Philosophy & Public Affairs
22
grounded in some sort of argument (such as the original position argu ment), it must become, for the deliberative intellect, an object of doubt. Rawls speaks forcefully to a parallel point about one's sense of morality : "Imagine . . . that someone experiences the promptings of his moral sense as inexplicable inhibitions which for the moment he is unable to justify. Why should he not regard them as simply neurotic compulsions? If it should turn out that these scruples are indeed largely shaped and accounted for by the contingencies of early childhood, perhaps by the course of our family history and class situation, and that there is nothing to add on their behalf, then there is surely no reason why they should govern our lives."J6 Transposing this to the problem posed for delibera tive thought by a publicly shared but un grounded conception of j ustice, the suggestion is that once again questions about whose interests are
thereby served are bound to arise, and these questions in turn will inev itably give rise to ones concerning the motives of participants.
In the modified version of the model proposed here, the original posi tion argument has been dropped, and this suggests that each partici pant's sense of the motives of others will play a much greater role in determining whether a particular arrangement is or is not stable. But this would seem to imply, in turn, that it is doubtful whether, at least within this revised version of the psychological model, there could be a general proof that this or that distributive principle of justice is more stable than another. That is, stability would appear to be a function, not of the content of a principle as such, but of each participant's perception of the motives of other participants. A perfectly egalitarian scheme that has emerged as the object of consensus simply in virtue of the balance of bargaining power is not necessarily stable. But it does not follow, I want to suggest, that there can be no argument from stability in favor of the egalitarian-efficiency principle. The model still implies, of course, that the requisite sense of justice will be engen dered and reinforced to the degree that participants have a manifest con cern for one another's well-being, and even more important, that for those who have not been ideologically indoctrinated, it is
only to the ex
tent that they regard themselves as forming such a community of shared interest that their society will be, in Rawls's sense, well-ordered. That is, the model that emerges here is decidedly communitarian in import. 36. Ibid., p. 5 1 4.
1 58
23
Justice and the Problem of Stability
Against this background, the q uestion is whether the egalitarian-effi ciency principle is more expressive of a sense of community than any other principle. In effect, since there are conditions under which consen sus on the egalitarian-efficiency principle will carry with it no assurance of stability, agreement on it cannot be shown to be a sufficient condition for a stable social contract. B u t consensus on this principle may still be, within the context of the sort of model of interaction we have been con sidering, a necessary condition of stability. The following consideration would seem to favor such a thesis. For a sense of community-a shared interest in one another's good-to exist, participants must not simply have a concern for one another, but make that concern manifest. N ow, in the context of deliberation and debate over alternative proposals, virtually any distributive proposal canies with it a problem of ambiguity concerning the intention of the proposer. But a proposal that accords Jess, or at least no more, to oneself than to others would seem to offer the best hope that the signal will not be destabiliz ing. And if all proceed in this fashion, only an egalitarian scheme will meet the req uirement that each person's concern for the good of others be expressed in the consensus.J7 Departures from strict equality can be defended in terms of the need to make concessions to human nature-a need to encourage persons to develop their talents and abilities in ways that work to mutual advantage.
37· Rawls himself appears to move in this direction in, for example, his discussion of the relation between the difference principle and what he calls the principle of fraternity; see tbtd., pp. 1 0 5-6. On p. 497 Rawls remarks that "to insure stability men must have a sense of j u stice or a concern for those who would be disadvantaged by their defection, preferably both." Adj usting this remark to apply to the somewhat enlarged sense of justice discussed here, one can take the argument of this article to be that the sense of justice itself will depend on a sense that others are concerned for us, and that therefore a mutual perception of communit arian concern is the ultimate basis for the stability of the social contract. It might seem that there is an obvious objection to this view: given certain marked in equalities in inhe rited capacities (such as strength or intelligence) stable cooperation will be possible among more than ust those who are prepared to treat one another as equals. j Some can be dispo sed without disadvantage to themselves to demand more than others if others perceive themselves to have little choice but to accept less. On the assumption that th ose who manag e to work out cooperative arrangements do better than those who do not, arrangements that are asymmetrical in this sense may still have significant survival value. ut where asymmet ries in cooperative relationships are not based on inherited capacities, 11 would seem that they must be explained in terms of cultural transmission, and that their endurance t estifies to the power of ideology. As I indicated in note 1 9, my concern here is With what can be defended when the effects of ideology have been bracketed.
�
1 59
Philosophy & Public Affairs
24
Insofar as they have any role to play, then, it can only be as they are required as part of a system of effective incentives. The argument, in effect, is that any inequality in treatment within a cooperative scheme places motivational strains upon that scheme, and that such tensions can be counterbalanced only to the extent that the inequalities can be rationalized by appeal to mutual advantage.38
m On this account, justice as fairness is to be understood as arising fro con a a need to find a compromise between persons who want to signal do cern that others benefit at least as much as, if not more than , they
themselves. Does this mean that persons who have no concern for one another, or only very limited concern, are not capable of effectively co operating with one another? One need not draw such a concl u sion. The
argument above addresses itself to intellect, but the psychological model
to which it appeals is deeply rooted in the Aristotelian conception of a
moral disposition as something that develops only by doing, not by mere intellectual concluding. The model implies not only that h uman beings have the capacity to develop affective ties to one another, but that these ties gradually develop and do so only as a result of specific and cumula tive interactions. The argument concerning the costs to each of a failure to achieve a stable cooperative arrangement is designed to provide per sons with a motive to present themselves as if they w ere motivate d by a concern for one another. The theory of developmental psychology su g gests that it is by interacting with others under these (and, it is to be
hoped, increasingly favorable) conditions that the requi site sen se of ju s tice, and a full-blown sense of concern for others, is engendered and reinforced. To adapt a metaphor of Aristotle's from the Posterior A na· lytics, the hypothesis is that by "making a stand" h ere, person s may be· gin developing the necessary affective ties to one another.39
8. THE IssuE OF CoMPARATIVE STABILITY Rawls argues that justice as fairness is superior to the principle of aver·
age utility with respect to its power to engender a sense of ju stice. In all,
38 . �is is.. of course, central to the way Rawls views the proviso about inequ alities; see . m particular 1b!d. , p. ti e See Posterior Analytics Iooa, where Aristotle gives an acc ount of how c um la _ kn wledge of pamc m le, ulars leads to knowledge of universals in terms of a battle simi which first one man makes a stand and then another, and so on.
39· �
78.
� �
1 60
25
Justice and the Problem of Stability
he offers some five considerations in favor of the greater stability of jus tice as fairness.
First, he suggests, justice as fairness expresses a stronger and more unconditional concern for the good of each person. Each is guaranteed
an equal liberty, and assured that his claims
will not be neglected or
overridden for the sake of a larger sum of benefits for others. This height ens the operation of the reciprocity principle.4° It must be recalled here, however, that on Rawls's own account this argument is addressed to per sons who are not behind the veil of ignorance. Thus it runs into the prob lem, already mentioned, of a potential complaint on the part of the more advant aged.
Second, in order to suppose that the utilitarian principle can effectively engender its own psychological support, one would have to suppose a modification in the underlying psychological process of attachment for mation. More specifically, one would have to suppose, for example, that according to the second law "persons tend to develop friendly feelings toward those who with evident intention do their part in cooperative schemes publicly known to maximize the sum of advantages, or the av erage well-being." But this, according to Rawls, is a questionable psycho logical principle: "suppose that certain institutions are adopted on the public understanding that the greater advantages of some counterbal ance the lesser losses of others. Why should the acceptance of the prin ciple of utility (in either form) by the more fortunate inspire the less ad vanta ged to h ave friendly feelings toward them? . . . No reciprocity Principle is at work in this case and the appeal to utility may simply arouse suspicion."41 Two thin gs worry me about this line of argument. The first concerns the sug gestion that for a u tilitarian theory of justice the psychological law s would have to be reformulated. On first consideration, it might
seem that this has to be incorrect, for the psychological laws are formu la ted, in sections 7 1 _73 of A Theory of justice, and again in section 75, in such a way as to permit one to read them in a more or less neutral w ay, that is, without reference to any particular conception of justice.
B ut if they were to be taken in that way, the argument from comparative stability would collapse: all coherent conceptions of justice would be
equ ally capable of engendering their own support. Thus the term "just"
40. Rawls, TJ , p. 4gg.
41. 161
Ibid., pp.
4�500.
Philosophy & Public Affairs
as it occurs in each of the three psychological laws is not merely a place holder that can be filled in different ways without substantially affecting the plausibility of the psychological principle that results. Rawls's point is precisely that the psychological principle that results when "j ust" is replaced by "satisfies the principle of average u tility" is a less plausible principle than the one that results when "just" is replaced by "satisfies the principle of justice as fairness." It seems clear from what Rawls argues in section 76, then , that the stability of a conception of justice is measured by the fi t between it and the other essential ingredient in the model, namely, the assumption that persons are disposed to react to one another in certain ways. This su g gests the following as a possible reading of the central question: When
persons are not behind the veil of ignorance, but face one anoth er with knowledge (or at least beliefs) concerning their relative prospec ts (given their natural talents, abilities, and so on), will j u stice as fairness as a principle for regulating the distributive effects of their cooperative ven ture tend to resonate with and reinforce the tendency to reciprocity more than, say, the average utility principle? I say that this is a possible reading, for there are passages (see, for example, the first one quoted below) in which Rawls seems to imply that each alternative conception of justice will call for a distinct theory of at tachment formation, so that, for example, the u tilitarian theory of j ustice will have to presuppose that persons have a strong natural tendency to sympathize with one another. On this account, each alternative concep
tion of justice will resonate with a different natural disposition, and the issue of comparative stability will tum on the comparative stren gth of these "natural" dispositions instead of on which principle m ost effec f tively resonates with a single natural tendency to reciprocity. In view o the nature of the worries I am about to express, I am not sure that it a makes a difference which interpretation one adopts , but this is again matter that deserves more exploration than I can give it here . My second, and deeper, worry is that it appears that one is still faced . With the problem that the difference principle, no less than, s ay, the av erage utility principle, will inadequately secure the allegianc e of some gz:oup. To be sure, on the assumption that the average utility principle will permit significant inequalities, those who are least adv antaged under such an arrangement can complain that the principle sacrifi ces their in terests for the sake of greater average benefits. Rawls describes the prob lem with the u tilitarian principle in a convi ncing manner: 1 62
27
Justice and the Problem of Stability
Why should the acceptance of the principle of utility (in either form) by the more fortunate inspire the less advantaged to have friendly feel ings toward them? This response would seem in fact to be rather sur
prising, especially if those in a better situation have pressed their
claims by m aintaining that a greater sum (or average) of well-being would result from their satisfaction. No reciprocity principle is at work in this case and the appeal to utility may simply arouse suspicion. The concern which is expressed for all persons by counting each as one (by weighing everyone's u tility equally) is weak compared to that con veyed by the principles of j ustice . . . . It is evident why the utilitarian stresses the capacity for sympathy. Those who do not benefit from the better situation of others must identify with the greater sum (or aver
age) of satisfaction else they will not desire to follow the utility crite
rion. N ow such altruistic inclinations no doubt exist. Yet they are likely to be less strong than those brought about by the three psychological laws formulated as reciprocity principles; and a marked capacity for sympathetic identification seems relatively rare. Therefore these feel ings provide less support for the basic structure of society.4' The problem, however, is simply that the most advantaged can com plain that the difference principle sacrifices their interests for the sake of maximally satisfying the interests of the least advantaged. Thus it could be argued that engaging their allegiance to justice as fairness will require a stronger sense of sympathy (for the less fortunate) than in fact exists. Prima facie, then, both principles pose a problem with respect to encouraging fellow feeling. That is, each would seem to have a drawback
With respect to how effectively it can resonate with what is presumed to be the stronger of our "natural" dispositions, namely, the tendency to
reciprocate; and if so, then each would require for its effective operation a stronger capacity for sympathy than human beings seem naturally to p ossess. A third con sideration is that "following the utilitarian conception tends to be d estructive of the self-esteem of those who lose out, particularly
when th ey are already less fortunate." justice as fairness is said to con trast favorably in this respect: "in a social system regulated by justice as fairness , identification with the good of others, and an appreciation of what the y do as an element jn our own good, might be quite strong. But this is possib le only because of the mutuality already implicit in the prin-
4 2. Ibid., p. soo.
1 63
Philosophy & Public Affairs
ciples of justice. With the constant assurance expressed by these prin ciples, persons will develop a secure sense of their own worth that forms the basis for the love of humankind."43 The problem here is that, once again, this speaks to the capacity of j ustice as fairness to secure the al legiance of the less advantaged, not the allegiance of the more advan taged. A fourth consideration is that the principle of j ustice as fairness com pares favorably with the principle of u tility as regards both the clarity of its conception and the attractiveness of the ideal to which it gives expres sion.44 Of course, the former of these two advantages disappears when the utilitarian principle is reformu lated in terms of the distribution of primary goods instead of the distribution of u tility. Rawls briefly ac knowledges the possibility of such an alternative form ulation of u tilitar ianism, but does not pursue the matter.4s As to the relative attractiveness of the ideal expressed by j ustice as fairness, this speaks once again to the matter of the congruence of different conceptions of j ustice with one's own good. Again, I cannot discuss this m atter in the detail it de serves, but the thrust of the argument for the
comparative superiority of
a sense of justice predicated on j ustice as fairness over a sense predi cated on the principle of u tility appears to be this: the principle of utility proves to be more demanding and exacting in the sacrifice s it can ask of us, and thus involves greater risks. It may, for example, call upon us to make sacrifices for the sake of greater benefits for those already more fortunate than us. In the light of this, Rawls concludes, "a rational per son, in framing his plan, would hesitate to give precedence to so strin gent a principle. I t is likely both to exceed his capacity for sympathy and to be hazardous to his freedom. "46 As a fifth, and last, point Rawls suggests that the principles of j u stic e as fairness "are closer to the tendency of evolution than the prin ciple of utility."47 Rawls contents himself here with a conj ecture, but All an Gib bard has more recently offered an expanded defense of this thesis .48 The problem as I see it with both the congruence argumen t and the evolutionary argument-and indeed with virtually all the com parative ar43 · Ibid., pp. 5oo-5o1.
44 · See in particular ibid. , p. 50 1 . 45· Ibid., p. 1 75. 4 6. Ibid., p. 573· 47· Ibid., pp. 503-4. 48· Allan Gibbard, "Human Evolution and the s in Sense of justice' " Midwest Studie Philosophy 7 (Ig!ll ).
1 64
29
Justice and the Problem of Stability
guments--is not that the tendency to reciprocate fails to be stronger than the tendency to sympathy (or altruism). The trouble is rather that the difference principle as well as the principle of utility seems to resonate with the (stronger) tendency to reciprocity less clearly than one might expect. That is, granting that stability is more likely to be secured by building upon the tendency to reciprocity, still it is open to us to wonder whether, given the problem of justifying it to the more advantaged, jus tice as fairness will be much more stable.
g. THE COMPARATIVE 1 SS UE RECONSIDERED Rawls himself acknowledges that he has only briefly considered the is sue of comparative stability, and that the points he has raised are sugges tive rather than dispositive.49 My own sense, as indicated above, is that even these suggestions have to be qualified. One might wonder, then,
whether the line of reasoning presented above in Section 7 provides an
alternative and more secure route to the conclusion that justice as fair
ness is superior to the principle of utility in its power to engender a sense
of justice and thus to underwrite a stable social charter. 1 believe it does.
Whatever attractiveness the principle of average utility might seem to have under the stringent conditions of the original position-as a rational approach to the distribution of benefits under conditions of radical un certainty-it cannot be said to meet the requirement that it provide a natural solution for persons who seek to have expressed in the consensus
their concern that others do at least as well as, if not better than, them selves. An advocate of the principle of average u tility might insist, of course, that in putting forth this principle he does signal his concern, for
he seeks to maximize the expected bene t of the average, and thu s in a fi sense any, participant. This arg ument might have some force if choice is to be m ade behind the veil of ignorance, but it has, 1 think, no such force
When persons face one another under ordinary conditions of differential life prospe cts. In this context, those with inferior life prospects can co gently obje ct that there has been no real taking up and equal bearing of risks. They can insist that in order for them now to have no complaint about their lesser share, they and those who now enjoy the superior pros J>ects mu st at least have all had the same real chances of ending up at
49. Rawls, TJ , pp. 4 53, 499. 1 65
Philosophy & Public Affairs
30
inequalities, any particular level of expectation. (A model of j ustifiab le from hav on such an account, would be, say, the inequalities that result service. ) ing a draft lottery to determine who shall be liable for military of risk, Where, however, there has been no real taking up and sharing be an the argument from the prospects of the average man ceases to objec argument about the prospects of any man, and thus is open to the tion that it is really expressive of vested interest. Rawls himself, in "Constitutional Liberty and the Concept of Justice," be a offered a version of this argument from the consideration that there real taking up and sharing of risk. He subsequently withdrew it, how
-as ever, on the grounds that it was inconsistent with his own claim to exemplified in the original position argument-that one could appeal hypothetical as opposed to real situations of risk taking. so B ut this argu ment comes back into its own again given a framework such as that pro
posed here, in which the argument turns on reactions to real, as distinct from purely hypothetical, social contracts.
1 0. CoNCLU SION I have sought to argue-if only too briefly and imperfectly-for two things. First, that within a society in which there is pluralism with re spect to final ends, and society is perceived as a common means to those
personal ends, a stable and hence truly efficient social contract will be possible only insofar as participants can develop and effectively e xpress to one another a communitarian commitment to their m utual w ell-bein g. Second, that for those who have achieved such a sense of commu nity, perhaps the mos t effective way to express it will be to regulate th eir af fairs by reference to Rawls's principle of j ustice as fairness, in terpreted as an egalitarian -efficien cy principle, while for those w ho seek to develop uch a sense of community, that same principle offers the m ost promis mg place to make their stand.
�
so .
See in particular Rawls, "Con stitutional Liberty and th e Concept of Justice," PP · 1 og-I 6?-68 and 1 68 n. 24.
14, and the remarks in TJ . pp.
1 66
THE STABILITY PROBLEM IN POLITICAL LIBERALISM BY
THOMAS E. HILL, JR. Despite its modest presentation and explicit limitations of scope, A
Theory of Justice has inspired many of us with its large vision, systematic methodology, and challenging arguments. Appearing in the midst of a
relatively stagnant period in political philosophy, the book seemed to highlight the inseparability of moral and political philosophy and to carry
both to a deeper level. Now, paradoxically, Rawls argues that we should see justice as fairness as a less comprehensive, less "deep," self-standing
political conception that does not compete with our many diverse (reasonable) comprehensive moral theories and religious traditions but rather appeals to these for its deep support. As before Rawls presents the political principles of justice as a
construction in a certain technical sense,
but, equally important, the continuous development of his work illustrates how good political philosophy can be "constructed" in a more ordinary
�nse: that is,
gradually and painstakingly built up, its initial vision and . Insights developed over time by persistent rethinking and adjustments responsive to the thoughts of others. Having already proposed a political
philosophy that has achieved acclaim unparalleled in our times, John
Rawls has continued his building project, patiently repairing, extending,
reconstructing, and, when necessary, deconstructing his previous fine work. Reasonable persons can disagree about the truth, and even the political merits, of his proposals, but on the proposition that John Rawls
�
hanged and enriched political philosophy as few others have there is a Just and stable overlapping consensus. Pacific Plrilosoplrica/ Quarterl 0031 -5621/9410200-0000 y 75 (1 994) 333-352 C 1994 Univcnity of Southern California. Published by Blackwell Publisben, 108 Cowley Road, Oxford OX4 11F, UK and 238 Main Street, Cambridge, MA 02142, USA.
1 67
334
PACIFIC PHILOSOPHICAL QUARTERLY
To tum from the obvious to the immediate, my aim here is just to raise some questions.1 For this I want to focus on the relations between Rawls's two books, A Theory of Justice and Political Liberalism.2 The new work, Political Liberalism, I take it, is meant to supplement A Theory ofJustice, not to replace it, much less to compete with it. In the essays published between the two books, Rawls offered many refinements, additions, explanations, and responses to critics, and Political Liberalism brings these together in a helpful and expanded form. In doing so, I think, Political Liberalism satisfies one of its main purposes, adding significantly to the power, subtlety, and persuasiveness of the main system of thought that underlies both works. For example, the liberty principle is somewhat altered and arguments for its priority clarified;3 the original position is now unmistakably presented as a device of representation;• the relation between the "primary goods" and the "two moral powers" of citizens is made more explicit;� the revised argument for concern for future generations seems less ad hoc;6 and now one can see more plainly what exactly, under "constructivism," is supposed to be constructed and what is not.7 These various revisions alone make Political Liberalism an important book, indeed an essential one for moral and political philosophers, quite independently of the particular issues on which my questions will be focused. As an admirer of A Theory of Justice, naturally I am most interested in the major systematic change that Political Liberalism makes in the ideas of earlier work. That basic change, from which Rawls says most other changes stem, is in the way justice as fairness is now presented. Rather than treating justice as fairness as a comprehensive (or, "partially comprehensive ') moral doctrine, he now proposes it as a political conception of justice that is no longer considered as competing for the same role in our lives as other reasonable comprehensive moral doctrines (such as utilitarianism and various traditional religions). At first the change may strike us, the admirers of A Theory of Justice, as disappointing, because a political conception is admittedly a less �nd �nd all encompassing thing than a comprehensive moral theory of jus�ce: lt ��supposes c mmon ideas in the public political culture, it gwd� ? �eclslons only Wlth respect to a restricted subclass of political issues; 10 . 1ts Rawlsmn version, it foregoes any claim to truth; and it must appeal �or �upport outside itself (to comprehensive doctrines) if it is to be deeply JUStified. Thus the larger questions we want to raise are: How radical are these changes? What was the problem that prompted them? Were they t really necessary? Are they successful in meeting the problem? And wha ble are the costs of making them? That is, which, if any, of the admira f�t�es of A. Theory ofJustice have been own d " to lost by the recent move . SIZe the claims for JUStice as fairness? C 1 994 Uni-.ity of Soutbena California
1 68
TliE STABILITY PROBLEM IN POLITICAL LIBERALISM
335
Like many readers, I suspect, I was struck by Rawls's remarks in the
Political Liberalism, suggesting that the major changes Theory of Justice stem from a problem (even an "inconsistency")
introduction to
from A.
in the way that work attempted to show that a society structured by justice as fairness would be
stable. Rawls explains the problem as follows:
[TJo understand the nature and extent of the differences, one must trying to resolve a serious problem internal to justice
as
see
them
as
arising from
fairness, namely from the fact that
the account of stability in Part III of [A] Theory [of Justice] is not consistent with the view
as a whole. I believe all the differences
are
consequences of removing that inconsistency.
Otherwise I take the structure and content of [A] Theory [of Justice] to remain substantially the same•
To explain: the serious problem I have in mind concerns the unrealistic idea of a well
ordered society as it appears in [A} 71reory [ofJustice]. An essential feature of a weD-ordered
society associated with justice as fairness is that all its citizens endorse this conception on
the basis of what I now call a comprehensive philosophical doctrine. They KQCPt, as rooted
in this doctrine, its two principles ofjustice. [Although not explicit in A Theory of Justice,]
one the question is raised, it is clear, I think, that the text regards justice as fairness and
utilitarianism
as
comprehensive, or partially, comprehensive doctrines.
Now the serious problem is this. A modem democratic society is characterized not simply
by a pluralism of comprehensive religious, philosophical, and moral doctrines but by a pluralism of incompatible yet reasonable comprehensive doctrines. No one of these doctrines is aflinned by citizens generally. Nor should one expect that in the foreseeable future one
of them, or some other reasonable doctrine, will ever be affirmed by all, or nearly all, citizens .'
The fact of a plurality of reasonable but incompatible comprehensive doctrines-the fact
of reasonable pluralism- shows that, as used in (A] Theory [of Justice], the idea of a weD ordered society of justice as fairness is unrealistic. This is because it is inconsistent with
realizing its own principles under the best foreseeable conditions. The account of stability of a well-ordered society in part Ill is therefore also unrealistic and must be recast. This problem sets the stage for the later essays beginning in The ambiguity of (A) Theory
(of JwticeJ is now removed and justice conception of justice .lo
as
1980.
fairness is presented at the outset
as
a political
My initial attempts to understand the changes from A.
Theory of Justice Political Liberalism turned crucially on this description of the project. Since the old problem was stability, 1 thought, the innovations of Political . Liberalism must be seen as proposed solution to this problem, m effect, to
as changes necessary to set up a new argument that, despite the fact of reasonable pluralism, a society structured by the principles of justice as fairness would tend to be stable. Many features of the text seemed to �upport this reading, for a major line of argument of Political Liber m
IS that once justice as fairness is presented as a merely a
��
polmcal
conception, then it can win the support of an overlapping consensus of
reasonable comprehensive doctrines, such as the major religio�
t:aditions
and the classic moral theories. This support, despite oppoSibon from
C 1994 Univenity of Southcm California
1 69
336
PACIFIC PHILOSOPHICAL QUARTERLY
"unreasonable" doctrines, would naturally have a stabilizing effect. Given this reading, it seemed correct tot say that changes in A Theory of Justice were necessary, but it remained far from obvious that the innovations in Political Liberalism would be successful in meeting the problem of stability. So long as the problem and its solution are seen as a simple stability problem. the changes in Political Liberalism, I suspect, will continue to strike readers as inadequate to its aim. And, given this, many admirers of A Theory of Justice are likely to see the changes as too high price to pay for the minimum increase in the likelihood of stability. On rethinking the issue, with the help of friends and some hints from Rawls, I am convinced that I have at least partly misunderstood the intent, or at least the current understanding, of Rawls's remarks about how the changes in Political Liberalism were designed to meet the problem of stability. If so, even if I am right to suspect that there will never in fact be an overlapping consensus (of reasonable comprehensive doctrines) on justice as fairness (conceived politically), or even on political liberalism, this may not be incompatible with Rawls's main point. There is another reading on which, even in the absence of actual consensus, the argument for the possibility of such a consensus of reasonable doctrines still serves an important purpose. In my remaining remarks, I will explain my initial reading of the stability problem in A Theory of Justice, its apparent solution in Political Liberalism, and my doubts about that solution; then I will briefly sketch an alternative, hopefully better, interpretation. My point is not to draw attention to my own struggles to understand Rawls's project but rather to articulate what I take to be a quite natural, though deeply subversive, understanding of that project, so that, if it is mistaken, we can get clearer a�out the alternative. Towards the end of my remarks I will also briefly ratse some other questions that, I suspect, are shared by a number of other readers of Political Liberalism. I.
The problem of stability in A Theory of Justice: initial reading
Political philosophy, according to A Theory of Justice, should distinguish and add�es� each of the following tasks. (1) It should provide a method . of td ntifymg, representing, and (in a sense) justifying fundamen . � �nll:ctp�es of j.ustice regarding the basic political and econoiJ11C se m�ttt�ttons; (2) tt should interpret and illustrate the application of tho pnnctples to familiar political issues, showing the implications of the �heory to be more or less in harmony with our considered moral a JUd�ents; and (3) it should offer considerations for believing that SOCiety well ordered by those principl A e. bl sta es of justice would be
�
C 1994 University of Soutbctu California
170
THE STABILITY PROBLEM IN POLITICAL LIBERALISM
337
scheme of cooperati on was defined as stable when it was "more or less regularly complied with and its basic rules willingly acted upon; and when infractions occur, stabilizing forces . . . exist that prevent further violations
and tend to restore the relation."1 1 A stable system is not merely one in equilibrium, like a ball resting on a flat table, but one which under stress, up to a point, tends to return to an equilibrium state, like a ball in a cup. The first two tasks were taken up in parts I and II of A Theory of Justice, while the last, the stability problem, was addressed in part III. Whether a society well ordered by Rawls's two principles of justice would be stable was a question that, in the end, members of the original
position were to consider, but Rawls did not contend that justice as fairness was the most stable conception of justice. For the most part arguments (in parts I and II) for the superiority of Rawls's two principles over alternative conceptions ofjustice presupposed that either Rawls's two principles, or some alternative, would be a basic public charter of a well ordered society and so, at least initially, complied with willingly by almost everyone. Part III was designed to check this initial assumption that the two principles could win the willing compliance of citizens and to argue further that their allegiance would tend to maintain itself, despite stresses, over time. The argument was that, given conjectured psychological principles and the benefits secured to each citizen under the two principles, citizens raised in a society well-ordered by Rawls's two principles would
naturally develop a sense of justice and find a just with their various conceptions of a
combined, A
life largely congruent good life. Thus, with its three parts
Theory of Justice argues that justice as fairness is more tenable than any other traditional conception of justice and, further, that a society well ordered by its principle would tend to stabilize itself under stress and thus to endure. I n Political Liberalism, as I said, Rawls acknowledges a serious internal problem in his previous argument for the stability of a society structured
by his two principles of justice. His earlier argument, Rawls now says, presupposed that citizens accepted justice as fairness as their "comprehensive philosophical doctrine" (or as their "partially comprehensive doctrine"). The argument took for granted, then, not only that other competing conceptions of justice had been rejected but also
�
t at citizens did not hold any general philosophies, religi ns, or wo ld � ? _ VIews that might conflict with their allegiance to the doctnnes of JUStice as fairness. The aim after all was to show that a (Rawlsian) just society
�
�ould be self-perpe uating
if its citizens were brou�t up to inte�
_ shape their Its far ranging conception of justice as fairness, lettmg thiS
thoughts, feelings, and relationships to one another. . But now, quite plausibly, Rawls grants that in any f� �ociety that we can realistically foresee; there will be a plurality of rehg�ons and other comprehensive philosophical doctrines. The very liberties of speech and C 1994
Univenity of Southcm California
171
338
PACIFIC PHILOSOPHICAL QUARTERLY
conscience that justice as fairness assures to each citizen make it almost inevitable that reasonable people will continue to disagree on deep and
broad issues of religion and philosophy. The problem, then, seems to be
that the old argument for the stability of a just society was unrealistic in
supposing that virtually all its citizens, in a climate of freedom, might
come
willingly
to
accept,
and
perpetually
maintain,
one
basic
comprehensive philosophical doctrine, namely, justice as fairness as presented in A
Theory of Justice. We now
society cannot be secured in that way . 12
see that the stability of a just
Even on this initial reading, I should stress, the issue of stability was
never simply the general question of how to make social order durable.
The issue was always how a just society could endure, and the aim was to show this is possible without lowering the standards of justice for the
sake of securing stability. Stability was to be based on the willing consent
of citizens, the idea of securing it by police-state methods being ruled out from the start. Moreover, mere durability was not the aim, but stability;
that is, durability was to be achieved in large part by the fact that the
citizens themselves would have developed attitudes towards justice that
would prompt them to restore the system whenever its just institutions were in danger. Now, supposing (as Rawls did) that in parts I and II of A
Theory of
Justice we already have a reasonable account of what justice demands,
why be concerned about stability? As Rawls notes, moral philosophers
have generally paid little attention to the question of whether their ideals will in fact become and remain widely accepted and used; so one may wonder, why should political philosophy be so concerned? A natural answer, I think, is that political philosophers typically put forward a package of ideas for public acceptance, arguing the merits of the system if it were to be generally accepted. (In contrast moral philosophy typically
offers standards to guide the conscientious choice of individuals, whether or not others will follow their lead.) Given the difficulties of winning general agreement, and the transition costs of reforming institutions, we are un�erstandably reluctant to bother with proposed political changes that will unravel as soon as instituted. Utopian writing is enjoyable to read, but of serious political proposals we demand feasibility and a gh reasonable likelihood that, once accepted, they will survive l ong enou ce to make the effort to institute them worthwhile . Thus insofar as justi as faim is put forward as a political proposal th t one hopes asive become Widely accepted, its appeal would be enhanced by a persu argument that a society just by its standard ent um arg would be stable. The would remove worry: the fear that s uch a society would only be just le for a moment, Its fine principles quickly giving way under predictab
�
stresses.
�
�
C 1994 Uniwnity of Southern California
172
�ill
II.
THE STABILITY PROBLEM IN POLITICAL LIBERALISM
The Solution in Political Liberalism: initial reading
339
Let us assume for the moment that the problem prompting the basic changes from A Theory of Justice to Political Liberalism was the stability problem as I have just interpreted it. The solution offered to this problem in Political Liberalism, then, seems to be not to reject or drastically revise
the content of justice as fairness but to reconceive it and present it in a new light. In sum, the main steps of the solution are: (1) to treat justice as fairness as a political conception rather than as a comprehensive moral
doctrine; (2) to develop a distinction, for political purposes, between reasonable and unreasonable comprehensive doctrines; and (3) to distinguish stability based on an overlapping consensus of reasonable doctrin es from stability based on either uniformity of doctrine or a mere modus vivendi among adherents of different doctrines. Relying on these
central ideas, the major step is (4) to argue that a (Rawlsian) just society can be stable in a pluralistic world because, once seen as merely political, justice as fairness can win an overlapping consensus of reasonable comprehensive doctrines. In effect, by making the claims for justice as fairness more modest and its scope more limited, thereby reducing its potential conflicts with reasonable religious and philosophical doctrines, the changes make justice as fairness more broadly acceptable. By these changes, then, and by stressing the political values secured to all under
justice as fairness, Rawls seems to hope that Political Liberalism will persuade people of various religious and moral convictions that they can agree that it is a reasonable set of political ideas within which they can work together. Finally ( 5) stability based in this way on an overlapping
consensus of reasonable comprehensive doctrines is supposed to be enhanced by the idea of public reason: that is, the idea that we are (voluntarily) to restrict our arguments, on matters affecting the basic structure of society, to political arguments drawn from within the limits of commo n reason and our working political conception of justice. In
other words, barring emergencies, we are not to appeal to our particular comprehensive religious and philosophical doctrines in public debates about the most fundamental political institutions. 13 Such restraint s ould reduce divisiveness on the basic political institutio ns and so contnbute further to stability.
�
The key ideas here may be explained briefly as follows. First, presenting justice as fairness as merely a political conception means at least three things: that its main ideas are drawn from the public political culture; that it is only a theory about the basic structure of a con stit tional . . em ratic regime; and that it is viewed as a "self-sta�d ng. tdea , � oc
�
?
mdependent of commitments to comprehensive religJOUS and philosophical doctrines.,. Insofar as justice as fairness is a form of 0 1 994 Univenity of Southern California
173
PACIFIC PHILOSOPHICAL QUARTERLY
340
"political constructivism," this means also that its claims are not presented as "true" but only as, in a sense, "reasonable. " 1 5 Second,
reasonable people
are conceived as those who are willing to reciprocate on fair terms of cooperation when others will, and who admit certain limits on our powers of judgment. They accept only
reasonable comprehensive doctrines, which
are more or less consistent, coherent doctrines, arrived at by use of theoretical and practical reason, covering "major religious, philosophical, and moral aspects of human life," and normally belonging to a tradition
overlapping consensus, as opposed to a modus vivendi, when a political conception (such as justice as
of thought. 1 6 Third, there is an
mere
fairness) gains the support of "all the reasonable opposing religious,
philosophical, and moral doctrines likely to persist over generations and
to gain a sizable body of adherents in a more or less just constitutional regime."l7
Ill.
Doubts about the solution to the problem, as initially conceived
This initial reading of the problem and the solution at least shows why changes were necessary in the earlier account of stability . Even if Part III of A Theory of Justice successfully established the likely stability of a society in which justice as fairness became the "comprehensive moral doctrine" of virtually everyone, this would be irrelevant to our practical
concerns because contemporary experience suggests that it is very . unlikely, for the foreseeable future, that any one moral, religious, or
philosophical doctrine will become accepted by virtually everyone. As
Rawls says, freedom of thought and discussion fosters a plurality of
doc
�es; and a somewhat stifled plurality seems to persist even in those
cou?tnes where freedom is suppressed. If a just society can be stable at
all, It must be so despite its containing a plurality of diverse doctrines.
��er, we �n readily grant that presenting justice as fairness as a polih� conception, rather than as a comprehensive doctrine, is at least a step lD the right direction towards a solution. Why? The task is to shoW how a society that is well ordered by justice as fairness can be stable, and ori y for w I ta e hat to mean that a society in which the substantial maj ?� � aSIC of Citizens Willingly accept and use justice b e th ng as fairness (regardi ve structw:) can be self-restoring and so ha y the enduring. But, given that ) many diverse comprehensive doctrines st o m , how could the citizens (al IS agree to accept justice as fairnes s es fairn s? Only ' it seem s if J'ustice as rendered . · comprehensive doctrines ; and presen�g mpati'bl e With · theu � . strce as lllmes as a political tng conception is a major step towar s � . compatible. S1nce tt as tice does not claim to be "true " the new JUs . s rin flllmess cannot, strictly speaking, contradict compr hensive doct e as
�
�
�
��
·
'
�
;
C 1 994 Uaivmity of Southern Calif ornia
1 74
�
�
THE STABILITY PROBLEM IN POLITICAL LIBERALISM
341
to what is truly just, truly the nature of persons, etc. Since its scope is restricted to the basic structure of society, the new justice as fairness does not conflict with the implications of comprehensive doctrines about the more specific political issues. Since it draws its ideas from he public political culture, the new justice as fairness does not require antecedent acceptance, or even understanding, of any comprehensive religion or philosophical doctrine. It should also be noted that the argument that there might be an overlapping consensus (of reasonable comprehensive doctrines) on justice as fairness is not merely that, when conceived politically, the view is so modest that it will not offend adherents of the diverse comprehensive doctrines. Importantly, the argument is also that justice as fairness secures to everyone, despite their doctrinal differences, certain basic liberties and an adequate floor of all-purpose means; moreover, it fosters civility and a variety of other political values.18 Many of these advantages have already been argued for in A Theory ofJustice, and the move to presenting justice as fairness as "political," as far as I can see, does not undermine those arguments. So far, so good. But, now, if the worry about stability is the practical concern that I have been assuming, is not enough to show that it is remotely "possible" that there could be an overlapping consensus (or reasonable comprehensive doctrines) on justice as fairness. If such a consensus is now seen as the stabilizing force of a just society, one wants some assurance that such consensus would be likely to develop and endure. If the point of looking for arguments for stability is to see, before we attempt reforms, whether the reforms would be lasting enough to be worth the effort, then a bare possibility is small comfort. Is it at all likely that justice as fairness will win an overlapping consensus of reasonable comprehensive doctrines in our pluralistic world, achieving the stability of a just society in that way? On this issue, it is hard not to be skeptical. And, if we continue to see the main point of Pol�t�cal Liberalism as an attempt to assuage the practical concern about stabdtty, then its arguments, I suspect, are bound to seem woefully inadequate. Why be skeptical here? A main reason, I suppose, is our experience of persistent conflict of ideas in our large heterogeneous democracies. Fortunately, we have somehow developed in the United States a considerable consensus on our constitution and associated political procedures, but when we raise questions of interpretation and philosophical justification the consensus seems to evaporate. Most people :ue not philosophers, and justice as fairness, even conceived P?litic�ly, . ctton ts a subtle and complex system of philosophical id�. Even the dts� . between a political conception and a comprehenstve moral doctnne ts a subtlety that may well escape the average citizen. (Think, for examp�e: of explaining to the larger public the distinction between polittcal C 1994 University of Southern California
1 75
342
PACIFIC PHILOSOPHICAL QUARTERLY
constructivism, which says "principles of justice may be represented as the principles agreed upon by free and . equal citizens," and Kantian constructivism, which says that the principles of justice are constituted by
the joint will of rational autonomous persons. 19) The core tenets of political liberalism, I suppose, are more readily understood and accepted than fully specified versions (such as justice as fairness), but I suspect that the same forces that generate divergence of moral and religious comprehensive doctrines will block consensus on the less comprehensive, but still philosophical, political ideas of liberalism . Moreover, the arguments of Political Liberalism seem quite inadequate, and of the wrong kind, to solve the stability problem as I have presented it. What makes for stability in society is largely an empirical question;
and Political Liberalism does not purport to offer empirical evidence that an overlapping consensus would be likely.20 One would expect, too, that to show there is a good chance for such a consensus, one would need to investigate the fundamental premises of the "reasonable comprehensive doctrines" with which we are acquainted and then to argue from within such systems that they have grounds to support justice as fairness as a political framework. But Rawls deliberately sets aside this approach. Many, if not most, people, in our society, I suspect, do not have any
effective commitment to a comprehensive moral, religious, or political doctrine. Perhaps a majority can name a religious affiliation, but this does not mean that they really understand and use the doctrines with which they associate themselves. Many people seem to be doctrineless ethical pluralists, with diverse opinions on particular matters but no "theory." If so, then even winning the allegiance of the major religions and pbil �ophical theories (for justice as fairness) would still not ensure � stability; the more or less doctrineless folk need to be convinced as well, and they are already averse to philosophical systems of ideas. Again, if a practical concern for durability and stabilizing forces is the focus of concern, then Political Liberalism seems to rely too heavily on appeals to rational judgment. The factors which stabilize various societies may in fact have relatively little to do with the systems of ideas that they espouse, and more to do with habit, reinforcement, and blind emotional attachmentS.21 These factors may also be highly contextual, and s� not transferable from one historical situation to another. Such nonrauonal factors are perhaps m ore prominent in tyrannical schemes that atternpt . ot a to ma_mpulate the citizens, but this does not mean that they are n force m democracies which contribute n pta ce. to citizens' "willing" acce rn Finally, if we understand the stability problem as the practical conce or that I have ssumed so far, there t is a danger that Rawls's argumen � an ?verlappmg nsensus will oWU18 be seen as nothing more than the foll � and cancature . Imagme representatives of all the traditional religions other (reasonable) comprehensive doctrines meeting, together with
�
C 1994
the
Um-.ity of Soutbcrn California
1 76
THE STABILITY PROBLEM IN POLITICAL LIBERALISM
343
many reasonable doctrineless folic. They had been assigned to read A Theory ofJustice, but, for one reason or another, they were not uniformly
ready to accept justice as fairness. Some, we imagine, had not understood it; others thought it denied features of their comprehensive doctrines; still
others refused to endorse any public political conception, for although they believed their own doctnne best for deciding issues about the basic structure, they reasonably acknowledged that no such doctrine should be
forced on an unwilling people. Now imagine that an enthusiastic (but misguided) Rawls student were to · try to sell the others on Political
Liberalism. "We need a common, working con�ption of justice for basic political question s," he says, "and it is unreasonable for any.of us to foist our comprehensive doctrines on anyone else, particularly since that would
sanction a public denial of the truth of all the other (reasonable) doctrines.
Much as we would each like to have our own doctrines control the issues
about the basic structure, we cannot reasonably insist on that." "So," he continues,
"I have a proposal: we Rawlsian liberals will downgrade our
comprehensiYe doctrine to a political one, shift our claims for it from 'true' to 'reasonable', forego terms not in the public culture, and restrict its use to the issues about the basic structure of society. Then the rest of
you, for your. part, simply agree to use this political conception as the standard that determines real issues about the basic structure. And so, when you all consept to this, we will at last have stability for a society that is just (by the standards we Rawlsians initially advocated)." This bargain, we can suppose, would not be very persuasive, for it asks
nonliberal groups to swap their share in the power to control basic decisio ns for mere modesty in the presentation of liberal doctrine. Careful readers will, of course, realiZe that this caricature of Rawls's liberal
proposal is not fair. He does not, for example, offer the new political conception of justice as fairness as a bargaining chip to entice support from those who would not accept A Theory of Justice . The desired
conse nsus is meant to be based o� the whole range of reasons offered for justice as fairness in both his books (the reasoning from the original po sition, the security ofbasic goods, the political values, etc.), not merely on its limited scope and willingness to forego claims to "truth." At best such a consensus, Rawls argued, would not be achieved in an explicit bargain but would develop gradually from a prior modus vivendi, as somewhat indeterminate comprehensive doctrines come to accommodate
themselves so they can lend their support to what they can see to be a fair and tolerant political framework. 22 Nonetheless, despite the distortions, the skepticism expressed in the caricature will, I fear, be widely shared, so
long as the stability issue is understood along the lines of my initial reading.
C 199-4 University of Southern California
1 77
344
PACIFIC PHILOSOPHICAL QUARTERLY
IV. Is there
a
better reading of the project!
Suppose that my skepticism is correct, that is, (a) it is very unlikely that there
will
be
ever
an
be
overlapping
consensus
of
reasonable
comprehensive doctrines on justice as fairness (or political liberalism),
and (b) the stability of (more or less) just societies, if ever achieved, is
likely to rest more on nonrational factors, relative to context, and empirically discoverable than on the prospect of philosophical arguments convincing a substantial majority of reasonable citizens. Is the main project of Political
Liberalism, therefore, a failure, and its down-sizing of
justice as fairness a waste of time? Was there no point in arguing for at least the possibility of a just and stable society?
If we reconceive the nature of the problem that led Rawls to reconceive
justice as fairness, I think we get a more attractive picture. Let us suppose that the concern with stability was not the practical concern I described
earlier. (Will a just structure unravel? Will it be too short-lived to be worth
instituting?) Perhaps establishing the possibility of stability "of the right
kind" is to be seen as a necessary move in a project of justifying the advocacy and use ofjustice as fairness as a standard . Now, however, what prompts the need for justification is not a fear that a just structure, as defined by Rawls's principles, would be short-lived but the suspicion that it would unfairly coerce reasonable people who hold other views 'about
justice and morality generally . Asking whether it is possible for reasonable people, committed to diverse doctrines, to form an overlapping consensus on justice as fairness is a way of checking to see
if there
are sufficiently
good reasons for making justice as fairness the determining standard regarding the basic structure, reasons which one could sincerely defend to others without denying their deepest religious and philosophical commitments. If one can make a cogent case that there are adequate re�ons for diverse reasonable people to join a consensus on justice. as fairness as their working political idea then certain liberal and Kantian conditions for the legitimate exercise o power over others have been met. If the case can be made, for example, then the maxim to use ju stice as . fairness as a policy would be one that in a sense we can will as a universal law. The argument, if successful, wo d show t at one can will the f�r everyone as reasonably acceptable from everyone's perspective, while self; still respecting most others publicly as no less reasonable than one e s or the (new, political) arguments that they should adopt justice as fairn � 10 n? way pre uppose that people should abandon their religion, tbei � �pecial. �nception of the good, etc., provided that they are "reasonable cy 10 a tnmiDl l sense. On this reading, the project is to show that advo� � n, o i d use of oerc JUStice as fairness, even though inevitably warranting c � to IS reasonable and respectful JD e of those whose use of rea son leads th
f
�
�
h
poli�Y
!
0 1 994 UniYCnity of Southern Calif
ornia
178
THE STABILITY PROBLEM IN POLITICAL LIBERALISM
345
disagree with us deeply. For this purpose it is unimportant whether in fact a majority of (reasonable) people will ever, in fact, come to agree on justice as fairness. Showing the possibility of stability of the right kind, on this view, would be analogous to the familiar Rawlsian strategy of justifying principles by hypothetical consent; here, though, the argument would be to allay doubts that the principles are worthy of actual consent by showing that they could, hypothetically, win (almost) universal agreement if everyone would consider them reasonably. Several passages suggest that the point of conceiving justice as fairness as political and arguing that it can win an overlapping consensus has less to do with practical concerns about the durability of a just society than with answering the questions, "What would be the most reasonable basic of social union?" and "What is, for liberals, a legitimate exercise of political power?" In the fourth chapter of Political Liberalism, for example, this new reading seems confirmed, at least as part of Rawls's understanding of his project and the need for changes. Here he introduces the "liberal principle of legitimacy", which says, [O]ur exercise of political power is fully proper only when it is exercised in accordance with a constitution the essentials of which all citizens as free and equal may reasonably be
expected to endorse in the light of principles and ideals acceptable to their common human reason.n
This is a constraint on justification which commitment to liberalism itself imposes. Two important assumptions are that (a) so far as we are entitled to assert for political purposes, human reason does not give us grounds to rule out the major traditional religions and philosophical doctrines as "unreasonable"2• and (b) the political issues that arise about the basic structure of society are such that, however these are resolved, the result will be a coercive use of political power. Given these assumptions, arguing that a political conception of justice can win an overlapping consensus of reasonable diverse doctrines is necessary to establish, from a liberal point of view, the legitimacy of using it. If there is no actual overlapping consensus, a working justice as fairness would lack the fullest justification and the deepest basis of social union, 2s but if such a consensus is not even possible, in the sense I have been discussing, then decisions based on justice as fairness would have no legitimacy at all . 26 If my current understanding is right, then, the stability problem Rawls addresses is not, or at least need not be taken to be, whether it is likely that a society based on justice as fairness will endure. Nor is the issue even whether it is likely that such a society will in fact win willing acceptance by all the reasonable adherents of all the reasonable comprehensive doctrines. The latter, surely, is an ideal for liberals who follow Rawls, but it is not necessary. Liberals, like anyone else, will C 1994 University of Southern California
1 79
346
PACIFIC PHILOSOPHICAL QUARTERLY
naturally want to believe that success in achieving lasting liberal reforms is likely enough to make it reasonable to work towards such reforms; but the point of Rawls's arguments that the appropriate overlapping consensus on justice as fairness is possible is not, or is not primarily, to provide assurance on that point. The primary aim, instead, is to give a
defense ofjustice as fairness against the charge that (even if had we power to do so) to use it as a standard for (inevitably) coercive political decisions
would violate the liberal principle of legitirnacy. That principle says, "our exercise of political power is fully proper only when it is exercised in
accordance with a constitution of essentials of which all citizens as free
and equal may be reasonably expected to endorse in the light of principles and ideals acceptable to their common human reason . " 21 Given that use of
"common human reason" will continue to leave us with diverse religions
and philosophies ("the fact of reasonable pluralism"), liberals are
constrained by their own principle of legitimacy not to use justice as fairness, or any other standard, in exercising political power unless they are convinced that there are sufficient reasons to defend justice as fairness to those who are reasonable but who reasonably disagree about the "true" answers to the deepest questions of religion and morality. To show that there are such good and sufficient reasons one needs (for example) to show, without challenging the truth or reasonableness of Catholicism, utilitarianism, etc., that adherents of those doctrines have adequate grounds to accept and use justice as fairness as a practical framework if limited in the ways implied by calling it a mere ''political conception."28 To do this, it is not necessary to prove that all or most C�tholics, utilitarians, etc., will, or even probably will, endorse justice as fcumess someday. Nor need one show that if there were such an overlapping consensus it would in fact last.29 What is needed is, first, a thoroughly articulated proposal of what it would be to view justice as fairness as a political conception and, second, a cogent statement of strong reasons why Catholics, utilitarians, etc., without abandoning thei r faith, could reasonably endorse this proposal. Rawls tries to provide both of . these lD Political Liberalism. But whether, in the end, his proposal and arguments are successful in meeting the concerns for Kantian justification, liberal legitimacy, and a reasonable basis for social union is a matter that ' will require further discussion.
V. Some related interpretative questions
?
Rawls's rich ut complex discussion re 1 raises 111a ny questions, but. he want to menbon two, rather prelimina on. ry questions of interpretab (1) How much of A Theory of Justice I wonder is now meant to be . ' mcluded in "justice as fairness" now that this is pre nted as a " political C 1994 Univenity
�
of Southern California
1 80
347
THE STABILITY PROBLEM IN POLITICAL LIBERALISM
conception" rather than as a "comprehensive doctrine"? Clearly parts of Part III of A Theory of Justice are meant to be set aside.3o No doubt the
suggestions that the framework might be extended into a general theory of the right, and also of virtues, are not part of the political conception (even though they may remain fruitful suggestions for moral philosophy).31 Also it is clear that explicit changes. e.g. in the principle of liberty and the account of primary goods, supersede the account of A Theory of Justice.32 But how much of the discussion of methodology, the alternative ways an Original Position might be defined, and the arguments for features of the Original Position are still included? Does commitment to a theory of justice as political imply acceptance of the method of reflective equilibrium? It is clear that as a political conception, justice as fairness includes the idea that principles of justice "may be representetf' as the outcome of a procedure of construction but not as in fact "made or constituted" by the agreement of free and equal persons, or their hypothetical representatives;33 but are the details of the Original Position, the previous arguments for features of the Original Position, and the arguments that members of the Original Position would accept the principles, part of the political conception? The reason for asking is this. To preserve as much of the force and substance of A Theory of Justice as possible, it would seem that Rawls should want to include in the "module" that is meant to be the focus of the overlapping consensus as many of these ideas and arguments from A Theory of Justice as is compatible with the move to a "political conception."34 This seems to be Rawls's intent.35 But then the more substantial, complex, philosophical, and controversial the political conception of justice as fairness is, the less plausibility there seems in the suggestion that it is a relatively simple, practical framework that can facilitate public discussion among people of diverse backgrounds and faiths. Also it will be more difficult even to show that all reasonable comprehensive doctrines can endorse justice as fairness, i.e. have adequate reasons, compatible with their doctrines, to do so. (2) The background for my second question is this. Justice as fairness, as a political conception, is just one fonn of po/itica/ liberalism;36 there are many liberalisms, presumably even many (possible) poli ical liberalisms.37 Political liberalisms need to identify " fundamental questiOns for which the conception's political values yield reasonable answers."38 These are the "constitutional essentials," which include principles regarding the structure of government and political processes and equal basic rights and liberties, but not the difference principle or "fair equality of opportunity" (as described in A Theory of Justice) . (Unless the difference principle appears as a guideline in a statute, for example, the idea of public reason does not allow the Supreme Court to appeal to it. 39) The general principle of legitimacy and the idea of public reason are
�
Cl 1994 Univenity of Southern California
181
348
PACIFIC PHILOSOPHICAL QUARTERLY
rights, essential to liberalisms; and their content must include basic and these, liberties, and opportunities, an assigned priority among "measures assuring to all citizens adequate all-purpose means to make
effective use of their liberties and opportunit ies."40 There is, then, not only a distinction between the old and new versions of justice as fairness but also a distinction between the latter (with the difference principle, fair equality of opportunity, and associated
arguments) and those minimal aspects of it that are needed for it to be a form of liberalism. We might call the latter "Rawls's basic political liberalism." Even this, the new political justice as fairness stripped of details inessential to liberalism, is apparently distinct from other (possible) forms of political liberalism.
Now Political Liberalism answers the basic question, "How is it possible that there may exist over time a stable and just society of free and equal citizens profoundly divided by reasonable though incompati ble religious,
philosophical, and moral doctrines?", largely by arguing that a liberal conception of justice can win an overlapping consensus of reasonable
In this extended argument are we to think of the (new) political "a theory ofjustice " (with the difference principle, etc. ) as winning the overlapping consensus, or are we to think of ( what I called) "Rawls's basic political liberalism " as achieving the consensus?
comprehensive doctrines.41
I raise this question, like the previous one, because it seems to me that the plausibility that a political conception of justice could become widely accepted, the focus of overlapping consensus of both adherents of reasonable comprehensive doctrines and other doctrindess folk, is reduced the more detailed, philosophical, complex, and controversial that political conception is. If the whole of the (new) political "a theory of justice" is meant to be the object of the consensus this naturally increases the difficulty of showing that there can be an ov rlapping consensus (�f r_easonable doctrine s) on it. If a stripped down conception ("R awls's basiC liberalism") is to be the object, there is more hope for showing that there are adequate reasons for a consensus but then doubts arise as to whether e�ough has n stabilized to call the scheme just. For example, a society f t extraord mary affluence and gross inequalities would fall far short o � nd a JUStice, I think, if it merely agreed to keep the basic liberties fixed still provided a minimum "safety net." 2 4 Rawls himself a pparently , but endorses he difference principle, which would oppose such injustice e of the question now arises whether aiven pluralism ' the liberal principl ' eo 1 . . eg�tunacy allows those in power, t. absent actual consensus, t o use i
�
�
�
�
C 1 994 Uai11enity of Southern Califo rnia
1 82
THE STABILITY PROBLEM IN POLITICAL LIBERALISM
VI.
Concluding remt�rks:
how much luzs
349
been lost!
Is justice as fairness now merely a political conception? Was it every really a comprehensive moral doctrine? What have we, the admirers of A Theory of Justice, lost in the redescription o f justice as fairness in Political Liberalism? There are two main points to keep in mind, I think, if one is inclined to mourn the loss of Rawls's earlier, apparently more ambitious and comprehensive, characterization of justice as fairness. First, many aspects of what Rawls now calls "the political conception" seem already implicit in A Theory of Justice, or compatible with it, and so the change may not be as radical as the introduction to Political Liberalism would lead one to believe. There is, to be sure, resolution of ambiguity on many points, and change on some; but A Theory of Justice was always limited in its focus, restrained in its metaphysical claims, and open in its use of ideas developed in modem Western political culture.43 It was rich with suggestions about an analogous theory of "right" and "virtue," but that the basic structure of society was its primary subject was clear from the outset. It asked us to engage in an thought experiment about choice under a "veil'' that called to mind Kant's abstract metaphysical "Ideas," but careful readers knew it was, even then, merely "a device o f representation." Compared to most other works in political philosophy of the time, it was bold, far ranging, and ambitious; but it appealed, repeatedly, to our contemporary intuitive understandings and judgments to generate a confirming "reflective equilibrium." That the current "political conception" is significantly different is undeniable; but many exaggerated accounts of the change, I think, are due to misunderstanding of the original account. Second, we should remember that the old unmodified A Theory of Justice is still on the table, as it were. To alter the metaphor, copies may still be used in the philosophy classroom, even if not, any more, in the Congress or the courts. My po!nt is not the trivial one that we have a right, according to Political Liberalism as well as in fact, to reject Rawls's second book and believe in the first . (Political liberalism, of course, tolerates and seeks support from much more radically distinct comprehensive doctrines than that espoused in A Theory of Justice.) The point, rather, is that Political Liberalism does not deny, in any wholesale way, that the principles and arguments of A Theory of Justice are true, important, and better grounded than competing "reasonable comprehensive" theories. There are, admittedly, some specific points in A Theory of Justice which Rawls now claims were unduly vague, needlessly ad hoc, and even inconsistent with its ot�er claims.44 Still, if, . as Rawls says, that earlier work expresses a (parttally) comprehensive doctrine, then surely it still counts, by his criteria, among the "reasonable" 0 1994 Univenity of Southern California
1 83
PACIFIC PffiLOSOPHICAL QUARTERLY
350
ones, the truth or falsity of which Rawls's political liberalism refuses to judge.
Political Liberalism does imply that, as with Catholicism and
utilitarianism, the truth or superior rationality of justice as fairness as a comprehensive doctrine should not be presupposed (or, normally, even
asserted)
in public debates and official decisions about fundamental
constitutional issues. Many will no doubt challenge this restriction, but it is important to see at least that Rawls's argument for this new restrictive proposal does not withdraw the claim that the old doctrine is supported by better reasons than are competing comprehensive theories of justice.
Also Rawls is still committed to the view that the old doctrine may be more profoundly grounded, in a sense, than the new political conception. Insofar as A
Theory of Justice proposed more comprehensive moral
ideas, a kind of partial K.antian liberalism, it remains a competitor for allegiance, and inspiration for further development, in the philosophical project that each person may have of trying to find, for him or her, the
most reasonable comprehensive moral/political theory. It may seem a bit
odd, but it is consistent, to treat the old justice as fairness as a reasonable
partially comprehensive moral doctrine that may, or may not, guide its advocates to join a overlapping consensus on justice as fairness as a restricted political idea.45 If it does, they may come to see Political
Liberalism not so much as rejecting or abandoning the old A Theory of Justice as partially cloning it, drawing from it the basic materials to build up a similar, but pared down, counterpart to be used in a different forum. University of North Carolina Chapel Hill
Notes o "Au� r
I pr changes, ar tho �oets �ntics�sesston ofthe American Philosophical ngs Association' Pacific Division, meeu
1
My �
ent� here, with only minor
c
se
esented at
an
s m April, 1994· In thinking about John Rawls' recent work I have been helped by discussion . number of people, including especially Eugene Mason, Henry West, M artin Carl Brandt, Andrews Reath, and a graduate student reading group at the . A.P.A . U of North Carolina. I am also grateful to Rawls for his response at the and for sending me some very helpful notes, prepared for his classes. d John Rawls, A '11u!ory of Justice Pre$$, 1971) an (Cam.bridoc ° ·· Harvard University Political Liberal'um (New York: Columbia 5hall I tes no er University Press, 1 993). In furth . abb 1 l'tVl&tc the former as TJ and the latter as PL. PL, pp. 291-371 .
��
�n, m�mty
�ton
PL, pp. 22-28, 3S, 7S. PL, p. 7 Sf, Hl6, 178f 1 86. ' PL, pp. 273-74. L, pp. 8�129, especially pp. 103 -4. L, p. XVI. A footnote follows
4
s
:;
.
noting some exceptions.
C 1994 UnAenity of Soutbem California
1 84
THE STABILITY PROBLEM IN POLffiCAL LIBERALISM
351
9 PL, p . xvi.
10 PL, p. xvii. II
12
TJ, p. 6.
PL, p. I S.
u From correspondence I understand that Rawls may introduce further qualifications on this restriction in subsequent writings. 14 PL, pp. 1 1-15. u PL, pp. 93-94. 1 6 PL, pp. 48-66, especially p. 59 . 17 II
19
20
PL, p. I S. PL. pp. 1 54-72. PL, pp. 90-107, 1 25-29, especially p. 93, 99, and 125.
Rawls offers a variety of considerations in defense of his view that such an appropriate
overlapping consensus is possible, noting both the advantages of justice as fairness to all
and the "looseness" in most comprehensive doctrines that might enable them to come to
accept iras a working political framework (PL, p. 160fJ). But he never pretends to offer empirical evidence that it is likely, and regarding even its possibility he realizes that, in the
end, we must wait to
see.
"Whether justice as fairness (or some similar view) can gain the
support of an overlapping consensus so defined is a speculative question. One can reach an
educated conjecture only by working it out and exhibiting the way it might be supported" see also PL, p. 167�8). 21 Rawls, of course, recognizes that such factors play a causal role in promoting de facto stability (see PL, p. 1 6 1 , for example); and, as I suggest later, his focus of concern is best understood as liberal justification of justice as fairness rather than assurance of durability of societies that accept it.
(PL, p. I S;
n 2.1
24
PL, pp. 1 58�8. PL, p. 1 37.
In his A.P .A. reply, Rawls reminded me and stressed the importance of the qualification
in the antecedent here. It is not that, as individuals or as philosophers, we must concede that all "reasonable comprehensive doctrines" are equally supported by good reasons. The point is that political liberalism does not permit one to assume, for purposes of deciding basic political questions, that comprehensive doctrines that meet its minimal tests for reasonableness arc unreasonable or less well grounded than one's own doctrine.
This is most explicit in the unpublished class notes, but see PL, pp. 1 34-72. Ibid, and p. 1 37fT. 27 PL, p. 1 37. (The italics are mine.) 21 Catholicism and utilitarianism are used here as familiar examples which, I suppose, most liberals would count as among the "reasonable comprehenaive doctrines" by Rawls's criteria, but it is not important for the main point to insist that either, or any other doctrine :u
ll
in particular in fact meets those criteria. 29 Liberals will naturally hope that both are true, i.e. that a lasting overlapping consensus with reasonable Catholics and utilitarians will someday come about and will prove durable. Also strong empirical evidence against a tendency towards such consensus, even under
favorable conditions, should make liberals worry that their "good and sufficient reasons" were, after all, weaker and less free from bias than they thought. But, as Rawls might note, we are now far from having such decisive evidence for or against the emergence of a
consensus; after all the relevant ideas of a "political conception" have only just recently been put on the table. Besides, if my reading is right, the primary concern regarding the possibility of overlapping consensus has not been the likelihood of de facto agreement but the adequacy of the reasons for it. 10 PL, p. xvi. 31 TJ, pp. 1 08-1 7, 433-39.
C 1994 Univenity of Southern California
185
352 n ll
l4
PACIFIC PHILOSOPHICAL QUARTERLY
PL, pp. 75f, 180-86, 290fT. PL, p. 93 and 99.
For the idea of the "module"
11
PL, p. xvi. 16 PL, p. xxix. l1 PL, p. 6, 226.
sec
PL, p. 1 2 .
• PL, p. 227.
l'l
40 41
41
PL, p. 237n. PL, p. 6 and 226. The basic question is stated at PL, p. xviii.
Much depends on how we interpret "adequate all purpose means" in the definition of
basic liberalism above. No doubt this is meant to be somewhere between what would be
wananted by the differenoe principle and what Ronald Reagan's advisors would count as a "safety net," but the further we move form the difference principle in economic justioe, I suspect, the harder it is to justify, or get reasonable agreement, on the strict priority of
liberties over economic issues.
The current political conoeption, however, is more severely restricted to such ideas . .. It is natural to suppose that anyone who accepts Political Libera lism must entirely
41
reject Part Ill of A Theory of Justice
changes, Part
Ill can be
as
an argument for stability. However, with
som:e
seen as making a reasonable argument for the proposition that if
a society adopted justioe as fairness as its comprehensive doctrine, then there are reasons to suppose it would generate strong stabilizing forces tending to maintain the citizen's willing acceptanoe of its principles. This, I take it, is somewhat more modest that how Rawls now
now sees the initial project of Part Ill, for it leaves open the possibility than, as Rawls believes likely, given the freedoms allowed in such a just society , moral and religious disagreements would (almost) inevitably develop. The modest thesis, then, does not resolve the problem that pluralism poses for the liberal, and so, from Rawls's perspective, the second
book would still be required. This is compatible, however, with the view that Justice as not an Fairness gave good grounds for the more modest thesis and that this thesis is insignificant one. Rawls's reply at the A.P.A. meetings treated this point briefly, perhaps as obvious, f an Sl}'lng that in principle the earlier book is still "on the shelr' and might be a part o overlapping consensus with other compreh ensive doctrines, "why not?"
4�
C 1994 Univmity of Southern
California
1 86
SAM UEL SCHEFFLER
M O R A L I N D E P E N D E N C E A N D T H E O R I G I N A L P O S IT I O N
(Received 27 February , 1 978)
In the first section of this paper I maintain that there is a serious conflict between John Rawls' account of the relation between personal identity and moral theory in 'The Independence of Moral Theory' , and his own argument against utilitarianism in A Theory of Justice . In the second section I explore
two ways of trying to resolve the conflict, and examine the implications of each of these possible resolutions for Rawls' work as a whole.
In 'Later Selves and Moral Principles' , Derek Parfit suggests that the plausibi lity of different moral theories may depend to some extent on the nature of personal identity . Parfit identifies a 'Simple View' and a 'Complex View' of personal identity . On the Complex View, such identity just consists in certain bodily and psychological continuities and connections (e .g., relations of me mory , intention , and character), 1 which continuities and connections can themselves hold to varying degrees. On the Simple View, personal identity is a further, independent fact beyond the holding of the relevant continuities.
Parfit maintains that if we adopt a Complex View, we acquire two beliefs as
corollaries. He says we come to believe that the identity of a person is in its nature a less deep fact than the Simple View takes it to be , and that it can
sometimes hold to reduced degrees. And he goes on to argue that, in view of th e corollarie s of the Complex View, moral principles that rely heavily on the separaten ess of persons will seem less weighty or important on the Complex Vie w than they do on the Simple View (he also suggests that they may be thought to have a different scope). For example, distrib utive principles that place constraints on maximization in the name o f fairness will be seen as less important than the Simple View takes them to be, and utilitarian principles favoring m aximization will be regarded as more plausible than they would be if a Sim ple View were correct. Ph ilosophical Studies 35 ( 1 979) 397 -403. 003 1 ·8 1 16/79/0354-03 97$00.70 Copyright © 1 9 79 by D. Reidel Publishing Co., Dordrecht, Holland, and Boston, U.S.A.
1 87
398
S A M U E L SC H E F F L E R
Thus Parfit's argument, described very generally , is that utilitarianism
seems more plausible on a Complex View of personal identity than it does on
a Simple View, and that what may loosely be called 'Kantian' normative theories seem more plausible on a Simple View than they do on a Complex View. As Parfit himself recognizes, however, this argument, if correct, in no way forecloses the possibility that Kantian theories m ay still b e more plausible
than utilitarian theories, even on a Complex View . Nevertheless, his argument raises the interesting question of whether Kantian normative theories are in fact compatible with a Complex View of personal identity .2
In the course of a general attempt to show "the independence of moral
theory", John Rawls comments on Parfit's paper. 3 Rawls wishes to defend the view that "the conclusions of the philosophy of mind regarding the
question of personal identity do not provide grounds for acceptin g one of the leading moral conceptions rather than another" .4 He begins by iden tifyin g the ways in which Kantian theories and utilitarian theories must each make
use of some criterion of personal identity . And he concedes that although both types of theory observe certain shared constraints on any adequate criterion of personal identity (which constraints include an agreement that "any criterion of personal identity is based ultimately on empirical regulari·
ties and connections" 5 ), nevertheless within those constraints "the utili·
tarian conception has less need for a criterion of identity than a Kantian view; or perhaps better, it can get by with a w eaker criterion of ident ity" . 6 In con· trast, "a Kantian view is more dependent on personal identitie s ; it relies, so to
speak, on a stronger criterion " . 7 Kantians "must conceive of identities as stretching over much longer intervals". 11 Rawls then interprets Parfit as maintaining that the "shifting and some· times short-term characte r of mental connections" 9 gives sup port to utili· tarianism , with its 'weaker ' criterion o f identity . In response , Rawls argues 1 0 and that the that "no degree of connectedne ss . . . is natural or fixed " , con· actual continuities that bind the lives of persons depend on what moral main· ceptions have been realize d in the societies in which they live . Thus, he ai s, if a society that fosters strong sible, intrapersonal identificatio ns is fea t be It IS no count against Kantianism that such identifications migh t no facts present in a society ordered by a different moral· conc eption . Since about the connecti ons that ocial· hold within lives are always "relat ive to the s _ be ly achieved moral concep tion" ' I I then so long as a moral conc ep tion can _ realized in society' there is no way that of lives facts about the con nec ted n ess
�
�
188
MORAL
I N D E PE N D E N C E
399
can give any reason to accept or reject that conception . Thus in response to the question whether Kantianism is compatible with a Complex View of personal identity , Rawls suggests that it is , provided it is possible to foster strong empirical connections within lives. Although I have reservations about the adequacy of Rawls' argument as a response to Parfit, I wish to focus here on a different feature of that argument . I f no one degree of connectedness i s natural, and if a 'Kantian society' and a 'utilitarian society' 1 2 are equally feasible, then Rawls' own argument against
utilitarianism in A Theory of Justice 1 3 appears to be undermined . There he
maintains that "utilitarianism does not take seriously the distinction between
persons" . 1 4 That it does not is alleged to be important in choosing prin ciples of justice that will regulate social institutions (and through them , the conduct of individuals), because "the correct regulative principle for any thing depends on the nature of that thing" . 1 5 Hence , he says, if we assume that "the plurality of distinct persons with separate systems of ends is an essential feature of human societies, we should not expect the principles ofsocial choice to be utilitarian" . 1 6 And indeed , he argues, utilitarianism would not be chosen by the parties in the original position , for they know that each of them has some rational plan of life , plus a long-term interest in striving to carry out that plan and in maintaining self-respect. And they are persuaded that utilitarianism does not sufficiently safeguard these interests . But if Rawls' argument in 'The Independence of Moral Theory' is correct, it does not seem that people in the original position can know that they have long-term life plans and interests, which they must know if this argument a�ainst utilit arianism is to succeed . For, ex hypothesi, people in the original p osition "do not know the particular circumstances of their own society " . 1 7
And so, if Rawls ' argument in response to Parfit is correct, they presumably don't know if their society has fostered strong, Kantian identifications in them , or only weak, utilitarian identifications. 1 11 They don't know if they are Kantian persons or utilitarian persons; they might be either. Moreover, they do know that they might be either, if that argument is correct . For the argument is based on the p urported general principle of social psychology that societies ordered by different moral conceptions will create different kinds of continuities and interests within persons. 1 9 And since people in the original position know the general laws of human psychology •20 they pre sumably know this law too. They know that "no degree of connectedness . . . is natural o r fixed ". They know that they may or may not have long-term
1 89
400
SAM UEL SCHE FFLER
interests in the fulfillment of a rational life plan. In A Theory of Justice, Rawls argues that utilitarian principles ignore the distinctions among people with long-term interests in the fulfillment of a rational plan of life (Kantian people). But in 'The Independence of Moral Theory', he concedes that there can be other kinds of people. It is not unnatural for a person not to be a Kantian person. One can just as easily lack a long-tenn interest in the ful fillment of a rational plan of life as have such an interest. So if people in the original position are to choose principles to regulate human conduct in society, and if "the correct regulative principle for any thing depends on the nature of that thing", and if people have no unique nature, how can a rational choice of principles in the original position ever be made? 2 1 II
I see two ways in which Rawls might try to resolve the conflict to which I have called attention (there may be other ways). The first involves resisting the idea that the connectedness of lives is altogether socially relative , and thus striving to retain the argument against utilitarianism while surrendering the argument for moral independence in its present fonn . The second involves generally conceding the relativity of intrapersonal connections, and consider ably modifying claims about what the argument against utilitarianism is sup· posed to show. The first attempt at resolution might proceed along the following lines. It certainly does appear to be the case that the degree of empirical connected ness within individual lives can vary from society to socie ty , depending at least in part on variations in the socially dominant instit utions, traditions, ideologies , mores, and so o� Jt would be hard to deny that the degree to which people are concerned with planning their own individual futures and reflecting on their own personal pasts is culture�ependent. It would be idle to pretend that there was no difference , in this connection, between growing up in Scarsdale and growing up in Shanghai. Nevertheless, it also appears to con· be the case that in every society we know of, certain kinds of e mpirical ast nectedness - of memory , intention, character, and le the like - hold to at rent s�me greater degree within most individual lives than they do among diffe ties h�es �is suggests that certain kinds of distinctive empirical continui : are Wtthin hves are resistant to the types of social variation with which we most familiar.
1 90
M O R A L IN D EPEN D E N C E
40 1
So long as some facts about the connectedness within lives appear to be independent of these familiar cultural variations, one might try to argue that these facts are relevant to the choice of a moral conception intended to re· gulate the conduct of people in the wide range of societies where the facts hold. Rawls might try in particular to show that these facts support the adoption of his two principles of justice in the original position over the adoption of utilitarianism. To do this, however, he would have to say some thing about the relation between the facts in question and the supposition that the p arties in the original position each know that they have some ratio nal plan of life. In particular, he would have to show that these facts render that supposition plausible rather than merely question-begging, and it is not clear how easy it would be to show this. For someone might argue that while the distinctive empirical connections within individual lives make it plausible to suppose that persons in a wide range of societies distinguish between the self and others in certain contexts, it remains an open question whether that distinction plays a central role in detennining the ends they pursue and the ways in which they allocate their energies . Indeed, one might argue that it is precisely at this point that the social variations alluded to by Rawls in 'The Independence of Moral Theory' become crucial . If that is right , then it might
plausibly be maintained that, despite the distinctive empirical connections
within lives, the argument in A Theory of Justice, with its assumption of in
dividual life plans, shows nothing more decisive than that Kantian people, preferring to live in a Kantian society , would choose Kantian principles of justice . The second imagined resolution would be appropriate if it proved impos sible to resist the idea that the connections within lives are substantially variable . Rawls might then retain the argument for moral independence , and con ced e both that the parties in the original position must be conceived as knowing that they have long-tenn plans (that they are Kantian people), and tha t this knowledge is, in view of the variability of intrapersonal connections, highly substantive and prejudicial of the parties' eventual choice of principles.
While he would then have to concede that an argument against utilitarianism that takes the fonn of showing that Kantian persons would choose Kantianism
over util itarianism is philosophically indecisive , he might nonetheless argue that it retains some interest . But the remaining interest of the original position construction would b e as a mod el highlighting certain basic features and presuppositions o f a n ideal
191
402
SAM UEL SCHEFFLER
of the person and of justice which has great intuitive appeal and coheres with many of our deepest convictions. And the remaining (not inconsiderable) interest of the argument against utilitarianism in the original position would be as an explanation of why a commitment to those basic features and pre· suppositions is incompatible with utilitarianism. The original position con· struction, on this view, would no longer be claimed to have independent justi· ficatory force. In other words, Rawls could retain that c onstruction , but only by conceding what many of his critics have urged on other grounds: "that the presumptions of the contract method Rawls employs are rather strong, and that the original position therefore offers less independent support to his con· elusions than at first appears".
22
This concession will seem less damaging if
one believes that "over the long term this book will achieve its permanent place in the literature of political theory because of the substantive doctrine that it develops so eloquently and persuasively " .
23
I do not of course know whether either o f the two kinds of resolution I
have imagined would be acceptable to Rawls , but I think that there is a genuine problem posed by the conflict between his two arguments, and that
some resolution is required .
University of California, Berkeley
NOTES
Parfi t ' s article is contained in : Philosophy and Personal Relations, ed . b y Alan Monte·
�o.re (Roupedg� a�d �egan Paul, London, 1 97 3 ) , pp. 1 3 7 - 1 6 9 . For Parfit 's precise deli· �JtJtns of continu ity and 'connectedness'," see pp. 1 3 9 - 140. . 1
ons of . �m grateful to Derek Parfit for correctin g some of my earlier misin terpretati h1s VIews on these topics. · , . ln 'Th� !' ndepe ndence of moral theory ', Proceedi s an d the Amen can ng Addres ses of !MT. Association XLVIII ( 1 975) , pp. 5 - 22 . Referred to herea fter as
!'h�;�,P:.���-
lMT, p. 1 9 . lMT, p. 1 9 . IMT, p. 1 9. IMT, p. 1 9. IMT, p. l 9. 10 IMT, p . 20. II IMT, p. 20. 12 IMT, p. 20.
5
' , ' '
d u Harvar University Press, Ca mbn·dge, 1 97 1 . Referred to hereafter as TJ.
1 92
M O R A L I N D E PEN D E N C E
14
IS
16
17 11
TJ, p. TJ, p. TJ, p . TJ, p .
403
27. 29. 29. 1 3 7.
Simply giving the parties this bit of information would of course appear to prejudice their choice of principles of justice. Toward the end of the paper, I discuss the impli cation of giving them this information even so. 1 9 In IMT, Rawls imagines a 'utili tarian society' as one in which the people are 'hedo
nistic and individualistic'. (p . 20) This is a bit odd, since it is Rawls himself who tells us in TJ that, despite what is usually though t, "utilitarianism is not individualistic, at least when arrived at by the m ore natural course of reflection, in that, by conllating all sys tems of desires, it applies to society the principle of choice for one man". (p. 29) If utili· tarianism is not individualistic, why would people in a society where utilitarianism was the socially achieved moral conception be individualistic? lD TJ, p. 1 3 7 . Rawls himself uses this kind o f argument t o rebut the claim that i t is inappropriate 31 for principles of justice to be at all contingent on the kinds of general facts about society that the parties in the original position are said to know. Denying them this knowledge, he says, "amounts to supp osing that the persons in the original position know nothing at all about themselves or their world . How, then , can they possibly make a decision ? " (TJ, p. 1 5 9 ) This is precisely the kind of problem I am calling attention to in connection with personal identit y. If intrapersonal connections vary from society to society, and if the partie s in the original position don't know what sort of people they are, how can they make any decision about principles of justice? But, on the other hand, how can they be given this knowledge without prejudicing their choice of principles? (The second half of this dilem ma supposedly does not arise in the case of the general knowledge of society that the parties are said to h ave, for that knowledge is alleged to be "true and suitably
general". { TJ, p. 1 6 0] ) T. Nagel , 'Rawls on justice', in : Reading Rawls, ed. by N . Daniels (Basic Books, New u York), p. 1 5 . ll
/bid, p. 1 5 .
1 93
RAWLS ON TH E I ND I VI DUA L AND TH E SOC IAL Wayne Proudfoot ABSTR ACT Three models suggested by Rawls ( 1 97 1 ) for conceiving the relation
between individual and society are described and critically evaluated . Special attention
is given to
Rawls's analogies of the problem of
mapping the moral sentiments with the problem of mapping linguistic competence
and
of
a social u n ion with participation
in a game.
Similarities are noted between the theory of j ustice as fairness and traditional religious conceptions.
Both aim to transcend particular
interests and both embody perfectionist ideals.
1 . Introduction
Th e contr act theory as developed by R awls ( 1 9 7 1 ) in h is analysis of the orig ina l position, the veil of ignorance, and the constraints which are p laced on ethical choice in that position is one which emphasizes the voluntary decision of the individual, u ncontami nated by altruism, envy or any interest in the choices or values of others. I t is an a ttempt to treat problems of social choice in terms of the choices of mutually disinterested individuals under certain carefu lly specified conditions. The a i m of the theory is to state questions of social ethics in such a way t�at th ey can be resolved into questions of prudential choice for the self-i nt erested i nd iv i du a l . It is to substitute for an ethical judgment a judgment of rati onal p rudence ( 44) � . Th e c h aracte rization of R awls's theory as one which resolves questions of socia l eth ics into questions of individual choice is not one with wh ich he would disagree. I t is the project whi cl:l is addressed by the tradition of the social JA E 212 1 1974), 107·1 28
195
108
PROUDFOOT
contract as well as by game theory and welfare econ om ics. The soc i a l contract tradition employs a myth which portrays the basic i nstitutions of society as the resul t of a contract between ration a l , autonomou s, self-i nterested and mutually consenting individuals. The theory of games consists of attempts to discover the best strategy for each of the several players in a game wh ich has been carefu lly specified according to rules of play and the i n i tial positions of the players. Rawls
announces that he is providing a version of the social contract theory, he draws heavily on the literature of game theory and welfare economics, and his choice of the word "fairness" to characterize h i s theory further suggests the game imagery. The basic model is that of institutions wh ich can be clearly resolved into autonomous, mutually d isinterested players for who m the i nitial positions and the rules of play can be carefu lly specified.
Recently the libetal tradition, emphasizing free and autonomous individuals, has come under critical review from several quarters for n ot possessing sufficient resources to account for the values of community. If the basic u n its of society are autonomous ind ividuals with no affective or soc i a l ties, how is it possible to account for, or to generate from such a basis, the values of com m u n ity ? The answer given in a number of v ariations by representatives of the l i be ra l tradition has been that individuals find it in their own self-interest to affirm certain comm unal values. That is, u pon rational reflection on thei r own i nterests, they voluntarily choose to enter into comm u n a l relations. Consequently, the rise of voluntary associations has been widely h ai led by l i beral theorists. The simJ:lle constituents with which l ibe ral theory begins are autonomous individuals. F rom these constitutents, the institutions of society are bu i l t up by mutual consent in the form of voluntary associations. There is reason to doubt, h owever, that basic communal ties can be accounted for or justified solely by refe rence to the voluntary choices of individuals. Similar attempts to red u ce langu age to a voluntaristic foundatio n, to ostensive definition and the v o l u n t ary esta blish ment of social convention, have led i nvestigators back to the soci a l basis for the autonomous individu al and for the possibi l ity of rational choice . The clearest statemen t of Rawl s's concern to account for the val ues of community on the basis of an individual istic conce ption a ppears i n h i s d iscussion of the Kantian i nterpretati on of justice as fairness. Th e prese n t essay maY be read as an extended comm ent on this statem ent. The essential idea is that we want to account for t h e social valu es, for th e intri nsic good of institution al comm u n i ty and associative activ ities , b y a conceptio n of i ustice that in its ns theoretical s i s is individua l istic. F o r rea so of clarit , among others, we do not want to rely on an undefined con cept of co m n ty, or to suppo se that society is an organic whole with a life of its d!sttnct from and superio r to that of all its members in their relat io ns ,· · · . · . Wtth one another• Th us th e contractu a l conceptton of the ong mal postt to n s rked ou t first. It is relatively simple and the problem of rational choi ce that tt poses is relatively precise. F rom this concept ion, however i n dividua l istic it . · might seem • we must · · eventua 11Y explam the value of comm umty . Ot h erW Ise the theo ry of justice cannot succeed. (264-2 5 ) . 6
�
�
ba
� � �
�
1 96
RAWLS ON TH E I N D IVI DUAL AND THE SOC I A L
1 09
At another point R awls states that the individual istic basis and the postulate of mutual disinterest i n the origi nal position are "meant to incorporate widely shared and yet weak conditions" ( 1 29). He wants to assume as l i ttle as possible for the foundation on which to build his theory. In this essay the notion that the individualistic basis with
its m u tually d isinterested individuals is a weak
condition will be q uestioned. Such a condition is weak only from the perspective of the trad ition in l iberal ph ilosophical ethics which begins with the isolated individual. F rom the perspective of descriptions of the genesis of affective and communal ties, the autonomous individual is a sophisticated and complex construction which presupposes an entire social and cu ltural fabric. 2. Three Models of the Relation Between the Individual and Society At least three m odels for conceiving the relation between individual and society are considered and discussed by R awls. The first is identified with classical utilitar ianism and is rejected, but serves as the primary foil over against which he establishes h is own position. The second is the model represented by justice as fairness, and is imag inatively portrayed in the description of the original position and in the constraints on ethical choice which are i m posed by that description. The third is the idea of a social union which is developed i n the third part of the book as an i nitial step toward the construction of a theory of the good, with particu lar emphasis on the values of community and on shared social end s. 2. 1. Utilitarianism The first m odel is
R awls's presentation of the position of classical
utilitarianism. H i s most tel ling argu ment against the utilitarian position is that it conflates the system of desires of all individu a ls and arrives at the good fo r a society by treating it as one large individua l choice. I t is a summing u p over the
fieI d
of
individual
desires.
U t i l itarianism
has often
been
described
as
individu a l i stic, but R awls argues convincingly that the classical u tilitarian
position does not take seriously the plurality and distinctness of i nd ividuals (2 7·29) . It applies to society the principle of choice for one man. R awls also observes that the notion of the ideal observer or the impartial sympathetic spect ator is closel y bound u p with this c lassical util itarian position. It is only fro m the perspe ctive of some such hypothetical sympathetic ideal person that th e vario us individu al interests can be summed over an entire society (27, 1 841 . The paradigm presented here, a n d rejected by Rawls, is one in which the
interests of society are considered as the i nterests of one person. P lu rality is ign ored, and the desires of individuals are conflated. The tension between indi vidu a l and society is resolved by subordi nating the individual to the social su m . The social order is conceived as a u nity. The principles of individual choice,
der ived from the experience of the self as a unity, are applied to society as a who le. R awls rightly rejects this position as bei ng unable to account for justice, ex cept perha ps by some administr ative decision that i t is desirable for the whole
1 97
PROUDFOOT
1 10
ss . B u t individua l to give i ndividua ls some m i n i m u m level of l iberty a n d happine sources or merely are They . ition pos persons do not enter i n to the theore tical di rections from which desires are drawn. the whole This position leads to a monism i n which the sum of u t i l ities over their desires society is the only relevant considera tion . Particu lar i n d i v iduals and value in or rights have no prima facie claim to justice apart from thei r possi ble s. A system social to specific not s i augmenting the total sum. Such a sum as a well as position this degenerate society of one member could be hand led by to priority given society of n members. There is no respect for persons wh i ch is have sts ethici the summing of accounts. As R awls ( 1 958) n otes, m ost util itarian the opposed slavery, but they have opposed it by su m m i ng up advant ages to
icient slaveholder and advantages to the slaves and j u dg i ng it to be an ineff ty. system which al lows for a lesser sum of total good than that of a fr ee socie to Rawls, on the other hand, would not a l l ow the be n efits to the slaveh older a in ir nfa u as zed figure i n such a l ist, because the institution of sl avery i s recogn i system in which l iberty takes precedence over a l l other goods. Rawls's own theory gives priority to l iberty . He says a t several points that his theory is susceptible of a Kantian interpretation . I t gives a h igh priority to respect for persons and to the treatm ent of each i n d i v idual as a n end in himself. A basic respect for the autonomy or freedom of the other is bu i l t into the foundation of the theory.
2.2. Justice as Fairness
The second paradigm is that which characterizes the origi n a l posi tio n . It has already been suggested that this is a p icture of an aggregate of ind ivi duals, y mutually disintere sted , conceived primarily a s w i l l . W h i l e not n ecessar il e f li egoistic, their interests are each of their own choosing. They have the i r own Y ghl la�s. They coexist on the same geographica l territory a n d they ha ve rou ong stmllar needs and interests so that mutu a l l y advantage ous coo pera tio n am
�
them is possi b le.
1 shall emphasize this aspect of the circu th at mstances of j ust i ce by assu m i ng � parties take no interest in one another's interest . . . . Thus, o ne can say, in bnef, that the circumsta nces of ested j ustice obtain mutually dis in ter
th
whenever persons PUt forward conflicting ges claims to the division of social adva nta under conditions of moderate scarcity ( 1 26·1 28 1 .
Here the tensio n between alitY, individual and society i s resolved i n fav or of plu r ce at of an aggregate of mutually pa disinterested individu a l s occu py i ng the sa m e s ·at · the same ttme. · s It IS resolved in favor of the plural wh i le givi ng u p a nY oc• . ' . a! · g n u mty wh tch mt'ght i · . The classical u t i l i ta r i an model and the ort 0btam . . ich h PQSitton as sketched by R aw ls prov ide paradigm s for two polar wa ys in w . tensto cw re n between the plurali ty of individuals and the u n i ty of soc i al stru mtght resolved. One resol ution litY· favors unity and the other fav ors p l u ra . . Tha s conception of th rson as e o ngmal positio n emphasizes a view of th e pe
th�
�
1 98
1 11
RAWLS ON THE I N DIVI DUAL AND THE SOCIAL
agent. R e ason functions primarily as a capacity for prudential calculus. It is used to calculate conseq uences. The ve i l of ignorance erases any differences in reasoning power and any particu l a r desires or affections. A consequence of the separation of the choosing w i l l from the affections and from specific knowledge of one's goals and plans is that the doctrine of the self wh i ch results is that of a solitary wil l . This is a view of the person which is widespread in contemporary treatments
of
eth ical
questions.
In
different
forms,
it
can
be
seen
in
existential ism, be h av iorism, in some forms of utilitarianism, and in the focus on intention and action i n contemporary analytical philosophy. The chief virtue of 2
man as pure unfettered w i l l is freedom or l i be rty.
If men and women are considered primarily as choosing individuals, the manne r in which they are brought together in order to form some com munal or social
orde r
is
through
voluntary
association.
Hence
the
social
contract.
Following the cl assical contractarians, Rawls describes the conditions of justice in such a way as to make clear that some sort of voluntary association is called for on the basis of rational prudence. A n u m be r of individuals are together on the same land at the same time, and resources are moderately scarce. The avai lable resources
are insufficient to satisfy everyone's individua l pleasures
without producing co nflict. Clearly some kind of association is indicated. R awls is obliged to construct a justification of the fundamental bonds of human community on volu ntary foundations. It is not clear that th is is an adequate justification. A case i n point i s that of self-respect. Rawls understands that self-respect is a primary good, that it is a necessary precond ition for the abi l i ty of an i n d iv i du a l ·to carry out any plan of l ife whatever. He a lso acknowledges that self-respect cannot be formed in isolation. One cannot have self-respect if he or she does not have the respect of anyone else. Here R awls
ack nowledges
that the individual istic basis of moral ity must be qualified. Th ough he sees that se lf-respect i s a condition of carrying out any l ife plan, or even of having a l i fe plan, R awls is obliged to try to root self-respect in volu ntary ch oice . He speaks of the duty of mutual respect. Now the reason why this duty would be acknowledged is that although the parties in the original position take no interest in each other's interests, they know that in society they need to be assured by the esteem of their associates. Their self-respect and their confidence in the value of their own system of
end s cannot withstand the indifference much less the contempt of others. Everyone benefits then from living in a society where the duty of mutual respect is honored . The cost to self-interest is minor in compariso n with the
support for the sense of one's own worth
( 338).
The treatment of mutual respect as a duty is a clear example of R awls' atte mpt to rest volunta rism f u ndament al ties on a volunta ry foundat ion. The in which the en ters at two points. The f irst is the model of the original position 3 es has a princip les of principl the of tion justifica The natura l duty are chosen . respect l vol untar istic mutua of ion descript the is foundation. The second point as a du ty wh ich is to be acknowledged as such and safegu arded by the
1 99
PROUDFOOT
1 12
institutions and obl igations of the society. The translation of mutual respect into terms of prudence and acknowledged duties is not convincing. R e spect is at least partly a matter of discovering a person o r an end to be admirable. I t cannot be accou nted for exclusively in voluntaristic terms. The description of the original position as one of a u tonomous and mutually disinterested choosers leads R awls to e m p hasize rational choice at the expense of affective response and the variety of i nterests which m otivate persons. Rawls suggests that such interests are included i n each i nd i v idua l 's conception of the good which is obscured by the ve i l of ignorance a n d does not enter into the choice of the principles of justice. B u t the ch ief p r i m ary good, a sine qua
non for
the pursuit of any conception of the good, is self-respect w h i ch is based on
mutual respect. The conditions for this primary good cannot be e x haustively described in terms of rational choice. The study of artifici al i ntell igence has produced i m pressive resu lts in the a b i l ity to simulate certai n cognitive processes w i th complex com pute r programming. A computer can evaluate relevant information and make choices that are rational in terms of its ove ra l l plan or progra m . Many of the choices put to individuals in the original position could be processed by such a computer. 4 But a computer could not ach ieve or understand the v a l ue of self-respect. The issue of self-respect is one that cannot be transl ated w i thout remainder into the language of pure choice and of rational p l a ns. Why does R awls set h i mself the task of accounting for such goods as self-respect on a theoretical basis that i s essenti a l ly individ u a l i stic? At various points he says that the condition of mutual d i si nterest i s an atte m pt to keep his assumptions weak and mimimal. T h e postulate o f mutual disinterest in t h e original position is made to i n sure that the principles of j u stice do not depend upon strong assum ptio ns. Reca l l that the original position is meant to incorporate w i d e l y shared a nd y e t weak
conditions . A conceptio n of j u stice should not presuppose, then, e xten sive ties e as of natural sentiment . At the basis of the theory one tries to assume as l ittl possible.
'
( 1 291 .
sh ared ? But how does one decide what assu m ptions are wea k , s i m p l e a nd wi dely arilY Rawls begins with the notion of an aggregate of persons, defi ned pri m , he IS choosers, with their own independe nt n t inte rests . G iven t h is star tin g poi fro s concerne d to keep the assumptio n ns weak and m i n i m a l . T h u s he ref rai b e aSIC assum ing altruistic ties of benevolence , envy or of a n y other sort. B ut th · · · d The Portrayal of th e perso n m the origin a l position remains unc nti CIZe assumpti ons are weak only if a noth er this basic portrayal is taken as g i ve n . F rom . perspective, they migh t be judge d to be very strong indeed .
�s
�
-
·
2.3. Th e Idea o f a Social Uni on
�
T e third paradigm is include ngru ence d u n de r Rawl s's di scuss ion of th e co . of JUSti ce and goodnes s, and of the proble m of stabi l ity. I t i s descr i bed as a good.
200
RAWLS ON THE I N DI V I DUAL AND THE SOCIAL
1 13
as an end in itself wh ich is a shared end. This paradigm is distinct both from the conflated application to the entire society of the principle of choice for one person and from the conception of society as an aggregate of mutually disinterested indiv iduals. The idea of a social u n ion is described in contrast to the idea of a private society. A private society is essentially the second model as realized in the actual world. It stems from a consideration of the conditions of the origin a l position as descriptive of a social order. Thus we are led to the notion of a p rivate society. I ts chief features are f irst that
the
persons
comprising
associations, have their
own
it,
whether
they are human
individuals or
p r ivate ends which are either compet ing or
independent, but not in any case complementary. And second, institutions are not thought to have any value i n themselves, the activity of engaging in them not being counted as a good but if anything as a burden. Thus each person assesses social arrangements solely as a means to his private aims. No one takes accou n t of the good o f others, or of what they possess; rather everyone prefers the most efficient scheme � hat gives him the largest share of assets
(521 1 .
Over against t h i s notion of private society, Rawls proposes his idea o f a social union. I t is one in which f i n a l ends are shared and communal i n stitutes a re valu ed. The social order of mankind is best seen by contrast with the conception of a private society. Thus human beings have in fact shared f i nal ends and they valu e their own common institu tions and activities as good i n themselves. We need one another as partners in ways of life that are engaged in for their own sake, and
the
successes
and
enjoyments of others are necessary for and
complementary to o u r own good
( 522 ·523) .
While Rawls speaks of shared ends, these ends might be shared for private reasons. The fact that persons need one another as partners, that they a l l have different and complementary abi l ities and skills, that they take pleasu re i n the successes a n d enjoyments of others, does not necessarily mean that we have left the realm of private ends. It may mean that there are certai n materials, in this case persons, wh ich are necessary i n order for us to reach our ends.
Rawls i nvo kes the i m age of the game, an image wh ich plays a major role in the notion of the justice as fairness. Fairness is interpreted as fai r play and theory is conceived apply as describa ble in terms of imperso nal rules which each. eq ually to a l l persons of position l itia n i the of ion descript and a precise n ion. Rawls no tes that social u of ristics characte the man forms of l ife possess he but hips, As exa m ples he l ists friends and ships, science a rt . family relation suggests tha all of these ca be dealt with by thinkin g through the simpler instance s of games he is pr posing hat ( 525) . It is i mportan t t o note here that _ tnterac ttons the an al ysis of soctal games can serve for the analysi s of comple x
�
�
�
�
201
�
PROUDFOOT
1 14
c re l ations, in fam ily and forms of l ife, including those i nvolved i n aestheti interaction, and in friendshi ps.
in playing a Rawls distinguis hes several sorts of ends that m ight be involved common the , game, the last and most important of which is "the shared end game" (525). This desire of all players that there should be a good p l ay of the if the common end can be realized only if a l l play fairly acco r d i n g to the rules, that sense he t have l l a ayers l p the if d n a sides are more or less evenly matched, of on descripti the of overtones they are playing wel l. This description h as strong kes ta e everyon the original position. When a game is p layed in this manner, the pleasure in the game itself. I t is a n e x a m pl e of a shared e n d . "A good play of all" of game is, so to speak , a collective achievement req u i r i ng the coo peratio n
(526) .
Rawls has offered a paradigm in wh ich the condition of m u tu a l disinterest is
transcended, not by some sort of be n evolence or a ltruism, but by common or shared ends. The essential thing is that there be a shared f i na l end and accepted ways of advancing it which allow for the public recognition of the attainments of everyone. When this end is achieved, a ll f i nd satisfaction i n the very same thing; and
this fact
together
with
the complementarity
individuals affirms the tie of commu nity
(526 1 .
of
the
good of
The burden o f this move is to avoid t h e resolution o f the tension between individual and society by collapsing a l l i nto the soc i a l u n ity , as i n util itarianism,
or by d isregard ing the u n ity and considering each indiv i d u a l as d i sinte rested in the interests of others, as in the description of the original positi on as rea lized in the private society. A sati sfy ing a n d fair p lay of the game beco mes the pa radigm for the sharing of social ends.
. Y is R aw I s e I a borates on th ·t s somewhat by declaring that a well ordered soctet . . iS ttself a social union. "I ndeed, it is a social union of social u n io n s " ( 527) . Th to is reminiscent of Josiah Royce's ( 1 9 1 4 ) notion of the h ighest good as "loya ltY t of loyalty," which was also an attempt to develop the v a l u e of co m mu nity ou r, h ad a n an essential ly voluntaristic notion of the i ndiv idu a l . R oyce , ho we ve · i of u n ·ty a d van tage tn that hts episte mological work on the notio n of a co mm · · · he · I nterpretation prov1ded an essentia l ly social p h i losoph ical stru ctu re on whiCh could bu ild. tu al The idea of sharing ends is itself a shared e n d . The co n d itio n of m u d.IS ·tnt rest ts o longer appl icable as a t , and the idea of a socia l u n i o n s e rves 1. 1 e socla paradigm . It I S a paradig m in which the relation of the indi vidu a l to th ith er a n whole is not dissolv ed at the expense of one or the othe r. I t is ne d ua ls undifferentiated u n ity nor an aggregate of i n d i v iduals. Rathe r' the i n div i n d wht c conceived as sharing certain ends or goa l s which bind the m togeth e r a . . th ir ts presu mab l Y enter mto their sense of themse l ve s as ind ivid u a l s . T h on the par digm , however, rests on the i m age of the game, and par ticu l a rl y The not1on of game wh tch · games. . ed wtth conte rs assoctat mpor ary the ory of . ·
�
�
·
·
h·rd
�
�r� d
·
·
202
RAWLS ON THE I N OI V I OUAL ANO THE SOCIAL
1 15
analogy between such games and comple x social interaction wil l be exam ined
more fully in the next secti on.
Three models of the relation between individua l and society have been descr ibed. The first is explicitl y rejected by R awls, the second is fundame ntal to his theory of justice as fairness, and the third is em ployed in his outline of a correspon ding theory of the good . Rawls is clearly aware of the issues involved in providing an adequate paradigm for conceivin g the relation between individu a ls and society. It is i m portant, howeve r, to lift these paradigms out of the vol u m e a n d to exam ine the ways in wh i ch they are interrelate d.
Of the three models, the first two are proposal s for understanding constr aints on the choice of principles in the initial situation. The third is qu ite differen t. It is an object of choice. The idea of a social un ion is a good which Rawls expects woul d be preferred by most persons. On the basis of a principle of
motiv ation which h e postu lates and labels the Aristotelian Principle, Rawls conclu des that persons choose to participate in cooperative e-ndeavors. Other thin gs equal, persons enjoy the exercise of th eir capacities, and this enjoyment increases with the realization of m o re complex capacities
( 426). Thus pe rsons
would more readily welcome a transformation of society as the result of com plex social interdependence and cooperation than as·the result of divine fiat or the benevolence of a despot. There is enjoyment to be had in the playing of a game, or in the participation i n a com munal effort, which will cause it to be chosen over less complex activities. The idea of a social u n ion belongs to the theory of the good. It describes a shared goal. It is in this manner that Rawls proposes to account for social and communal values. Rawls is not introduci ng a social model at the foundation of his theory. He . Is not provid ing a social analysis of the original position. He is introducing social activ ity as a good which one m ight expect to be chosen withi n the context of
ju stice as fairness. In this sense, shared social ends are contingent and not necessary to the theory. The idea of social union is considered in the discussion of the stab i l ity of t h e theory of justice and of the congruence between justice as
fai rne ss and a theory of the good. The Aristotelian Principle sugge sts that the the ory of justice as fairness w i l l lead to the creation of a society in which shared 5 ends and com m u n a l values wi l l be sought by i ndividuals. Rawls considers socia l relations both as descriptive of human life and as an ideal that would be chosen as part of the good of each individual. The descriptive point, that social l ife is a presupposition of think ing, acting and valuing, he considers to be trivial for the pur poses with wh ich he i s concerned
(522). Such a statement is also true of pure egoists and of persons who view
thei r rel ati ons instru menta l ly ; i t possesses no discriminatory power . The �cond pOint at wh ic h the social character of hu man life enters is i n the conception of the ideal of a soci al union. N eitt:ler of these considerations affects the basic model of the theory which is the descriptio n of the original position. The trivial ' character of hum an social ity is presupposed , but need not be considered. The val ue of comm u n ity and social cooperation must be justified on the basis of
203
PROUDFOOT
116
choices made by individu als after the p r i n c i p l e s h ave bee n chose n . Thus the original position remains i n divid u a l i st i c and assu m e s m utual d i s i nterest on the part of the parties involved.
3. Two Analogies
Two analogies employed by R awls at crucial poi n ts w i l l be discu ssed in this
section. The first is his analogy between the p r oble m of mapping the moral sentiments and the problem of mappi ng l i ngu istic competence. The second is his use of the theory of games to describe soc ial a n d com m u n al i nteraction.
3. 1 . Language and Morals
While some difficu lties may be presented by t h i s a n a l ogy ( for i n stance, the
linguist is attem pting to describe the speech of native speakers, wh i l e the ethicist may come to conclusions which suggest the reform of moral intuition). it i l lum ines the model of the original positi o n . Ju stice as fairness i s proposed as a theory of the moral sentiments which can be tom pared in its com plexity w ith the description of the sense of gramaticalness of a n ative speake r ( 4 7 ) .
I f the analogy between theories of l a nguage and m oral theo ries i s pressed ( beyond R awls's empl oyment of it) , it wou l d seem that the charac teri zation is the original position is simi l a r to a situat i on i n wh i ch gr ammatic al stru ctu re
of
accounted for by an hypothetical situation in w h i c h m u tu a l ly disinterested individuals are portrayed as com ing together i n order to agree on conventions and rules to form a language. Each has needs, i n ne r th oughts, p o i n ts t o e xp ress, and they need only agree on conventions by which to neg otia te and communicate. The model is a nominalistic o ne . I t i s r e m i n i scent of d i scussi ons ostensive definition and of conventions ex p l i ci tly agreed upon . We k now, course, th at such a situation never actu a l l y occurs, any more tha n the o ngm a I situation of choice which R awls describes cou l d actu al ly occu r . Both are hypotheti cal situatio ns imagined in such a way as to provi de app ro pri ate constraints in order to develop a theory of language o r morals. T h e issue is n ot su ch a whether such a situation m ight have occurred but the adeq u acy of .
���.
I
.
of of
�f
Accounts of the structure of langu age which i nvolv ed the an aly sis nts In language into simples , the ostensive def i n ition of th e sim p l e co m po ne bin ing terms of atomic facts i n the world, and the articulation of ru les fo r co m m ong these simples were given by the early work of R u ssell and W i tt ge nst e i n, a so cl·aI ot h ers. They were also x concerned to prov ide an accou nt of c om ple was relation s (of the gram mar of human l a ng u a ge ) on a fou n dati on which a nd s theoret_ically individ ualistic t . The simpl e bu i l d i n g blocks were ato m i c fa c names. Name s provided ucted the basic u n i ts out of wh ich langu ag e w as constr and they mirrore d individ were ual atomic facts i n the wo r l d . These p h i l oso ph ers heY al so concern ed to keep their assumpti ons weak, si m ple and wide l y s ha red. T . oke did not w nt t invoke nv i the fabric of language as R aw l s does not wa nt to any orgam c not ion of society .
�
�
204
RAWLS ON TH E I N D I V I DUAL AND THE SOCIAL
117
The names w h i c h m a ke up l a nguage were connected to the facts which make up the world and to each other by the i nv ocation of wi l l or choice. I t was through naming, throu g h ostensive defin ition and the choice of conventions, that words were l i n ke d with one another and with the world. There is here an analogy with the way in which R awls i nvokes choice or wi l l as the basis for association and co m m u n ity. I n orde r to keep their assumptions weak, the logical ato m i s t s
did
n ot
want
to
assume
connections
between
denotations,
connotations, and various levels of meaning an d use of language. They hoped eventual l y to e x p l a i n such connections on the basis of the simple paraphernalia with wh ich they began . F rom our present vantage point i t is possible to see that the assumptions of the logical atom ists were n e ither simple nor wea k , though they may h ave been
widely shared among m e m bers of the V ienna Cir �l e . To assume that language
and the wor l d are com posed of mutually independent sim ples is to make a strong assu m ption. M o st contemporary students of language woul d claim that the theory which resu l ts from such an assumption is unable to account adequately for l i nguistic competence. It is not clear that names and ostensive defin itions are pri mitive elem ents of w h i ch l a nguage is composed. They function in certain sectors of l i n g u istic u sage, but are narrowly restricted and rather soph isticated uses. They are not p r i m itive, and they shed relatively little l ight on the structure or acqu isition of l a nguage. S i m i larly, it m ight be the case that R awls's assu m ptio n of m utu a l ly disinterested i nd i vid u als in the original position of ch oice may be a strong, rather than a weak, assu mption . It might have the appearance of be i n g weak because it assumes that the i n itial situation is composed of si mples. It m ight rather be the case that the abi l ity to choose rests on a fab r i c of shared associations, affections, i n terests, s k i l ls, ru les and rituals ra the r than that such rituals rest on a foundation of individual choices.
Th e a n a l ogy betwee n the analy sis of language into simples which m irror the w orl d and out of wh i c h more complex construction s are bu i lt, and the analysis of com plex soci a l forms i n to contracts between i ndividuals who form Vt)luntary associatio n s is not a new one. B oth find an early and clear articulation in the
w or k of J oh n Locke . Locke, as e pistemologist, gave an account of language begin ning with the tabula rasa and the reception of simple impressions from the w orl d . Lo ck e , the l iberal theorist, provided the state ment of contractarian
th eory on wh i ch R aw l s draws. B oth were attempt s to account for complex social Phe no men a by a n a l y z i ng the m , without rema i nder , into the i r sim ple and m u tua l ly i nd e pe ndent constituents. M u ch of the later work of Wittgenste i n was concerne d to show that the atom ists h ad been m isgu i d e d i n the assumption that analysis of language into si m pl e co m ponents wou ld lead to fundame ntal and weak assum ption s on wh ich a the ory of lan guage m igh t be bu i lt. A certain vagueness , i m precision or i nabil ity to p rov i de ru l es m ay not be a defici ency but part of the structure of language. l i ng u i sti c co m pe te nce may not be vague or im precise at a l l , bu t may appear to be on ly w hen i m ported from measur ed agai nst some standar d which has been
205
PROUDFOOT
1 18
u sage. Often the outside l i ngu i stic usage or from some paroch i a l corner of that i s a n a lyzed i nto age u ng a l of use the ch i wh n i attem pt to prov ide clear models that use. about nctive sti i d is sim ple componen ts leads to the l oss of what obscured, but , e n i m u l l Wittgenstei n argued that many such d i st i nctions d i d not i the forms of l ife which are embedded i n gra m m a r .
or T h e poi nt h e r e i s not t h a t R awls subscr i bes to Locke's the ory of l a ngu age on of that he would countenanc e an atomistic or nom i na l i st i c i nte rpretati of problem the ng i r compa n I not. d l l i ngu i stic usage. I t is qu ite clear that he wou of problem e h t h t i w describing the sense of gra m matical ness of a native s pe a k e r ky's Choms to descr i bing and systematizing moral se ntiments, R awls a ppe a l s work on syntactical structures ( 47 n ) . G iven t h i s a n a l ogy with Chomsk y's attempt to prov ide a structu ral model that w i l l syst e m at i ze the i ntu i t i ons of the
in native speaker, the model offered by R awls is i n d i v i d u a l i stic a n d nom inal istic a manner which appears to have its cou n terpart i n e a r l i e r approa ches to the
study of language and not i n transform ational gra m m ar. T h e com parison is offered only in order to dramat i ze the i nd i v i d u a l istic charact e r of Rawls's portrayal of the i n itial situation. 3.2. Games
I n the opening l i nes of h i s consideration of the i d e a of a soci a l u n io n, Rawls question s whether, given the indivi d u a l i stic fe atu res of j u st i ce as f a i r ness, th e contract doctri ne is "a satisfactory framework for u n dersta n d i n g t h e valu es of . com munity , and for choosing among social a r ra ng e m e nts to rea l i ze the m" (5 20) He argues that a private society, in w h i ch i nterests are not shared , is not enta iled by the contr act view. R ather we are led to the i d e a of a soc i a l u n i o n . While m any forms of l ife possess the characteristics of soc i a l u n i o n , R aw l s sugg ests th at it
wil l be sufficien t to think through the s i m p l e r i nstance of g a m es ( 5 2 5 ) . A t this point the s i m i l arity betwee n the paradigm o f t h e origi n al position cia l. 1 and the analyses of game theory and of welfare eco n o m ics beco m es cru th at have argued that the original position is no m in a l i stic and i n d i v i d u a l i stic a nd w ls Ra i� shares these characteri stics with ts the theory of games. At seve ra l p o i n tes the 5 Ci Pr isoner's D ilemma a s a classic i nstance of a s i m p l e game ( n, 269 771· all I n the theory of games , the r u l es a re clearly stated the i n i t i a l p osit ion s fo r �layers are described , the , and players are assu med to m u t u a l l y d i s i n terested he •t rs possible to calculat e the most rational strategy for e a ch p layer. T . . Pnsone r's D ilem m a . . . ro m th e presents a case 1n wh1ch self-i ntereste d d ec1s ro ns f . m pomt of view of each lead to a situ ation i n which both a r e worse o ff. Th us, fro in . the perspective of the goo · ted . a d o f each , some sort of coopera t i o n ·IS 1 nd 1c orde r to achie ve st a b"l be can 1 ·lty. Th e parameter s are clear a n d t h e situ at io n . rat ional ly anal ysed in ord er t0 sh ow that coope ration i s the best stra tegy. . There ts, h owever, another u� sense of the word "game " i n con tem p or ary . es " wh 1 ch is less rest nc m · te d and forma ga ge ' l . Wittgenstein refers to " l a ngua . tical others spe ak of 9 ames . i l 0 P or th at c h aracter rze certai n perso n a l i nteracti on s . · of relat ion sh ips 1 n th · b IS roa d er sense of the word , "ga m e " can refer to a van etV
�
·
•
•
206
RAWLS ON THE I N DI VI DUAL A N D TH E SOCIAL
1 19
social rel ationshi ps. One can speak of the games engaged in by phi losophers, or by parents and chi ldren, or by men and women. I n each case we understand what is meant. The notion of game can be taken as a shorthand way of referring to patterns i n soci a l rel at ionsh i ps. B ut we would not say, in th is context, that game theory could be appl ied in each of these cases or that the Prisoner's Dilemma is a simplified representation of a l l games. It is one very specialized kind of game in the same way in wh ich ostensive definition is one very special ·use of l anguage. Neither can stand for games or l i nguistic usage as a whole. R awls seems to trade on the broader sense of the word "game" when he suggests that the idea of socia l u n ions such as families, friendships and other groups can be analyzed by t h i n k i ng t hrough the sim pler instance of games. Then he returns to the notion of game which is dom i nant thoughout the work, and to the theory of g a mes. The broader connotation is ordinarily used to designate social relationships, while the narrower one is i nd ividualistic. Th is sl i P.page in the meani ng of the word "game" aids h i m i n his attempt to account for social and com munal values on a foundation wh ich is theoretically individual istic. 6 I n the theory of games, each i ndividual rationally calculates the most pruden tial plan to adopt from h is or her position . The game consists primar i ly of a nu mber of decisions or choices. It requires of the players only rational calcul ation and cho i ces between clear a l ternatives. This is in contrast, for instance, to the games children play when they "pl ay house," or even to the game of basketball . 1 n the latter cases, creativity and i magination are requi red in order even to pose choices wh ich m ight be made. Decisions are conti n u a l ly
invo lved, but they are certai n ly not the salient aspects of the game of playing house. T he Prisoner's D i lemma, however, and sim i lar examples are the types of games wh i ch form the parad igms for the contemporary theory of games, employ ed exte nsively by welfare economists and by some social psychologists. By way of contrast, consider Wittgenstei n's d iscussion of games in the
Inves tiga tions. Wittgenste i n sees no possibi l ity of giving an exact defin ition of a
game, or of clear ly specifying the cond itions of one. It is in the attempt to articu l ate simi larities between games that he arrives at his notion of a family rese mbl a nce. There may be no feature which is common to a l l games, but they may rese mble one another as do members of a family . F o r how is the co n cept of a game bounded? What st i l l counts as a game and what no longer does? Ca n you give the bounda ry? No. You can draw one; for no n e has so far been drawn. ( Bu t that never troubled you before when you used t h e word "game ." )
·
·
play wi th It IS " B u t t h e n t h e use of the word is u n regulat ed, the 'game' we are unr eg u l ated ." -It is not everyw here ci rcumsc ribed by rules; but no m o re yet hard; how r o th ere an y r u les for how h igh one throws the ball i n tennis, 68). ten nis is a game for all that and has rules too ( Wittgen stein, 1 953: 1 •
§
Wi ttge nstein focuses on the differen c�s between games. We k now what games are, bu t they cannot feature s. H e suggest s that be specifie d i n terms of commo n
207
PROUD FOOT
1 20
although we m ight say that a l l games h ave r u l e s , s o m e of these r u les are only tacit, are not set out in any l i st of rules, a n d p e r h a ps cou l d not be so set out. Some games, language games, forms of l ife, d e p e n d u p o n r i t u a l action, tacit knowledge and patterns of activ ity that a r e n ot m atters of conscious attention. Such ru les as e x ist are neither e x p l icit, p u b l i c or f i x e d . The P r i so n e r ' s D i lemma would be no game at a l l i n the context of the theory o f games if t h e re were an added rule that the players cou l d m a k e u p r u l e s as they w e n t a long. Yet Wittgenste i n suggests that some games a l l ow, a n d e v e n d e m a nd, j u s t that. The point i s that the notion of game
is i m p r ec i se .
For h i s a n alysis of
decision, R awls d r aws on the game of game t h eory . E x p l i c i t p u b l i c ru les, m u tual di sinterest, and the poss i b i l ity of ra t i o n a l p r u d e n t i a l c a l c u l a t i o n as we l l as self-i nterested cooperation a r e marks of the theory o f j u stice as f a i r ness as wel l a s o f the Prisoner's D i l e m m a . W h e n he e m pl oy s t h e n o t i o n o f g a m e a s an example of a social union, howeve r, he i s a b l e to trade on a less precise notion of game and on the recent use by W i ttgenste i n a n d o t h e r s of the a n a l og y of game with various com p l ex forms of soc i a l aff i l i a t i o n and c oope r a t i on . I n the model represented b y the theo ry of games, each p l aye r i s a n isol ated agent. M utual d i s i nterest is a condition of t h e g a m e . B enevo l e nce an d e n v y are elim inated. E ach p layer is bou nd by no ties to the oth e r p l aye rs, e x cept those ties which are expl icitly articu l a ted in the pu b l i c r u l e s of the game. Wittge n ste i n's conception of game, howeve r , e x t e n d s to m a n y a reas of life in w h i ch there are no pu b l i c l y articu lated r u l e s . H i s we l l - k n own l i st near th e beginning of the Investiga tions suggests a v a r i e ty of l a n gu ag e g a m es, i n c lu ding
giving orders and obe y i n g the m , m a k i n g a j o k e a n d te l l i n g it, ask i ng, thanking,
cu rsing, greeting and pra y i n g ( W ittgen ste i n , 1 9 5 3 : 1 , § 23) . M a n y o f these are examples of forms of l ife and action w h i c h are g ov e r n e d by r i tua l s that a re taci t, not clearly bounded and seldom consc i o u s . S u ch a v iew of t h e v a r i e t y of ga m es suggests numerous associative bonds on the p e r s on a l , soci a l a n d cu l tu ral levels.
to act O e must know how to speak the l a nguage, what pray i ng i n v o l ves, how � _ arl y Wt h t n a particu lar ritual c o ntext cle . The k n ow l e dg e req u i red can not be � a to arttculated in a l i st of r u l e s or g strate g i e s . I t is a m atte r of be l o n g i n ial so com m u n i t , of learning a way � of l ife, or partic i p a t i n g i n a l i n g u i st i c an d � nstv e ord r wh tch provi des the ste O e s . gam conte x t for t h ese r i t u a l s a n d : _ efo re ef tm_ t t on , thou gh it appea rs to be pr i m i t i v e , presu p poses a soc i a l co n t e xt b n of tt can be u nderstoo d . W ittge nste i n 's notion of game d e m an d s a con ce pti o perso ns re lated to one another u als, by many a n d v a r i e d soc i a l t i ssu e s a n d rit while the Prisoner' s D i le m m a red by suggests a game which i s v o l u n tar i l y e nte agent s with the power of ratio n a l calcu lati o n . m es , I t i s i m possibl e to conceive o f a person sta n d i n g outs i d e o f al l ga . . . und e rstood in the b roa th e d er sense, a nd then v o l u n t a r t l y dect d ng to e n te r t ·s to real m of ga mes • T be come h u m a n , to d ev e l o p f ro m neonate to p e r so n, 1 · . mat u re with i n the re a 1 m o I t is s. f vart ou s game s, acti ons r u l e s a n d con ve n t t o n . • • er s QU ite conce ·tva bl e , though n , that a pe rson cou l d e x i st o u t s i d e of th e P riso . D i le m ma, and v o l u nta of r i ly e nter the re a l m of such a g a m e . T h e f i r st n ot io n
?
0
208
RAWLS ON tHE I N DI V I DUAL AND THE SOCIAL
121
game enters i nto the definition of a person , whi l e the second presupposes the existence of the person prior to the beginn ing of the game. The games and forms of l ife in which one is i nvolved p rovide the criteria for what a person is taken to be. I f R awls actu al ly em ployed the Wittgensteinian notion of game, it would
strongly qualify the i n d ividual istic bias that is present in h is description of the original position. H e descri bes that position in such a way, however, that it portrays i n d ividual agents who come together in order to cooperate to ach ieve ends which are sh a red, but wh i ch are stil l individual ends. Man remains an isolated i nd ividual agent. Wittgenstei n 's d i scussion of games and forms of l ife suggests that rather than viewing
social
institutions through the contractarian paradigm of sim ple
mutu a l ly d isi nterested i ndividuals choosing cooperative strategies, it i s i m portant to recognize that it is the theory which creates the i ndividual, and which furnishes the criteri a within wh i ch individuals are isolated as choosers, agents and w i lls. Then the individual agent is taken to be a fundamental simple entity out of wh i ch a contractarian notion of social justice can be bu ilt. The theory enters into a description of what an i ndividual is at least as much as the individual provides a starting point for the development of the theory. When W i ttgen stei n suggests that there are games i n which the pl ayers m ake up rul es or change the rules as they go along, he is poi nting to the element of
imagination, interpretation and creativity wh ich is involved i n l i ngu istic usage and social i nteraction. I n terpretation presupposes com munity, creativity and
improvisation, and precludes a f i n a l deductive solution. I nte l l igence and imagina tion m ay be involved in articu lating the maxi min strategy in welfare econ omics, but i t is the sol ution of a game where a l l the constraints are given.
The ration a l soluti on can be shown to be such, with i n the constraints of the gam e. Rawls u n derstands the theory of justice to be of this type, so that it has a rati onal so lution wh ich can , i n t i me, be translated into a purely deductive sol ution arrived at by prudential calcul ation, as the game of tic-tac-toe, or the Prisoner's D ilemm a , h as a solution wh i ch is final for each of the players, for each cond ition from w h i ch one starts.
O ne shou ld note a l so that the acceptance of these principles is not conj ectured as a psyc hological law or probab i l ity. Ideally anyway, I should like to show that their acknowledgemen t is the only choice consistent with the f u l l
description of t h e original posit i o n . T h e argument a i m s t o b e strictly deductive
(121).
This a i m demands a notion o f game or a description of the orig i n a l position Which can be precisely and finally articulated, exactly the notion of game and social i n teraction which is rejected by Wittgenstein and by others who have attempted to describe complex social activity. 4. The Righ t and the Good R aw ls is develop i ng a notion of justice as fairness, and we have seen that
209
PROUDFOOT
1 22
fairness is closely correlated with fair p l ay i n the th eory of games. While he
carefully sets the boundaries of h i s task, h e suggests at several points that the
theory of justice as fairness' can be e x panded i nto a theory of rightness as
fairness, and that the resu l ting ethical position is o n e i n wh ich the concept of right is prior to that of the good
( 1 7 , 1 1 1 , 1 96). I t w i l l be argued in the next
section that the model of the origi n a l position i ncludes e le ments of a conception
of the good, or of a particu lar perfe ct i o n i st ide a l . It is sometimes unclear how much scope Rawls is cla i m i n g for the theory. Justice as fairness is not a complete co n t ract theory. F o r i t i s clear that the contractarian idea can be extended to t h e choice of more or less an entire ethical system, that is, to a system including principles for all the virtues and not only for justice
( 1 7) .
The considerations adduced above suggest that t h e contract notion may be
adequate to a rather more narrowly conceived conception of justice and may describe the concern for fair play which is o n e aspect of the moral sense, but cannot serve as a paradigFJl for mapping the moral sentiments more generally. The analogy with theories of language may be h e l pful here. Ostensive defin ition and clearly articulated rules are characteristic of some subsystems of linguistic usage such as the coining of new terms i n a tech n i c a l d iscip l i n e or the agree men t upon rules in a formal game. Each of these su bsyste m s , howe ver, presupp oses a l larger realm of l inguistic usage. It should not be c l a i med that the hy poth etica er ith volu ntaristic paradigm is adequ ate for a comprehen sive acco u nt of e linguistic usage or ethical justification.
ns Certain situations, such as the d i stribu ti o n of an i nd ivis i b l e ben e fit by mea of a lottery, the establ ishment of laws that d o not d i scri m inate between individu als, and the fair opportu n i ty for everyone to enter e ach po siti o n in society are central to the theory arad ig ms and may be proper a n a l og u e s to the p cannot of game theory. But R awls has said that the theory of justice as fa i r ness If succeed if it canno t explain the value of commu n ity . Why is tliis th e case ? e r o Rawls were con cerned only to tice m prov ide an accou n t of the sen se of ju s val ue of narrowl y conceiv ed, why shoul d he be responsi b l e for an acco u n t of th e we h ave com m u nity? In the section on the idea of t h e soc i a l u n i o n he cla im s, as . soc• , seen, to u nderstand the value s of art, scie n ce , frien dsh i ps a n d oth e r sh a red ends by analogy with an spect 15 analysi s of games a n d of ration a l c h o i ce . Se lf- re also justi fied by appe al to rat iona l se lf-in te rest . as th eo rY Argum ents have been adduced e a r l i e r to show that R aw ls's presented i n the des . . fo r the · · On of the ongmal position c a n n o t acc ou nt cnptJ . w•th . vak.t es of co mmu n ·1 t Y an d that the t. de a of a soc i a l u n i o n IS deve 1 0ped . reference to the f ra in the mewo rk of the o r i g i n a l position The i n d i vtdu a l se If . . unted ongma l positi on pres upp oses associa tions a n d v a l u e s w h i ch are n ot a cco for by the theory . th e th eOrY• In sofar as the se soctal supports a r e dealt wt. th b Y Y are they are Justifi ed o n the basis of voluntary choice. I n fact, h o we ve r, th e presupposed by the possibility of v o l u n tary c h oice .
�
•
.
·
210
RAWLS ON THE I N D I V I DUAL AND THE SOCIAL
123
Rawls consider s the socia l context needed to develop and to mai ntai n the moral personal ity which is presupposed by the original position . But these concerns are considere d u nder the ru bric of the condition s necessary for the stabi l ity of the sense of justice . They are not addressed in the theory proper, but as part of the fu rnishings that are required in order to prov ide for the appl icab i l ity of the theory . U nder the rubric of stabi l ity, they appear to be mora l ly neutral . I n fact these conditions are not neutral, but reflect the premium on autonomy, l i berty a nd mutual i ndependence wh ich is characteristic of the liberal trad ition. W h i l e I believe these values to be worthy of their position in the theory, they are exemplified by the theory rather than derived from it. 7 Agai n , presuppositions which R awls takes t o b e weak conditions may i n fact b e strong ones. Accord ing to R awls, a just society is one that approx i mates a social structu re wh i ch would result from the rational choices of the indiv iduals who com pose i t. But a certai n soci a l context of mutual respect, l inguistic patterns, rituals, supporting i n st i tutions and aff i l iative ties is a necessary cond ition for a n indiv idu al capable of rationa l choice. 8 I t m ight be argued that R awls has provided an hypothetical construction of the original position and thus is not responsible, with i n the theory itself, for a genetic accou nt of the conditions of social existence. The issue is not whether Rawls is aware of the social genesis of the individual , but whether or not contractarian theory contai ns sufficient resources to enable him to accomplish his task. Of course, the descri ption of the original posit ion is hypothetical. The l inguistic paradigms provided by the atomists were also hypothetica l . It was not suggested that langu age had actually developed by explicit volu ntary consent. The paradig m was rejected not because it could not account for the h istorical dev elop ment of l a nguage, but because i t was shown to be inadequate for a
description of the complex i ties of l i nguistic usage. Claims for the analysis of language i nto simples and for the assumption of weak conditions were later discove red to obscure rather strong conditions and controversial assumptions cont ained in the para digm .
5. Religion and the Contract Theory
T hree points a t w h i ch the study of rel igion may be relevant for a n appraisal of th e theory of j ustice as fairness merit comment. They are ( 1 ) the social character of the self as observed i n the symbol ic and ritualistic activity of rel igio n , ( 2 ) the affi n ity of the attempt to d iscover a neutral Archimed ean poin t i n the origi nal position with religious understan dings of transcend ence, especially
in theistic tradi t i o ns, and (3) the "perfection ist" character of most re l igious visi ons. O ne of the reason s for the success with which R awls's enterpr i se has been greeted is that he brings togethe r systema ti cally two diverse traditio ns of that of recent conceiv i ng of the i nd ividual self in society. The f irst is hics. W i le a e Anglo- A merican phi losophy and particu larl y philoso phical ly as w1ll has co nception of the self as indi v idual agent understood priman
�
21 1
�
PROUDFOOT
124
predominated i n this tradition, especially i n analytic a nd ex istential ist ethics, the
need for a social conception has been recogni zed. Analysis of the nuances of
lingu istic usage is itself a form of attention to the rituals a n d i nstitutions of
social l ife, but a socia l conception of the self h as not been characteristic of this tradition. The second tradition is that of pragmatism in t h i s coun try and of sociology
and social psychology i n Europe. Durkh e i m and Weber i n i t i ated a tradition of
sociological theory which has been influential i n contemporary eth ics and which
emphasizes social facts and structures as p r i m ary . Dewey, Mead a n d others in the pragmatist tradition emphasized the social ch aracter of a l l k n owledge and action and described the development of the a u tonomous choosing i n d ividual from lm differentiated forms. I n this tradition, individual sy m bo l s and actions are
understood by considering the social a n d cultural context a n d the variety of resources which are available to the individu a l . F ro m t h i s perspective, the
isolated i ndividual is not conceived as a simple entity from which the social order is constructed , but
is described as d iffere ntiati ng h i mself from his
environment. E xcessive isolation is characterized a s a l ienation, and there is a longing for community, and for the i n tegrat ion of the soci a l order through shared cultural values. Attempts to provide general theoretical accounts of rel igi on have seldom met with widespread agreement. I t i s difficult to prov i de c r i teri a whi ch a re a t once specific enough to be meaningf u l , and genera l e n ough to e m brace a l l of the f phenomena which most of us wou ld describe a s rel igiou s. B u t the study o religion reveals shared symbols, goal s, visions of the a d m i rabl e and th e worshipfu l , and concepti ons of that wh i ch makes a pe rson worth y of respect, which are prior to and more fundmental than the autono m ou s sel f or the notio n d of � contractual relationsh i p between persons. Religious e x per ien ce , sy m bols a n lelY so actions all express the recognition that the individu a l cannot be c once i ved or as a rational ego and wil l . Forms of rel igious l ife m a y be asce tic, an ti- social ms ecstati c, bu t a l l express admiration , worsh i p , conception s of the sac re d, in te r the which are communal. Fulfil l ment is conceived as the u n ity with natu re or of cosmos, overcom mg I l l u sory boundaries , bu i ld i n g a new soc1. ety or k''"gdom ch su God, the reu nion of a l l persons i n the l a n d of the drea m i n g, and often _ ze visions are dramatize d in the systems of sym bo l s and action w h ich characten s se s a religious l ife. As the rituals of hu man l anguage are soc i a l , so reli gio n ex pre some context of values, goals and actions in wh ich the i n d i v i du a l i s situ ated i n ·
·
broader and more fund amen tal cont ext. ontex t '" . t h e rational Even Wtt h m · pu ritani sm of Kant the com pre h e nsive c e d bY th which the individua l is set i s provided by the i d als of pu re re as on an ov postu lates of the second Critique. When Kant descri bes a n d a dvocates the rn . eiVes thiS from an et h .teal state of nature to an eth ical com m o nwe a l th , h e c �m c he s. and comm onwealt h as the concept of a people of G od u n der eth ica l l aw . _ . untttn9 describe s re rtglon wtt ' the l t. m .tts of reason . h tn °f as serving the f u n ctto n ca1 e eth t persons under a common lawgiver . Kant draws t he not i o n o f th ·
·
·
�
�
.
.
212
RAWLS ON TH E I N DI VI DUAL AND TH E SOCIAL
1 25
commonwea lth from R ousseau and the social contract tradition, but he pl aces it in the context of his critiq ue of religion. The commonwealth is one of i ndiv id u a l agents united i n t h e i d e a o f a people o f G od which can only be realized i n human orga n ization through some f o r m of rel igious institution. A feature which differentiates the theory of justice as fairness from the ethical theories of pragmati sm and of many who have emphasized the social character of the sel f i s its claim to have di scovered an Archimedean point, a neutral ground f rom which principles m ight be evaluated and chosen. The pragmatists considered the construction of good to be the resultant of confl icts between various i n terests in the society. They rejected any possi b i l i ty. of transcendi n g that field of confl ict in order to discover some neutral point from which the field cou l d be surveyed. The origina l position provides such a transcendent point from which to generate principles that can be used i n the resolution of soc i a l confl icts and in the reform of institu tions. This is part of the Kantian l egacy in the doctrine of justice as fai rness. The function of the original position as one which provides a point of transcendence i n the theory is s i m i l ar to that of the notion of God i n theistic trad it ions. By i maginatively adopting the perspective of the origin a l position, one. can adopt the moral point of v iew, or a point of view which transcends those of i n d iv id u a l self-interests. The sim ilar attempt to achieve a transcendent perspective by the e ighteenth century notion of the d isinterested sympathetic spectator has often been closely related to thei stic thought ( see Reynolds, 1 970). O ne of the fu nctions of doctrines of God h as been to affirm th is k i nd of transce nde nce . Doctrines of G od attempt to achieve a transcendence of paroch i a l intere sts, but they a lso reflect a conception of what is thought to be adm irable or worshi pfu l . They i nclude an element of what Rawls cal l s "perfecti onism," the
cha racterizat ion of a particul ar sort of human excellence as good or adm ira b le. God i s transcendent, but i s not neutral with respect to human ideals or pe rfection s. Doctr i nes of G od have served n ot only to provide transcendence, bu t h ave a lso embodied conceptions of the supremely good or worthy. Even attempts to avoid any such reflection of h uman ideal s by def i n i ng G od as "whol ly other" in neo-orthodox writings meant that the perfections of freedom and autono my, surely human ideals themselves, were glorified and taken to be
m ost wort hy of our admiration. Freedom was a perfection that was dramatized in the doctrine of G od 's utter transcendenc e.
R awls notes that different i n terpretations of the i n itial situ ation y ield d iffere nt conceptions of justice. H e conjectures that each trad itional conception of justice m ay be correlated with an i nterpretati on of the i n itial situ ation ( 1 2 1 ) . This cor rel ation may be due not o n ly to the process of calcul ation of the pri ncipl es w h i ch wou l d be chosen u nder different conditions, but to the fact th at each i n terpretation em bodies different ideals or val ues which are taken to be of unque sti o ned worth. The d istinctive character of the original position, for . i nstan ce, in contrast to the theory of the ideal observer i s the priority it g ives to
213
PROUDFOOT
126
l iberty and to the autonomy of each i ndivi d u a l . T h e i nterpretation of t h e original position i n the theory o f j u stice as fairness is not a neutral one. It contains certai n ideals or i m ages of what a person is or should be. Rawls notes that the theory, as an ideal-regardi ng theory, occupies an
intermediate position between perfection ism and u t i l itarian i sm ( 32 7). It shares
with perfectionism the characteristic that a certai n ideal is bu i l t i nto the theory.
In the case of justice as fairness, this ideal i s that of the autonomou s individual. Whi le it is certa inly a worthy i deal, it m u st be recog ni zed as such . It may be more admirable but is not necessarily a weaker or assumption than wou ld be some orga n i c notion of soci ety.
less
perfectionistic
The conditions of the original position cou l d be descr i bed in such a way as
to elim inate the postulate of mutu a l d is interest. Such an i nterpretation would be more complicated in the sense that the procedures of s i m ple game theory would
not be as readi ly appl icable as in the present case, but it i s not clear that an organic conception of society wou ld i nvolve stronger assu m ptions than that of mutual disinterest. I t wou ld involve d iffere nt assu m ptions, and a different conception of what is admirable. Both the assum ption of mutual d i s i nterest with the priority of l iberty and the assu mption of an orga n i c mode l with i ntertwining i n terests incorporate certai n ideals or notions of pe rfection. Rawls has chosen the former and has provided good reasons for h i s choice , but it is m isleading to suggest that he has remained neutral , reject i ng perfecti o n i s m a nd keeping his assumptions weak. The original position can be seen as the dramat izatio n of a particu lar human idea l . Religious tradition s each contai n conceptions of t h e goo d a n d ideaIs of . perfection . For many trad itions, the l i berty a n d a utonomy of the indi v id ual has not been a central part of such concepti ons. I n m a n y cultu res i n t he past, i ndividual liberty was not held essential for self-respect a n d m ay , i n some cases, er n have been inimical to it. I t may wel l be , a n d I wou l d suspect, that in the m od of world the priority of l ibe rty cannot on be d i sregarded w ithout the destructi r fo self-respect. But the conceptio n of a person as an a uton o m ou s ind ividu al an hom liberty is primary and essenti a l to h i s self-respect i s a statem ent of l na Ideal, of what i s admirable in gi ori the human l ife . The i nterpretatio n of posi tion in justice as fairness th e fu nction s both to ach ieve transcenden ce of parochia l and to embody a perf ecti onist idea l.
�
214
RAWLS ON THE I N DI V I DUAL AND THE SOCIAL
127
1 Page references in the text refer to R awls, 1 97 1 , unless otherwise designated. 2 3
For a critique of this i m age of the person, see Murdoch, 1 970:9ff.
R a w l s disting u ishes between duties and obligations ( 1 1 4, 333ff.). Obligations rest u pon actual voluntary acts of consent, but duties apply to us
without regard to voluntary acts. Both duties and obl igations, however, are derived from the contractarian point of v iew. They are principles th at would be chosen in the original position , though it is not necessary that they be actually chosen i n a particu lar society .
4 The example of artificial i ntel ligence is not as irrelevant as it m igh t seem in this context. In i ntroducing the notion of rational plan which is central to his conception of rational choice, R awls refers to M i l ler, Galanter and Pribram, 1960, for a d iscussion of p lans. Th is volume has been influential in cognitive psychology for its attempt to describe human planning and behavior by analogy to compu ter programs. For a consideration of the lim itations of some of this work, see D reyfus , 1 9 7 2. 5
The Aristotelian Principle, postu lated by R awls as a basic principle of
motivation m ust, of course, be justified empirical ly. R awls appeals to our sense that "it seems to be borne out by many facts of everyday life, and by the behavior of children and some of the h igher animals" (43 1 ) . The princi ple seems int uitivel y sou nd, but he does not cite any specific data in support of it. 6 For a classification of different k inds of games and a differentiation between formal and informal inte l l igent activities, see Dreyfus, 1 972: 204-208. D reyfus notes that many com plex human activities are regu lar but not rule-governed. 7 Thom as N agel h as m ade a sim ilar point: "I have attempted to argue that the pres u m ptio ns of the contract method R awls employs are rather strong, and that t he original position therefore offers less independent support to his
conclusi ons than a t f i rst appears. The egal itarian l i beralism which he develops and the conception of good on wh ich i t depends are extremely persuasive, but th e origi n a l position serves to model rather than to justify them" ( 1 973: 233). 8
Stuart H ampsh i re makes a similar point when he refers to "the psychology of mo ral sentime nts, which suggests that gu i l t or shame about injustice and unfa irness, and natural respect for the ir opposites, extends to more primitive rel ations between people than are imagined by his [ R awls's] hypothesis of a ratio nal choice of an unbiased social order" ( Hampshire, 1 97 2 : 38) .
21 5
PROUDFOOT
128
R E F E R E N CES Dreyfus, Hubert L. 1 972 What Computers Can't Do: A Critique of Artificial Reason. New York: Harper a nd R ow. Hampshire, Stuart 1 972 "A new philosophy of the j u st society . " The New York Review of Books 1 8 ( Fe bruary 24) : 34-39.
M i l ler, George; Galanter, Eugene; and Pribram , Karl H . 1 960
Plans and
the
Structure
o f Behavior.
N ew Y ork: Holt,
R inehart and Winston . Murdoch, I ris 1 970
The Sovereignty of Good. New York : Schocken Books.
Nagel, Thomas 1 973
"Rawls on justice." The Philosophical Review 82 (April): 200-234.
Rawls, John
"J ustice as fairness. " The Philosophical Review 67 (April) : 1 64- 1 94. 1 97 1 A Theory of Justice. C a m bridge : H arvard U n ive rsity Press. Reynolds, Charles H . n 1 970 "A proposal for u nderstandi ng the pl ace of reason i n Chri stia 8. ethics." The Journal of Religion 50 ( Apri l ) : 1 55- 1 6 Royce, Josiah 1914 The Philosophy o f Loyalty. N e w York : The M acm i l lan 1 958
Company. Wittge nstein , Ludwig 1 953
T he Philosophical Investigations. Ansco m be, tr. N e w York :
Macm i l lan Company.
216
THE PROCEDURAL REPUBLIC
AND THE UNENCUMBERED SELF
MICHA EL J.
SANDEL
Harvard University
OLITICAL PHILOSOPHY seems often to reside at a distance from the world. Principles are one thing, politics another, and even our best efforts to "live up" to our ideals typically founder on the gap between theory and practice. 1 But if political philosophy is unrealizable in one sense, it is unavoida ble in another. This is the sense in which philosophy inhabits the world from the start; our practices and institutions are embodiments of theory. To engage in a political practice is already to stand in relation to theory. 2 For all our uncertainties about ultimate questions of political philosophy-of justice and value and the nature of the good life-the one thing we know is that we live some answer all the time. In this essay I will try to explore the answer we live now, in contem porary America. What is the political philosophy implicit in our practi ces and institutions? How does it stand, as philosophy? And how do tensions in the philosophy find expression in our present political cond ition? It m ay be objected that it is a mistake to look for a single philosophy, that we live no ..answer," only answers. But a plurality of answers is itself a kind of answer And the political theory that affirms this plurality is . the theory I prop ose to explore.
A UTHOR 'S
NOTE: An earlier version of this article was presented to the Political Philosophy Colloquium at Princeton University, and to the Legal Theory Workshop at ColumbiQ LDw Scho ol. I am grateful to the participants, and also to the Editor, William Co1111olly, for helpful comments and criticisms. l would also like to thank the Ford Fo dot o i 11 for support of a larger project of which this essay is a first installment. U1J
PoLmCAL THEORY, Vol. 12. No. I , February 1 984 8 1 -96 0 l984 Sage Publications, Inc.
217
82
POLITICAL THEORY
/
FEBRUA RY 1 984
THE RIGHT AND THE GOOD We might begin by considering a certain moral and political vision. It is a liberal vision, and like most liberal visions gives pride of place to justice, fairness, and individual rights. Its core thesis is this: a just society seeks not to promote any particular ends, but enables its citizens to pursue their own ends, consistent with a similar liberty for all; it there fore must govern by principles that do not presuppose any particular conception of the good. What justifies these regulative principles above all is not that they maximize the general welfare, or cultivate virtue, or otherwise promote the good, but rather that they conform to the concept of right, a moral category given prior to the good, and independent of it. This liberalism says, in other words, that what makes the just society just is not the telos or purpose or end at which it aims, but precisely its refusal to choose in advance among competing purposes and ends. In its constitution and its laws, the just society seeks to provide a framework within which its citizens can pursue their own values and ends, consistent with a similar liberty for others. The ideal I've described might be summed up in the claim that the right is prior to the good, and in two senses: The priority of the right means first, that individual rights cannot be sacrificed for the sake of the general good (in this it opposes utilitarianism), and second, that the principles of justice that specify these rights cannot be premised on any particular vision of the good life. (In this it opposes teleological conceptions in general.) !his is the liberalis m of much contemporary moral and politi�al phllosophy, most fully elaborated by Rawls, and indebted to Kant for ItS philosophical foundations. 3 But I am concerned here less with the lineage of this vision than with what seem to me three striking facts about it. First, it has a deep and powerful philosophical appeal. Second, despite its philosophical force, the claim for the priority of the righ� over th� g�od ultimately fails. And third, despite its philos ophical fatl�re, this hberal vision is the one by which we live. For us in late twenu�tb ed �entury America, it is our vision , the theory most tho roughly embodi d n A m the practices and institutio ns most central to our public life. o seeing how it goes wrong as philosophy may help us to diagnose �r present political condition . So first its philosophical power; second, Its · ent �h"l1 osophical failure; and third, however briefly, its uneasy embodun m the world. '
218
Sandel / THE UNENCUM BERED SELF
83
But before taking up these three claims, it is worth pointing out a central theme that connects them. And that is a certain conception of the person, of what it is to be a moral agent. Like all political theories, the liberal theory I have described is something more than a set of regulative principles. It is also a view about the way the world is, and the way we move within it. At the heart of this ethic lies a vision of the person that both inspires and undoes it. As I will try to argue now, what make this ethic so compelling, but also, finally, vulnerable, are the promise and the failure of the unencumbered self.
KA NT/AN FOUNDA TIONS The liberal ethic asserts the priority of right, and seeks principles of JUStice that do not presuppose any particular conception of the good. 4 This is what Kant means by the supremacy of the moral law, and what Rawls means when he writes that "justice is the first virtue of social institutions .s Justice is more than just another value. It provides the framework that regulates the play of competing values and ends; it must therefore have a sanction independent of those ends. But it is not obvious where such a sanction could be found. Theories ofjustice, and for that matter, ethics, have typically founded their claims on one or another conception of human purposes and ends. Thus Aristotle said the measure of a polis is the good at which it aims, and even J .S. Mill, who in the nineteenth century called "justice the chief part, and incomparably the most binding part of all morality," made j ustice an instrument of utilitarian ends. 6 This is the s olution Kant's ethic rejects . Different persons typically have different desires and ends, and so any principle derived from them can only be contingent. But the moral law needs a categorical founda tion, not a contingent one. Even so universal a desire as happiness will not do. People still differ in what happiness consists of, and to install any particular conception as regulative would impose on some the concep tions of others, and so deny at least to some the freedom to choose their own conceptions. In any case, to govern ourselves in conformity with desires and inclinations, given as they are by nature or circumstance, is not really to be self-governing at all. It is rather a refusal of freedom, a capitulation to determinations given outside us. According to Kant, the right is ''derived entirely from the concept of freedo m in the external relationships of human beings, and has nothing •
•
.•
219
84
POLITICAL THEORY
/
FEBRUARY 1 984
to do with the end which all men have by nature [i.e., the aim of achieving happiness] or with the recognized means of attaining this
end.'" As such, it must have a basis prior to all empirical ends. Only when I am governed by principles that d o not presuppose any parti cular ends am I free to p ursue my own ends consistent with a similar freedom for all.
But this still leaves the question of what the basis of the right could possibly be. If it must be a basis prior to all purposes and ends, uncondi tioned even by what Kant calls "the special circumstances of human nature,"11 where could such a basis conceivably be found? Given the
stringent demands of the Kantian ethic, the moral law would seem almost to require a foundation in nothing, for any empirical precondi tion would undermine its priority. "Duty ! " asks Kant at his most lyrical, "What origin is there worthy of thee, and where is to be found the root of thy noble descent which proudly rejects all kinship with the inclinations?"' His answer is that the basis of the moral law is to be found in the
subject,
not the object of practical reason, a subject capable of an
autonomous will. No empirical end, but rather "a subject of ends,
namely a rational being himself, must be made the ground for all maxims of action. "10 Nothing other than what K ant calls "the subject of all possible ends himself" can give rise to the right , for only this subject is
also the subject of an autonomous will. Only this subject could be that "something which elevates man above himself as p art of the world of sense" and enables him to participate in an ideal, unconditioned realm wholly independent of our social and psychological inclinations. And only this thoroughgoing independence can afford us the detachment we need if we are ever freely to choose for ourselves, unconditioned by the vagaries of circumstance. 1 1 e Who o r what exactly is this subject? I t is, i n a certain sense, us. t. moral law, afterall, is a law we give o urselves; we don 'tfind it, we w i/l l d That is how it (and we) escape the reign of nature and circu m stance an o h w " merely empirical ends. But what is important to see is that the ••we or do the willing are not "we"qua particular persons, you and me, each f " a ours�l�es -the moral law is not up to us as individuals-but "we q� C rtl a p qua �artlCl �ants in what Kant calls "pure practical reason, " "we"
�
lpants m a transcendental subject. of N �� what is to guarantee that I am a subject of this kind, capa le no exerclsmg pure practical reason? Well, strictly speaking, there . 13 • a lS guar�t�; the transcendental subject is only a possibility. B ut 1t l a r o posslblhty I ·must presuppose if I am to think of myself as a free m of agent. Were I wholly an empirical being, I would not be capable
�
220
Sandel
/
THE UNENCU MBERED SELF
85
freedom, for every exercise of will would be conditioned by the desire for some object. All choice would be heteronomous choice, governed by the pursuit of some end. My will could never be a first cause, only the effect of some prior cause, the instrument of one or another impulse or inclination. "When we think of ourselves as free," writes Kant, "we transfer ourselves into the intelligible world as members and recognize
the autonomy of the will. ''1 2 And so the notion of a subject prior to and independent of experience, such as the Kantian ethic requires, appears not only possible but indispensible, a necessary presupposition of the possibility of freed om. How does all of this come back to politics? As the subject is prior to its ends, so the right is prior to the good. Society is best arranged when it is governed by principles that do not presuppose any particular concep tion of the good, for any other arrangement would fail to respect persons as being capable of choice; it would treat them as objects rather than subjects, as means rather than ends in themselves. We can see in this way how Kant's notion of the subject is bound up with the claim for the priority of right. But for those in the Anglo American tradition, the transcendental subject will seem a strange foundation for a familiar ethic. Surely, one may think, we can take rights seriously and affirm the primacy ofjustice without embracing the
Critique of Pure Reason. This, in any case, is the project of Rawls.
He wants to save the priority of right from the obscurity of the transcendental subject. Kant's idealist metaphysic, for all its moral and p olitical advantage, cedes too much to the transcendent, and wins for
just ice its primacy only by denying it its human situation. "To develop a viable Kantian conception of justice," Rawls writes, "the force and co ntent of Kant's doctrine must be detached from its background in transcendental idealism" and recast within the "canons of a reasonable empiricism. " 1 3 And so Rawls' project is to preserve Kant's moral and political teaching by replacing Germanic obscurities with a domesti
cated metaphysic more congenial to the Anglo-American temper. This is the role of the original position.
FROM TRA NSCENDENTAL SUBJECT TO UNENCUMBERED SELF The original position tries to provide what Kant's transcendental argument cannot-a foundatio n for the right that is prior to the good,
221
86
POLITICAL THEORY
I
FEBRUARY 1984
but still situated in the world. Sparing all but essentials, the original position works like this: It invites us to imagine the principles we would choose to govern our society if we were to choose them in advance, before we knew the particular persons we would be-whether rich or poor, strong or weak, lucky or unlucky-before we knew even our interests or aims or conceptions of the good. These principles-the ones we would choose in that imaginary situation-are the principles ofjustice. What is more, if it works, they are principles that do not presuppose any particu lar ends. What they do presuppose is a certain picture of the person, of the way we must be if we are beings for whom justice is the first virtue. This is the picture of the unencumbered self, a self understood as prior to and independent of purposes and ends. Now the unencumbered self describes first of all the way we stand toward the things we have, or want, or seek. It means there is always a distinction between the values I have and the person I am. To identify any characteristics as my aims, ambitions, desires, and so on, is always to imply some subject "me" standing behind them, at a certain distance, and the shape of this "me" must be given prior to any of the aims or attributes I bear. One consequences of this d istance is to put the self itself beyond the reach of its experience, to secure its identity once and for all. Or to put the point another way, it rules out the possibility of what we might call constitutive ends. No role or commitment could define me so completely that I could not understand myself without it. No project could be so essential that turning away from it w ould call into question the person I am. For the unencumbered self, what matters above all, what is most essential to our personhood, are not the ends we choose but our capacity to choose them. The original position sums u p this cent ral claim about us. "It is not our aims that primarily reveal our nature ," writes Rawls, "but rather the principles that we would acknowledge to govern the background conditions under which these aims are to be formed · · · We should therefo re reverse the relation between the right and the good proposed by teleological doctrines and view the right as prio r. "14 On_Iy if the self is prior to its ends can the right be p rior to the g ood· Only if my identity is never tied to the aims and interests I may h ave at any moment can I think of myself as a free and independent agent, capable of choice. · no . ThIs tIon of ·mdependence carries consequences for the k'md of community of which we are capable. Understood as unencumbered
222
Sandel
I
THE UNENCUMBERED SELF
87
selves, we are of course free to join in voluntary association with others, and so are capable of community in the cooperative sense. What is denied to the unencumbered self is the possibility of membership in any community bound by moral ties antecedent to choice; he cannot belong to any community where the self itself could be at stake. Such a community-call it constitutive as against merely cooperative-would engage the identity as well as the interests of the participants, and so implicate its members in a citizenship more thoroughgoing than the unencumbered self can know. Forjustice to be primary, then, we must be creatures of a certain kind, related to human circumstance in a certain way. We must stand to our circumstance always at a certain distance, whether as transcendental subject in the case of Kant, or as unencumbered selves in the case of Rawls. Only in this way can we view ourselves as subjects as well as objects of experience, as agents and not just instruments of the purposes we pursue. The unencumbered self and the ethic it inspires, taken together, hold out a liberating vision. Freed from the dicates of nature and the sanction of social roles, the human subject is installed as sovereign, cast as the author of the only moral meanings there are. As participants in pure practical reason, .or as parties to the original position, we are free to construct principles of justice unconstrained by an order of value antecedently given. And as actual, individual selves, we are free to choo se our purposes and ends unbound by such an order, or by custom or trad ition or inherited status. So long as they are not unjust, our conceptions of the good carry weight, whatever they are, simply in virtue of our having chosen them. We are, in Rawls' words, "self-originating sources of valid claims. "1 5 This is an exhilarating promise, and the liberalism it animates is perhaps the fullest expression of the Enlightenment's quest for the self-defining subject. But is it true? Can we make sense of our moral and political life by the light of the self-image it requires? I do not think we can, and I will try to show why not by arguing first within the liberal project, then beyond it.
JUS TICE AND COMMUNITY We have focused so far on the foundations of the liberal vision, on the ay w it derives the principles it defends. Let us turn briefly now to the
223
POLITICAL THEORY
88
I
FEBRUARY 1 984
substance of those principles, using Rawls as our example. Sparing all but essentials once again, Rawls' two principles of justice are these: first,
equal basic liberties for all, and second, only those social and economic inequalities that benefit the least-advantaged members of society (the difference principle).
In arguing for these principles, Rawls argues against two familiar alternatives-utilitarianism and libertarianism. He argues against utili· tarianism that it fails to take seriously the distinction between persons. In seeking to maximize the general welfare, the utilitarian
treats society as whole as if it were a single person; it conflates our many, diverse desires into a single system of desires, and tries to maximize. It is indifferent to the distribution Gf satisfactions among persons, except insofar as this may affect the overall sum. But this fails to respect our
plurality and distinctness. It uses some as means to the happiness of all, and so fails to respect each as an end in himself. While utilitarians may sometimes defend individual rights, their defense must rest on the calculation that respecting those rights will serve utility in the long run. But this calculation is contingent and uncertain. S o long as utility is "6 what Mill said it is, "the ultimate appeal on all ethical questions, 1 individual rights can never be secure. To avoid the danger that their life prospects might one day be sacrificed for the greater good of others, the parties to the original position therefore insist on certain basic liberties for all, and make those liberties prior. If utilitarians fail to take seriously the distinctness of persons, . . hbertanan s go wrong by failing to acknowledge the arbitrariness of fortune. They define as just whatever distribution results from an efficient market economy, and oppose all redistributio n on the grounds that people are entitled to whatever they get, so long as they do n ot cheat wls or steal or otherwise violate someone's rights in getti ng it. Ra opposes this principle on the ground that the distributio n of talents an s 15 ass ts and even efforts by which some get more and others get les � To a�blt rary from a moral point of view, a matter of good luck. . to t o dlstnb n ute the good things in life on the basis of these differences is e h do us ti e, but simply to carry over into human arrangements . � l· lV arbltranness of social and natural contingency. We deserve , as ind the duals, neither the talents our or good fortune may have brought, n benefits that flow from them. lents We should therefore regard these ta as common assets, and rega es of rd one another as commo n beneficiari
�
�
j
224
Sandel
I THE UNENCUMBERED SELF
89
the rewards they bring. ''Those who have been favored by nature, whoever they are, may gain from their good fortune only on terms that improve the situation of those who have lost out . . . Injustice as fairness, men agree to share one another's fate. "17
This is the reasoning that leads to the difference principle. Notice how
it reveals, in yet another guise, the logic of the unencumbered self. I cannot be said to deserve the benefits that flow from, say, my fine
physique and good looks, because they are only accidental, not essential facts about me. They describe attributes I have , not the person I am, and
so cannot give rise to a claim of desert. Being an unencumbered self, this
is true of everything about me. And so I cannot, as an individual, deserve anything at all.
However jarring to our ordinary understandings this argument may be, the picture so far remains intact; the priority of right, the denial of desert, and the unencumbered self all hang impressively together. But the difference principle requires more, and it is here that the argument comes undone. The difference principle begins with the thought, congenial to the unencumbered self, that the assets I have are only accidentally mine. But it ends by assuming that these assets are therefore common assets and that society has a prior claim on the fruits of their exercise. But this assumption is without warrant. Simply because I, as an individual, do not have a privileged claim on the assets accidentally residing "here," it does not follow that everyone in the world collectively d oes. For there is no reason to think that their location in society's province or, for that matter, within the province of humankind, is any less arbitrary from a moral point of view. And if their arbitrariness within me makes them ineligible to serve my ends, there seems no obvious reason why their arbitrariness within any partiuclar society should not make them ineligible to serve that society's ends as Well. To put the p oint another way, the difference principle, like utili tarianism, is a principle of sharing. As such, it must presuppose some Prior m oral tie among those whose assets it would deploy and whose efforts it would enlist in a common endeavor. Otherwise, it is simply a formula for using some as means to others ends, a formula this liberalism is committed to reject. But on the cooperative vision of community alone, it is unclear what the moral basis for this sharing could be. Short of the constitutive
225
90
POLITICAL THEORY
/
FEBRUARY 1984
conception, deploying an individual's assets for the sake of the common good would seem an offense against the "plurality and distinctness" of individuals this liberalism seeks above all to secure. If those whose fate I am required to share really are, morally speaking, others, rather than fellow participants in a way of life with which my identity is bound, the difference principle falls prey to the same objections as utilitarianism. Its claim on me is not the claim of a constitutive community whose attachments I acknowledge, but rather the claim of a concatenated collectivity whose entanglements I confront. What the difference principle requires, but cannot provide, is some way of identifying those among whom the assets I bear are properly regarded as common, some way of seeing ourselves as mutually indebted and morally engaged to begin with. But as we have seen, the constitutive aims and attachments that would save and situate the difference principle are precisely the ones denied to the liberal self; the moral encumbrances and antecedent obligations they imply would undercut the priority of right. What, then, of those encumbrances? The point so far is that we cannot be persons for whom justice is primary, and also be persons for whom the difference principle is a principle of justice. But which must give way? Can we view ourselves as independent selves, independent in the sense that our identity is never tied to our aims and attachments? I do not think we can, at least not without cost to those loyalties and convictions whose moral force consists partly in the fact that living by them is inseparable from understanding ourselves as the particular persons we are-as members of this family or commu nity or nation or people, as bearers ofthat history, as citizens of this republic. Allegiances such as these are more than values I happen to have, and to hold, at a certain distance . They go beyond the obligations I voluntarily incur and the "natural duties" I owe to human beings as such. They allow that to some I owe more than j ustice requires or even permits, not b y reason of agreements I have made but instead in virtue of those more or less enduring attachments and commitments that, taken together, partlY define the person I am. To imagine a person incapable of constitutive attach ments such as these is not to conceive an ideally free and rational agent, but to i magine a person wholly without character, without moral depth. For to have character is to know that I move in a history I neithe r summo n nor command, which carries consequences nonetheless for my choices an conduct. It draws me closer to some and more distant fro m others; tt
�
226
Sandel
I
THE UNENCUMBERED SELF
91
makes some aims more appropriate, others less so. A s a self-interpreting being, I am able to reflect on my history and in this sense to distance myself from it, but the distance is always precarious and provisional, the point of reflection never finally secured outside the history itself. But the liberal ethic puts the self beyond the reach of its experience, beyond deliberation and reflection. Denied the expansive self-understandings that could shape a common life, the liberal self is left to lurch between detachment on the one hand, and entanglement on the other. Such is the fate of the unencumbered self, and its liberating promise.
THE PROCED URAL REPUBLIC But before my case can be complete,
I need to consider one powerful
reply. While it comes from a liberal direction, its spirit is more practical than philosophical. It says, in short, that I am asking too much. It is one thing to seek constitutive attachments in our private lives; among
families and friends, and certain tightly knit groups, there may be
found a common good that makes justice and rights less pressing. But with public life-at least today, and probably always-it is different. So long as the nation-state is the primary form of political association, talk of constitutive community too easily suggests a darker politics rather
than a brighter one; amid echoes of the moral majority, the priority of
right, for all its philosophical faults, still seems the safer hope.
This is a challenging rejoinder, and no account of political com
munity in the twentieth century can fail to take it seriously. It is challenging not least because it calls into question the status of political philos ophy and its relation to the world. For if my argument is correct, if the libe ral vision we have considered is not morally self-sufficient but parasitic on a notion of community it officially rejects, then we should expect to find that the political practice that embodies this vision is not
practically self-sufficient either-that it must draw on a sense of community it cannot supply and may even undermine. But is that so far from the circumstance we face today? Could it be that through the original position d arkly, on the far side of the veil of ignorance, we may gli mp se an intim ation of our predicament, a refracted vision of ourselves? Ho w does the liberal vision-and its failure-help us make sense of our p ublic life and its predicament? Consider, to begin, the following
227
92
POLITICAL THEORY
/
FEBRUARY 1 984
paradox in the citizen's relation to the modern welfare state. In many ways, we in the 1980s stand near the completion of a liberal project that has run its course from the New Deal through the Great Society and into the present. But notwithstanding the extension of the franchise and the expansion on individual rights and entitlements in recent decades, there is a widespread sense that, individually and collectively, our control over the forces that govern our lives is receding rather than increasing. This sense is deepened by what appear simultaneously as the power and the powerlessness of the nation-state. One the one hand, increasing numbers of citizens view the state as an overly intrusive presence, more likely to frustrate their purposes than advance them. And yet, despite its unprecedented role in the economy and society, the modern state seems itself disempowered, unable effectively to control the domestic eco nomy, to respond to persisting social ills, or to work America's will in the world. This is a paradox that has fed the appeals of recent politicians (including Carter and Reagan) , even as it has frustrated their attempts to govern. To sort it out, we need to identify the public philosophy implicit in our political practice, and to reconstruct its arrival. We need to trace the advent of the procedural republic, by which I mean a public life
animated by the liberal vision and self-image we've considered. The story of the procedural republic goes back in so me ways to the founding of the republic, but is central drama begins to u nfold arou?d the turn of the century. As national markets and large-scale enterpnse displaced a decentralized economy, the decentralized political fo rms of the early republic became outmoded as well. If democracy was to s��ive, the concentration of economic power would h ave to be met by a stmllar concentration of political power. But the Progressive s under stood, or some of them did, that the success of democ racy required more !ha� the centralization of gove rnment ; it also required the national IZation of politics. The primary form of political commu nity had to be a �ec�t on � national scale. For Herbert Croly, writi ng in 1 909, he _ mg of natiOnahz American political, economic, and social life" was an _ essentially formative and enlightening political transformation. " We wo�ld become more of a democracy only as we became "more of a natto� . . . in ideas, in institutions, and in spirit. " 9 1 Thts nationalizing project would be consummated in the New l)eal, tion but for the democratic traditi on in America the embrace of the na rt of was 8 decl· s · ve depa ure. From Jefferson · to' the populists, the �a Y � � g democracy m Amencan political debate had been , roughly speakm , tbe
�
228
Sandel f THE UNENC UMBER ED SELF
93
party of the provinces, of decentralized power, of small-town and small scale America . A nd against them had stood the party of the nation first Federalis ts, then Whigs, then the Republicans of Lincoln-a party
that spoke for the consolida tion of the union. It was thus the historic achievement of the New Deal to unite, in a single party and political program, what Samuel Beer has called "liberalism and the national idea. ,,;zo
What matters for our purpose is that, in the twentieth century, liberalism made its peace with concentrated power. But it was under stood at the start that the terms of this peace required a strong sense of
national community, morally and politically to underwrite the extend ed involve ments of a modern industrial order. If a virtuous republic of small-scale , democratic communities was no longer a possibility, a national republic seemed democracy's next best hope. This was still, in principle at least, a politics of the common good. It looked to the nation,
not as a neutral framework: for the play of competing interests, but rather as a formative community, concerned to shape a common life suited to the scale of modern social and economic forms.
But this project failed. By the mid- or late twentieth century, the national republic had run its course. Except for extraordinary moments, such as war, the nation proved too vast a scale across which to cultivate the shared self-understandings necessary to community in the forma
tive, or constitutive sense. And so the gradual shift, in our practices and inst itutions, from a public philosophy of common purposes to one of fair procedures, from a politics of good to a politics of right, from the national republic to the procedural republic.
OUR PRESENT PREDICAMENT A full account of this transition would take a detailed look: at the changing shape of political institutions, constitutional interpretation, and the terms of political discourse in the broadest sense. But I suspect _ we would find in the practice of the procedural repubhc two broad tende ncies foreshadowed by its philosophy: first, a tendency to rowd � out democratic possibilities; second, a tendency to undercut the kmd of
cornmu·nity on which it nontheless depends. . Where liberty in the early republic was understood as a funct10n of democratic institutions and dispersed power/ 1 liberty in the procedural
229
POLITICAL THEOR Y
94
I
FEBRU ARY 1 984
republic is defined in opposition to democracy, as an individual's guarantee against what the majority might will. I am free insofar as I am the bearer of rights, where rights are trumps. 22 Unlike the liberty of the early republic, the modern version permits- in fact even requires concentrated power. This has to d o with the u niversalizing logic of rights. Insofar as I have a right, whether to free speech or a minimum income, its provision cannot be left to the vagaries of local preferences but must be assured at the most comprehensive level of political association. It cannot be one thing in New York and another in
Alabama. As rights and entitlements expand, p olitics is therefore displaced from smaller forms of association and relocated at the most universal form-in our case, the nation. And even as politics flows to tbe
nation, power shifts away fro m democratic institutions (such as legislatures and political parties) and toward institutions designed to be
insulated from democratic pressures, and hence better equipped to
dispense and defend individual rights (notably the judiciary and bureaucracy). These institutional developments may begin to account for the sense
of powerlessness that the welfare state fails t o address and in some ways doubtless deepens. But it seems t o me a further clue t o our condition recalls even more directly the predicament of the unencumbered self lurching, as we left it, between detachment on the one hand, the enta glement on the other. For it is a striking feature o f the welfare state � th t It fers a powerful promise of individual rights, and also de�ands � � of Its c1t1zens a high measure of mutual engagement. But the self-un88'=
�
that attends the rights cannot sustain the engagement. As bearers of rights, where rights are trumps, we think of o urselves as f�eely choosing, individual selves, unbound by o bligations antecedent to nghts, or to the agreeme nts we make. A n d yet, as citizens of t e proced �ral r�pu lic that secures these rights, we find ourselves i�pli· cate Wllly-mlly m a formidable array of dependencies and e xpectauons we did not choose and increasin gly reject. In our public life, we are more entangled, but less attached , thaD y tbe ever befor . It is as though the unencumbered self presupp osed b � . hberal eth1c had begun to come true-les wered, s liberated t han dise mp o e�tangled in a network ia� of obligati ons and involveme nts unassoc dentifi Wlt any act of will, and yet unmediated by those com m on i le. AJ catiOns or expansive self -definition s that would make them t olerab rnore the scale of social and political organiza tion has become rnore comprehensive the terms , of our collective identity have become
�
�
�
�
230
Sandel I THE UNENCUMBERED SELF
95
fragmented, and the forms of political life have outrun the common purpose needed to sustain them. Something like this, it seems to me, has been unfolding in America for the past �alf-century or so. I hope I have said at least enough to suggest the shape a fuller story might take. And I hope in any case to have conveyed a certain view about politics and philosophy and the relation between them-that our practices and institutions are themselves embodiments of theory, and to unravel their predicament is, at least in part, to seek after the self-image of the age.
NOTES I . An excellent example of this view can be found in Samuel Huntington, American Politics: The Promise of Disharmony (Cambridge: Harvard University Press, 1 98 1 ). See
especially his discussion of the "ideals versus institutions" gap, pp. 10- 1 2, 39-4 1 , 61 -84, 22 1-262. 2. See, for example, the conceptions of a "practice" advanced by Alasdair Macintyre and Charles Taylor. Macintyre, After Virtue (Notre Dame: University of Notre Dame
Press, 198 1), pp. 1 75-209. Taylor, "Interpretation and the Sciences of Man," Review of Metaphysics 25, ( 1 97 1 ) pp. 3-5 1 . 3. John Rawls, A Theory of Justice. (Oxford: Oxford University Press, 197 1 ). Immanuel Kant, Groundwork of the Metaphysics of Morals, trans. H. J. Paton. ( 1785; New York: H arper and Row, 1 956). Kant, Critique of Pure Reason, trans. Norman K.emp Smith ( 1 78 1 , 1 787; London: Macmillan, 1 929). K.ant, Critique of Practical Reason, trans. L. W. Beck ( 1 788; Indianapolis: Bobbs-Merrill, 1 956). K.ant, "On the Common Saying: 'This May Be True in Theory, But It Does Not Apply in Practice,' " in Hans Reiss, ed., Kant 's Political Writings. ( 1 793; Cambridge: Cambridge University Press, 1970). Other recent versions of the claim for the priority of the right over good can be found in Robert
Nozick, A narchy, State and Utopia (New York: Basic Books, 1 974); Ronald Dworkin, Taking Rights Seriously (London: Duckworth, 1 977); Bruce Ackerman, Social Jwtice in the liberal State (New Haven: Yale University Press, 1 980). 4. This section, and the two that follow, summarize arguments developed more fully in Michael Sandel, Liberalism and the Umits of Justice (Cambridge: Cambridge University Press, 1 982). 5. Rawls ( 197 1 ), p . 3. 6. J ohn Stuart Mill, Utilitarianism, in The Utilitarians ( 1 893; Garden City: Do ubleday, 1 973), p. 465. Mill, On Uberty, in The Utilitarians, p. 485 (Originally published 1849). 7. Kant ( 1 793), p. 73. 8 . Kant ( 1 785), p. 92. 9. Kant ( 1 788), p. 89. 10. Kant ( 1 785), p. I 05. I I . Kant (I 788), p. 89.
231
96
POLITICAL THEORY
/
FEBRUARY 1 984
1 2. Kant ( 1 785), p. 1 2 1 . 1 3. Rawls, "The Basic Structure as Subject," American Philosophical Quarterly ( 1 977), p. 1 65. 14. Rawls ( 1 97 1 ), p. 560. 1 5. Rawls, "Kantian Constructivism in Moral Theory," Journal of Philosophy 11
( 1980), p. S43. 16. Mill ( 1 849), p. 485.
17. Rawls ( 1 97 1 ), pp. 1 0 1 -102. 18. The account that follows is a tentative formulation of themes requiring more detailed elaboration and support.
19. Croly, The Promise of American life (Indianapolis: Bobbs-Merrill, 1965), PP· 270-273.
20. Beer, "Liberalism and the National Idea," The Public Interest, Fall ( 1966), PP· 70-82. 2 1 . See, for example, Laurence Tribe, Foundation Press, 1 978), pp. 2-3.
American Constitutional Law (Mineola: The
22. See Ronald Dworkin, "Liberalis m," in Stuart H ampshire, ed., Public and Private
Morality (Cambridge: Cambridge University Press, 1 978), p. 1 36.
Michael J. Sandel is Associate Professor of Government at Harvard University. He is the author of Liberalism and the Limits of Justice.
Rawls
and Individualism
o B L A N B Y , Chairman, Department of Philosophy, University of Notre Dame, Notre Dame, Indiana 46556.
c. F.
his The most characteristic strand of Rawls' A Theory of Justice is etical ypoth strategy of justifying certain principles of justice by way of h ce. choices made by self-interested individuals behind a veil of igno ran ve ha s) xist (Mar Critics from both the right (Aristotelians } and the left t focused on this feature of his theory in mounting arguments to the effec e t that Rawls' account of justice is simply the narrow individualism of � modem liberal tradition in contemporary dress. The pivo tal notion m
wan! �0
the multi-faceted polemic is "individualism. " In this paper I ex�lore the extent to which Rawls' position is and is not individuah � tic pnor to examining what can be said for and agains t the individualism his position actually embodies.
I. gument Rawls' critics from both the right and the left share an ar f gr form that calls for closer scrutiny . Allowing for a certain de ee ? if d o tw simplification, the same form can be seen to be emb odied in
ferent lines of attack: A.
B.
on Rawls' view of society is narrowly individualistic because e ar n io his account the people in the decisive choice situat characterized 11socially and ahistorically ; on �wls' view of persons is narrowly individualistic be�use e ar on his account the people in the decisive choice situati sts. re e int characterized as having no interest in one another's
1 12
234
As an additional complication, in most discussions these arguments are not kept distinct and are often run together as if they were ma·king the same point. The first observation to be made about these generalized arguments is that they bear on two quite different senses of " individualism." Argu ment A bears on metaphysical individualism (atomism) which in this case is the view that social wholes are simple aggregates of their com ponent individuals ; whereas argument B bears on ethical individualism (egoism), in this case the view that human action is/ought to be guided solely by self-interest. This being accurate, the thrust of argument A is that Rawls' position is vitiated by a metaphysically " abstract" view of persons while the thrust of argument B is that Rawls' position is vitiated by an ethically "empoverished" view of persons. I t should be apparent that these are two quite different senses of "individualism" ; rutd, however they may be related historically or even conceptually, they should be kept distinct. The second observation to be made about both of these generalized arguments is considerably more tendentious ; I think they both involve the same fundamental mistake, a mistake which completely undermines them as arguments against Rawls' position. In the examination of any argued position one must clearly distinguish between the substantive conclusions of the position and the justification strategies employed in recommending those conclusions. This is not to say that these two ftatures of a total account are unrelated; but however closely related they are, it is surely still an error to move directly from characteristics of arguments for a position to characteristics of the position itself. The general character of this fundamental error can be appreciated by con sidering other domains. Does a probabilistic justification of a set of epistemic principles thereby commit you to only probabilistic epistemic Principles? Does a historical defense of certain moral principles thereby COmmit you to merely historical moral principles? And, does a prudential justification of certain political principles thereby commit you to only �rudential political principles? I would submit that the answer to each m this series of questions is "no" because characteristics of the way in �hich a position is justified do not automatically translate into character Istics of the substantive content of the position as justified. If in a given case, there is a positive transition to be made, a specific argument to that effect has to be adduced. Nothing in the general form of argument licenses the inference. Arguments A and B above embody just such an unjustified transition . Both presume that because individuals in the initial choice situation are characterized abstractly and/or self-interestedly, then it must be the case that a society informed by Rawlsian substantive pri�ciP_les would be _ volves a a omistic and/or egoistic. As it stands, this line of . obJectiOn m _ that Simple misunderstanding of Rawls' position, a misunderstanding seems to be without excuse given Rawls' explicit attempt to set the matter s traight :
�
Notes and Discussion
235
1 13
ized The fact that in the original position the parties are character ' entail that as not interested in one another s concerns does not be persons in ordinary life that hold the · principles that would T motiva he . . . · agreed to are similarly disinterested in one another.
tion of the persons in the original position must not be confused the with the motivations of persons in everyday life that accept sense g correspondin the have and chosen be principles that would of justice ( p.
148).1
nt expres· Rawls here explicitly anticipates and rejects the lines of argume responds tly sed in A and B. Again, at the very end of his book, he explici alistic individu to the objection that the contract view embodies a narrowly doctrine for the above reasons with the comment that " once the point of the assumption of mutual disinterest is seems misplaced" ( p. 584 ).
understood,
the objection
How then does one explain this basic misconstrual of Rawls, a
mis·
construal that revolves around the role of " the people in the original position" ?
It is my contention that the misunderstanding grows out
of
a false concretization of his " people in the original position." Concretely, there simply are no people in the initial choice situat ion. This is simply a figurative way of talking about constraints that Rawls maintains are reasonable to put on any argument for one view of justice ove r another:
in To say that a certain conception of justice would be chosen the original position is equivalent to saying that rational delibera· a tion satisfying certain conditions and restricti ons would reach d certain conclusion. If necessary, the argument to this result co be set out more formally. I shall, however, spea·k throughout m . terms of the notion of the original position. It is more economical and suggestive (p . 1 3 8).
�
All
�
repltraS e talk about people in the original position could ea sily be . ct our ed m terms of a rubric n t res will like, " In this discussion we argum ents to thes� kinds course, of premises for these reasons." And, of ange there may be many reason arr s for restricting arguments for a social do . ment to those of a to hin g · certam k md, ' . which reasons may have not , . Ut bo a With one s VIews about r o ty the relations of real individuals to socie . what motivates real indi viduals. Rawls is , after all, attemp ting stice �d to const ruct a theory of ju the ge eral canon s cert� at h t of � theory construction may w ell urge constramts be put Moreover' m on the kinds of premises employed . adelitl· onal constraints se t on the premises. The aim is not si mply to pre � that 1t a correct account y of the matter but to present it in such a wa thUS will be convmc . mg . as opposed to other accounts of the matter and •
I Rawls, A Theory of Jus tice (Cambndge : Harvard University Pre ss,
Jo�
1 14
1971 ).
rences All the page refe paper are to this book.
236
in
thiS
function as an effective guide for action in the public forum.
Given
these various considerations, then, there would seem to be more to a
good argument in political philosophy than that the premises be true and that the conclusions follow from them.
For theoretical reasons we
would want the premises to be as minimal as possible and for practical
reasons we would want them to be acceptable to those to whom the argument is directed. So, even from a theoretical point of view there is much to recommend that arguments for an altruistic conclusion not be based on altruistic premises, and from a practical point of view there is much to recommend that arguments directed against utilitarians assume no more than self-interest in the premises. These concerns for theoretical power and practical effectiveness come together to recommend minimal istic assumptions. It would be a mistake to infer from this that Rawls' conclusions must be similarly minimalistic. opposite effect.
The whole strategy is to the
Furthermore, the inference from these restrictions on the premises of his argument ( that is, his characterization of the people in the original position) to claims about what must be his views on the nature of society and the motivations of persons seems doubly gratuitous in the face of Rawls' suggestion that he could derive the same conclusions from social and altruistic premises : The combination of mutual disinterest and the veil of ignorance achieves the same purpose as benevolence. . . . Furthermore, this pair of assumptions has enormous advantages over benevolence plus knowledge. As I have noted, the latter is so complex that no definite theory at all can be worked out. . . . The combination of mutual disinterest plus the veil of ignorance has the merits of simplicity and clarity while at the same time ensuring the effects of what are at first sight morally more attractive assumptions. And if it is asked why should one not postulate benevolence with
the veil of ignorance, the answer is that there is no need for such a strong condition. Moreover, it would defeat the purpose of ground ing a theory of justice on weak stipulations (pp. 14849). Here Rawls suggests that it is not even necessary that he characterize the ini tial choice situation in an "individualistic" manner to get his con clusions. The conclusions would follow from a characterization in terms of either "kno wledge and altruism ," or "ignorance and altruism," or "ignorance and self-interest." But the first characterization would be unwieldy and the second unnecessarily strong. So for both theoretical ' and prac tical reasons it is methodologically preferable to characterize the initial choice situation in the third manner. It is clear that this is
a metho dological stipulation made with good reason rather than the _ expression of substantive beliefs about social reality or human motiva tion. In fact, one might say that the whole point of he �roject s to show that even from s uch " individualistic" premises one IS driven to non-
�
Notes and Discussion 237
!
115
individualistic" conclusions. Hence, to point to his individualistic pre mises and to infer from these that he must therefore have individualistic
conclusions is to condemn the project without a hearing and a fortiori without an argument.
II. Of course, it does not follow from the unacceptability of arguments
A and B that Rawls' substantive position is not individualistic in the two relevant senses.
The question is still open as to whether or not it em·
bodies an atomistic conception of society and/or an egoistic conception
c
of human motivation.
I would like to argue in this section that in fa t the position is not individualistic in either of these senses in that the structures and values of community are important ingredients in his view.
That it is Rawls' intention to provide a central place for these dimen·
sions of community in his account he ma'kes quite explicit:
Justice as fairness has a central place for the value of community. . . . The essential idea is that we want to account for the social
�d
values, for the
intrinsic good of institutional, community, associative activities, by a conception of justice that in its theore�I� basis is individualistic. . . . From this conception, however mdi·
vidualistic it might seem, we must eventually explain the value of community.
264-65 ).
Otherwise the theory of justice cannot succeed (pp.
Far from ignoring the social nature of man and the altruistic features de a ?f �m� motivation, the intention seems clearly to be to prov e JUstlflcatiOn for these that does not beg the crucial questions at Is.su : O lO u ec But of ourse, there may be many slips between intent ion and ex � � : ten� � so It might still be argued that in spite of these express ed in Rawls' substantive view may in fact remain narrowly individualis tic 10 o�e or both senses. In my the text opinion a careful consideration of wlll not support this inte rpretation. wls take With regard to the charge of atomism not only does Ra ds the cognizance of the social and historical nat re of man but he fin an hum co munitarian claims � made in reference to these feature s of existence totally uncont roversial :
�
�
o
;
'
u'
CI' tizens The soci'al system shapes the wants and aspir ations that ItS t e come to have. It determi th Y w nes in part th e sort of persons to be a well as the sort of persons they are. Thus an � eXIS �g system IS not only an in stitu tional device for satis fying ants wants and needs but a way of creating and fas hionin g w ese the . future. How men work togeth er now to satis fy t heir �r f s d in desires affects the desires they will have later on, the k persons they will be. prefectlY These matters are, of cou rse , ·
� eco�o�nc
� :
1 16 238
obvious and have always been recognized. They are stressed by economists as different as Marshall and Marx (p. 259). Rawls thinks that most theories of j ustice accept as obvious any general statement of the socio-historical nature of man. One of the points of his project is to show that his theory (as opposed to other accounts of justice) takes these features of man and society seriously and develops them in a thoroughgoing manner. He begins by distinguishing between instrumental and essential con struals of the sociability of human beings, construals which result in what we might term shallow and deep communitarianism respectively: Now the sociability of human beings is not to be understood in a trivial fashion. It does not imply merely that society is necessary for human life, or that by living in a community men acquire needs and interests that prompt them to work together for mutual ad vantage in certain specific ways allowed for and encouraged by their institutions. Nor is it expressed by the truism that social life is a condition for our developing the ability to speak and think, and to take part in the common activities of society and culture. No doubt even the concepts we use to describe our plans and situa tion, and even to give voice to our personal wants and purposes, often presuppose a social setting as well as a system of belief and thought that is the outcome of the collective efforts of a long tradi· tion. These facts are certainly not trivial ; but to use them to characterize our ties to one another is to give a trivial interpreta tion to human sociability. For all these things are equally true of persons who view their relations purely instrumentally (p. 522 ). The contention here is that on his account the social nature of man is taken essentially and deeply in that the common institutions and activities are valued as goods in themselves and not merely as instrumentalities for the increase of more basic individual goods. From Rawls' perspective "we need one another as partners in ways of life that are engaged in• for their own sake, and the successes and enjoyments of others are neces sary for and complimentary to our own good" (pp. 522-23). The historical nature of man is taken as seriously as the social : Our predecessors in achieving certain things leave it up to us to pursue them further ; their accomplishments affect our choice of endeavors and define a wider background against which our aims can be understood. To say that man is a historical being is to say that the realizations of the powers of human individuals living at any one time takes the cooperation of many generations (or even societies ) over a long period of time. It also implies that this cooperation is guided at any moment by an understan ing of � at has b een done in the past as it is interpreted by soc1al trad1t10n
�
?
( pp. 524-25).
Notes and Discussion
239
1 17
Man is seen as essentially an historical being whose very potentialities are fashioned by the historical process. Rawls draws an interesting con· trast here between men and lower animals, tying the potentialities of the former to history in ways that those of the latter are not linked. The range of realized abilities of individual men is very much a function of
historical development whereas that of lower animals appears relatively impervious to the passage of time. Finally, Rawls underlines the point that this historicity of man is not to be construed merely instrumental but is to be seen as an essential feature of human nature. It is precisely this deep interpretation of the socio-historical nature of man that Rawls' contract theory is designed to take into account.
The question of egoism Rawls discusses in much greater detail. Returning to the question of the difference between the motivation of the hypothetical persons in the original position and that which would actually follow from the acceptance of his two principles, Rawls draws the distinction sharply: Clearly the two principles of justice and the principles of obliga· tion and natural duty require us to consider the rights and claims of others. And the sense of justice i s a normally effective desire to comply with these restrictions. The motivation s of persons in the original position must not be confused with the motivations of
persons in everyday life who accept the principles that would . be chosen and who have the corresponding sense of justice. In practi cal affairs an individual does have a ·knowledge of his situation and he can, if he wishes, exploit contingencies to his advantage. Should his sense of justice move him to act on the princip les of right that would be adopted in the original position, his desires and aim s are surely not egoistic. He voltmtarily takes on the li mitations exp res· sed by this interpretation of the moral point of view (p. 148).
two T�e �otiv�tion of actual persons in a society informed by Rawls' �e pnnclples . lS clearly not that In . ors of the hypothetical cont ract hl�h actual SOCiety people would w e ustic f j be motivated by the sense o is would be the embodiment of the two principles and in virtue of th on sense of justice the peopl cti e e would have " ties of entiment and aff and wo�d want to advan en�s ce the interests of others and see their . thiS attamed (p. 129). Far h wit from being egoistic then indi viduals In sense of justice would y. wa c be motivated in a ecide ly alt ruisti al of �act, R�wls sees this sense e id al sic of justice as embodying the clas :atermt� to the degree bers that one can reasonably expect between mem f a S�lety at �arge (compare p. 106 ). . nus pomt lS further nt of driven horne in Rawls' explicit treatme . the psychology of . s of . rual . m t'IVatlon Ou1te contrary to egms tlc const . ple JUs tice, Rawls maint . . . peo a ams that 1f 1t were a psychological fact th t could only purs ue their own interests they would never develop an . effect1ve sense shoW · t'lee. of JUS Accordingly he takes it upon himself t0
�
d
·
0
·
·
'
l l8
240
d
how one could learn to adopt an altruistic perspective as a condition for the stability of his sense of justice.
Toward this end he articulates
an account of moral development that involves the three stage sequence of the morality of authority, the morality of association, and the morality of principles.
In brief, after the introduction to precepts in the caring
environment of the family, the child gradually learns to take the point of view of others and to see things from their perspective.
Against this
background the development of a genuine sense of justice can be under stood: Once the attitudes of love and trust and of friendly feeling and mutual confidence have been generated in accordance with the two preceding psychological laws, then the recognition that we and those for whom we care are the beneficiaries of an established and enduring just institution tends to engender in us the corresponding sense of justice.
We develop a desire to apply and to act upon
the principles of justice once we realize how social arrangements answering to them have promoted our good and that of those with whom we are affiliated. In due course we come to appreciate the ideal of just human cooperation (pp. 473-74). In this way the virtue of justice takes form and it is this virtue which then informs our actions.
Another consideration that weighs heavily against any egoistic inter pretation of Rawls is his treatment of the moralities of supererogation. He views these not as fundamentally different orientations from that of justice but as continuous with it: Thus the moralities of supererogation, those of the saint and the hero, do not contradict the norms of right and justice ; they are marked by the willing adoption by the self of aims continuous with these principles but extending beyond what they enjoin (p.
479).
If the concern for justice were merely a matter of self-interest, then the moralities of supererogation would have to be seen as a totally dif ferent matter. But Rawls does not see supererogation this way ; it is simply a matter of the concern for others being carried to a further stage. The picture of Rawls has of a society informed by the principles of justice is certainly not the picture of an accidental union of competing self-int erested individuals. It is the picture of a genuine community: Thus we may say following Humboldt that it is through social union founded upon the needs and potentialities of its members that each person can participate in the total sum of the realized natural assets of the others.
We are led to the notion of the community of human
kind the members of which enjoy one another's excellences and individuality elicited by free institutions, and
they recognize the
Notes and Discussion
241
1 19
good of each as an element
in the complete activity the whole all (p. 523}.
scheme of which is consented to and gives pleasure to
It is in the system of justice as fairness as he construes it that "persons will develop a secure sense of their own worth that forms the basis of the love of human kind" ( p.
501).
III. Having argued, first, that the standard arguments for the view that Rawls' position is individualistic in one or both of the pejorative senses of that term are fallacious, and, secondly, that his position is not in
fact individualistic in these senses, I now want to conclude by articulat· ing some senses in which hi'i position is individualistic-and defensibly so . There is a minimal sense in which the theory could be viewed as methodologically individualistic, and two further senses in which it could
be seen to be substantively individualist ic. There is a modest kind of methodological individualism embodied in the constraints Rawls puts on his arguments for his theory of justice ; namely, his general adoption of the contract model and, more specifically, his minim alist characterization of the people in the initial choice situa· tion. However, on my reading of Rawls this does not commit him at . all to a full-blown methodological individualism ( the view that all social phenomena are to be explained in terms of facts about individuals ) but can be construed simply as the adoption of the most effective strategy ot for mounting an argument against utilitarianism It m ay or may � . on be � ffective base for arguing against perfection ist or Marxis t tbe es � of � ustice, but this is beside the point. Utilitarianism is the only fully tly �n· articulated theory of justice in the field, and thus Rawls is righ cerned ttr focus his present ation so as to provide a genuine and a acttve ult. We alternative to the view that will otherwise hold sway by defa ook: should take seriously his stated b the aim for the argument of
�0
offers, And �bile, of course, it is not a fully satisfac tory theory, it r so fo I beheve, an alternative as to the utilitarian view whi ch h h�ve I lo?g held the pre-eminent phy . place in our moral philoso doctrme tned to pres ent the theory of justice as a viable sys tematic b so that the idea of maximiz sw Y y ing the good d oes not hol d � t ui default. e criticis m of teleological theories cannot fr er proceed Piecemeal h . We must attempt to construct anoth t u view which has the same virtues of clarity and sys tem b s . ns ibilltie ' Yields a more d Iscnmm atmg mterpretation of our mora1 se (p. 587).
�
!h
0�
·
?I.e �·�
·
·
�
The ov rall thrust of theorY the book is to show that a finally adequate � . . of JUStic s . . . s than ·m e e (if such exists Will look more hke JUS tice as fai ) . d . . hke util anatecnJ itari�msm. eJ stra G'Iven this aim, the general contract . . tu . si the mim m ahst characterization ice of the people in the ini tial cho ·
·
.
•
120 242
tion seem to be constraints on argument forms ideally suited to the project.
They need not be read as involving any general commitment
even to methodological individualism. There are at least two senses in which Rawls' theory can be seen to be substantively individualistic.
In the first place, the theory is an attempt
to give an account of the just state that accommodates the possibly different conceptions of human fulfillment of its individual members rather than an account that prescribes one overall ideal of human fulfill ment.
Many . see this ·kind of substantive pluralism as individualistic,
and it may well be ; but it seems to me to be a !kind of individualism that is perfectly defensible on both theoretical and practical grounds. People are different; they have different life projects and different kinds and degrees of satisfaction.
The structure of society should acknowledge
rather than ignore this fact of life.
Moreover, the recognition of this
fact does not involve the relativistic view that all conceptions of human fulfillment are equally good but simply the acknowledgment that given the kinds of beings we are there is a range of plausible ideals of human fulfillment and that it is part of the common respect and concern wo owe each other to foster self-determination rather than prescribe con formity with regard to overall life projects.
But even if the case were
otherwise, that is, if there was but one rather specific ideal of human fulfillment that was philosophically defensible, the individualism under discussion here would still be recommended on practical groWids.
We
do in fact live in a pluralistic society ; so our account of justice, if it is to be an effective force in this society, should address itself to this reality.
We should, of course, simultaneously engage the more basic
moral questions that may underlie this issue ; but all reflection on justice need not wait upon the resolution of these deeper problems.
This being
the case, should we want our philosophical reflection on justice to play an active role in the improvement of the human condition rather than be merely idle speculation, then our philosophical theories at this level should speak to the pluralistic realities of our concrete situation.
Not only do I see this kind of individualism as defensible, but I am incline d to think that the only serious objection in this area is to the effect that in this regard Rawls is not individualistic enough. It has been objected that the minimal list of primary goods is not neutral vis-a vis alternative life projects but rather implicitly prescribes a single ideal of human excellence. Specifically, the charge is that in its focus on libe rty (particularly political liberty) and in its employment of what
might be termed the acquisitive principle (one would prefer more rather th an les s ) Rawls' theory comes up short of genuine pluralism and involves the prescription of one ideal of human excellence. This objection has to be taken seriously, but I think that there is on balance an adequate Rawlsian reply. The list of primary goods can be plausibly defended as providing simply the necessary conditions for the realization of any
Philosophically defensible ideal of human excellence and the operation on them ( more rather than less ) regarded as the most reasonable attitude
Notes and Discussion 243
121
to adopt "in the original position• ; and, as we have seen, this in no way involves the recommendation of acquisitiveness as a virtue in the real world. Again, it can be defended on methodological rather than sub stantive grounds.
There is a second sense in which Rawls' account is substantively individualistic, namely, the sense in which any " rights theory" will defend the moral inviolability of the individual person against all collectivist conceptions of the just society whether they be of the classical perfec tionist, utilitarian, or Marxist variety. Rights theories are individualistic in the sense that they maintain there can be no justifiable trade-offs of the life prospects of some just so others can be better off. Far from being an objection, however, this seems to me to be an
Surely the fundamental objection to all "common-good" doctrines of whatever variety has been their disenfranchisement of the individual as over against the totality, whether that totality be a matter of the "numbers" in democracies or of "vision" in the various perfection· ist ideologies. Rawls' project is an attempt to articulate a conception of justice that takes into account the dimension of social responsibility without violating the dimension of moral autonomy. Again I am inclined to think that the only serious objection in this area is one to the effect that Rawls ' is not individualistic enough. The locus of complaint is the difference principle, and the specific concern is that Rawls' suggestion that we think of individual talents as common holdings of the society drags him back into the collectivism of his 0P" ponents. I believe that this concern can be allayed, however, by a closer e scrutiny of the text. What Rawls says is that " the difference principl ral represents, in effect, an agreement to regard the distribution of ?a� talents as a common asset and to share in the benefits of this distnbu·
endorsement.
claimin!
tion h � �t�ver it turn s out to be" ( p . 101 ) . Rawls is not here that mdivtdual natural talents are common assets but simply that on way of thinking about the difference principle is in terms of "an agree the ment to regard• natural talents It is still as a common asset. agree is indivi�ual who autonomously agrees to so regard them, and h n t ment IS a manifestation of the conviction that individual rights are � e th restricte d but must be tempered so as not to undermine � i r�ghts 0 others . Rawls' ce whole program is an attempt to balan md · VIdual nghts and social respo nsibility . An overall evaluati. on of Rawls' project would involve the en'tical . in evaluation of numerous facets of the theory not even mentioned tandar paper. What I hope to have shown however t' s that many of the s ' aW on bJ�C dr ' rtons to it are �nd�entally misguided. Objec tions which vertY po rwsm e � �bout the socio-historical nature of man and/or about th ways of egoistic motivational he t in theories can, I believe, be blocked . n t S I have suggested here. This is not to say' of course' that there I r . r o in de · some profounder obJectio n even in this very neighborhood . But ch for the · · . . 0bJectton to be relevant it will have to be made mu specific-and thereb Y much more contentious.
�as��
�
�
thi�
'
o
Dlore
122
244
AMY GUTMANN
Commun itarian Critics of Libera l ism
We are witnessing a revival of communitarian criticisms of liber al political theory. Like the critics of the 1 g6os, those of the 1 g8os fault liberalism
v w_� e
for being mistakenly and irreparably individu alistic. B ut the new of criticism is not a mere repetition of the old . Whereas the earlier cnucs were inspired by M arx, the recent critics are inspired by Aristotle and Hegel. The Aristotelian idea that justice is rooted in "a commu nity primary bond is a shared understanding both of the good for man an.d hiS the good of that community" explicitly informs Alasdair M aci ntyre in 1 rt s criticism of John Rawls and Robert N o zick for their neglect of de e ; nd f d and Ch arles Taylor in his attack on "atomistic" liberal s who " try to e e e e h T the priority of the individual and his rights over soc iety. "" . I p elian conception of man as a historic ally condition d eing irn �C � � be mforms both Roberto Unger's and M ichael Sandel's reJe cuon of the li
whose
·
·
·
?
view of man as a free and rational being. 3
��
J
Li Th.ts revtew essay concentrates on the arguments presented in Michael S ande1• · "MoralltY 1 982 ); Sandel, beralism
!,��:
and the Limits of]ustice (New York: Cambridge University Press , c and the Uberal ldeal ," The New Republic, May 7. 1 g84. pp. I S- I ? ; Alasd ait M a : After Vt. rtue (Notre Dame: Notre Dame University Press 1 98 1 ) · and Macintyre, h sop Y: triotism a Virtue?" The Lindley Lecture (University of K sas: D partm ent of Ph ilo ocra : 0em ng Stro March 2.6, 1 98 r, arbe 4). Other works to which I refer
�
�
�
'
Benjamin B Participatory Politics for a New Age (Berkeley : University of California Press. 1 g84 ssays reedams ·· E�1• · Charles TayIor, "Atonusm. "m · Alkis · Kontos, ed., Powers, Possessi ons and F . tn Honor of C. B. Macpherson (Toronto: University of Toronto Press , 1 979), PP· 3 ard e and "The Diversily of Good s," in Utilitarianism and Beyond, ed. Am artya Sen and B jra ng Williams (New York: Cambridge University Press, 1 982) , pp. 1 29-44; Roberto Ma a��eres Unger Knowledge and Politics (New York: Free Press, 1 975) ; and Michael Walzer. Sp ofJushce (New York: Basic Books, J g83). I. Macintyre, After Virtue, pp. 232_ 33. 2. "Atomism," p. 39. are
�
•.
J. Knowledge and Politics, pp. 85, I Q I-23 1 Limits, ; pp. 1 7g-8o.
246
309
Communitarian Critics of Liberalism
The political implications of the new communitarian criticisms are correspondingly more conservative. Whereas the good society of the old critics was one of collective property ownership and equal political power, the good society of the new critics is one of settled traditions and estab . lished identities. For many of the old critics, the role of women within the family was symptomatic of their social and economic oppression; for Sandel, the family serves as a model of community and evidence of a good greater than justice. 4 For the old critics, patriotism was an irrational sentiment that stood in the way of world peace; for Macintyre, the par ticularistic demands of patriotism are no less rational than the univer salistic demands of justice.s The old critics were inclined to defend de viations from majoritarian morality in the name of nonrepression; the new critics are inclined to defend the efforts of local majorities to ban offen sive activities in the name of preserving their community's "way of life and the values that sustain it. "6 The subject of the new and the old criticism also differs. The new critics recognize that Rawls's work has altered the premises and prin ciples of contemporary liberal theory. Contemporary liberals do not as su me that people are possessive individualists; the source of their indi vidu alism lies at a deeper, m ore metaphysical level. According to Sandel, the problem is that liberalism has faulty foundations: in order to achieve ab solute priority for principles of justice, liberals must hold a set of im plausi ble metaphysical views about the self. They cannot admit, for ex ample, that our personal identities are partly defined by our communal
attachments. 1 According to Macintyre, the problem is that liberalism lacks any foundations at all. It cannot be rooted in the only kind of social life that provides a basis for moral judgments, one which "views man as having an essence which defines his true end. "8 Liberals are therefore
bou nd eith er to claim a false certainty for their principles or to admit that morality is merely a matter of individual opinion, that is, is no morality at all . The critics claim that many serious problems originate in the foun
dational faults of liberalism. Perhaps the most troubling for liberals is 4· Sandel, Limits, pp. 3o-3 1 , 33-34, 1 6g. 5· "Is Patriotism a Virtue?" pp. 15-1 8 and passim. 6. Sandel, "Morality and the Liberal Ideal," p. 1 7. 7· Limits, pp. 64�5. 1 68-73. 8. After Virtue, p. 5:-l.
247
Philosophy & Public Affairs
3IO
their alleged inability to defend the basic principle that "individual rights
cannot be sacrificed for the sake of the general good. "g Because Sandel
and Macintyre make the most detailed and, if true, devastating cases against believing in a liberal politics of rights, I shall focus for the rest of this review on their arguments. The central argument of Sandel's book is that liberalism rests on a series of mistaken metaphysical and metaethical views: for example, that the claims of justice are absolute and universal; that we cannot know each other well enough to share common ends; and that we can define our personal identity independently of socially given ends. Because its foundations are necessarily flawed, Sandel su ggests in a subsequent ar ticle that we should give up the "politics of rights" for a "politics of the common good. " 1 0 Macintyre begins his book with an even more "disquie ting suggestion":
that our entire moral vocabulary, of rights and the common good, is in
such "grave disorder" that "we have-very largely, if not entirely-lost our comprehension, both theoretical and practical, of morality. " u To ac
count for how "we" have unknowingly arrived at this unenviable social condition, Macintyre takes us on an intriguing tour of moral history, from Homeric Greece to the present. By the end of the tour, we leam that the internal incoherence of liberalism forces u s to choose "Nietzsche or Ar istotle," a politics of the will to power or one of commu nally defined virtue. n THE LIMIT S OF COMM UNITARIAN CRITI CISM
Do the critiques succeed in undermining liberal politics? If the onlY eed foundations available to liberal politics are faulty, then perh aps one n e not establish a positive case for communitarian politics to establish iS th laim that liberal politics is philosophically indefen sible . 1 3 Alth ou h lS the logic of Sandel's claim concerning the limits of liberal ju s uce, he . . gives no general argument to support his conclusion that lib . eral ngh�
�
�
�
9· Sandel, "Morality and the Liberal Ideal ' " p . 1 6. 1 0 . Ibid., p . 1 7. 1 1.
After Virtue, pp.
1_5.
. . . . then San.delg ' 3· I say "perhaps" beeause if defens1bili ty is relative to our alternatives, . aiJnlll still would have to establ ish the positive case for communitarian politics before ci that the faulty foundation s of liberal politics render it indefensible . 12. Ibid.,
pp. 49, 1 03-1 3 , 238-45 ·
248
31 I
Communitarian Critics of Liberalism
are indefensible. '4 He reaches this conclusion instead on the basis of an interpretatio n and criticism of Rawls's theory, which he reasonably as sumes to be the best theory liberalism has yet to offer. Sandel argues that despite Rawls's efforts to distance himself from Kantian metaphysics, he fails. Sandel attributes Rawls's failure to his acceptance of the "central claim" of deontology, "the core conviction Rawls seeks above all to defend. It is the claim that 'justice is the first virtue of social institutions . ' "•s As Rawls presents it, the "primacy of justice " describe s a moral requirement applicable to institutions. Sandel interprets Rawls as also making a metaethical claim: that the foundations of justice must be independent of all social and historical contingencies Without being transcendental. •6
Why saddle Rawls's moral argument for the primacy of justice with this meaning ? To be sure, Rawls himself argues that "embedded in the Principles of ju stice . . . is an ideal of the person that provides an Archi
medean point for j ud ging the basic structure of society. " • 1 But to translate this pass age into a claim that the grounds ofjustice can be noncontingent
14. The general argument that can be constructed from Sandel's work (using his con ceptual framework) is, I think, the following: ( x) To accept a politics based on rights entails believing that justice should have absolute priority over all our particular ends (our con ception of the good); (2) To accept the priority of justice over our conception of the good entails believing that our identities can be established prtor to the good (otherwise our conception of the good will enter into our conception of justice); (3) Since our identities are constituted by our conception of the good, justice cannot be prior. Therefore we cannot consistently believe 1n the politics of rights. But each of the steps in this argument are suspect; ( 1 ) We may accept the politics of rights not because justice Is prior to the good, but because our search for the good requires society to protect our right to certain basic freedoms and welfare goods; (2) Justice may be prior to the good not because we are "anteced ently individuated," but because giving priority to justice may be the fairest way of sharing the goods of citizenship with people who do not accept our conception of the good; (J) Our identities are probably not constituted, at least not exclusively, by our con· ception of the good. If they were, one could not intelligibly ask: "What kind of person do I want to become?" Yet the question reflects an Important pan (although not necessarily the whole) of our search for identity. If, however, we assume by definition that our identities are constituted by our good, then we must consider our sense of justice to be pan of our identities. My commibnent to treating other people as equals, and therefore to respecting their freedom of religion, is just as elemental a part of my identity (on this understanding) as my being Jewish, and therefore celebrating Passover with my family and friends. 15. Limits, p. 1 5. Emphasis added. See A Theory of Justice (Cambridge, MA: Harvard University Press, 1971 ), pp. 3-4 . 58 6. 1 6. Limits, pp. 16-1 7. Rawls must, in Sandel's words, "find a standpoint neither com Promised by its implication in the world nor dissociated and so disqualified by detachment." 17. Rawls, A Theory of justice, p. 584; see also pp. 26o-62.
249
Philosophy & Public Affairs
312
ignores most of what Rawls says to explain his Archimedean point, the nature of justification, and Kantian constructivism. ' 8 "justice as fairness is not at the mercy, so to speak, of existing wants and interests . It sets up an Archirnedean point . . .
without invoking a priori considerations."'9
By requiring us to abstract from our particular but not our shared inter ests, the original position with its "veil of ignorance" and "thin theory of the good" avoids reliance on both existing preferences and a priori con
siderations in reasoning about j ustice. The resulting principles of justice, then, clearly rely on certain contingent facts: that we share some interests (in primary goods such as income and self-respec t), but not others (in a particular religion or form of family life ) ; that we val ue the freedom to choose a good life or at least the freedom from h aving one imposed upon us by political authority. If we do not, then we will not accept the con
straints of the original position. Rawls's remarks on j ustification and Kantian constructivism make ex plicit the contingency of his principles of j ustice. The design of the orig inal position must be revised if the resulting principles do not "accom
modate our firmest convictions. "'10 Justification is not a matter of deduction from certain premises, but rather "a matter of the mu tual support of many considerations, of everythin g fitting toge ther into one coherent view."2' Since Rawls accords the view "that j ustice is the first "l2 virtue of social institutions" the status of a " common sen se conviction , wls this view is part of what his theory must coherently combine. Ra therefore does not, nor need he, claim more for j ustice as fairness It that "given our history and the traditions embedded in our public life,
th�
'
ry of 1 8· In interpreting Rawls I rely (as does Sandel) on passages from both A Theo �jl " Tht · · res J ust zce and "Kantian Constructivism in M oral Theory: The Dewey Lectu 1 �0• 1 Journal of Philosophy 77. no. 9 (September J g8o), pp . 5 1 5-72- Someone mi ght reaso a fen nl argue that not until "The Dewey Lectures" does Rawls consistently and clearl� � po tion on Justification that I attribute to him. Had Sandel directed his cnttcJsm 0 e5 agamst A Theory of Justice, his interpretation would have been more credible. But h more could not have sustained his central claim that awls's principle s and liberalism R generally must rest on Implausible metaethical grounds. , esp. 1 9 · A Theory of Justice, p. 261 . Emphasis added. See also "The Dewey Lectures, pp. 564-67. �� 20· A �ory of Justice, p. 20. The reasoning is circular, but not viciou sly 50• 51 SJ'dered con must also be prepared to revise our ur o ch weaker judgments when principle s mat convtcti�s. until we reach "refle ctive equilibrium. .. 2 1 . lb1d., pp. 2 l , 579 22. Ibid p. 5ss.
�
�
.
.•
250
� �! �
3 13
Communitarian Critics of Liberalism
is the most reasonable doctrine for us. We can find no better charter for our social world. "23 Rawls could be wrong about our firmest convictions or what is most reasonable for u s. But instead of trying to demonstrate this, Sandel argues that Rawls must show that the content and claims of justice are inde
pendent of all historical and social particularities. 24 If this is what con
stitutes deontological metaphysics, then it is a metaphysics that Rawls explicitly and consistently denies. What metaphysics must Rawlsian liberalism then embrace? Several commentators, along with Rawls himself, have argued that liberalism does not presuppose metaphysics. 25 The major aim of liberal j ustice is to
find principles appropriate for a society in which people disagree fun damentally over many questions, including such metaphysical questions as the nature of personal identity. Liberal justice therefore does not pro vide us with a comprehensive morality; it regulates our social institutions, not our entire lives. It makes claims on us "not because it expresses our deepest self-understandings, " but because it represents the fairest pos sible modu s vivendi for a pluralistic society. 26 The characterization of liberalism as nonmetaphysical can be mislead
ing however. Although Rawlsian justice does not presuppose only metaphysical view, it is not compatible with
one all such views. Sandel is
correct in claiming that the Kantian conception of people as free and equal is incompatible with the metaphysical conception of the self as "radically situated" such that "the good of community . . . (is) so thor
ough goin g as to reach beyond the motivations to the subject of motiva tions. "27 Sandel seems to mean that communally given ends can so totally
constitute people's identities that they cannot appreciate the value of justice. Such an understanding of human identity would (according to 23. "The Dewey Lectures," p. 5 1 9. Cf. Sandel, Limits, p. JO. 24. Limits, p. 30. Sometimes Sandel comes close to making a more limited but potentially more plau sible argument-that Rawls derives his principles of justice from the wrong set of historical and social p articularities: from (for example) our identification with all free and rational beings rather than with particular communities. Such an argument. if suc· cessful, would establish different limits, and limits of only Rawlsian liberalism. 25. See Rawls, "The Independence of Moral Theory," Proceedings and Addresses of the American Philosophical Association 48 ( 1 975). pp. 5-22 26. Charles Larmore, "Review of Liberalism and the Limits of justice," The journal of Philosophy 8 1 , no. 6 (June 1g84): 338. See also Rawls, "The Dewey Lectures," p. 542. 27. Sandel. Limits, pp. 2o-: u ; 1 49.
251
Philosophy & Public Affairs
constructivist standards of verification) u ndermine the two principles.l8 To be justified as the
political ideals most consistent with the "public
culture of a democratic society,''29 Rawlsian principles therefore have to express some (though not
all) of our deepest self-understandings. Rawls
must admit this much metaphysics-that we are not radically situated selves-if justification is to depend not on "being true to an order ante· cedent to and given to us, but . . . [ on] congruence with our deeper understanding of ourselves and our aspirations. "3o If this, rather than Kantian dualism, is the metaphysics that liberal j ustice must admit, Sandel's critique collapses. Rawls need not (and he does not) claim that "j ustice is the first virtue of social institutions"
in
aU societies to show that the priority of j ustice obtains absolutely in those
societies in which people disagree about the good life and consider their ·
freedom to choose a good life an important goodY Nor need Rawls assume that human identity is
ever totally independent of ends and relations to always command our moral alle
others to conclude that j ustice m u st
giance unless love and benevolence make it unnecessaryY Deontological justice thus can recognize the conditional priority of j ustice without em· bracing "deontological metaethics" or collapsing into teleology. Sandel has failed therefore to show that the foundations of rights are mistaken.
MISSING fOUNDATIONS? Macintyre argues that the foundations are missing : The best reason for asserting s o bluntly that there are no su ch rights is indeed of precisely the same type as the best reason whi ch we possess for assertin g that there are no witches . . . : every atte mpt to give good reasons for believing there are such rights has failed.33
The the
The analogy , properly drawn, does not support M acinty re's position. t b t reason that people can give for believin g in witc hes is th a noJJl· eXIsten ce of witche s explains (supposed ly) observed physical phe
e:'
� :��= �:
�J;
�w�
tic s l Th e · Lectures," pp. 534-3 5, 564 -67. S ee also A Theory of ]us ·. 2 h tn t . be." ry of JU s tice does, indeed, presuppose a t heory of t he good, bu t Wl t limi ts this does not PreJu dge the choice of the sort of persons that men w ant to (EmphasJ.s . added.) lQ. Rawls, "The Dewey Lectures," p. 5 1 8. 3 o. Ibid., p. 51 9· Ibid., pp. 5I 6-24 . Cf. Sandel, Lim its, pp. 2g_4 0 . Theory of Justice, pp. wls , 56o-77 . Cf. Sandel , Limits, pp. 4 7-65 . · �' er V trtu e, p. 67.
3�· 333·��
�
252
315
Communitarian Critics of Liberalism
ena. Belief in witches therefore directly competes with belief in physics, and loses out in the competition. The best reason for taking rights seri ously is of a different order: believing in rights is one way of regulating and constraining our behavior toward one another in a desirable manner. This reason does not compete with physics; it does not require us to believe that rights "exist" in any sense that is incompatible with the "laws of nature" as established by modern science.J4 Macintyre offers another, more historical argument for giving up our belief in rights. "Why," he asks, "should we think about our modern uses of good,
right, and obligatory in any different way from that in which we
think about late eighteenth-century Polynesian uses of taboo?"Js Like the Polynesians who used
taboo without any understanding of what it meant human right without understanding its
beyond "prohibited," we use
meaning beyond "moral trump. " If the analogy holds, we cannot use the idea c orrectly because we have irretrievably lost the social context in which its proper use is possible. But on a contextualist view, it is reasonable for
us to believe in human
rights: many of the most widely accepted practices of our society�q ual ity of educational opportunity, careers open to talent, punishment con ditional on inten t-treat people as relatively autonomous moral agents. Insofar as we are committed to maintaining these practices, we are also committed to defending human rights.J6 This argument parallels Mac Intyre's contextualist defense of Aristotelian virtue: that the established practices of h eroic societies supported the Aristotelian idea that every human life has a socially determined
telos. Each person had a " given
role and status within a well-defined and highly determinate system of roles and statuses," which fully defined his identity: "a man who tried to withdraw himself from his given position . . . would be engaged in the enterprise of trying to make himself disappear. "37 If moral beliefs depend upon supporting social practices for their va lidity, then we have more reason to believe in a liberal politics of rights than in an Aristotelian politics of the common good. In
our society, it
does not logically follow that: "I am someone's son or daughter, someone 34· I am grateful to Thomas Scanlon for suggesting this reply. 35· After Virtue, p. I 07. 36. We need not be committed to a thoroughly deontological moral apparatus. Sophis· ticated consequentialist theories justify these same practices and are consistent with be lieving in rights. 37· After Virtue, pp. 1 1 7, I J g.
253
Philosophy & Public Affairs
316
else's cousin or uncle; I am a citizen of this or that city, a member of this or that guild or profession ; I belong to this clan, that tribe, this nation[ , ]
hence what is good for me has to be TH E good for one who
inhabits these roles. "38 One reason it does not follow is that none of these roles carri es with it only one socially given good. What follows from "what is good for me has to be the good for someone who was born female, into a first-generation American, workin g-class I talian, Catholic family"? Had Geraldine Ferraro asked, following Sandel, "Who am I ? " instead of"What ends should I choose?" an answer would not have been any easier to come by.39 The Aristotelian method of discovering the good by inquiring into the social meaning of roles is of little help in a society in which most
roles are not attached to a single good. Even if there is a single good attached to some social roles ( as caring for the sick is to the role of a
nurse, or searching for political wisdom to the function of political phi· losophers, let us suppose), we cannot accurately say that our roles de· termine our good without adding that we often choose our roles because of the good that is attached to them. The unencumbered self is, in this sense, the encumbrance of our modem social condition. But the existen ce of supporting social practices is certainly not a suf ficient condition, arguably not even a necessary one, for believin g in liberal rights rather than Aristotelian virtue. The practic es that su pport liberal rights may be unacceptable to u s for reasons that carry more moral weight than the practices themselves ; we may discover m oral reasons (even within our current social understandings) for establishi ng ne practic es that support a politics of the common good. M y point here 15 not that a politic s of rights is the only, or the best, possible politics for our society, but that neither M acintyre's nor Sandel's critique su cceeds in undermining liberal rights because neither gives an acc urate acco u t possi· of their foundations. Macintyre mistakenly denies liberalism the need bility of foundations; Sandel ascribes it to liberalism fou ndations not have.
�
�
THE TYRANNY OF DU ALI SMS
· ee the The en· u·cs mterpr · etive method is also mistaken. I t invite s us to s dent moral universe in dualistic terms: either our identities are ind epen '
· 38· Ibid. , pp. 204-S (emphasis added). Limi ts, P Sandel makes a very similar point in 179· 39· Limits, pp. sB-
sg.
254
317
Communitarian Critics of Liberalism
of our ends, leaving us totally free to choose our life plans, or they are constituted by community, leaving us totally encumbered by socially given ends; either justice takes absolute priority over the good or the good takes the place of justice; either justice must be independent of all
historical and social particularities or virtue must depend completely on the particular social practices of each society; and so on. The critics thereby do a disservice to not only liberal but communitarian values, since the same method that reduces liberalism to an extreme metaphys ical vision also renders communitarian theories unacceptable. By inter
preting Rawls's conception of community as describing "just a feeling,"
for example, Sandel invites us to interpret Aristotle's as describing a fully constituted identity. The same mode of interpretation that permits Sandel to criticize Rawls for betraying "incompatible commitments" by uneasily combining into one theory "intersubjective and individualistic images" would permi t us to criticize Sandel for suggesting that community is "a mode of self-understanding partly constitutive" of our identity. 40 Neither Sandel's interpretation nor his critique is accurate. Macintyre's mode of interpreting modern philosophy similarly divides the moral world into a series of dualisms. The doomed project of modern philosophy, according to Macintyre, has been to convert naturally ego istical men into altruists. "On the traditional Aristotelian view such prob lems do not arise. For what education in the virtues teaches me is that my good as a man is one and the same as the good of those others with
whom I am bound up in human community."41 But the real, and rec ognized , dilemma of modern liberalism, as we have seen, is not that people are n aturally e goistical, but that they disagree about the nature of the good life. And such problems also arise on any (sophisticated) Aristotelian
view, as Macintyre himself recognizes in the context of distinguishing Aris totelianis m from Burkean conservatism: "when a tradition is in good
order it is always partially constituted by an argument about the goods the pu rsuit of which gives to that tradition its particular point and pur pose. "4� The duali stic vision thus tyrannizes over our common sense, which
rightly rejects all "easy combinations"-the individualism Macintyre at40. Ibid. , p. 1 50. When Sandel characterizes his own preferred "strong" vtew of com munity, it is one in which people conceive their identity "as defined to some extent by the community of which they are a part. " (Emphases added. ) 4 1 . After Virtue, pp. 2 1 2-13.
4 2. Ibid., p. 2o6.
255
Philosophy & Public Affairs
tributes to Sartre and Coffman "according to which the self is detachable from its social and historical roles and statuses" such that it "can have no history,"43 as well as the communitarian vision M acintyre occasionally seems to share with Roberto Unger according to which the "conflict between the demands of individuality and sociability would disappear. "44
Because the critics misinterpret the metaphysics of liberalism, they also
miss the appeal of liberal politics for reconcilin g rather than repressing most competing conceptions of the good life.
BEYOND METAPHYsic s : CoMMUNITARIAN PoLITICs Even if liberalism has adequate metaphysical foundations and consid· erable moral appeal, communitarian politics might be morally better. Bu t
Macintyre and Sandel say almost nothing in their books to defend com· munitarian politics directly. Sandel makes a brief positive case for its
comparative advantage over liberalism in a subsequent article. "Where libertarian liberals defend the private economy and egalitarian liberals defend the welfare state, " Sandel comments, "communitarians worry about the concentration of power in both the corporate economy and the bureaucratic state, and the erosion of those intermediate forms of com· munity that have at times sustained a more vital public life." But these worries surely do not distin guish communitarians from most contem· porary liberals, unless (as Sandel implies) communitarians therefore op· pose, or refuse to defend, the market or the welfare state . 45 Sandel makes explicit only one policy difference : "communitarians would be more likely
than liberal s to allow a town to ban pornographic book stores, on the ground s that pornography offends its way of life and the values �at sustain it. " His answer to the obvious liberal worry that such a policY opens the door to intolerance in the name of commun al s tandards is that "intolerance flourish es most where forms of life are dislocated , roots un·
those
settled, traditions undone. " H e urges us therefore "to revitalize in our civic republican possibilities implicit in our tradition but fadin g . tlme."46 bli· What exactly does Sandel mean to imply by the sort of civic repu 43- Ibid., p. 205. See also Sandel, Limits, pp. 40, 1 50. Cf. p. 1 80. 44· �nowledge and Politics , p. 220. 45· Morality and the Libe ral Ideal' " p . 1 7 . 46 . I bid.
256
Communitarian Critics of Liberalism
canism "implicit within our tradition"? Surely not the mainstream of our tradition that excluded women and minorities, and repressed most sig nificant deviations from white, Protestant morality in the name of the common good. We have little reason to doubt that a liberal politics of rights is morally better than that kind of republicanism. But if Sandel is
arguing that when members of a society have settled roots and established
traditions, they will tolerate the speech, religion, sexual, and associational
preferences of �orities, then history simply does not support his opti
mism. A great deal of intolerance has come from societies of selves so "confidently situated" that they were sure repression would serve a higher cause. 47 The common good of the Puritans of seventeenth-century Salem commanded them to hunt witches; the common good of the Moral Ma jority of the twentieth century commands them not to tolerate homosex uals. The enforcement of liberal rights, not the absence of settled com munity, stands between the Moral Majority and the contemporary equivalent of witch h unting. The communitarian critics want us to live in Salem, but not to believe
in witches. Or human rights. Perhaps the Moral Majority would cease to be a threat were the United States a communitarian society; benev olence and fraternity might take the place of justice. Almost anything is possible, but it does not make moral sense to leave liberal politics behind on the streng ths of such speculations. 48 Nor does it make theoretical sense to assume away the conflicts among
competing ends--such as the conflict between communal standards of sexual morality and individual sexual preference-that give rise to the characteristic liberal concern for rights. In so doing, the critics avoid discussing how morally to resolve our conflicts and therefore fail to pro4 7· Sandel may be correct in claiming that more intolerance has come-in the fonn of fascism-from societies of "atomized, dislocated, frustrated selves. " But the truth of this claim does not establish the case for communitarian over liberal politics unless our only ch oice is to support a society of totally "atomized" or one of totally "settled" selves. This dualistic interpretation of our alternatives seems to lead Sandel to overlook the moral value
of establishing some balance between individualism and community, and to underestimate the theoretical diffi culty of determining where the proper balance lies. 48. Sandel might want to argue that societies like Salem were not "settled. " Perfectly settled com munities would not be repressive because every individual's identity would be
fully constituted by the community or completely compatible with the community's un derstanding of the common good. This argument, however, is a truism: a perfectly settled . SOCiety would not be repressive, because perfect settlement would leave no dissent to repress.
257
Philosophy & Public Affairs
3 20
vide us with a political theory relevant to our world. They also may over look the extent to which some of their own moral commitments presup pose the defense of liberal rights.
CoNSTRUCTIVE PoTENTIAL
Even if the communitarian critics have not given good reasons for aban doning liberalism, they have challenged its defenders. One should wel come their work if for no other reason than this. But there is another
reason. Communitarianism has the potential for helping us discover a politics that combines community with a commitment to basic liberal values.
The critics' failure to undermine liberalism suggests not that there are
no communitarian values but that they are properly viewed as supple menting rather than supplanting basic liberal values. We can see the extent to which our moral vision already relies on communitarian values
by imagining a society in which no one does more or less than respect everyone else's liberal rights. People do not form ties oflove and friendship (or they do so only insofar as necessary to developing the kind of character that respects liberal rights). They do not join neighborhood association �, political parties, trade unions, civic groups, synagogues, or churches. This might be a perfectly liberal, arguably even a j u st society , but it is certainl.Y not the best society to which we can aspire. The potential of commuro tarianism lies, I think, in indicating the ways in which we can strive to realize not only justice but community through the many s oc ial unions
of which the liberal state is the super social union. What might some of those ways be? Sandel su gge sts one possibility: states might "enact laws regulating plant closings, to protect their co� munities from th e disruptive effects of c apital mobility and sudden m s gives dustrial change. "49 This policy is compatible with the priority Rawl of the to liberty and may even be dictated by the best interpre tation n difference principle. But the explicit concern for preventin g the disrup o of local communities is an importa riaJllSIIl nt contrib ution of com munita e to li beralism. We should als o' as S andel su ggests be " troubled by th ' of s tendency of liberal progra ms to displace politics from smaller form h fore asSOCiaUon to more comprehe ere nsive ones." But we should not t
�
·
·
49· "Morality and the Liberal Ideal ," p. 1 7.
258
321
Communitarian Critics of Liberalism
oppose all programs that limit-or support all those that expand-the jurisdiction of local governments. We may be able to discover ways in which local commun ities and democracy can be vitalized without vio lating individual rights. We can respect .the right of free speech by op posing local efforts to ban pornographic bookstores, for example, but still
respect the values of community and democratic participation by sup porting local (democratic) efforts to regulate the location and manner in
which pornographic bookstores display their wares. Attuned to the dan gers of dualism , we can appreciate the way such a stand combines uneasily-liberal and communitarian commitments.
Some ways of fostering communal values-I suspect some of the best ways-en tail creating new political institutions rather than increasing the power of existing institutions or reviving old ones. By restoring "those intermediate forms of community that have at times sustained a more vital public life, " we are unlikely to control "the concentration of power in both the corporate economy and the bureaucratic state" that rightly worries both communitarians and liberals.so If large corporations and bureaucracies are here to stay, we need to create new institutions to prevent them from imposing (in the name of either efficiency or expertise) their values on those of potentially more democratic communities . Re alizing the relatively old idea of workplace democracy would require the
creation of radically new economic institutionsY Recently mandated cit izen review boards in areas such as health care, education, and com munity developmen t have increased interest in democratic participation.
Wholehearted political support of such reforms and others yet untried is probably necessary before we can effectively control bureaucratic po wer.s:. Although the political implications of the communitarian criti cis ms of liberalism are conservative, the constructive potential of com
munitarian values is not.
Had they developed the constructive potential of communi tarian val ue s, the critics might have moved further toward discovering both the so. Ibid. 5 1 . For a communitarian defense of economic democracy that is not based on a rejection of li beral values, see Michael Walzer, Spheres of justice, pp. 161 and 291-JOJ. 52. For a suggestive agenda of democratic refonns, see Benjamin Barber, Strong De· mocracy, pp. 26 1_307. Although Barber attacks liberal theory as fundamentall flawed m · • the first nine chapters, the aim of his agenda for reform in the last chapter rs to re ent . ty, not to raze It (p. liberal democracy toward civic engagement and political commuru 308).
�
259
��
322
Philosophy & Public Affairs
limits of Rawlsian liberalism and a better charter for our social world. Instead, Macintyre concludes that we should be "waiting not for a Godot, but for another-doubtless very different-St. Benedict. "sJ The critics
tend to look toward the future with nostalgia. We would be better off, by
both Aristotelian and liberal democratic standards, if we tried to shape it according to our present moral understandings. At the end of his book,
Sandel urges us to remember "the possibility that when politics goes well, we can know a good in common that we cannot know alone. " But he has neglected the possibility that the only common good worth striving for is one that is not "an unsettling presence for justice. "54 j ustice need
not be the only virtue of social institutions for it to be better than anything
we are capable of putting in its place. The worthy challenge posed by
the communitarian critics therefore is not to replace liberal justice, but to improve it.
53· Ibid., p. 2 4 5. Roberto U nger similarly concludes Knowledge and Politics waiting for God to speak (p. 235). 54· Cf. Sandel, Limits, p. 1 83.
I am grateful to Robert Amdur, Michael Doyle, Steven Lukes, Susan M oller Okin, Judith Shklar, Dennis Thompson, Michael Walzer, Susan Wolf, and the Editors of Philosophy & Public Affairs for their helpful suggestions.
RAWLS, HEGEL, AND COMMUNITARIANISM SIBYL A. SCHWARZENBA CH
Baruch College, City University
F
ROM
ITS
ofNew York
ORIGINS in Moore and Russell 's revolt against the British
idealists McTaggart and
F. H.
Bradley, analytic philosophy has defined itself
in opposition to the Hegelian speculative and metaphysical tradition. By "analytic philosophy" is meant that twentieth-century philosophical move ment which may be characterized (roughly) by a number of salient features: by an emphasis on the analysis of language and meaning; by the employ
ment of mathematical logic as a tool or method or as the method of philoso
phy; and by the fact that many of its practitioners have held a set of broadly
empiricist assumptions, while viewing science (especially physics) as a paradigm of human knowledge.' Moreover, it is a relatively uncontroversial fact that this new philosophy had its origins, at least in part, in Moore and Russell's so-called refutation of central British Hegelian positions: in their wholesale rejection, for example, of the doctrine of " internal relations" and of "organic wholes," of knowledge considered as "synthesis" or dialectic, and of reality conceived as fundamentally monistic, one and absolute. Moore an d Russell, for their part, simply argued for the opposed positions.2 As we approach the end of the twentieth century, however, it is not
altogether clear who has won this debate. Ample evidence exists that despite more than fourscore years of disparagement and ridicule, the influence of
Hegel (and many of the idealist positions) has not only not died but may even be gaining in strength. 3 In particular, I hope to reveal Hegel's influence in an
AU H R T O 'S NO TE: Th.is research was supported in part by a grant from the City University of New York PSC-CUNY Research A ward Program. I would als like to tltanJc the rrviewers Professor G. Doppelt, and audiences at the University of Zurich, where parts this article were presented, numerous helpful comments and suggestions. A similar article, entitled "Ziige der Hegelschen Rechtsphilosophie in der Theorie von Rawls " will be appearing in a specia l vol ume devo�d to a discussion of Hegel �
Political Theory, of
Right
o
of
for
ofHegel-Studien (VoL 26)
PotmcAL lliEORY, \\11. 19 No. 4, November 1991 539-511 C 1 99 1 Sage Publications, Inc.
261
Philosophy of
POLITICAL THEORY I November 1991
540
area where one might not yet have suspected it: in the thought of the
paradigmatic theorist of justice in the Anglo-American world John Rawls.
At first sight, this claim seems preposterous. Rawls's thought is generally
regarded as the epitome of contemporary, game-theoretic contract theory,
firmly grounded in the Anglo-American tradition, with its essentially anti
metaphysical intentions and (some believe) its fundamental, atomistic indi
vidualism. Hegel's metaphysical system, on the other hand, still represents
the paradigm of pretentious Continental system building, with its abstruse language, its speculative talk of one "world spirit" realizing itself through
history and its claims to " absolute knowledge," while in the political domain
(our primary focus in this discussion), Hegel was not only a n ardent critic of social contract doctrine but the inspiration of modem com munitarianism . So how, one might justifiably ask, could two thinkers stand further apart? I hope
to show, however, that there is an important sense in which one can apply, without significan t distortion, the term "Hegelian" to important aspects of
Rawls's theory.4
An exposition of the close links between Rawl 's thought and that of Hegel serves several purposes. For one, it serves the purely h isto rical interest of vindicating Hegel in the face of m uch unjustified vilification in the Anglo world, assuming, of course, one bel ieves that A Theory ofJustice deserves its present high regard. More important, the exposition shou ld work to dispel something of the aura of unacceptable "individualism" surro unding Rawls s
:
wo rk, while simulta neously revealing the flawed nature of recent com mon!· . tanan accounts (those of C. Taylor, Sandel, Macinty re, and Walzer, tn particular). Such accou nts are, in my view, sympathetic but fundam entally ague and misgui ded attacks offering a paucity of clear-cut viable altema· � hves . Should it em erge that Rawls in fact sides with the com m unitari an Hegel ?n numerous coun ts, the ground of contemporary debate should shift signif· •cantly. I hope to show that the true conflict cannot be conc eived in terms of her, the a simplisti c dichotomy between "liberals" and "comm unitarians"; rat importan t issue regards the kind of comm unity we seek. Final l y, the follow in� at u u study su gests tha t an adequate conception of "comm unity " - of wh � mately bi ds a j ust society together - may j ust be possible in a Rawl�•an � language : m a politi cal language, that h siCS· is, without a full-blown rnetap � discus;, Allow me to ad dress one last preliminary point. In the s ubse que nt . eo Sion, I assume the legitimacy of Rawls's distincti on between " moral th matiC and "moral philosophy. " By te "moral theory" Rawls inte nds the sys oral "m · ' compariso n f hIstonc ally prominent moral conceptio ns, whe reas blelll philosophy" (w h 1c ' h Inclu · des moral theory) has as its major issue the pro . . . oral m of JUshfication .5 In what follows, I shall primarily be concern ed w ith
�
�
•
0
262
Schwarzenbach I RAWLS AND HEGEL
54 1
�heory. Rawls has made a plausible case for provisionally setting aside the
Important questions of the truth or falsity, or even the reasonableness or unreasonableness, of various moral and political conceptions in order to proceed first with a systematic and comparative study of them. Only after we have clarified the nature of "Hegelianism," that is, its relation to Rawls's theory, and the consequences to be drawn for contemporary debates, are we in a position to ask which view is "best justified."
POLITICAL, NOT METAPHYSICAL
I shall begin my comparison of the political thought of Rawls and that of Hegel by stressing what remains, no doubt, the fundamental difference between them: their respective stances in regard to metaphysics in general. My thesis shall be, in broadest outlines, that A Theory ofJustice retains much of the fundamental structure of Hegel's political theory while detaching this structure from its background metaphysics of absolute idealism - from Hegel 's monism and his talk of "world spirit," from the doctrines of absolute knowledge and concrete universals, from the concept of the self as alienation and return, and so on. Many will here surely object that such a reading will result in but an evisceration of Hegel. In defense of my project, however, I shall try to show that Rawls nonetheless retains the import of many of the most significant strands of Hegel 's metaphysics; Rawls does this, however, in w hat he now takes to be a practical, and no longer metaphysical, form . I d o not mean to minimize the profound differences between the two thin kers. It is well known, for example, that Hegel viewed his political philosophy as but one subpart of his more comprehensive metaphysical system, as set forth (in skeletal form) in his Enzyklopaedie (1 830). Although scholars dispute the sense i n which Hegel claims his Philosophy ofRight can actually be "deduced" from the more general system, agreement does exist that some form of "necessary connection" is being propounded between the general metaphysics and the political theory.6 More recently, of course, scholars have begun to question whether any such necessary connection de facto exists, but this was clearIy not Hegel's problem.7 Hardcore "Hegelians," moreover, continue to stress Hegel 's uncompromising holism; on the Conti nent, at least, it is considered improper to study Hegel's political thought without first spending semesters, if not years, on the Science ofLogic. ' The Hegelian horse piJJ, it seems, must be swallowed whole or not at all.
263
542
POLITICAL THEORY I November 1 991
John Rawls, on the other hand, posits an expl icit separation between his political philosophy and any comprehensive, "metaphysical" system .9 1n this respect, Rawls is decidedly "un-Hegelian"; he stands closer here to the positivist, or more accurately, to the A merican pragmatist tradition. For Rawls's claim is not so much that metaphysical systems ultimately reduce to "nonsense" (that metaphysical claims are without purpose or meaning), but rather that such systems generally underdetermine (they may support, but they do not entail) one's substantive position in ethics or political philosophy. In regard to his own theory, Rawls writes, If metaphysical presuppositions are involved, . . . they are so general that they would not distinguish between the distinctive metaphysical views - Cartesian, Leibnizian, or Kantian; realist, idealist or material ist - with which [modem) phi losophy traditionally has been concerned. In this case, they would not appear to be relevant for the structu re and content of a political conception of justice one way or the other.
(PNM, 240)
Rawls's insigh� a lthough not altogether novel, 1 0 is important, for it acknowl edges that one might well be an ontological materialist (as was Hobbes) or an absolute idealist (as was Hegel) and yet still be a political monarchist rather than a democrat in both cases . M oreover, there appears to be no inconsistency involved. By taking such a normative, "practical " approach to the study of pol itical issues, the metaphysical similari ties or differences between any two theorists will be minimized for the express purpose of focusing on the structure and content of their substantive, eth ical positions. In rejecting Hegel's extreme holism, however, and in clai ming that m oral theory retains a certain "independence" from further questi ons of m etaphys· ics, o�to logy, or semantics, Rawls denies a major tenet of Hegelianism .: 1 �e questiOn thus remains as to the respect in which (if any) his theory is st mtl ar to Hegel's . I propose to ide ntify three areas in which Rawls 's position may be.c�nside�ed typically "Hegelian." By this I mean that in each case, the move on gm ally m tr �uced by Hegel and accepted by Rawls di ffers marked�Y. not ? ?nly fro� positiOns held within the A nglo-A me ri can , predo minantly utth tar· tan trad t tton but from positions held by Kant. The three are as I assemble under th� .hea dings of the task of political philosoph y (inclu ding i.ts meth . tftcatlo . n), the conception th and JUS of the political person, and fmally, � eo th conception of hum an community and the state. If I am correct, Rawls's ry may appropriately be labeled "Hegelian" in these importan t areas, once we have granted, that is, the possible separation of political theory from 3 full -blown metaphysics.
�
264
Schwarzenbach I RAWLS AND HEGEL
543
THE TASK OF PHILOSOPHY: DIALECTIC AND REFLECTIVE EQUILIBRIUM
For Hegel, the task of philosophy in general is "reconciliation" (Ver soehnung); it is a reconciliation of the individual, by means of reason, not to that which "merely exists" but to the real and the "actual" (das Wirkliche).12 From the 180 1 Differenzschrift onward, Hegel stresses that the need for philosophy begins in "bifurcation" or "conflict" (Entzweiung); its aim is to
surmount and to comprehend such fundamental dichotomies as the one and the many, the finite and the infinite, subject-object, or mind-body (to name but a few).13 Political philosophy, for Hegel, is no exception; it too aims at a comprehension and resolution of the deepest cultural conflicts and aspira tions of its time. In our time, Hegel believes, the conflict is one between the claims of an ancient communal ethical life (Sittlichkeit), on one hand, and that of the modem principle of individual freedom, on the other. 14 His Ph ilosophy ofRigh t defends the position that only in the modern state -with its rational rule of law and its system of individual rights - is such a recon ciliation between apparently diverse interests possible. Although Rawls views philosophy as the attempt ultimately to "render coherent" our considered moral judgments (TJ, 2 1 ), I believe it does his thought no injustice to stress that it too aims at a "reconciliation by reason" in fact, Rawls uses this exact phrase numerous times.1s In Hegelian fashion, political philosophy not only begins in conflict for Rawls but "justice as fairness" takes as its starting point a historically specific conflict: what Rawls calls the "impasse" reached in the modem period between the claims of freedom (in the tradition of Locke and Mill), on one hand, and those of greater equal ity (as represe nted by Rousseau or Marx), on the other.16 The task, as Rawls sees it, is to formulate "a deeper underlying basis of agreement" not only regardi ng the values of freedom and equality but of"fratemity" as well, and it is for this purpose that the two principles of justice are designed (TJ, 1 05) . It thus turns out that A Theory ofJustice - essentially a theory of the mod ern state - attempts to reconcile the conflicting tendencies of nothing less than what Hegel calls the animating principles of the modern epoch: the princi ples of liberty, equality, and fraternity for all men. Interestingly enough, this is al so expressly how Hegel conceives his own task; for both, philosophy is "its own time apprehended in thoughts" (PR, "Preface," 1 1). The simila rities, however, run far deeper. Even though Hegel is, strictly speaki ng, a "moral realist" and Rawls only a "constructivist" in ethics (I shall
265
544
POLITICAL THEORY I November 1991
return to this point in a momene7), both perceive the fundamental moral principles of the modern epoch as, to a large extent, already "embodied" (implicitly or explicitly) in contemporary political institutions and social practices as well as in the "traditions of their interpretation"; they are already embodied, that is, in what Hegel calls "objective spirit" and what Rawls terms "our public political culture."18 Whereas for Rawls, philosophy aims for a "reflective equilibrium" between our most deeply held moral principles and a theory which purports to generate them ( TJ, 48ff.), Hegel 's philosophical
method - the political employment of the infamous "dialectic" - may simi· larly be so described. Hegel writes, for instance, in the preface to the
Philosophy ofRight: After all, the tru th
about Right, Ethics, and the state is as old as its publ ic recognition in the law of the land, in the morality of eve ryday li fe, and in religion. What more does this tru ly require - since the thinking mind is not content to possess it in this ready fashion? It requires to be grasped in though t as well. (P. 3)
and formulation
According to both thinkers, what is needed is not some radical new beginning for ethics but rather that the moral principles and values latent in our everyday practices be "grasped in thought" as well - be made conscious and explicit, rendered consistent with each other, and their implicit rationality (or irratio nality) grasped. For both thinkers, the reconciliation of principle and value is to be achieved by means of"reason" - not by the inexorable march offaith
or by class struggle or violent revolut ion. Of the two major tendenci es within ethical writing, it is thus clear into which camp both Hegel and Rawls fall. The first tende ncy attempts to tell us what we should do; it claims we need a radical reconstructio n of our first-o der duties. Both utilitarianism (with its principl e of utility) and at least � some mterpretations of Kantianism (with its categorical impe rative) call for such a radical revision of our morality. Hegel and Rawls, on the other hand, (together with Aristotle) fall into the second camp; moral philo sophy is the ng. !n attem t to clarify and systematize what we have "all along " bee n doi eahst Raw stan lang uage, moral philosophy is " Socratic" (TJ , 49) ; in id ains an ermmology, it aims at ethical "self-knowledge ." For bo th, there rem if pre· tmportant contrast with, say physics. , To take an extreme example s th at sented, with an accurate account of the motions of the heave nly bodie more we do not find appealing, we cannot change that motion to conform to a ntrast, attractive theory (TJ, 49). 1n the case of theories about ourselves, by co we may well wish to alter our be, once views, actions, even who we desire to . thetr underlying regulative principles have been brought to light.
�
�
�
266
Schwarzenbach I RAWLS AND HEGEL
54S
Noting similarities between Hegel and Rawls in regard to the self conception of political philosophy sheds light on what I shall now call "the first communitarian criticism" of Rawls. This criticism runs roughly as follows: Rawls's theory purports to present us with a universal, ahistorical account of justice supplied by the notion of pure rationality itself.19 Such a notion, however, is an illusion; all reasoning is in fact "situated," dependent on various empirical assumptions, perceptions, and cultural practices of definite historical pe riods, or (as Macintyre stresses) on particular cultural
"traditions." Not only is A
Theory ofJustice's claim to "objectivity" a sham,
but the work furthers the illusion that mankind itself is to be conceived on the model of modem Western bourgeois individualism and its instrumental, market rationality.
A central
concern of communitarians here is that by ig
noring the cultural variation between concrete, historical human communities - by ignoring their alternative conceptions of personality, say, or of reason, or their subtle use of "thick" ethical concepts that tend to bind people together - resources or po tentials for community are lost. Although this crit icism is not restricted to those who call themselves "communitarians," som e version of it does unite them as a group.20 As we have j ust seen, however, and as Rawls 's later works make abun dantly clear, "justice as fairness" is intended to resolve an impasse reached in the modern Western tradition. A
Theory ofJustice
is already explicit in
maintaining that the method of reflective equilibrium begins with the data from com mon sense and "our moral tradition," that the two principles of j ustice are to be judged against the leading contenders of this tradition, that they are "contingent" in the sense of being subject to revision in light of new e mpirical facts, and so on (TJ, 578). Already in A Theory ofJustice, that is,
p rac tical reason and its conclusions are conceived as "empirically condi tio ned" (a rather "unKantian" move). The real issue between Rawls and
Macinty re or Walzer is not whether practical reaso n is conditioned by time a nd space in its origins and functioning; Rawls never denies this. The real i ssue is w hether practical reason is thereby rendered "relative" and stripped
of all "transcen dent" critical function: a position Walzer, until recently at leas t, has tried to hold .21
Once again, a comparison with Hegel is helpful, for Hegel was the first to argue seriousl y not only that our ideas are historically conditioned but that this fact does not rob reason of its "objectivity"; in its "dialectical" operation,
�
�
�
at least, reason can perfonn both an immanent an a ritical tran ndent . . function (see The Philosophy of History, "Introduction ). I beheve It can be
show n that Rawls, with the method of "reflective equilibrium," attributes to practi cal reason a similar, if mitigated function. 22
267
546
POLITICAL rnEORY I November
1991
Reflective equilibrium is that process whereby reflection seeks a "mutual adjustment" between particular considered judgments (formed through con crete observation and practice) and general moral principles until a satisfac tory "fit" is reached - first within one's own moral position (narrow reflec tive equilibrium) and then between one's own view and that of an ever widening circle of others (wide reflective equilibrium; TJ, 46ff.). It is thus a method which, like the Hegelian dialectic, not only essentially entails the movement of thought "back and forth" between concrete particularjudgment and general principle (Hegel would say it aims at the "concrete universal"), but importantly, is a conception of thought whereby a novel awareness develops through the emergence of conflict or contradictio n and the over coming of such conflict. In this respect, reflective equilibrium, like the Hegelian dialectic, may be compared with the idea of a metalanguage.23 Briefly, a metalanguage is one in which we can say things about some other language (an object language) that cannot be said in that object lan guage itself.24 In this way, we might draw an analogy between a hierarchy of ever richer languages (object-language, metalanguage, meta-metalanguage, and so on) and what Hegel terms different "stages" of the dial ectic; the "higher" stage will be a richer metalanguage in which problems posed in terms of the previous object-language (problems the object-language itself could not solve) are "resolved" (versoehnt}, while the insights of the previous stage "preserved" (aufgehoben).25 If we focus on a number of Hegel's own examples, the analogy (although limited) is apt enough. Hegel argues, for instance , that the Ancient Greek world lacked the language as well as the political institutions of individu al subj ective rights; the Athenian way of life rested on the secure foundation of a shared public religion and century-old filial duties. Hence when Athe ns w as confro nted with Socrates ' criticism, the city responded in accord with the only options open to it - either silen ce Socrates or be destroyed itself.26 By contrast, the mode�n state expressly has at its disposal this conceptu ally enriched schem e _ subjective rights. ?f 1_.n�tvtdual In ac kn owledging the universal princi ple of m�t�td�al conscience, for instance, the state has expre ssly incorporated � tthm .ttself o! "reconciled" a domain of conflicting perspectiv es on th� good �tfe. ';1thout tts own unity being threatened. Further, the recogniuon of mdt�tdual liberty of conscience, according to Hegel, is a sign of the modem stat� s moral superiority: increased tolerance, a greater universality, and a . .shm . dtmtnt ent m the severity of punishment of critics (PR, para. lOOA) . I_do not mea� to imply that there are not important disanalogies between the tdea. of a senes of metalanguages and the Hegelian dialect ic. For one, �e former tdea (unlike the latter) carries w ith it no requireme nt that the confltcl 268
Schwarzenbach I RAWLS AND HEGEL
547
resolution proceed in only one way nor that the series culminate in "absolute knowledge" (a full awareness of the whole process) rather than proceed to infinity.27 The issue for the moment is whether reflective equilibrium too may productively be viewed as a process whereby we achieve an ever richer or more comprehensive ethical "overview," and I believe it can. Rawls claims, for instance, that any adequate account of justice will be one that reconciles by "a higher principle" what he calls our "common sense precepts of justice" (TJ, 305ff.). Such ordinary-language precepts (Rawls mentions five) will inevitably conflict when measured against each other. For instance, the precept "to each according to his ability" (elevated to a first principl e by many libertarians) conflicts with "to each according to his need" (stressed in turn by Marxists); giving someone what they need is hardly identica l to rewarding ability. Both precepts, in tum, conflict with "to each according to his effort" and so on. For Rawls, any adequate theory of justice will not elevate one commonsense precept at the expense of all others (in effect, ignoring or suppressing the conflict) but will be capable of "preserv ing" the basic insights behind each. Thus we find that in Rawls's well-ordered society, the two principles of justice will be interpreted by four branches of government, each of which recognizes, as its special responsibility, one of the commonsense precepts (TJ, 275ff.). A balancing of the precepts and a "reconciliation" here occurs at the level of (and in the language of) the modern state. Only the scheme taken as a whole, writes Rawls, comes close to preserving the insight behind our most basic moral belief that justice requires "giving to each his due" (TJ, 3 1 3). At this point one may be tempted to ask what detennines, in Rawls's theory, the weighing of these precepts within the state, given not only that com mon sense is undecided and may be systematically distorted but that mutual ly incompatible overarching accounts would seem possible? To come to grips with this problem, Hegel presents us with a rather thick, univocal theory of the "cunning of reason": his metaphysical philosophy of h i story. Rawls nowhere, of course, attempts any such theory and presumably again eschews all such attempts. Nonetheless, 1 believe Rawls's answer to this problem retains critical aspects of Hegel's stance. That is to say, while jettisoning all talk of one world spirit, of the in exorable march of reason, or of the end of history, Rawls's theory nonethe less retains certain similar but far weaker assumptions: that there is such a thing as moral progress in history (since the days of slave�), that � st�dy of m an discloses a strong desire for freedom and for the e� erctse of ��s ht�e� t powers (Rawls's Aristotelian principle) and that our SOCial and poht1cal InSti tutions indeed reveal a minimal rationality and coherence. In Rawls's theory 269
548
POLITICAL THEORY I November 1991
also, that is, how we weigh the wisdom of the ages and balance the precepts of common sense (deciding which to reject and so forth), is essentiaHy tied to a larger (if stiJJ partial) systematic conception, not only of the kind of beings
we have historically been but of the kind of persons (consistent with this history) that we aspire to be. For Rawls, as for the German idealist tradition in general, the normative conception of the person plays a fundamental role in determining the content of the principles of j ustice
("DL," 559).
Before turning to Rawls's conception of the person, however, (merely another target of communitarian criticism), a H ow me to consolidate the position adopted thus far. My excursion into the dialectic and reflective
equiJibrium was meant to show how practical reason, working on the material of a particular historical tradition, can yet achieve a certain "objecti vity." Not merely for Hegel but for Rawls too (in explicit contrast to the position attributed to him by his critics), the historicaJJy conditioned nature of prac· tical reason is acknowledged, while our capacities for re flection and self·
criticism are affirmed. In Rawls's view, if a theory exhibits an internal
coherence of a high order, if it better than its competitors matches our normative judgm ents in reflective equil ibri um, and if, importantly, it exhibits a greater "adequacy" or "comprehensiveness" (i f it can account for its competitor's position and not vice versa)/8 then together these criteria work to make one conception of justice, if not unequivocaHy true, at least "more reason able" for us to hold than another (TJ, 577ff.). FinaJ J y, if this mit igated holistic and "idealist" conception of justification is acknowledg ed as Rawls's own, the import of the first communitarian criticism vanishes ; "justice as fairness" can claim greater "objectivity" - in the sense of fulfi lling th e above crite ia - without loss of historical specifici ty. And Rawls, in light of he � restncted practical aim he has set for himself (the practi cal aim of reacbmg mora) a reemen t on principles regulating the basic structure of a modem . plurahshc democracy), needs claim noth ing more . . To be sure, emphasizing the similarities between Hegel and Rawls 10 wh at 1 regard to philosophical reason as "reconciliation" is not to overlook ave called their fundam ental difference: general their respective stan ces in the 1 reg rd to a full-blo wn metaphysics or ontology. For Hegel, r c l, _ abso dtalectJc •s not only a doctrine of "rational necessity " culm ina ting m ly lute knowledge" but is conceived as an ontological category ; presuma v ti ec e eve� th ngs in nature operate "dialectically" (Phen, "Preface"). Refl . equlltbnum, by contrast, is ought. an open process of fallible, all-too-human th So too, for Hegel, the princip y to be les of freedom ' equality' and fr atern it ·•..-If reconct'J ed are ultimately the ideas of one "world spirit" insta nti ati ng through finite minds and coming to know itself in histo ry (PR, 341ft'.). Tbt
�
�
�
�
�
�
�
��
�
1 ""'
270
Schwarzenbach I RAWLS AND HEGEL
549
principles are conceived as universal and absolute; in an important sense,
they are "discovered" by us.
� awls, on the other hand, refrains from making such further, strong cla1ms. Whether his two principles of justice are in fact universal, for in
stance, remains as much a matter for future empirical investigation to decide
as for reflection to ponder (TJ, 578). It is not even clear that his two principles
de facto underlie our tradition; they may well be nothing more than the "best
interpretation" yet, a "construct" out of the pool of deeply held and widely
shared moral and political considered convictions (see note
1 7).
By taking
this "constructivist" position, however, Rawls is not denying that there may
be such a thing as a growing worldwide consciousness and recognition of
individual freedom, as well as the possibility of absolute moral truths. It is
only that a conception of political justice in his view (one appropriate for a
modem, pluralistic democracy) cannot rest on the truth or falsity of such
strong theses and must even be compatible with a number of confl icting
positions concerning them. Hence for Rawls, unlike for Hegel, political
philosophy m ust be set free from the anchor of a metaphysical foundational
ism; in self-conception, its task comes closer now to the revamping of
Neurath 's ship cast out on the open sea. In regard to some of the most ancient
debates in philosophy, "justice as fairness" wishes to remain "agnostic."
THE CONCEPTION OF THE PERSON Perhaps the leading criticism which the German idealists have leveled at the utilitarians is that the latter operate with an inadequate conception of the
person and of human dignity; utilitarianism conceives of the person as little more than a container of homogeneous desires bent on maximization.29 Rawls, of course, reiterates a version of this critique: Utilitarians operate with
a "co nsu mer person, .. underestimate the possibility of a rational restructuring of de sire and motivation itself, and fail properly to acknowledge the "distinc
tions between persons" (TJ, 23ff., t 85ff.). Ironically, Rawls's own notion of th e person has come under attack by communit arians, most notably, by
M. Sandel .30 Once again, such criticism at least appears to echo Hegel 's fam ous attack on the individual ism of Kant.31 Sandel argues that Rawls presents us with a "hyper-Kantian," "denuded,"
or "abstract" conception of the person, conceived as an agent of choic , which � is "prior to" and separate from its particular ends, attnoute , commltments, � . and even concrete character (LU, t5ff.). Since Rawls conce1ves the plurahty
and separateness of individuals as ontologically "prior to their unity," his
27 1
550
POLITICAL THEORY I November 1991
position leaves no room for communal values or social commitments in the "constitution" of the individual's identity and self-understanding (LU, 147ff.). Rawls operates with a fundamentally "thin" and fl awed notion of the person and with an inadequate understanding of man 's social n ature. Sandel accuses Rawls, in short, of a metaphysical "atomism." As Rawls's later writings have clarified, however, the basic mistake of the preceding criticism is that his use of the term "person" is intended not as a comprehensive account of personality but as a "political conception," that is, as an appropriate conception of the person for the limited purpose of deciding on principles of justice for the basic structure of society ("PNM," 231ff.). What I wish to show here is how close in fact this "political conception" of the person is to Hegel's account of the person in part 1 of Philosophy ofRight, entitled "Abstract Right." (No one accuses Hegel of atomism.) Of further significance is how similar Rawls's original position and Hegel's abstract right are in general. Rawls's A Theory of Justice stands squarely in the tradition of the German Rechtslehre. Hegel's concern in part 1 of Philosophy of Right, in the section titled "Abstract Right," may be stated thus: What are the conten t and l imits of relations among persons respecting and treating one another according to the single norm that each is a person? This concern reveals Hege l 's con tinuity with the natural rights tradition; the section is a form of methodological abstraction from the immediate and concrete social bonds between persons similar to that in state-of-nature methodologies . 32 Unlike the latt er, however, Heg� l . is explicit that he seeks the principles o f person hood underly ing the specifically modern period (PR, paras. 40A, 57R). His use of the term ·person" is thus narrower than that of his natural right predecessors, who use It to refer either to the universal individual in the state of n ature (Hobbes and Locke) or to the indivi dual conceived as moral subject (Kant). For H egel, by con�rast, the term "person" refers to the individual qua h is capacity to be the subje�t of modern political rights; it is only in the modern state th at the capacity for citizenship, in principle at least, is extended to all m en (PR•. �a�; 40R). Furthe r, Hegel considers such personhood only the "first condltton for f�eedom and human flourishing (PR, paras. 1 , 33). For a more compre he�siVe good, the individual must be able to conceive of him self, not only as a nghts-bearing legal person but as an autonomous moral subject and as a par i�ipating member of a rational community as well.3 3 t IS of interest to note that on each of these counts Rawls fol lows Hege1 ' (rat er than Kant or other social contract theorists). Part 1 of A Theory of lu.sttce, for instance, sets forth the original position c le arl y a form of · bonds ' "method0Ioglcal · · , from the richer abstraction and more con crete soc1al
�
;
�
272
Schwarzenbach I RAWlS AND HEGEL
55 1
between persons. This abstraction, however, is from our "post-Reformational public culture," and its aim is to arrive at substantive political principles expressin g men 's respect for each other (TJ, sec. 40). Similarly, part 1 gives us only a "thin theory of the good" (TJ, 396); it is not until part 3 that we receiv e a fuller conception, as well as, finally, an account of man's social nature (TJ, 520ff. ). My point is that for neither Hegel nor Rawls is any claim being made about the ontological "priority" of the individual to the group. Instead, modem political personhood (entailing the individual rights of conscience, free speech, various political liberties, the right to contract, to hold at least personal property, and so on) is claimed by both thinkers to be a necessary condition for human flourishing in post-Reformational circum stances, never a sufficient condition as Sandel's critique implies. If one looks more closely still, it emerges further that for both Hegel and Rawls, modem political personhood presupposes two minimum "moral powers," capacities, or competencies of individuals (TJ, 505; "DL," 525). In Hegel, personhood entails, first, what he calls the capacity for"self-conscious universality" (PR, para. 35). The human subject is unique insofar as it can recognize itself as "universal" - in the first instance, as "indifferent to par ticularity" (PR, para. 3 7). Hegel here refers to the ego's ability to negate or distance itself from anything in particular - its own determinate thoughts and desires included (i.e., it can revise, reject them, and so forth). This "unre stricted capacity" for abstraction, Hegel believes, is presupposed not only in the person 's ability to perceive its likeness to the ego of others (abstracting from particular differences) but for its ability to grasp, and to determine itself to act in the world from, universal rules and principles (PR, para. 258R). Man alon e, Hegel stresses, can consciously sacrifice everything particular, his own aims and life included; this capacity will be tested particularly in times of w ar. The capacity for "self-conscious universality" in Hegel corresponds quite cle arly, I belie ve, to Rawls's first moral power of personality: to what Rawls calls "an effective sense of justice" and which he describes as "the capacity to understand , to apply and to act from (and not merely in accordance with) the p rinci ples of justice" ("DL," 525). (In Rawlsian terminology, this capac ity refers to the "reasonable" in us, in contrast to the merely instrumental and self-interested "rational" part.) But, so too, Rawls's second moral power, "the capacity to form, to revise and rationally to pursue a particular conception of the good" ("DL," 525), may be seen to correspond to what Hegel calls the ego 's second fundam ental capacity for "self-determination" (PR, para. 6). The claim of both thinkers here is that fundamental to the modern notion of free personal ity is the ability not simply to follow rules nor merely to negate
273
552
POLITICAL THEORY I November 1 99 1
or choose between de facto given alternatives
(Willkuer) but to form plans,
posit particular goals, and in general "express" a self-conception or a plan of
life in an external, publicly recognized sphere (PR, para. arrive at Hegel's important concept of "expression"
6). At this point, we (Entaeusserung). I hope
to show that the concept is also central to Rawls's account.
As C. Taylor has rightly pointed out, Hegel was influenced by the German
"expressivist" movement of the 1 770s, whose members revived the old Aristotelian notion of the good life as the expression of purpose or the realization of "form."34 Hegel, for example, p raises Aristotle's view of the soul as "self-organizing form" inseparable fro m a particular organic body
(Enzy, para. 378 and LHP, II,
1 80ff.), and he, like Aristotle, imbues person
ality with the motive force of bringing its distinctively human capacities to
fruition. Nonetheless, Hegel explicitly departs fro m Aristotle when it comes to questions ofpolitical personality. Unlike for the Ancients who, according to Hegel, viewed the form of an individual's l ife as fixed "by nature" or independent of the subject who receives it, the purpose of a subject's life must be given to it
by itself,
it must be its
own
conception. This Hegel calls the
"principle of subjective freedom," and he considers it to be the distinguishing mark of modernity; in the modern state, this principle has been acknowledged
for the first in the universal right of free personality (PR, paras. 1 82A, 185R).
Hegel's concept of "expression" is of further interest because it signals an
important aspect of his departure from Kant. With this co ncept, Hegel clearly attempts to overcome the rigid Kantian dualities between mind and body, reason and desire, and so on. Hegel is, in the e nd, a monist; it is of the esse�ce of mind (Geist) that it express itself in space and time . So, too, imbmng personality with this motive force entails for Hegel (unli ke for Kant) that sensuous desire and impulse be considered "intrinsic to freedom" (PR, para. 6) · Thus in cont ast to Kant's view, where the physical, mechan ical world forev r _ remams a fod to our transce ndental freedom, the sensuous material world 10 Hegel (including our own sensuous desire) is viewe d as the necessary d. me tum m wh1ch our freedom is embodied and revealed. A number
�
�
·
of
·
important subtheses follow from this altere conception. d For one, politically speaking, Hegel now attributes to the ri ght of mod ern d the personality what may be called a material content; free person ality, an equate development of h uman powers is i mpossible severed from an ad . ' �0r matenal substratum (PR, para. 41). Thus we find in Hegel's theory, n under examp le, that the state will play a far more extensive welfare role tha n, tradition al liberalism; the "public authority" should provi de educatio ubhc see public utilities (e.g., r street lighting and bridge building), care fo p
ove�·
274
Schwarzenbach I RAWLS AND HEGEL
553
�
health, price aily necessities, and alleviate poverty among other things (PR, 236ff.). 5
para.
Second, for any individual to exist qua person, for him to embody his plans and express a self-conception in an objective public sphere, it is essential that others be able to recognize him as well
as such
as acknowledge his desires and aims (PR, para. 71). This presupposes, in Hegel 's view, not just a material
substratu m but a social background of shared understandings, expectations, and the "reciprocal recognitions" entailed by his notion of Geist.36 Without such a prior cultural formation (without, in contemporary language, an understanding of the "fonn of life"), the individual could never adequately express his intentions or even come to know them. The minimal knowledge of such expectations Hegel terms
Bildung
(culture or education) and he
describes it as our "second nature" (PR, para . 3). The important point is that for a person to obtain "substantive freedom" in Hegel (and not just Willkuer or choice), it is not enough simply to overcome the alien and compelling character of the natural physi ca l world
(achieved through labor and property), nor is it enough to bring order to the chao tic inner world of one's desires (achieved primarily through moral reflection); one must also overcome the compulsory nature of human, com
munal l ife. The latter is achieved, in Hegel's view, through educating oneself to an awareness of universal ends as well as by participating in the construc tion of the rational character of public social life
(PR,
paras.
149,
260).
"Substantive freedom" can only be fully instantiated, he claims, with a commu nity wide "reciprocal recognition" of freedom as "lived social prac tice ." The latter Hegel terms Sittlichkeit or "rational ethical life."
I
have emphasized Hegel 's notion of personality as "expression," for I
believe the thrust of his departure here from Kant is fully accepted by Rawls. Rawls, too, views his own thought
as an attempt to overcome the many dualisms of Kantian philosophy ("DL," 516). So too, in A Theory ofJustice,
Rawls explicitly criticizes Kant's theory of the person for lacking "the concept of expression"
(TJ, 255).
Further, l ike Hegel, he revives an essen
tially Aristotelian notion of the good life
as the realization of purpose or "a
plan of life" (which must now be given to the individual by himself); the minimal motivation attributed to persons in Rawls's theory is the "Aristote lian principle," which claims that other things being equal, humans enjoy the exercise of their distinctive capacities, and this enjoyment increases the more the capacity is realized or the greater its complexity
(TJ, 426). As a conse
quence, Rawls, again like Hegel, stresses those minimal background condi
tions (both material and social) necessary for the realization of the powers of
personality: Rawls's list of "primary goods"
275
("DL,"
525ff.).37 And i mpor·
554
POLfTICAL THEORY I November 1991
tantly, among these goods, Rawls i ncludes the "social basis of self-respec�" a prerequisite for which is that the individual participate i n "social union" (TJ, 44 1 ). As we shall see in the next section, with Rawls's idea of a "social
union," critical aspects of Hegel 's notion of Geist reemerge in contemporary political theory. For the moment, however, it suffices to note that Rawls, too, distinguishes between "rational" and "full autonom y"; the latter, he writes,
can be "realized only by citizens of a well-ordered society in the course of their daily lives"
(TJ, 528).
i If the preceding analysis is correct, we can see how wide of the mark s Sandel's criticism of Rawls's "thin" and "denuded" conception of the self or person . In the original position, Rawls is not giving us an account of the "self' or "subject" at all but is (as is Hegel's abstract right) presentin g us with the
In em· ly mere is phasizing the moment of individual choice, moreover, Rawls orma· articulating, in political terms, an insight already won during the Ref tion: that the good the individual seeks should ideally be ob tained via that individual's own choice, consciousness, and will, not imposed from without by the dominant religion or the state or even by "the majority. "38 Here, Rawls simply sides with the communitarian Hegel for whom the political " princi�le m of subjective freedom" remains the distinguishing mark of modernity. Aga , ien neither thinker is claiming that such political personhood is a suffic t condition for human flourishing, only that it is a necessary one given post· · Reformational circumsta nces. Sandel's "communitar ian" acco unt (and I be l a vidu l ieve Walzer 's, too, as well as Macintyre's critique of the notion of ind i rights) misses this crucial point completely.39 , One might attempt to defend Sandel by cla iming that, in criticizing R�wls hts g e h�s somethi ng closer to the Marxist critique of bourgeo is ind ividual n. . posttton 10 mtnd and not Hegel's position after all . Yet I belie ve Sandel 's would also be a misreading of the Marxist stance.40 Nor am I denying that both Hegel and Rawls operate with an impoverished notio n of the person, . e that s, of the political person or citizen.4 1 1 have only tried to stress that th ! er corn· partl ul r critiqu e of the Rawlsian person made by Sandel (an d oth � � mumtanans) takes us back to pre-R eform ational thinking. l F'maIIy, not only are Hegel 's and Rawls's conceptions of the poI'10'ca person si gni ficantly similar but it i c s lear th at both are employing the concept " 42 That person" not as a natural kind term but as a socio political construct. . ts as IS, for both thinkers, a conception of man 's individual l iberty and righ . e c "metaph ysically " given (by tin tiV God, say, or Nature) misses their dis J 't " and as , pect as h ard -won h uman ach i evements; universal "free pe rson a 1 Y . . . 'tndtvldu long and al rights" are cultural products, not starting points, of a
minimal conditions for modern, politi cal personhood or
�
•
276
citizenship.
Schwarzenbach I RAWLS AND HEGEL
555
arduous historical struggle. Moreover, it is only if one views them as such - as presupposi ng this larger cultural effort - that their full significance is recognized and the responsibility for maintaining them in existence acknowl edged. Thus it is that one finds in the works of Hegel and Rawls the con ception of the person embedded in a further account of those particular background social, economic, and political institutions which alone allow free personality to flourish.
SI1TLICHKEIT, SOCIAL UNION, AND THE WELL-ORDERED SOCIETY
Whether one is reading Hegel's Philosophy ofRight or Rawls's A Theory ofJustice, the movement is from "abstract to concrete," from the minimal moral requirements of political personality (set forth in abstract right and the original position) to an account of those background economic and political institutions supporting such a conception until both works end, finally, with a reading of man 's "social nature" (part 53). This fact is repeatedly over looked, however, in communitarian criticisms of Rawls. Sandel argues, for instance, that given Rawls's "denuded" conception of the person, his theory cannot justify the strong other-directed tendencies of its own difference principle.43 Such a justification would require the notion of a "group" or "community subject" (an idea often attributed to Hegel).•• C. Taylor 's concern also focuses on the notion of community rights and he, l ike Sandel, faults Rawls for neglecting "background" considerations (spe cifically, considerations of the human good} i n relation to which all questions of distributive justice and of individual "desert" must be situated !5 This last criticism of Taylor's is particularly surprising, given that Rawls from the beginning has insisted that his two principles of justice are to apply to the background "basic structure " of society and not to individual actions; Rawls is well aware that there is no "context i ndependent" notion of desert. 46 Again, I believe such communitarian criticisms are wide of their mark, but in spirit at least, they do revert back to Hegel: to Hegel 's basic stance that social contract theory is an inadequate approach for an understanding not only of hu man comm unity but of the modern state. Allow me to turn to this last point first. Why, according to Hegel, is the ideal of the social contract inadequate to an understa nding of the modern state? It is important to realize that Hegel distingu ishes three different senses of the term "state" (der Staat):' Hegel
277
556
POLmCAL THEORY I November 1 99 1
distinguishes between, first, the political state in the narrow or " strict" sense.
This refers to the state qua its "internal organization": whether it is a
democracy, a monarchy, etc., as determined by its poli tical constitution and
exp l icit laws (PR, para. 274). Second, there is "the external state," which refers to the organization of the judiciary and the pol ice, including the
concrete physical manifestations of government, such as the courthouse and
jails (PR, para.
1 83ff.). Finally, Hegel refers to the "state proper" (PR, para.
267ff. ), which encompasses both of the previous senses as well as something more; the state proper includes the customs, manners, and moral conscious
ness of a people historically united together in a tradition. The strictly
(the of a people's prior
political state, in Hegel 's view, is thus conceived of as the expression legal articulation or the making explicit and consistent)
ethical practices. Granted, this is a rather broad conception of the state, but
under imple· ments the two principles of justice) or, for that matter, under R. Dworkin's notion of "law"; in each case, the domain of the state, of justice, or of ''the
it is no broader, we might note, than what is being encompassed Rawls's notion of "the well-ordered society'' (in which government
l aw" includes the tradition's underlying moral principles.�8 Keeping these different senses in mind it becomes clearer why for Hegel the modern pol itical state could never be raced back to an original "histori· cal" contract between individuals in the state of nature. The strictly political state is the legal expression or articulation of a people's historically prior ethical practices. But so, too, the ideal of the state as nothing more than a contract between individuals - a mere modus vivendi' as it were - is al.so . cont act· � �nadeq�a�e in Hegel's view. This is the case, he argues, because all mg acttvtty must take place against a background of shared assumpuons, ct; ��st and social practice, which themselves cannot be the subj ect of contra � act o n l l ts I mpossible that everythin g be open to contract at once .�9 Again, wn s it of ntract of (social or otherwi se) can generate the cond itions .o � ce an h v hdtty, ut presupposes background norms, � rules, or prin ciples, co mp �
;
back· Wtth whtch confers legitimacy on the contractual transaction. This ustom groun � of shared moral assumptions Hegel terms Sittli chkeit (ethical c ed as or soctal life) and is that element which the model of the state conceiv
a mer sel - i terested "compact" fai l s properly to acknowledge. � � � . Tins cnttctsm of the contract model has become comm onp lace (altho�gb . o whtch 11 as � �?t so, of course, in Hegel's time). The issue here is the extent t . �hts cnhctsm touches on Rawls's social contract theory. Interestingly enou�b. ftt tt does n touch o� Rawls's theory at all; Rawls ful ly acknowledges the pol . � es Early m the Philosophy ofRight, Hegel sets forth three essenti al featur of the modem notion of con tract :
278
Schwarzenbach I RAWLS AND HEGEL
557
I . The contract must arise from the "arbitrary will" (Willhler), i.e., from the free initiative
of private contracting panics (and not from some higher public authority, say).
2. A common will (or shared purpose of the panicular wills) is brought into existence by their mutual consent or agreemenL
3. The object(s) with regard to which the contract is made, must be single "external"
thing(s). (The modem practice of oontract presupposes non-alienability of personhood and its essential characteristics.) (PR, para.
is)
Hegel 's general point is that the contractarian tradition has confused such norms brought into being and having binding validity within the sphere of private transactions with those norms governing the public rights of political bodies, such as the state (PR, para. 75). It is of the essence of modem individual rights, for instance, that .they are not, properly speaking, private property (unlike various rights in the medieval period); modern political rights cannot be alienated to others at will but are universally secured by the impersonal and general norms of the rule of law. Hence the realm of private contract could never lead to, or account for, the public, universal character of the modem state but instead must presuppose it. What is of interest here, however, is that (in contrast to the social compact in Hobbes or Locke) the agreement in Rawls's original position clearly violates Hegel's first condition for the existence of a modem contract: that the contract be the result of the "private" arbitrary will. The veil of ignorance expressly excludes parties from the possibility of acting on such a private will insofar as it excludes all particular knowledge to them; the veil forces the parties instead to focus on common and universally shared characteristics (TJ, 1 36ff.). So, too, as a direct consequence of this veil, the parties in the original po sition "agree" that certain types of primary goods (the basic liberties and equal opportunities) are, in effect, to be withdrawn from the scope of future contract and universally granted to all.� What transpires in Rawls's original position is thus no ordinary contractual agreement. It is in fact an agreement which (to use Hegelian language) "transcends the stand point of contract" itself; it acknowledges a far more substantive union between persons and political institutions.' 1 The original position, which serves as a "means of public reflection and self-clarification" ("PNM," 236), concentrates as much on those aspects "beyond contract" as it does on the moment of contracting itself. Again, part 1 of A Theory ofJustice may be seen to perform a function sim il ar to Hegel's abstract right. In Rawls's original position, "the rational" (each person 's legitimate rational advantage as represented by the parties) is acknowledged and granted a certain legitimacy, but it is ultimately framed and subordinated to "the reasonable": to the fair background terms of a 279
558
POLITICAL THEORY I November 199 1
system of free and equal cooperation as a whole and to the capacity of individuals to honor such terms ("DL," 529ff.). The central difference between Hegel and Rawls in this respect lies, I believe, in the fact that, whereas Hegel envisions the fair cooperative terms to be those in accordance with the universal principle of substantive freedom, Rawls attempts to make this principle more specific; the well-ordered society entails realizing the difference principle as well. The fact that "justice as fairness" does not conceive of our political life (or the state) on the model of a social contact but rather acknowledges (as does Hegel) a realm of private contractual transactions as an essential aspect within the modern state is even clearer in part 3 of A Theory ofJustice, where Rawls speaks of the well-ordered society (one in which his two principles of justice are implemented by government) as a "social union of social unions" (TJ, 527). In this third part, Rawls proffers his account of man's "social nature" as well as his account of the "good of community" (TJ, 395). It is important to understand what he here has in mind, consi�ering his many communitarian critics. In claiming that man's nature is fundamentally "social," Rawls is not merely claiming that society is necessary for human life or that social life is a condition for the individual to develop speech and language and to acquire certain sorts of needs, interests, and so on (TJ, 522ff.). Nor is his point merely the Wittgensteinian one (repeatedly stressed by Taylor, for instance) that only in a community of speakers are certain conditions met whose satisfaction is necessary for us to hold justified beliefs or even to express our individual beliefs and thoughts in the first place.52 These facts are not trivial, but to characterize our social ties to one another by reference to these facts alone is to "trivialize" our sociability (TJ, 522). Why? Because all these things are equally true of persons who view their relations to one another in instrumental terms. They are all true of a group of egoists, say, who could not have developed language, voiced their egotistical needs, or justified their selfish desires without a prior social life, and so forth. Rawls's point is stronger (as is Hegel's): Only by actively cooperating with other humans and by sharing important, moral ends with them can certain of the individual's distinctively human powers reach fruition. Fur thermore, only by doing so can the individual participate in many of the realized capacities of others (TJ, 525ff.). Rawls defines a "social union" as that form of cooperative activity whereby individuals share final ends, participate in common activities valued for their own sake (a version of Aristotle's notion of praxis), and agree on a scheme of conduct leading to a complementary good for all (TJ, 525). Rawls contrasts this notion with that
280
Schwarzenbach I RAWLS AND HEGEL
559
of a "private society," where individuals not only have independent or conflicting conceptions of the good but where they regard their social institutions i n purely instrumental terms. "Private society," Rawls explicitly notes, corresponds to Hegel 's notion of "civil society," and the idea's natural habitat, he claims, is economic theory (TJ, 521 ff.). A paradigm of social union, by contrast, is that of musicians playing together in an orchestra; a requisite in this case for the individual to develop his capacities is that others also develop theirs and that certain rules and principles are accepted by all from the start (TJ, 524). 1n a successful play of the music, the distinction between personal and communal well-being, at least temporarily, collapses. Insofar as individual players identify with the ends of the group (its goals have become their goals), each player not only shares in the responsibility of the group activity but is eligible for pride or shame with regard to it. I am thus eligible for pride or shame with regards to how you, another member, play. Unlike in private society, that is, in social union, members win or Jose together; social union is not a zero-sum game. It is important to stress, however, that for individual players to "identify" with the group does not entail that they share all ends with other members of the orchestra (this would be an altogether illiberal model). As Dworkin has recently emphasized, a good performance of the music does not entail that members share all cultural aims nor that they all believe in one God nor that they all participate in a common sex Jife.53 In order to "identify" with the group here, it is only necessary that individuals share the end of the union in question. Establishing the fact that Rawls views the well-ordered society on the model of a "social union of social unions" with justice as a defining aim (and not on the model of a modus vivendi) is important for a number of reasonS. 54 For one, the idea affords a way of attributing a certain "primacy" to group activity without committing ourselves ontologically to the notion of a "group subject" (or to Bradley's idea, say, of a "moral organism"). An established orchestra can be, and legally is, treated as a "unit of agency" in its own right; it has an internal organization and interests peculiar to it, decisions made in its name, schedules to meet, and so forth. Such a scheme is not only "temporally prior" to any new member entering in but "conceptual1y prior"; the new individual's activities will gain their significance against this back ground scheme. Further, such agency is not adequately comprehensible in terms of a "mere sum" of isolated actions; it expressly concerns the manner or way in which these actions are weighted and organized. An "integrated orchestra" or a "community between members" seem genuine enough, everyday phenomena. Nonetheless, one may still wish to maintain that this 28 1
560
POLITICAL THEORY I November 1991
"group spirit" or "way of life" is created by, and resident in, certain
human
attitudes and practices and nothing more. Second, the introduction of the idea of social union means that Rawls's conception of the political state (which implements the two principles of justice in the well-ordered society) not only explicitly departs from Kant's position but is far closer to Hegel's view than is commonly recognized.55 Kant, that is, continued to view the state on the model of a modus vivendi: as a nonmoral consensus founded on the convergence of self-interest.56 As
we have seen, however, Hegel conceives of the political state as the deeper articulation of a people's moral practices; the state is conceived on the model of Geist or Sittlichkeit. In pointing to the similarities between Hegel and Rawls in this respect,
I
am not, of course, attributing to Rawls Hegel's
particular views on such topics as democracy, nationalism, war, or civil disobedience.57 I simply wish to emphasize that for both thinkers - unlike for the "Hobbesian strand" of liberal thought - the state is not conceived as a mere neutral "umpire" between competing interests; it plays a fundamental role in the articulation and education of shared moral interests.58 Finally, the fact that Rawls views crucial aspects of the political life of a people on the model of a social union means that he too recognizes the important "good of community": that people lead better l ives when they do not draw a sharp distinction between their own welfare and that of the community to which they belong. 59 Although, for Rawls, citizens no longer share comprehensive religious or moral conceptions of the good life, they do share important ends in common; they share a desire for justice and they value their political institutions and activities as goods in themselves
522). At least
in the well-ordered society, citizens are viewed as
(TJ, identifying
with the political community and hence eligible for praise or blame regarding its actions. At this point,
notion of "liberal commu nity" (although an advance over traditional liberalism) inadequate after all. I wish to emphasize, however, that it is not for the reasons that Sandel, Maci ntyre,
I must
admit that
I find this
or Taylor cite. Uberal community is not inadequate because person s no longer hold comprehensive conceptions of the good life in common. Given
the diversity of religious and ethnic backgrounds which together mak e u p the modem state today, shared conceptions of the good life must inevitably,
it seems, be partial in the future. Nor is it necessary that members of the modem state share a common ethnic culture or racial characteristics; any
suggestion to the contrary should strike us as reactionary and even dangerous. In my view, it is the fundamental flaw of recent communitar ian accoun
ts th at, although rightfully stressing identification with one's society's fundamental
282
561
Schwarzenbach I RAWLS AND HEGEL
institu tions as an i mport ant good, they so little concern themselves with ance. This mistake neither Hegel nor Rawls mak es. My own view i s that the thinness of Rawls's conception of liberal community lies not in his rej ection of compr ehensive conceptions of the good or of religious, racial, or particular cultura l features of humans as bases for founding commu nity today; this is, in fact, liberalism's strong point. The thinness of l iberal commu nity lies, rather, in its continuing to look away from i m portan t, shared moral ends we in fact hold in common, namely, economic, and I would also argue, "reproductive" ones. In Rawls's theory, that is, "social union " never enters into the economic sphere, at least on the day-to-day level .60 Social union is defined in explicit contrast to the idea of private society, whose proper home, Rawls claims, is the economic domain. Simi larly, when Rawls enumerates examples of social unions (he mentions families, games, the arts, science, sexual love, friendships, and the well ordered society itself), the work relation, and any mention of the firm or productiv e relati on, is consp icuously absent (TJ, 5 25ff.).6 1 Finally, when he gives his account of the learning of the social virtues as well as of the i mportant "a rt of perceiv ing the pe rson " (the art of d iscerning their beliefs, intent ions, and feel ings) , Rawls actually seems to relegate such learning to extra econom ic activit ies - to the family, games, school, friendships, and so on (TJ, 465 ff.). It is at this point, I believe, that the "thinness" of liberal com munity is revealed. Eight hours of the average person's day is spent at "work" or in the domai n of so-called private society; social institutio�s an relatio ns are vie wed i nstrumentally, and other perso ns are seen as havmg, tf not com peting , then at least independent ends. . Elsew here, have ory, indeed, has not hbe�ated the I Rawls's that argued . Itself from the model (reaching back to Locke and Adam Smith) of a pnva e appropriati ng indiv idual bent on maximization when it comes to the domam o,· ' labor, des pite ht 62 And I ' strongly opposed other tende nctes of h ts thoug · in fact su t e �am ch ggesting that this model must be rejected for mu am 6 reason t dtgmty. hat R aw ls reJ· ects it elsewhere : It is simply beneath human · · F.JOally, one odel way of deflating the power that this pnvate appropnaftve 10 holds ov er dted em !abo� us is to take seriously the altern ative form of an the tradit chtld, , ion al activity of women outside the mark�t (in famdy . horne ca re to the cate ).64 Such "reproductive" activi ty (in exphctt contrast . parttcu ar of n gory of "pro ctio ductive" labor) ai ms directl y at the reprodu human re " but retatns ' lationships· it is not only essentially "other dtreeted . ttons perCritical featu res of s cial union such as shared activities and ac · ) So too sueh tormed · totl e •s praxiS for their own sake (a vers ion of Ans
which institu tions and ends are worthy of our allegi
�
�
·
�
�
� .
�
!
�
'
·
283
'
�
'
562
POLITICAL THEORY I November 1 99 1
reproductive activity and work fundamentally entails, develops, and exer cises what Rawls calls the "art of perceiving the person." For the present, however, it is enough to note that what I am considering the "legitimate" criticism of Rawls's theory is not the contemporary communitarian criticism at all but a new version of the old socialist one. The greatest (moral) threat to "community" does not lie in our religious, cultural, or racial diversity; the peril lies in that expanding, commodified market relations threaten us all. In concluding this section, it is important to note that Hegel conceived of numerous institutions that were to foster communal values and keep the atomizing tendencies of the market and modem civil society in check. Hegel mentions in this regard the family (with its caring relations), the state (with its imposition of universal rights and duties), the i nstitution of primogeniture
(which was to keep all land from alienability, a suggestion Marx rightly
mocks), and finally and importantly, the economic "corporations," which
organize isolated workers into powerful econom ic communities, thus form· ing an intermediate community between private individual and universal state
(PR, paras. 231-56).
Many of these options, however, are not open to Rawls. Rawls cannot appeal to older feudal institutions, which run contrary to principles of the market (even if this were helpful); the United States, at least, never had such institutions. So, too, Rawls 's theory has no equivalent to Hegel's notion of the economic corporation - no intermediate form of community between individual and state as might be found, for instance, in the idea of workplace democracy - and for this Rawls has been criticized.65 Finally, even the family (that presumed haven in a heartless world) is undergoing a transformation as never before. With the twentieth-century movement of women into the market, those strong filial and caring bonds between family members, rather than remaining a check to instrumental market relations, are in danger of being invaded by them. So the threat to "community" - the threat to the possibility of a genuin e identification of the critical interests of the i ndividual with any larger, significant community, much less with the political community of the state
seems real enough. As I have been suggesting, however, the solution does not lie in a yearning for the past.
THE OWL OF MINERVA We have seen that Rawls's theory is not so far from the original mouth· piece of modem communitarianism as is commonly believed . In rega rd to
284
Schwarzenbach I RAWLS AND HEGEL
563
political philosophy 's aim as conflict resolut ion or reconciliation, its method as the attempt to "bring order" and to gain an "overview" of our moral life' . " 1ts expressive" conception of the politica l person with two minimal moral powers, and its vision of the we11-{)rdered society as a "social union of social unions," Rawls's theory (like Hegel's) has greatly distanced itself from the Hobbesian strand of liberalism. In the reading presented here, Rawls and Hegel even share a common weakness: Both still allow the Hobbesian strand too unbridled a rein in the economic real m - in "private" or "civil society." This brings me to a final similarity betwee the two. n In Hegel 's view, the owl of Minerva - philosophy - spreads her wings only at dusk; only when an action has already been completed or a way of life grown old is it possible to grasp it fully in thought (PR, "Preface"). This conception of philosophy as a "looking backwards" holds, with some
qualification, for Rawls's theory as wel1.66 That is, just as Marx criticized Hegel for failing to recognize, at the beginning of the nineteenth century in
Gennany, the movement toward democracy in the political domain, so I
believe Rawls's theory has not taken seriously the call for democracy in the
economic domain in this century.67 Certainly, the radical implications of the
women's movement have yet to capture his attention.68 And I believe in both cases, the reason is the same: Similar to the political employment of Hegel 's
dialectic, reflective equilibrium starts from the data of our philosophical
tradition and "public political culture" (see p. 543 above). Although Rawls
here intends to highlight our shared political tradition (in contrast to the individual istic economi c doma in), examining this tradition alone would appear insufficient ; radical new developments may emerge elsewhere first, for example, within the workplace. Or again, examining our philosophical tradition and public political culture, although necessary and important,
cannot be su fficient; the realm until recently has been composed entirely of mal es . Th is suggests that for a more "adequate" account of the well-ordered society, reflective equilibrium must be "radical ized" and extended into new ( in part icu lar, into the so-called private) domains.69 My own view is that if we are to think deeply about community (about �hat it is that hol ds a just society together), we can no onger overlook the . Important comm unal activities which women have traditionally performed within the pri vate sphere, for instance, interpreting and r�s�nding to the concrete needs of others, an activity that goes far toward bmdmg peo� l� to one anoth er. F urther as women move into the public sphere (and as fem1msts have begun to s argu ) a new demand emerges that our political institution hen eforth state," "modem e acknowle ge this activity. A conceptio� of th � . . for m stance traditiona11y conceived in terms of mamtammg law and order, ,
�
�
d
285
POLITICAL THEORY I November 1 99 1
564
a military prepared for war, and a policing of citizenry and competition, could give way to a different conception whereby the state is fundamentally con ceived as a flexible provider of services, an educator and satisfier of need. In conclusion, I believe much of the contemporary "communitarian"
attack on Rawls is a red herring; either the attack is misconceived or Rawls from the start acknowledged the point. Perhaps this much the comparison of his thought with Hegel's has clarified. But so, too, my comparison has hopefully suggested that, particularly in the area of moral and political philosophy, many of the slandered Hegelian notions appear to have a certain appropriateness; the tradition of analytic philosophy, in discarding
all
of
Hegel, threw out the baby with the bathwater. The doctrine of "internal relations," for instance, far from proclaiming its legitimacy across the board, has a definite appeal when dealing with relations between persons.70 When I hear of a child abused next door, the death of a loved one, or of a peoples' rights being systematically violated,
I
am (or at least I should be) altered.
Similarly, the idea of"dialectic" or "synthesis," that knowledge of the person is to be attained not simply by reflection (as in Descartes) nor by mere empirical observation (as for Hume) nor by the direct intuition of some mysterious faculty (Moore) but rather (like reflective equilibrium) is
ated
medi
and indirect, the hard-earned result of concrete experience, subtle
reflection, and the interaction between a variety of particular, historically situated selves - this complex approach leads
away,
in my view, from the
smug self-certainty that accompanies all dogmatism. Finally, the fact that the focus is again on the person, not on the person conceived as an i solated organism (as in, say, a biological reading) but on the person considered as a political and "cultural" being, as one whose desires and actions have an essential connection to the background institutions and social conditions amid which it was schooled; that the focus is again on the person - not merely in the sense of focusing on what kind of beings we are but on what kind of persons we aspire and ought to be - this is only a part of the legacy of Hegel which remains alive and well in the thought of Rawls.
NOTES 1 . Admittedly, this is only a rough, working sketch. For a more extended discussion, see
P. Hylton's Russel� Idealism and the Emergence of Arralytic Philosophy (Oxford: Clarendon, 1 990), 44ff.
2. Moore and Ru ssell argued for a doctrine of "external relations," for a conception of the
whole
as
"reducible
to the sum of its parts," for
286
1
philosophical method as "analysis," for
Schwarzcnbach I RAWLS AND HEGEL
565
�
kno lc gc as immediate or "by acquainta nce," and for a view of reality as fundamentally ": atomtsllc. See Moore, "Refutation ofldealism" (1903), "External and Internal Relations" (1919), and Hylton (1990).
3. The works of J. N. Findlay, C. Taylor, A. Danto, N. Goodman, and the later Putnam are only a few of those in the Anglo tradition that reveal strong idealist tendencies. There is, m reov r, a growing interest in such themes as "holistic justification,"and Bradley's thought is � � Wttnessmg a revival, at least in England. See the recent collection of essays The Philosophy of F. H. Bradley, edited by Manser and Stockton (Oxford: Clarendon, 1986).
4. The actual historical influence of Hegel's philosophy on Rawls will not be a topic of this es say. Rawls clearly read Hegel 'sPhilosophyofRightalready in the 1960s (cf. references to Hegel's work in A Theory of Justice); however, much of Hegel's influence on Rawls seems to be more "indirect"- by way of the ethics of F. H. Bradley, for example, and the work of J. Dewey.
S. See "Kantian Constructivism in Moral Theory: The Dewey Lectures 1980," Jour111JI of
Philosophy,
September 1980, 554 (hereafter "DL"). For a further, secondary discussion, see
A. Davidson, "Is Rawls a Kantian?" Pacific Philosophical Quarterly 66 (19&5): 49.
Hegel :r Philosophy of Righi, translated by T. M. Knox (Oxford: Oxford University 1977), para. 2. All further references to this text will be indicated by PR followed by the
6. See
Press,
paragraph number. "R" after a paragraph number refers to Hegel's remarks immediately
following the main paragraph, "A"to later additions culled from notes taken at Hegel's lectures.
7. Professor U. Steinforth, for instance, recently defended such a separation between Hegel's
metaphysical doctrines and his political theory (lecture at Columbia University, Spring
8. WlSsenschaft der LogiJ< (Berlin,
1987).
1812). This work espouses what the Anglo-American
world would call Hegel's "metaphysics." 9. And this is the case, Rawls intcnds, for any of the standard meanings of the term
"metaphysics." Sec "Justice as Fairness: Political not Metaphysical," Philosophy and Public A/fairs 14, no. 3 (Summer 1985): 223-51 (hereafter "PNM"). 10. Wittgenstein, for instance, makes a similar point: "But the idealist will wish to teach his children the word 'chair, • after all, for of course he wants to teach them to do this and that, e.g.
to fetch a chair. Where then will the difference lie between how the idealist-educated clJildrcn speak and the realist ones? Won't the difference only be one of bailie cry?" (Zettel, para. 414,
translation mine). Whereas Wittgenstein's is a pragmatic point, Rawls's theory may be viewed as extending this insight into the normative domain. 11. See "The Independence of Moral Theory," Proceedillgs and Addresses of the American
Phil osophical Association 48 (1974-75): 5-22, where Rawls argues that moral theory (the study of structures, as ihe se relate to our moral sensibilities and natural attitudes) is independent of the theory of meaning, epistemology and philosoph y of mind.
12. "Pre face,"in Philosophy of Righi, J0-12. For Hegel, "the actual" (das Wvkliche) is not the same ca tegory as "the existing" (das Dasein); the former is essentially "rational" nue nftig), whereas the latter frequently is not. This distinction is crucial to Hegel's famous cia1m
(v�
that "what is rational actual and what is actual is rational," which does not mean that whatever is no w exists is rational. In this famous line, Hegel instead stresses the "power" of reason: that it has the ability to have Wirkung (effect or actuality) in the world.
13. See Hegel, The Difference Between Fichte's and Schelling's System of Philosophy, translated by Harris and Cerf (Albany: State University of New York Press, 1977), 89. 14. This theme is already explicit in Hegel's early Essay 011 Natura/Law (1802-3), translaled by T. M. Knox (Philadelphia: University of Pennsylvania Press, 1975). . . 15. See, for instance, Rawls,A Theory ofJilSiice (cambridge, MA: Harvard Un1vers1ty Press, 1971), 580 (hereafter TJ), or "PNM,'' 226.
287
566
POLITICAL THEORY I November 1991
16. See "DL," 517, or "PNM," 225. The historically specific nature of Rawls's theory continues, almost miraculously, to escape the notice of his critics. 17. By "moral realist" I intend the position that moral claims are granted cognitive status and that at least some moral statements are true by virtue of their reflecting the true "nature of things"; some version of this position Hegel surely holds given that the ultimate reality is absolute spirit (Geist). Rawls, on the other hand, has replaced the traditional epistemological search for moral truth with an essentially political, "practical task" of reaching agreement on principles of justice in accordance with a particular democratic, moral conception of ourselves ("DL," 517ff.). This "constructivist" approach will entail a "suitably constructed social point of view" that all can accept and in terms of which"moral objectivity" will now be conceived. Thus Rawls, unlike Hegel, makes no claims as to the ultimate nature of"moral facts" or to whether, strictly speaking, there even are such. 18. See Hegel, Enzyklopaedie, paras. 469-487ff. and Rawls, "PNM," 225ff. A detailed discussion of the practical similarities and differences between these two concepts would lead us too far afield. Allow me, however, to mention that in Hegel's political thought, "objektiver
Geist" refers to the world of concrete political institutions, customs, and social laws, in which a people's "spirit" (their fundamental moral principles and ethical self-conception) is publicly embodied or "objectified." Rawls's idea of a "public political culture" similarly refers to a people's shared moral conceptions, including "the basic intuitive ideas that are embedded in the political institutions and the public traditions of their interpretation" ("PNM," 225). A5 I shall argue, Rawls assumes, as in Hegel's case, that the core of these institutions are, at least minimally, "rational" and "coherent." The two conceptions thus reveal important similarities. The differ ences between them will again have to do with the strong metaphysical and ontological implications of Hegel's "objektiver Geist" which Rawls's idea does not entail. 19. See, for instance, A. Maclntyre,After Virtue (Notre Dame, IN: University of Notre Dame Press, 1981), 20, where the author includes Rawls among those who attempt to show "that the notion of rationality itself supplies morality with a basis"; C. Taylor, "The Nature and Scope of Distributive Justice" (Philosophy and the Human Sciences, 1985, 303) and "Justice after Virtue" (presented at Princeton, April 1988) 25, where Taylor interprets Rawls's view as purporting to be a "timeless, context-free theory"; M. Walzer, Spheres ofJustice (New York: Basic Books, 1983), 4ff., where Rawls is accused of resisting the "displays of history" and ignoring the "particularism of history, culture, and membership," as well as the more recent "A Critique of Philosophical Conversation" (The Philosophical Forum, Vol. 21, nos. 1-2, Fall-Winter, 198990), where Walzer faults Rawls's construction of the original position for being in "asocial space," for thinking itself presuppositionless and so on. 20. See previous note. For a similar criticism of Rawls by a thinker not usually considered "communitarian," see V. Held, Rights and Goods (1984), chap. 4. B. Williams has also voiced similar concerns, although the author now (Ethics and the Limits of Philosophy, 1985, chap. 5) exempts Rawls's theory from his attack. 21. See Walzer, Spheres of Justice, where the author clearly holds the "relativit y" thesis. More recently, however, Walzer has distanced himself from this earlier stance and admits to what he calls a "minimal universalism" across cultures in a paper presented at the New York University Law and Philosophy Colloquium, Fall 1988. 22. There are many points of contact between Hegel's dialectic and Rawls's method of reflective equilibrium, as well as important differences; to pursue these similarities and differ ences would entail an essay in itself. I wish to note here, however, that both the dialectic and the reflective equilibrium are proffered as alternative philosophical approaches to the modem methodological dichotomy of"rationalism," on one hand, and "empiricism," on the other. This point alone makes them worthy of comparison. Uke Hegel (PR, para. 2), Rawls's method rejects
288
Schwarzenbach I RAWLS AND HEGEL
567
any form of "Cartesian" a ppeal to intuition and to the self-evidence of first principles, on one hand, and to mere empirical generalization, on the other {cf. TJ, 518ff., also "Outline of
a
Decision Procedure for Ethics" [1951 ]). For both, the practical goal is a form of ethical
�If-knowledge, �
conceived as necessarily "mediated," indirect, and so forth. The primary
d ffcrcnccs between the two methods will pertain, again, to what I have called the "fundamental dafferencc" in regard to their background stances on ontology and metaphysics.
23. Sec J. N. Findlay, Hegel: A Re-examinlltion, (Nav York: Oxford University Press, 1959),
chap. 3, and M. J .. Inwood, Hegel {London: Routledge & Kcgan Paul, 1983), part 2, chap. 5, where both authors compare the Hegelian dialectic with the idea of a metalanguage. 24. According to Copi, the term "metalanguage" originates with Russell who writes in 1922: "These difficulties suggest to my mind some such possibility as this: that every language has, as Mr. Wittgcnstcin says, a structure concerning which, in the language, nothing can be said, but that there may be another language dealing with the structure of the first language, and having itself a new structure, and that to this hierarchy of languages there may be no limits" (quoted in I. M. Copi The Theory of Logical Types [London: Routledge & Kcgan Paul, 1971) 107-8). The
idea of a metalangu age, however, has come to have both a "technical" sense (as used by Tarski) and an "intuitive" one. By the " intuitive" sense (the mpari n with which is alii attempt to so co defend here), I have something closer to the later Wittgenstein's notion of an Uebersicht
(comprehensive "overview") in mind. 25. Sec Hegel, Enzyklopaedie, part 1, paras. 79-83. 26. Hegc� Lectures m the Hislory of Philosoplry, translated by Haldane (LonOOn: Routledge & Kegan Paul, 1982), 426ff. (herea fter LHP). 27. Unlike the idea of a hierarchy of metalanguages, the stages of the dialectic culminate in a full a wareness of the whole process- in a language, as it were, in which we can speak about the whole hierarchy of languages- and this is the stage of "absolute knowledge" as set forth in the Sci ence of Logic (sec Phen, 479ff.). So, too, it is not the case with the idea of a series of mctalanguages that there is only one metalanguage in which we can speak a bout a given
object-language; a given object-language may have two metalanguages with respect to it-in principle, even numerous incompatible ones. Hegel's dialectic, by contrast, (see introduction to Scien ce of Logic) is meant to generate a uniq� series of categories; the resolution of "contra diction" proceeds according to "rational necessity" or in one way only. Significantly, these last two aspects of the H egelian dialectic (aspects whereby it diverges from the idea of a series of
mctalanguagcs) also tend to be the aspects w hich Hegelian sympathizers find among the most difficult to swallow. (Sec Inwood, Hegel, 128ff.). 2 8. T hat the criterion of greater comprehensiveness is central to the superiority of Rawl 's
�
the maJOr position in his own eyes can be seen as well from the fact that in A Theory of Justice eth ical alternatives are revealed paraded before the parties in the original position and ultimately l22f .). (TJ, as inadeq utilitariani , uate, incl uding "justice as fairness" 's chief competitor an Hence one way o f fo contractan has whereas that state to is k rmulating the claim of Rawls's boo pos ition can acco case. the not se is rever the insight, unt for the utilitarian 29. For Hegel's criticism of utilitarianism, sec Enzyldopaedie, paras. 473-82, and PR, paras. 18-2 1; for Kant's, sec his Groundworlc, chap. 2. . • . 30. See Sandel Liberalism and the Limi.J.s of Justice (Cambridge: Cambndgc Unavcrsaty 21. after Justice" (1988). Press, 1982; hcre r Ll.J). For a similar criticism by Taylor, sec "Virtue 31. Sec PR, paras. 133-40. . . 32. Sees. Benhabib, "Natural Right and Hegel," (Ph.D. diss., Yale Untvci"Slty, 1975 . . 3 3. These arc the three increasingly rich and ooncrctc modes of self-chara tcn�taon hfc. corresponding to the three sections of Hegel's work: abstract right, morality, aod ethacaJ 34. Ta ylor, Hegel (Cambridge: Cambridge University Press, 1975), 13ff.
�
�
a&
�
289
?
568
POUTICAL THEORY I November 1991 35. The public authority has the duty to alleviate poverty, in Hegel's view, because the family
was originally the basic economic unit. Hist�rically, however, civil society "tears the individual
from [the soil and inorganic resources] and family ties," forcing the person to become dependent on civil society. For this reason, it is the duty
today of the public authority to "protect its (PR para. 238).
members" by providing subsistence, job training, accident insurance and the like
36. Hegel's paradigm of Geist (mind, spirit) minimally entails two consciousnesses acknowl edging common characteristics and shared ends, including the "reciprocal recognition" of this acknowledgement. Cf.
Phenomenology of Spirit, translated by A. V. Miller (Oxford: Oxford
University Press, 1979), 11lff. Depending on the specific nature of the relationship, of course (whether the relationship is between two contracting parties, two friends, marriage partners, citizens, and so on), the ends shared and the characteristics acknowledged will vary.
37. The Rawlsian idea of "primary goods" can itself be traced back to Aristotle's notion of
choregia or "props" for the good life. See my On Civic Friendship (Ann Arbor: University of Michigan Press, forthcoming), chap. 2.
38. By contrast, Sandel, at times, seems to hold a simple "majoritarian" view on questions
Republic, 7 May 1984, 17). Perhaps R. Taking Rights Seriously (Cambridge, MA: Ha!Vard University Press, 1979), Laws
of morality (see "Morality and the Liberal Ideal," New Dworkin,
£mpire (Cambridge, MA: Harvard University Press, 1986), chajP't . has argued most carefully against the altogether unsatisfactory nature of such a position. 39. E. Baker, "Sandel on Rawls,"
University ofPennsylvania Law Review 134(Aprilt985):
895ff, and A. Gutmann, "Communitarian Critics of Liberalism," Philosophy and Public Affairs 17(Fall 1985): 308-22, make related criticisms of Sandel's and Macintyre's positions. 40. That Sandel's position cannot be the Marxist one (but in fact seems to revert back to
political personality) can be intimated by the following: Even Karl Marx in On the Jewish Question explicitly claims that modem individual rights "certainly represent[] a great progress," in Marx-Engels Reader, edited by Tucker, 35. Marx's argument is not that individual Aristotle on
rights (with the exception of the right to private property in the means of production) are unimportant or even unnecessary at this stage of history but merely that they are not sufficient for "true human emancipation." Even as late as his
Critique of "The Gotha Program" (1875),
Marx argues that political rights will become "superfluous" in the last stages of socialism, not that they can be discarded now (in
Marx-Engels Reader, 531ff.). do operate with an impoverished notion of
41. I in fact believe that both Hegel and Rawls
the political person (or citizen), but for none of the reasons contemporary communitarians cite. See my "Are There Only Two Moral Powers? Ambiguities in Rawls's Concept of the Person," in Analyse&:
Kritik, (forthcoming 1992), where the author argues that even modem citizenship
requires a third "emotional" power overlooked by Rawls (and Hegel); cf. also author's forth coming
On Civic Friendship, chap. 3.
42. Rawls notes that the term "person" derives from the Latin
persona, which originally
referred to the mask worn by actors in Greek tragedy; since ancient times "a person" refers primarily to those capable of playing a public, political role ("PNM," 233). 43. See Sandel, UJ, 66, 70-78, 101-3.
44. The attribution of a personified "group subject" to Hegel was abetted, I believe, by (see Ethical Studies, 1876). Hegel indeed uses the term
Bradley's notion of a "moral organism"
"organism" in reference to the state at numerous points in the Philosophy ofRight (e.g., paras. 267, 271 ), but his usage appears largely metaphorical; elsewhere, he repeatedly stresses that the notion of
a
biological organism is an inadequate model for rational mind or Geist
£nzyklopaedie). It is on the paradigm of Geist, of
COIIISe,
that Hegel models the state.
(see
esp.
45. See Taylor, "The Nature and Scope," 291ff., and "Justice after Virtue" (1988), 21, 25.
290
Schwarzenbach I RAWLS AND HEGEL
569
46. For the confusions surrounding Sandel and Taylor's criticism of Rawls s notion of ' "desert,� see Baker, "Sandel on Rawls," 907ff. The author argues that Rawls's theory presu� poses no "pre institutional" basis for determining desert. What people "deserve" is precisely the problem of justice to be solve d. 47. See Z. A. Pelczynski, "The Hegelian Conception of the State," in Hegel's Political . Philosoph y, edited by Pelczynski (Cambridge: Cambridg e University Press, 19 71), 1-29. 48. See Dworkin, Taking Rights Seriously (1979), 47, and Low's Empire ( 19 86), chap. 1. lt shout also be noted here that whereas Hegel uses the term "justice� (Gerechtiglceit) in a narrow, lechmcal sense (cf. PR, para. 99), Rawls normally uses the term in the wider sense of the German
�
Recht.
49. See S. Benhabib, "Obligation, Contract and Exchange: On the Significance of Hegel's Abstract Right," in The State an Civil Society, edited by Pelczynski (Cambridge: Cambridge d University Press, 1984), 159 .
50. This is the case, at least, under what Rawls calls the "special conception of justice," which (in contrast to the "general conception" in which the difference principle applies to all the primary goods) pertains only as "social conditions improve" or to our specifically modem
epoch
(TJ, 83). 5 I· Hegel speaks of "transcending the standpoint of contract" (PR, 163R) in referring to the marriage agreement , which he (in contrast to Kant) denies is simply one contract among others. In Hegel's view, althoug h the modem (bourgeois) marriage agreement "begins" with a decision t�e arbitrary will (and although Hegel acknowledges the right of divorce due to practical difficultie s), the agreement has as its basis the recognition of shared moral principles and the "surrendering of the arbitrary will in a substantive union"; marriage's aim in principle is to be "inherently ind issoluble" (para. I63R). A similar reasoning holds for our relation to the state conceived now as an even more fundamental form of "rational social life"; in the state proper, is even m ore difficult f or an individual to "opt out." As we shall see, in Rawls's view, political . tnslltu tions are also not conceived on the model of contract but on that of a "social union." 52. Cf. Taylor, Human Agen cy and Language (Cambridge: Cambridge University Press, 1985), chap. 3. 53. See Dworkin, "Liberal Community," CaliforniJl Law Review 77(1989): 479-504." 54. See "The Id ea of an Overlapping Consensus," Oxford Journal of Legal Studies 7.ii (198 7): Sff., where Rawls explicitly denies that our political life can adequately be conceived on the model of a modus vivendi (hereafter "IOC").
0�
�1
55. Both Gutmann, "Communitarian Critics," for instance, and Larmore in a recent review of Sand el, Jour nal of Philos ophy 81, no. 6 (1984): 338, continue: to interpret Rawls's view of the state on the mode l of a "modus vive ndi." 56. See Kant's "Perpetual Peace" (1796). Acoording to Rawls, Kant here remains within the by "Hobbesian strand" of liberalism which conceives of "ordered liberty (as] best achieved skillful constitutio for to ts work group-interes and ) nal design framed to guide self- (familysocial purposes by the use of various devices such as balance of powers and the like" ("IOC'p. 2). ted Rawls' centra l criticism of this Hobbesian conception of the state is that such a self-interes s consensus is and "inevitably fragile" and temporary -that it could never achieve the stability
Social u nit y which marks a well-ordered political regime (ibid.). 57. Hegel, for instance, in his later years was not a democrat. For his view on democracy, see PR, paras. 273 R' 279R. 58. In Hegelian language the state must further the education of its citizens to an aware lan age, g vemment, by of " un ive rsality" or shar ends (PR, para. J87R). In Rawlsian tf not untversal shared m plement ing the ronstruct or two principles of justice, helps maintain tnterests, then them. regarding " at least an "overlapping consensus
�ess �
�
�
291
�
570
POLmCAL THEORY I November 1991
59. This formulation is Dworkin's in "Liberal Community." 60. This claim needs some qualification. Economic "social union" applies to Rawls's well ordered society
as a whole insofar as the society realizes the difference principle- that principle
which specifies that no one shall institutionally gain unless all are benefited (including the worst oft). In Rawls's theory, however, this principle is implemented from the "top down" by the various branches of government
(TJ, 247ff.); it is thus not a principle which holds within the
firm, say, or at the day-to-day level in the realm of work, where individuals may still live in "private society." See my "Rawls and Ownership: The Forgotten Category of Reproductive Labor,"
Canadian JourMI of Philosophy, Supp., 13 (1988): 139-67; also G. Doppelt, "Rawls's
System of Justice: A Critique from the Left," Nous 15(September 1981): 259-307.
61. At first sight, it looks as if Rawls exempts the work relation because it is believed to
presuppose an element of "compulsion," whereas social union is described (at one point) as something all can "freely participate" in (TJ, 529). On closer observation, however, whether or not an arrangement constitutes a social union for Rawls cannot rest on the presence or absence of all "necessity" in the type of activity, for the fact that man must eat and labor in any particular way is no more "forced" than that he must live in families or in society, and yet the latter two are among the class of social unions. The answer as to why Rawls exempts economic activity from the list of social unions must lie elsewhere; my own view is that Rawls simply accepts a major tenet of neoclassical theories of production and continues to operate with the model of a privately appropriating individual in the realm of labor. See my "Towards a New Conception of
Ownership," Ph.D. diss., Harvard, 1985, chap. 3: llOff.
62 See Schwarzenbach, "Rawls and Ownership." . 63. G. Doppelt, "Rawls's System of Justice," cf. ft. 6 1, taking a more standard Marxist line,
also criticizes Rawls for restricting "community" to the political domain and for leaving "no room" in his theory for "the model of dignity through self-affirming labor in community with others" (p. 277). 64. Women have been "mixing their labor," after all, for centuries in the domestic realm, whatever else they have been doing. To distinguish this form of activity clearly from the model of "productive" labor, I have called the form "reproductive" labor. Reproductive activity aims not, in the first instance, at the production of physical objects, exchange value, or even "human services" but at the "reproduction" of a set of concrete human relationships-in the best case, on my analysis, specific relationships of friendship or
philia. See Schwarzenbach, "Rawls and
Ownership." Moreover, with women moving en masse into the marketplace in this century, the time would seem ripe to retain aspects of reproductive labor in the public sphere. See the author's
On Civic Friendship (forthcoming). 65. Most recently again by Gutmann, "Communitarian Critics," 321. 66. I believe it holds generally, although one finds such "utopian" or forward-looking passages in Rawls as "until we bring our.;elves to conceive how this [a public understanding of mutual respect} could happen, it can't happen" ("PNM," 231). 67. This is not to say that certain "left-wing Rawlsians" have not made such a call; I include
the work of 1. Cohen and J. Rogers, On Democracy (New York: Penguin Books, 1983), chap. 6, Doppell (1981), and myself ( 1988) in this category.
68.1 might here add that the women's movement of the past century, and all the gains women have thereby won, lends renewed credence to Hegel's view of history
as
the struggle for "the
realization of freedom." It may just be that Rawls's theory, although implicitly operating with certain assumptions about the nature of historical progress, needs to elaborate such assumptions more fully to better ground his own position. 69. Why, for instance, should our common "shared" precepts about family life, the treatment of children, animals, principles of friendships, trust, and so on not be elaborated and critically
292
Schwanenbach I RAWLS AND HEGEL
571
relevant for our p ublic political life? An approach developing such a reflective equilibrium of
the private realm,
as
well
as
its implications for our public life, is developed in the author's
forthcoming On Civic Friendship.
70. Things are "internally related" to each other if (as Bradley typically expresses it) the
terms are "altered necessarily" by the relations into which they enter. Again, if it is necessary of
me that I stand in a certain relation to a certain object, so that I would not be what I am if I did not, then this relation is "internal" to me. See Hylton, Russel� Idealism, 44ff.
Sibyl A. Scltwarzenbach is Assistalll Professor of Philosophy at Baruch College of tlu! City University of New York. Site obtllined Iter undergraduate degree in philosophy
from Cornell University, was a Fulbrigltl Scholar and spent two years at the Universily of Heidelberg in Germany, and received Iter Ph.D. from Han>ard under John Rawls in social and political philosoplry. Site has published a review and numerous articles in moral and political theory in Ethics, Canadian Journal of Philosophy, Social Theory and Practice, Metaphilosophy, and New York University Review of Law and Social Change. Site is presently at work on a boolt entitled On Civic Friendship (Univeniry of Michigan
Press, forthcoming).
CANADIAN JOURNAL OF PHI LOSOPHY
Volume 23, Number 1, Mar ch 1993, pp. 75 - 100
Rawls's
Cammunitarianism
ROBERTO ALEJANDRO University of Massachusetts, Amh erst Amherst, MA 01003 USA
�os� di�cussions of Rawls's philosophy tend to neglect the strong com
urutarian strand of his theory: so much so that in the debate between liberals and co mrnunitarians Rawls's account of community has been for the most p art intriguingly absent.1 This article is an attempt to fill in the g p by of ering f a discussion of the Rawlsian understanding of commu ? �ty as it w as prese nted in A Theoryoffusticeand its possible implications or a plur alist society. 2 At the same time, I want to take issue with one of the most influ ential critiques leveled against Rawls's conception of the �If: namely, Sandel's critique of the 'individuated subject' that, in his VIew, un derlies justice as fairness. Rawls's constructions, so Sandel argu es, rest on an unencumbered self that is individuated in advance and whose identity is fixed once and for all. --
�erald Doppelt argues that 'Rawls's framework can be understood as a "commu
liberalism"' (281), focus is different from mine. See his 'Beyond Liberalism and Comrnunitarianism: Towards a Critical Theory of Social Jus tic e,' Philosophy and Social Criticism 14 (1988) 271-92. Susan Moller Okin discusses the role of feeling in Rawls's account of justice, she does not address Rawls's vision of community. See her 'Reason and Feeling in Thinking about Justice,' Ethics 99 (1989) ity. See 229-49. James W. Nickel concentrates on Rawls's view of political commun his 'Ra wls on Political Community and Principles of Justice,' uw and Philosophy9
mtarian
but his
but
(1990) 205-16.
2 Though Rawls's articles after A Theory of Justice include important de �elopmen � and, in some cases, modifications of his previous arguments, I do not think that his account of community has been substantially altered.
295
75
76 Roberto Alejandro
Sandel's critique has largely gone unchallenged. Some liberals seem 3 to accept it; others attempt to improve on Rawls's formulation;4 and still 5 others simply ignore it. Actually, the thrust of liberal arguments against Sandel is consistent in avoiding a critical examination of Sandel's attack
?
on the Rawlsian self, while confusin , in Charles Taylor's words, issues
of ontology and issues of advocacy. Those arguments tend to concen trate on Sandel's vision of community, which is, in my view, the weakest part of his analysis. My goal here is to examine Rawls's text to argue that another reading of the Rawlsian self is possible. Needless to say, I follow here the hermeneutic principle that a text goes beyond its author's intentions, and so I attempt to reconstruct Rawls's argument along lines that, to the best of my knowledge, have been unexplored. I begin by discussing Rawls's communitarianism (sections I-IV), then
I present Michael Sandel's critique of Rawls's notion of community
(section V), and conclude with some remarks about what I take to be the disturbing uniformity that emerges from a Rawlsian community (sec tion VI). I will suggest that some central assumptions of Rawls's theory
of justice are either contradicted or completely abandoned in his com
munitarianism. If my argument is correct, the first casualty of a Rawl
sian community may come as a surprise. Yet it is the case that one of his
central assumptions, the priority of the self over its ends, is either denied or substantially modified.
I Rawls's account of community is anchored in the goals of cooperation,
stability, harmony, and transparency. Cooperation entails mutuality and reciprocity, which means that members of a Rawlsian community are 7 going to share in the distribution of benefits. Stability implies that the members' cooperation with one another is expected to be one over a
3 Willi am Galston, 'Pluralism 4 Will Kymlicka, 'Liberalism
5
18 (1988) 181-204
and Social Unity,' Ethics 99 (1989) 711-26
and Communitarianism,' Canadian Journal of Philosophy
Amy Gutmann, 'Communitarian Critics of Liberalism,' Philosophy and Public Affairs 14 (1985) 308-22
6 'Cross-:Purposes: The Liberal-<:ommunitarian Debate,' in Nancy L. Rosenblum, ed., LiberaliSm and tht Moral Lift (Cambridge, MA: Harvard University Press 1989) 159-82
7 John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press 1971), H. Subsequent references will be integrated into the text. 296
Raw/'s Communitarianism i1
complete life. Harmony and transparency mean that individual plans are comp lemen tary (563) and, more importantly , the individual is trans parent to himself to the extent that his ends cohere with each other,8 and he is transparent to others to the extent that his plan of life is part of a larger social plan just as individuals, through their institutions, are part of 'a social union of social unions' (527) 9 . These goals inform Rawls's vision of community which possesses what seems to be a neglected feature of Rawls's philosoph that is, his y; comrnunitarianisrn does not depend upon the original position and its parti es: it relies on his understanding of associations, institutions, and moral psychology. In all these areas, justice is the principle that, like a red thread, orders and regulates both the Rawlsian individual and the Rawlsian community. I will address these issues in tum. Rawls conceives of associations as institutional settings that comply with the precepts of justice and provide a space for mutual recognition and apprecia tion of the person's abilities Associations socialize indi . �iduals into the principles of trust and friendship, strengthen the mdividual's self-esteem, and provide a 'secure basis' for the worth of their members (442). Associations thus occupy a central place in the Rawlsian universe since self-esteem, in Rawls's theory, is 'the most import ant primary good' (440). Or to put it differently, since t�e good of self-esteem requires that our person and deeds be appreciated by other s; and since 'associative ties' strengthen this aspect and 'tend to reduce the likelihood of ailure and to provide support against the sense f of self-doubt when mishaps occur ..' (441), the indi idua 's member � . . /her ship in ass ociations is not an attribute, but a substantial trait of his character. What is im portant is that in his descriptio ns of associative ties, Ra�ls not only pres ents a picture of moral personality which is far from bemg the unen cum bered self so often ascribed to his theory, but also, and m�re . un portantly , his reasorung underrrunes th e pnon'ty of the self over Its . ends (560), whi see tion ch m s to be one of the core elements of his concep
�
.
·
.
·
·
·
.
8 J Rawls, 'I
Journal of Philosqphy 77 (1980) 512-72, at 529; hereafter q uoted as '[)eweY Lectures ·
f
.
·
9 The cy which assumes Rawlsian community is a space o harmony and transparen . ars as a natural Jus ce inclination toward ��· insists, a 'stable . capacity, a built-in mechanism for human 'ty and ts conception of justice ... elicits men's natural sen 0 he says : (502). It is thus possible to apply to Rawls p there is a1 5 describing John Stuart Mill's theory: '[o]ne of a w citizenS' (502). should be harmony between his feelings and those 0
that men have a
natural
..
��
sooa�ility, an ' fellow feeling �� f when : what � wants that
perso;'�':
297
78 Roberto Alejandro
of justice. Let us explore why. In Rawls's account, there are three princi ples which I will call the principle of
external confirmation, and
mutual recognition, the principle of dependence. Mutual confirma
the principle of
tion means that the conviction of the worthiness of the individual's endeavors is placed, not in an unencumbered subject of possession, but in a historical individual who is guided by social standards of judgment. '[U]nless our endeavors are appreciated by our associates,' he says, 'it is impossible for us to maintain the conviction that they are worthwhile ... '
(441). External confirmation means that the person's endeavors need to be confirmed by his/her associates in a community of shared interests: 'what is necessary is that there should be for each person
at least one community of shared interests to which he belongs and where he finds his endeavors confirmed by his associates' (442, my emphasis). It is the princi
ple of dependence, however, that is central to understanding the impli cations of Rawls's communitarianism.
In his view of associative ties, the
individual has to obey the norms regulating his/her group. If the indi vidual acts wrongly, Rawls argues, 'he has failed to achieve the good of self-command, and he has been found unworthy of his associates
upon 10 whom he depends to confirm his sense of his own worth' (445, my emphasis). I suggest that the principle of mutual recognition, the external confir mation individuals need for their endeavors, and the dependence the individual has on others to confirm his/her own worth deny the priority of the self over its ends. There are two reasons to explain why this is so.
First, if our endeavors, and for instance the ends they pursue and the
identity they shape need to be appreciated by others, and if it is this social
appreciation that determines the worth of our endeavors and ends, we are no longer prior in any meaningful sense to our ends. Those ends are determined in important ways by social (principles accepted by society) and communal (principles recognized by a 'community of shared inter
ests') standards of worthiness. Second, if the standards to confirm the individual's endeavors are provided, not by himself as an unencurn-
10 Along the same lines, he also argues, that the 'soundness of our convictions' depends upon a 'common perspective.' 'The acceptance of the principles of right and justice forges the bonds of civic friendship and establishes the basis of comity amidst the disparities that persist. ... But unless there existed a common perspective, the assumption of which narrowed differences of opinion, reasoning and argument would be pointless and we would have no rational grounds for believing in the soundness of our convictions' (517-18, my emphasis). This assertion suggests that though the theory is individualistic, the conception of rationality informing it is t �·' the soundness of our convictions' depends upon a 'common per
socia�. TI_'a
spective, which turns out to be a set of beliefs accepted by a community. This is another instance of Rawls's communitarianism.
298
Rilwl's Communitarianism 79 bered self, but by a 'community of shared interests' (a claim that Rawl sian liberals tend to neglect), 11 and if the individual confirms his/her own worth, not by standards he/sh e has created, but by norms and criteria accep ted by his/h er assoc iates, then, the Rawlsian self is not so prior to its ends, after all.lts self-e steem is not anchored in values the self derives from itself, but in values that its associates accept. But if the ends that my plan of life pursues requir e the approval of my peers to confirm my sense of worth, namely, if I consider as my primary concern the approval my ends may receive from my associates, I am no longer prior to my ends. In an importa nt way, the ends I choose are determined by others' approva1.12
11
12
See, for example, Will Kymlicka, ' Liberalism and Commu nitarianism '. This tum of Rawls's communitaria ism shows how mistaken is the attempt to n present the liberal commun itarian debate as a conflict between society and the individu al's judgment. For this misconstruction, Will Kymlicka, 'L�bera�m �d Communit arianism.' I t could be argued, however, that the self is still pnor to Its ends in the sense that it can revise them. Raw ls himself claims that 'free persons
see
con ce1ve of themselves as bemgs who can re vJ.se an d alter their final ends and who . e fust . giV . · these rnatters' ('Reply to Alexander pnority to preservm Ul . . g ther. r liberty . d an Musgr a ve,' Quarterly Journal of Economics 88 (1974), 641). I
·
.
.
uses
·
.
.
.
ens10ns ·
·
.
·
•
.
are
·
.
•
::!" �
.
·
·
.
0�
ru:
rrns conf
�andards
�us
299
.
.
� u:::k .
.
80 Roberto Alejandro
Rawls's view of associative ties thus suggests that the Rawlsian indi
vidual is not the autonomous self that, in Dworkin's description, leads
his life from the inside13 and that, in Kateb's account, seems to be suspicious of social standards.14 It is rather an individual whose very capacity for judgment may be compromised by his membership in a community of shared interests.15 Better still, this individual may be
unwilling to revise his ends in order to maintain his associates' approval and, for instance, his self-esteem.16If the individual comes to disapprove
of the values of his associates, and there is no other association to which he may belong, he might prefer to go along with his peers, thereby avoiding any damage to his self-esteem. 'He is apprehensive lest they reject him and find him contemptible, an object of ridicule'
(445).
This
conception of the self shows how misleading are the fixed boundaries that are often found in liberal arguments against communitarian dis courses. In open contrast to those arguments, the Rawlsian individual appears as one who needs a community of shared interests which provides standards of worthiness and allows him to preserve his self-es teem: associations and communities provide 'a secure basis for the sense of worth of their members'
(442).
It may be argued, however, that though Rawls's communitarianism
emphasizes mutual recognition and communal standards of worthiness, he still provides enough room for the individual's judgment by insisting that, 'for the purposes of justice,' citizens are to 'avoid any assessment of the relative value of one another's way of life'
(442). But this argument
external so u rce of his self-esteem. A Millian or an Emersonian self would be willing to stand up for its moral independence regardless of what a community of shared
interests may do. But it is not clear that a Rawlsian self is equally willing to risk its self-esteem in order to preserve its moral independence. Rawls's arguments, then,
suggest a tension between the self's moral independence and its self-esteem, and
the latter, after all, is the most important good. Kymlicka's analysis does not explore these tensions in the Raw !sian construction of the self.
13 Ronald Dworkin, 'In Defense of Equality; Social Philosophy and Policy 1 (1983) 24-40 14 George Kateb, 'Democratic Individuality and the Meaning of Rights,' in Nancy L.
Rosenblum, ed., Liberalism 1md the Mor11l Life (Cambridge, MA: Harvard University
15
Press1989)183-206
It is worth exploring the similarities between Rawls's comrnunitarianism and John
Dewey's vision of commw\ity. This is a problem that is beyond the scope of this paper.
16
It would be worth exploring whether a Raw!sian community contributes to the same docility that George I
300
Raw/ 's Communitarianism 81
is unconvincing and Rawls's contention is short-lived . He contradicts it to the extent that the individual 'regards the virtues, or some of them anyway, as properties that his associates want in him and that he wants in himself' (444 ) . If no one assesses 'the relative value' of ways of life, why should we care about virtues or propert ies that our associates want in us? If they want certain properties as traits of our character, they are making judgments on the value of our way of life. To sum up, in the associations Rawls describes the individual is no lo nger
prior to his ends. Quite the contrary, presumably those associations have their ends already and possess a clear understanding of the virtues required by them.
The individual may be prior to his ends before entering an association. Once he enters it, that priority is blurred. For in associations he needs the presence of others to confirm his worth and endeavors, and it is his membership in a 'community of shared interests' that strengthens s
�
self-esteem, not the other way around. If he finds himself in an associa
tion (like the family), and he is shaped by its values, he has never been
prior to them. Those values and the attachments they carry have formed his character .17
II
We may have a better assessment of Rawls's view of associations by examining it against the backdrop of the principles of moral � s�chology. It is not my inte bu ntion to discuss here the validity of th�s� prmCiples, ch rather to show approa that the parties of the original position the sense how Varus of on · rsi ve s • hin hi g uces pomt as soon as Rawls mtr · od . f· . pnn 0 JUStice 1s the 5 a ting t In . c1ples SOCiety. acquired in a well-ordered of oral psy es in t e sechology' Rawls's model assumes three stag m quence of moral development, and these stages stress 'the formmg o f
:
.
·
·
�
. . groups, thus showing 1 7 It IS possib le to argue that a person may dec!de to . . other
JO� � if�� ; J7e 1firs� !' which .
two self-esteem. There are that she or he is prior to communal stan�ards rm affi to others on nds f de �rguments to reply to this contentiOn. F�t, group that helps a ave an t rath tts worth, it may be willing to comp rorruse there is one from it to COrJStitute its identity. Second, even � the self le�v� 'ts ou ' rience with the e pe 1ts 1t ve ga which tt · cannot escape, the group the poss1biiity to open principles of justice, namely, its family. A se.If that always line with John in more . is that 15 on of leaving its group is not a Raw!sian self; Jt IS �s. Will Kymlicka, for example, Stuart Mill's account of individuality than ' g to examine its ends, · self IS ways illin subscribes to this notion that the Rawls'an 1 on in n. 12. discUSSi my and, for instance, its membership in a group. See
w � alls
.
30 1
w
.
82 Roberto Alejandro
attachments as final ends'
(495). These stages are the morality of author (462-79).
ity, the morality of association, and the morality of principles
In the first stage, the child develops her conception of justice in the
family, and, in the second stage, the individual deepens her sense of justice through her association with others. It is the third stage, however,
that is really crucial to understand Rawls's views. The third stage, the
morality of principle, assumes that individuals develop allegiance to the
principles of justice regulating their society, and become attached to the highest order principles expressed in a public conception of justice (473).
The morality of principles holds that if 'we and those for whom we care
are the beneficiaries' of just institutions, those institutions and the bene fits we derive from them will 'engender in us the corresponding sense
(474). Accordingly, 'we want to do our part in maintaining these arrangements' (474). 1f we betray our sense of justice, we are likely
of justice'
to experience 'feelings of guilt by reference to the principles of justice'
(474). Once we arrive at this stage, the 'complete moral development has .' (474). In Rawls's view, the morality of principles is
now taken place
..
the final and highest stage in the individual's moral development. For
the morality of principles does not depend upon our relationship with our parents. Nor does it depend upon ties of friendship and mutual trust.
It depends upon allegiance to the principles of right. If we violate these principles, we will feel guilty, not because we have harmed our parents
or friends, but because we have harmed people whom we do not even know. This morality does not depend upon personal attachments, but upon moral attachments to certain principles which allow us to love human beings regardless of our relation to them. The sense of justice, as Rawls says, 'is continuous with the love of mankind'
(476).
The principles of moral psychology are 'reciprocity principles'
(453),
and justice itself is anchored in reciprocity. That is, justice is the capacity
to answer in kind, which means that it depends on everyone's willing
ness to do his share. In an important way, justice requires the activities
of other selves to be preserved. If others do not do their fair share, their attitude may weaken our commitment to justice.18 Though the principle of reciprocity is central to the acquisition of a
sense of justice, it finally swallows up the Rawlsian self. For this self is supposed to be the only source of aims and ends (the self is prior to its ends), but the Rawlsian self develops sentiments and attachments, not
18 It is possible to argue that the principles of reciprocity, as Rawls understands them, do not necessarily apply to other virtues like love, excellence, courage. Love, for example, does not depend on the other's willingness to reciprocate. People tend to love their relatives and friends, even when they do not appreciate that sentiment. 302
Rawl's Communitarianism 83
out ofitself, but out of the influence it experiences in its dealing with other selves. The 'three laws' of moral psychology governing the development of the Rawlsian self 'assert that the active sentiments of love and friend ship, and even the sense of justice, arise
from the manifest intention of other persons to act for our good. Because we recogn that they wish us well, ize
we care for their well-being in return ... . The basic idea is one of reciproc ity, a tendency to answer in kind' (494, my emphasis). So justice does not rely on altruistic feelings, but on our expectations of a response in kind, whereb y the agency of the self, a self that is supposed to be prior to its ends, appears as reactive; namely, the self reacts to 'the actions of others' (494). The self needs both institutions and 'the actions ofothers' to develop a ma ture sense of justice.1 9 How, then, can the self be prior to its ends when its sense of justice, the sense that allows it to be prior, is not prior (itfoiiows from the actions of others)?20 I want to suggest two further problems which are related to Rawls's notion of the priority of the self over its ends, and his claim that the 'essential unity of the self is already provided by the conception of right' (563). First, if a developed sense of justice requires the presence f o er , � ? the self may find th at in the process of acquiring this se� of Justice, It . may also acquire other ends without ever having the pnonty Rawls ascribes to it. Second, if this is so, the self is not going to follow a sequence in �hich, first, it develops its sense of justice, and then jus ce p vides � uruty to its ends. The self may be open to several overlaps � w ch the e�d it m ay acquir � e in the process of developing its sense of JUStice may Sigruficantly affect the nature of that sense. . Seen from another perspective, it is possible to a that eve� assummg � � that the uni ty of the self depends upon the pnnc1ples of nght, th� th Principles rely on the presence of others, and that presence along �I the associative consti may res �te ties that the good of self-esteem requi a part of th e t so is e individ ual's conception of the good. If this : of the self, = in Rawls's own terms, cannot be given by the p�c•p . de right alone. The m individual's good may play an important role that unity. denv Stated differently' since the principles of right are �ot b i lated · teractions' those Y so individuals, but acquired through soc1·a1 m
�
�
�
�
�� J
-a19 <:" as I characterize transform Rawls puts it: '!the three Jaws of moral psych0I ich in wh anner m the . . e from our rec gruzmg tions of our pattern of final ends that ans institutions and the actions of others affect our good ' (494)·
�Y.
20 This see p an intriguing problem: ming twist of the Raw !sian discourse may b . do �� �u position when they how can selves choose principles of justice in ttw: on d the insti � tions an not even have a developed sense of justice, �hich this problem here. influence of 'the actions of others' to arise? I will not a ress
�\IU'e5
303
84 Roberto Alejandro
principles are acquired through social institutions and associations that may become part of the self's conception of the good, and that may 21 contribute to shape its unity. To sum up: the self cannot be prior to its ends since a developed sense of justice requires the actions of others, and in order for us to value those actions we need a 'common perspective'
(517), which assumes commu
nal standards of worthiness to judge our endeavors and ends. Accord
our sense of justice and our ends overlap in the community of 'shared interests' where individuals confirm their worth and feel recognition. If this
ingly,
argument is valid, it suggests that the priority of the self over its ends
either vanishes or ought to be viewed in a more complex relationship.
III What is important, for our purposes, is that the principle of reciprocity and the Rawlsian conception of associations and institutions it informs bring to the fore a different picture of human personality. The priority of the self Rawls describes in Part I of Theory ofjustice is no longer present.
Now he abandons the mechanical view of a self that is always prior to its
ends, and proposes a far more complex assessment of the individual's character. 'Moreover,' he says, 'the social system shapes the wants and aspirations that its citizens come to have' (259). 'It [the social system]
determines in part the sort of persons they w�t to be as well as the sort of persons they are' (259, my emphasis). This view suggests that before
individuals choose the sort of persons they want to be, they have already
been shaped by standards and values which in part determine the kind
of persons they are. This is precisely Sandel's reply to Rawls, but he does
not acknowledge or fails to notice that it is already present in Rawls's philosophy. To put it differently, in Rawls's account of community, we don't find abstract parties and principles that are beyond any kind of contingencies. We find historical individuals who are part of a social tradition
(525),
21 Rawls himself suggests this reading when he writes: Thus, a conception of the good
normally consists of a more or less determinate scheme of final ends, that is, ends we want to realize for their own sake, llS well as of attllChments to other persons and loyaltin to various groups and associations. These attachments and loyalties give rise to affections and devotions, and therefore the flourishing of the persons and associations who are the objects of these sentiments is 11lso part of our conception of the good' (J. Rawls, 'Justicus Fairness: Political, not Metaphysical; Philosophy tmd Public Affirirs 14 [1985) 233-4, my emphasis). 304
Raw/'s Communitarianism 85 and require cooperation and social union
for their completion: 'It is only
in active cooperation with others that one's powers reach fruition. Only
in social union is the
individual complete' (525, my emphasis). Thus, a well-ordered society is not only a social union of social unions pursuing 'the good of community' (520), but also its members 'have the common aim of cooperating together to realize their own and another's nature in ways
allowed by
the principles ofjustice' (527, my emphasis). This is a claim that considerably expands the scope of justice since a well-ordered society is not only a rna tter of a llegiance to the principles of justice, but a commu nal enterp rise to realize each member's nature. This vision of community �uggests that the notion of pluralism underlying a Rawlsian community IS not the sam e pluralism that characterizes actual liberal societies. Rawls's understanding of pluralism assumes a strong similarity of inter
ests and values in such a way that each member is able, through a cooperation regulated by justice, to realize his own and another's nature. The present understanding of pluralism which Rawls himself charac terizes as grounded in incompatible comprehensive doctrines and vi sions of the h uman good is replaced by a new pluralism defined by complementarity .22 We are approachi ng an intriguing metamorphosis in Rawls'� argu ment: what began a s a conception of justice as the fundament } vt e � e ba sic structu re of society, ends up as a 'social union of �tal . 10 which jus tice is the most fundamental virtue of the mdlVtdual s life. '... [I]t follow the s,' he says, 'that the collective activity of justice preeminent fo ons, condtti favorable rm of human flourishing. For given it by mainta ining these public arrangements that persons b�t expre
0� � �0 �
�
.�
�
�;
thezr natu re and achieve the widest regulative excellences of whzch each capable' (529, my expre� our emphasis). Or better still: 'But the desire to nature a s a free and equal rational being can be fulfilled only by actzng on . the Principles of right and justice as havingfirst priority' (574, my emp�as�).
�
oly JUSti�e Rawls 's comm unitarianism turns out to rest on the mono exerci s acter. And thisf se over institutions and the individual's char Illon op oly ent o elem an is anything but plural. Moreover, there is dogmatis t best a e o m in the claim that one virtue, justice, is the on y � a act, expres s f se the indi vidual's nature. This seemmg dogmatism 1s, m ·
·
!
�
-22
Se,e for example, ibid
.
rt of a moral view ' •one pa 15 claims that a 'conception of JUstice component of ant rt . Rawls's philosop {512) But m . . ti'ce is the most JJI\Po . hy, JUS t of . �d fines 'the moral pom . e hat)ushce ys morality. Actually, t h e sa at other moments
23 It is true th at he
.
view'
. . . but
(491).
305
86 Roberto Alejandro
central trait of the Rawlsian construction of human nature. The self's nature, and by this I mean here the self's core, is not a complex of different virtues that may place different demands and conflicting claims on the individual's character. Nor is it a space of painful dilemmas. It is a uniform dimension which only requires justice to best express itself. One of Rawls's early formulations about the sense of justice may explain why justice is so central to the individual's nature. It is so because, in the self's core that Rawls proposes, the foundation of our humanity is justice. Our true nature and our true self require us to be just, and if we disobey the precepts of justice, we ' disfigure' ourselves. 'Put another way,' he writes, one who lacks a sense of justice lacks certain fundamental attitudes and capacities included under the notion of humanity. Now the moral feelings are admittedly unpleasant, in some extended sense of unpleasant; but there is no way for us to avoid a liability to them without disfiguring ourselves. This liability is the price of love and trust, offriendship and affection, and of a devotion to institutions and traditions from which we have benefited and which serve the general interests of mankind?4
The 'plurality' that Rawls's well-ordered society seeks to protect turns out to be more problematic than he suggests. The individual's attach ments (associations) and aims (final ends) are plural. But the individuals who, in Rawls's account, are behind those attachments and aims, are the same. They all regard justice as the regulative principle of their lives. They all aim at full cooperation over a complete life. They all realize their natures in the activities of other selves. They all belong to one or another association. I suggest that Rawls's conception of pluralism is restricted to what his account understands as external features of the self: namely, those traits, attachments, etc., that his philosophy tends to conceive of as external characteristics of the self. The self's core, by contrast, is anything but plural. It conjures up an image of sameness that turns out to be the necessary requirement for the goal of harmony Rawls relentlessly pur· sues. My argument, however, is meant to propose a stronger claim than the view that his communitarianism undermines his pluralism. I suggest that the monopoly justice exercises in Rawls's theory offers an impover ished vision of society. That is, 'the fact of pluralism' that Rawls rightly invokes, suggests a more complex view of the ordering of the virtues in both the individual's character and in associations.25 Let me spell this out.
24
J. Rawls, The Sense of Justice,' Philosophical Review 72 (1963), 299, my emphasis
25 When Rawls refers to associations as 'an institutional setting'
306
that is 'just' (491), it is
Rllw/'s Cammunitarianism
87
. H��n nature is something extremely complex. Through a complete . life, mdtvtduals may find themselves with different orderings of the moral excellences that define their character. To assume, as Rawls does, that the place of j ustice in the individual's character must be fixed namely, it must occupy the first and foremost place - is to deny the complexities accompanying the changes of the individual's character and its quest for self-understanding. This fixity that Rawls attributes to justice in the individual's character is hardly compatible with 'the fact of pluralism,' unless one assumes that pluralism does not refer to the individual's character and its willingness to give justice an absolute priority, but to traits the individual chooses once he has accepted the primacy of justice. Needless to say, such a vision of pluralism as a space of external traits surrounding, like appendages, a uniform human na ture would render pluralism something quite different. The complexity of the human condition makes it clear that human nature may require a plurality of orderings of the virtues as well as a plurality of ass ociations, or even the absence of them to realize itself. If we assume that individ uals need a plurality of associations to realize their nature , it is likely that individuals will rank the importance of those associations on the ground of how they contribute, in t eir j u g ment, to realize their nature . A devout individual will certainly think that his chu rch, and the particular ordering of the vi es it � r�up _ o � ses, is what best expresses his nature. Yet a Rawlstan mdtvtd�al inhabiting a 'well-ordered society' would challenge hi�. Deep �ommit ents rooted in comprehensive moral, religious, or philos ophical d � trines, he would say in line with Rawls's articles after A Theory of belong to the private sphere of individuals, not to the public sphere. Those deep com es are not mitments arising from comp rehensive d . llo � wed to play any role in deliberations concerning the pnn�tples of Justice J wlstan con · depends upon 'mtultive . ushce · · 1·deas.' What the Ra · struction cle exp�ess and arly suggests then is that individuals best . realiZe th ' . that · tnnes . oc d ive eir nature' not by following the comp rehens , 1deas e uitiv some pn. va int te associations require, but by accepti·ng th e ' . ith the rules . which const . itute the foundation of J UStice and comp1Ying w · dly, IS of public allege ins titutions since the primary concern of justice,
zt:u
�
�
?
fustlc�
�
'
-er . . Ies ofu not clear whether he is refernn _l stic . e,or wheth . two pnnap . g to his . associations embody is different from Rawls's two pnnaples
thejustice
Philosophy and Public Ajfoirs 26 J . Rawls, The Priority of Rights and Ideas 0� the Good us , xford Journal of Legal O 1 7 (1988) 251-76; 'The Idea of an Overlappmg consens ' Studies 7 (1987) 1-25 '
307
,
88 Roberto Alejandro
the basic structure of society.27 As he says: ' ... the desire to express our nature as a free and equal rational being can be fulfilled only by acting on
the principles of right and justice as havingfirst priority' (574, my emphasis).
This is consistent with the monopoly Rawls ascribes to justice, but it is certainly doubtful that many individuals will accept this account as representative of their character and dispositions. That is, many indi viduals may question that compliance with the principles of justice and support for public institutions is more central to best express and realize their nature than their private associations and their comprehensive philosophical, moral, or religious doctrines.28
The monopoly justice exercises in both a Rawlsian personality and
community explains Rawls's view that a genuine mature morality
requires a relation, not to persons, but to institutions that embody principles of justice.29 A genuine mature morality is thus a question of
impersonal relations. This view, in Rawls's argument, is axiomatic. It
is equally problematic. Many people may argue that a mature morality requires a relation to, and a concern for, people they know and care deeply about, not people who are strangers. Rawls opposes this view
and stands for 'the love of mankind.' But though Rawls accords a substantive status to this notion, it does not deliver what it promises.
Actually, the 'love of mankind' ends up as a very meager offer. For on close scrutiny, it is nothing more than the individual's willingness to
pay his/her taxes so that public institutions will take care of distribut
ing the social product equitably and tax payers will feel that their taxes
have benefited the least advantaged members of the community;
members whom they do not even know. This is the epitome of a higher
morality. The Rawlsian 'love of mankind' and the 'complete' moral development it purports to represent does not assume an immediate
relation between individuals and strangers, but between individuals
27 In his articles after A ThLory of Justice, Rawls has not modified this position. Justice is a highest-order interest that ought to regulate our character and public life. See his 'The Idea of an Overlapping Consensus.'
28 In his articles after A Theory ofJustice, this problem has become more complicated. Now individuals are presented as having both a public and a nonpublic identity. Justice must define the individual's public identity, while comprehensive doctrines may define his private identity. But those doctrines have to comply with the principles of justice. Again, many citizens may question the priority justice contin ues having in the definition of their nonpublic identity. 29 This position, it should be said in passing, is in tension with his conception of justice as reciprocity: we expect the other person to respond in kind. But in a Rawlsian society, our expectation is mediated by institutions.
308
Raw/'s Cornmunitarianism 89
and the sta e, and thro�gh the state, individuals relate to strangers � . The . Rawlsian higher morality thus seems to be a statist conception: it needs to assume the centrality of the state as an aggregate of institution s that connect, through the distributi on of social products, individuals with one another.
IV A P�sible liberal reply might take issue with the views advanced in the previous section and claim that only by accepting the priority of justice, both in social institut ions and in our lives, do we avoid oppression, and that those who invoke the complexity of human nature to dispense with �t priority may be willing to engage in or defend oppressive practices. This possible reply seems to be unaware that Rawls himself may offer an argument to dispute it He might leave aside his view of the develop . ment of the sen se of justice and argue, in keeping with the first part of A Theory ofJustice, that the centrality of justice refers to social institutions, not to the ind ividual character. When those institutions are regulated by e Principles of j ustice, he might claim, it is irrelevant whether some mdiv� duals may be prone to oppress others. A frame or� of ri�ts �d � . the distrib ution of the social product according to pnnCiples of JUStice Would prev ent them from carrying out their designs. . . Though this resp onse, from a Rawlsian perspective, IS v�lid, I prefer to put it aside and address the possible reply on its ow� ments. I assume t most con temporary liberal theorists would not diSpute the follo� mg assertio n: many citizens of liberal societies do not accept the cia� that justice is the virtue that best expresses their nature. In keeping with this asse ion rt , I suggest that throughout her life, a person may order �d reorder th e virtues orienting her character, without thereby advocating oppressl· On. A person may place a higher premium on 1°ve or truthfulness , and even though j ustice is not the first VI•rtue of her character, that . erson is not advocating injustice. Let me thus leave aside the pia�e of . JUstice in our character, and address it in the context ?f ��tut�ons. Rawls c1a un ons, · s that justice ought to be the first virtue of soctal instituti . . e and the rep nty ly I am considering asserts that without that pn� Wo ld be e, � Willing to defend oppression. From a liberal pers��� u conten . . mdIVIdua Is om . fr tion 1gno ing s . an res some pervasive conflicts . ch0I• ct · the context of liberal societies . . That · t say there IS a confli ces � �litical rights and When a li beral discourse stands for personal f, ding' the right rt 0PP0 unities to dev ise a (rational) plan of life, while ' · of 10 dIVtduals to ge m practices that' participate in groups or enga
�
�
·
p
�
15 0 an�
·
·
309
�� en
90 Roberto Alejandro according to liberal standards, are oppre�sive.
30
A religious group, for
example, may defend the subordination of women and hold principles that are oppressive in the light of liberalism, but individuals have the
right to join, if they wish, that group, and the state should not interfere
with their choices. This suggests that the priority of justice in social
institutions may coexist with oppressive groups and practices provided they are organized and carried out on a voluntary basis. The priority of a framework of rights and the distributive paradigm insinuates another problem that is more closely related to Rawls's con ception of justice. Suppose a society whose social product is primarily derived from military industries. This society has a framework of rights and complies with Rawls's second principle: inequalities have to benefit
the most disadvantaged members of society. This society is, then, just
according to Rawls's standards. For in Rawls's account, the strategy that creates the social product is not at stake: what is morally and politically relevant is the distribution that strategy makes possible. Yet some people may find oppressive a society where the moral character of its central economic activities is not deemed important to decide whether that 31 society is just. A society may become wealthier by producing and selling missiles, and its wealth may be distributed along Rawlsian prin ciples, but the moral character of that society and its justice would be dubious. Along similar lines, a society may derive its wealth from the activities their industries perform abroad. These industries may pay low salaries to, say, Guatemalan women or Mexican citizens, and low wages abroad may increase the profits those industries make and the taxes they pay at home, the same taxes that are part of the social product the state distributes according to Rawls's principles. Once again, the content of the practices underlying the production of the social product is irrele vant. And that society, whose social product depends on what many may consider the oppression of foreigners, would be just according to
the Rawlsian paradigm.
My response to the possible liberal reply is thus threefold. First, the complexity of human nature may require different orderings of the
virtues without thereby implying that if love replaces justice as the
preeminent virtue of our character, we would be prone to oppress other
30 This is not necessarily Rawls's case, but it is
societies.
a
well-known scenario in liberal
31 A possible liber� �ly is that such a conception of oppression is not rational, and that such a
SOCiety
oppression is
a
IS not oppressive. But the Rawlsian view of rationality and among others, and, even in liberal societies, there is hardly
one view
consensus on it.
310
Raw/'s Communitarianism 91
pe�ns. �cond, a liberal order anchored in the priority of justice may coe�t With oppressive group s and practices which ought to be toler ated m the name of indiv idual freedom. Third, a Rawlsian society may carry out a just distribution of the social product in the light of Rawls's model, but that society may depen d on strategies which some members may consider oppressive. So the claim that only by accepting the prior ity of justice (I take this to mean the Rawlsian conception of justice) we avoid oppression, is unp ersuasive.32 Suppose, however, that in effect, the complexity of human nature may entail a particul ar ordering of the virtues which invites oppression. In this case, the Rawlsi an claim would be right, and justice must be the first virtue of our chara cter to thwart any inclination on our part toward oppressing others. But if this is so, if the Rawlsian argument is that "':i�ou t accepting the priority of justice in our character, without recog ruzmg that justice, and nothing else, is the virtue that best expresses our nature, and with out acknowledging the transparent harmony of Rawls's commwti tarianism, we would be inclined to oppress others, then tha t arg ument would be propounding a metaphysical conception of �e self as well as an intriguing notion of choice. Justice, the most _ unportant virtue of our character, would be beyond our choosmg fa�ul ties sin ce, if we choose, we might be inclined to choose op�resslOn rather than justice.33 The self, in this view, would have to be JUSt, for otherwise it would disfigure its humanity. But this contention would presuppose a conception of the self that is antecedently given, and f�r tnst�ce, it would ratl� not be political. It is not pa� of �e �em� _ tradition fro m which Rawls derives his version of political hber�lis':'. . ,. For m · that tradi. tion · · I"deas ' claun · ing that JUShce there are not mtuitive must be the first to prese�e �ur want we v i�tue of our character, if humaruty, ing that JUstice hold · I·dea and even less is there an mtwtive requires ha ag�ment ersal rmony. If those ideas exist, there is no univ on them. So with the deal t the possible reply leads to a trap. If I"t canno . . It Complex vtrtues · the ity of human nature and the different ordenngs of .
·
·
•
•
--
32
the one propounded, does
. w It IS . of soci. ety such as . orth noticing that an alternative not · an George J(ateb' respectivel .1chae1 Walzer' y, by Alasdair Macintyre, M' society a such that . rely on the fixed y means clear place of justice, and It IS not by an would be an opp ressive settin g.
vtew
d
tice e,explidtly, (l) thatj� 33 H my a rgument is correct, Rawlstan . rals ought toargu . libe accept the fixed we after o b nJ .d is not an o ject o choice; (2) that diversi� is vall f acter; and (3) that the ti �f place justice must occupy as the central virtUe b 0�th the Rawlsian critique of WI e centrality of justice in our character is compa perfectionism .
�
311
92 Roberto Alejandro may require, it has to redefine its conception of choice and say, explic itly, that justice is beyond our choosing faculties. If justice must be the first virtue of our character, it smacks of a metaphysical conception of the self, and thus contradicts the Rawlsian version of 'political liberal ism.'
v My arguments thus far suggest that Rawls's analysis offers many com plexities which in tum explain how one-sided is Sandel's critique of Rawls's paradigm. Sandel's critique concentrates on the original position and the social contract Rawls derives from it. Yet Rawls's account of
moral personality as is presented in his discussion of associations, vir tues, and moral psychology, are aspects conspicuously absent from
Sandel's analysis. I now turn to address Sandel's critique of the Rawlsian self and the Rawlsian community.
In characterizing Rawls's view of community, Sandel argues that it
'describes a possible aim of antecedently individuated selves, not an ingredient or constituent of their identity as such. This guarantees its
subordinate status .... As a person's values and ends are always attributes
� is only an
and never constituents of the self, so a sense of communi
attribute and never a constituent of a well-ordered society.'
Sandel's critique of Rawls's view of community relies on his under standing of the Rawlsian self, a self that, according to Sandel, is individu ated in advance, whose identity cannot be engaged, and whose ends never fully constitute it. Just as these ends are external attributes that the
self chooses, so also community is an external trait preceded by individu ated subjects. The 'Rawlsian self,' Sandel writes, 'is not only a subject of possession, but an antecedently individuated subject, standing always
at a certain distance from the interests it has. One consequence of this distance is to put the self beyond the reach of experience, to make it invulnerable,
to fix its identity once and for all' (LLJ, 62, my emphasis) .
This is the key to understanding Sandel's argument: his claim that the
Rawlsian community presupposes 'the antecedent individuation of the subject'
(LLJ, 149) whose identity is fixed
'once and for all. ' 'But a self so
thoroughly independent as this,' he insists, 'rules out ... the possibility
of any attachment (or obsession) able to reach beyond our values and
34
Michael Sandel, Liberalism md the Umits of Justice (Cambridge: Cambridge Univer sity Press 1982), 64. Subsequent references will be integrated in the text as LLJ. 312
RDwl's Commrmitarianism 93
sen��nt to e�gage our identity itself. It rules out the possibility of a � pubhc hfe m which, for good or ill, the identity as well as the interests of the participants could be at stake' (LLJ, 62). I think Sande l's interpretation is mistaken. My discussion of Rawls's comm unitar ianism argued that, in Rawls's theory, individuals are shaped by institu tions and comm unities of interests, and if this is so, the 'anteced ent ind ivid ua tion' Sandel sees is not what RawIs suggests. Better still, this 'individuation' is one of the moments of Rawls's philosophy concerning one aspect of the self, which, contrary to Sandel's claim, is not its identity. Let me clar ify this. In Rawls's view, Sandel argues, the person's identity is given in advan ce; it is 'antecedently individuate d' in such a way that it can never be reached by the self's attachments.35 But Sandel conflates two different dimensions of the Rawlsian self. Rawls's analysis suggests a distinction between the selfs core and the selfs i dentity.36 The former is constituted by the self's two capacities: a capacity for a sense of justice, and a capacity for a conception of the ood g (505). The latter (the self's identity) is formed by th ose values, attachments, and ends that are acquired through asso ciations and through a plan of life that is formed gradually; namely, it is
formed by those values that, in Sandel's view, cannot engage 'our identi ty itself.' Stated differently, it is the self's core, not its identity, that is independent of values, attachments, etc., and, for instance, it is the self's core that is individuated in advance and fixed once and for all. If my argum ent is correct, the identity of individuals, in the eI � have mentioned, is neither prior to their ends nor incapable of bemg engaged and constituted by values and attachments. Quite the contrary: Rawls 's account of institutions shaping the individual's attaclunents
suggests that our identity is not beyond our values and attaclun�nts, since those values and attachments help to constitute it, � d smce agam, what div iduals shape their rational plans gradually (561). IS beyond our But how es. reach the self's core and its two capaciti
�
On��
is
that the 'individuated' is the self's core? A further exploration shows s me self's core, that is, its 'moral personality,' is just a 'p�te�tiality,' � here defined thing that San del neglects altogether: 'moral personality 1S (505). The as a p otentiality course' due in that is ordinarily realized --
her aim is to defend the 35 Susan Moller Okin presents a critique 0t this 1'dea.. but . _ hallengeSandel's original position, not to see how Rawls's co��ta · ' us u a Thinking interpretation. See 'Reason and Feeling in
�� m.a�;, Z45-6
. . he does not develop 36 Sandel hints at this distinction in his discussion °f desert' but it (LLJ, 82-95). 313
94 Roberto Alejandro original position relies on the 'potentiality' of the parties; namely, their moral personality 'refer[s] to a capacity and not to the realization of it'
(509),
whereby it is not clear why Sandel calls this 'potentiality' an
'individuated self' which not only possesses an identity but which is beyond the reach of experiences and values. 'Moreover,' Rawls goes on, 'regarding the potentiality as sufficient accords with the hypothetical
nature of the original position, and with the idea that as far as possible
the choice of principles should not be influenced by arbitrary contingen cies'
(509). That is to say, this potentiality is necessary to explain the
original position, not to justify a Rawlsian community.
When Rawls points out that the social system shapes the kind of persons individuals want to be and the sort of person they already are; when he insists that individuals develop the sense of justice in associa tion with others; and, finally, when he argues that individuals need at least one community of shared interests to confirm their worth, he
displaces the emphasis on a self that is always prior to its ends in favor of a community which contributes to constitu te the individual's identity. Sandel disagrees. His critique of the Rawlsian community is predi
cated upon a distinction which he never spells out between 'attributes' and 'constituents.' In Rawls's view, he argues, community is an 'attrib ute,' not a 'constituent.' But if my argument is correct, this distinction is
as doubtful as 'the antecedent individuation of the subject' Sandel sees in Rawls's theory. For a Rawlsian community is far from being a mere attribute. It is rather an arena of institutions, associations, and moral principles that make possible harmony and stability and shape the sort of persons individuals want to be and the sort of persons they already
are. Put differently, if the community provides standards of worthiness to judge and confirm the individual's self-esteem, namely , the most important good, this community is hardly an attribute; it is constitutive of the individual's identity. Furthermore, by participating in a 'commu
nity of shared interests' Rawlsian individuals engage in a process of self-understanding. They come to know their nature as moral persons and, more importantly, they come to develop their sense of justice and the good of self-esteem in association with other individuals. 'We need one another as partners in ways of life that are engaged in for their own sake, and the successes and enjoyments of others are necessary for and
complimentary [sic] to our own good' (522-3). In addition to his treatment of the Rawlsian self, a substantial part of Sandel's analysis relies on oppositions that he absolutizes without ex ploring the porousness of their boundaries. Thus, he opposes 'attributes' to 'constituents,' 'feelings' to 'self-understanding,' 'choice' to 'discov ery.' In many instances, however, the boundaries separating these terms are not as sharp as Sandel leads us to believe. This is important since a substantial part of Sandel's argument is his claim that the Rawlsian self
314
Rslwl's Communi/arianism 95
does not represent what we really are. Neither does Sandel's description. For a non-be�ever, for example, religion is an attribu te he may certainly choose. But tf he becomes a devout convert, the status of this attribute changes altogether. It is now a constituent, perhaps the most important one, of his identity. Likewise, if he is born into a religious fundamentalist group, religion is not going to be an attribute, but constituen a t of his character and identity. But if he, later in life, abandons the beliefs of his fundamentalist group and becomes indifferent to religious matters, religion as a constitutive element of his identity is transfonned into an attribute from which he now exercises distance. This does not mean, of course, that his religious background and its implications are now erased from his life. This is, perhaps, impossible. It means that the individual has reordered the components of his identity. For all his criticisms of a Rawlsian self whose bounds are allegedly fixed, Sandel's conception of
attributes and constituents which, again, he never spells out, participates
in the same 'fixity' he ascribes to Rawls's view of the subject. That is,
Sandel does not recognize that the boundaries between attributes and constituents are porous, not fixed, whereby an attribute may become a constituent, and vice v r . e sa The same argument holds for his distinction between choice and discovery. The same individual of the previous example may discover
the attachments that tie him to his religious group. But if he decides to live by values other than those his group advocates, he is making a choice. Along the same lines, the individual may discover an attachment to his group on the basis of his socialization, and not necessa rily an a�ac�ent to the religious beliefs of that group .. I� is a well.-�own pnnctple of Christian fundamentalist groups, that It ts not suffioent to be born into a religious family. The individual has to experien�e his own _ conversion, and this is a choice he makes, not a discovery he inhents.
Just as he opposes 'attributes' to 'constituents,' Sandel opposes 'the capacity for choice' to 'the capacity for reflection' (LLJ, 153). And he says: 'But on Rawls's � ou�d moral epistemology, the scope for reflectio� . app ear seriously posst�tlity a be limited. Self-knowledge seems not to m gtven as the relevant taken sense, for the b ounds it would define are �t antec�de advance, unr of eflectively, once and for all, by a principle individuation' (LL], 153). Again, the core of Sandel's ar�ent claim of an 'antecedent individuation.' We have seen that this claim ts not convincing Sand�l presents . But, more importantly, the opposition ts that the between ch . ts d ub"tous. His argument otce and reflection . . this the self, voluntarist dimension of agency requires choice, and . . chmces. The cogru·tive dtmenswn of vt·ew, appears as external to tts agency, by "ty °f the subJect apcontrast, requires reflection. The 1"denti , pears 'as the prod uct rather than the premise of its agency (LLJ, . kind of chmce. The se Thus the reflection Sandel proposes rules out any
�
��
·
�
·
·
.
315
152�
%
Roberto Alejandro
that is engaged in the cognitive dimension of its agency reflects upon itself, inquires 'into its constituent nature,' and acknowledges 'its purposes as its own' (LL/, 58). There is no doubt that there is a reflection at work here. But we might distinguish between two types of reflection, which, while being active, have different goals. Let me call the first type confir matory reflection and the second type critical reflection. In the first type, the individual reflects upon himself, his identity, his nature, to confirm and acknowledge what is already there. This reflection aims at a better self-understanding, not at a transformation or a reordering of the constitu ents that make him the sort of person he is. In the second type, the individual reflects upon himself, his identity, his nature, both to acknow ledge the attachments he wants to keep and to discard those he no longer finds relevant. In a confirmatory reflection, the individual acknowledges his identity. In a critical reflection, the individual is open to reconstitute it. Sandel's analysis, with its insistence on acknowledgment and discov ery stands for a confirmatory reflection, thus weakening his arguments. If that is the kind of reflection Sandel defends, it is better, so the liberal argument goes, to accept the unencumbered self of Rawls's philosophy which at least provides a space for critique and distance of inherited values. Sandel leads us to believe that in exercising reflection the individual does not exercise choice?7 He does not explore the possibility of choice through reflection; that is, a reflection that allows the individual to choose new attachments that may question the values he already has, or to consciously discover the principles that have constituted his identity. Finally, and in keeping with a line of reasoning whose coherence seems to depend on absolute oppositions, Sandel opposes will to self understanding38 (LLJ, 58, 152). Self-understanding, presumably, does not
37
According to him, the self cannot choose 'that which is already given (this would be unintelligible)' (U], 58). But if the individual reflects upon what is given and reaffirms it, he is clearly making a choice, though. for Sandel, this is 'unintelligible.' It is so if we assume the fixity of the oppositions he presents.
38 Sandel is consistent in insisting on these kinds of oppositions. In a recent article he poses the following question: 'What then is the resemblance between heterosexual intimacies on the one hand, and homosexual intimacies on the other, such that both are entitled to a constitutional right of privacy?' And he answers: 'This question might be answered in at least two different ways - one vol\U\tarist, the other substantive. 'The first argues from the autonomy the practices reflect, whereas the
second appeals to the human good the practices realize' (534). Thus Sandel opposes 'the autonomy the practicee reflect' to the human goods they realize. This opposition
presupposes that autonomy is not a human good and this presupposition is prob lematic. Instead of opposing both categories, it is better to see autonomy as a human
316
Raw/'s Communitarianism
97
a�empt to make choices. It rather seeks a reexamination, while being driven by an unexamined goal: to acknowledge ends which are already given
(LLJ, 58)
and attachments that are 'found'
(LLJ, 158).
It is thus
possible to tum Sandel's analysis against his own conceptions. He claims that the Rawlsian self is individuated in advance. As I have argued, that
is not the case with Rawls's theory, but it is what really happens with the Sandelian self. The Sandelian self is one individuated in advance by the attachments and constituents of its community. So it does not exercise ch"lice; it reflects upon what is already given. It does not construct
attachments; it finds them. It does not use its will to choose traits that are outside it. It uses self-understanding to acknowledge what is already there in its inne r dimension. Its will is confined to the inner life of the
self, not to its external dimension. But even in the inner life, this will is one that acknowledges, not one that transforms. The 'unencumbered self' is thus replaced by one so encwnbered by its attachments that it
seems to be incapable of exercising distance from them. It remains to be
seen whether the Sandelian self is capable of offering a meaningful account of agency.
VI My inquiry is not meant to suggest that a Rawlsian community !s free of all the problems Sandel ascribes to it, but rather that Rawls s commurutarianism suggests other problems which h�ve not been fully explored, and which require additional �alyses. It IS clea�, howeve�, . that for all Rawls's suspicions of the notion of soc1ety as an org�c whole with a life of a� lts of its own distinct from and superior to tha� soctety s s members in their relations with one another' (264}, Rawl m�t al resembles just that: an organic whole, and since all individ� �39 this life, e be 'fully cooperating member(s) of society over a complet
F example it is doubtful that good that conbibutes to define other hwnan goods . · , realized without can be onships, the h · personai relati uman good expressed , say, Ill · uld contribute to assuming the autonomy of the participants. A forced c ider it a human procreation d even to a stable f�Y� and some :u' iduals who found gOOd. Bu t this good would be realized at the expense 0 us choice · Could o nom auto · exercised an themse1ves in a marriage without havmg . . . aJs? (See Michael Sandel, human goods be realized without autonomo� m�vldu &view " [1989) 521-38.) 'Moral Argument and Liberal Toleration,' Califorma LAw
0�
�ge � peof;..:��
litical, Rawls, 'Justice as FairneS:S Po Priority of Right and Ideas of the Good,' 270.
39 See
John
31 7
· al ' 233· not Metaphys!C I
I
and 'The
98
Roberto Alejandro
argument suggests, Rawls's denial to the contrary, that society pos sesses what he seems to fear the most: namely, 'a life of its own distinct
from and superior to that of all its members in their relations with one
another.' If society is neither distinct nor superior, why should indi viduals be fully cooperating members over a complete life? Why is there no place for distance from society's institutions, even assuming that they are just (in the Rawlsian sense)? Why is Rawls's society a space without dissidents (the principle of 'full cooperation' seems to exclude this); without deviants (the principle that the individual's plan has to be in harmony with the social plan seems to imply this); and without people who may reject the Rawlsian assertion that their nature is completed through other selves, or that their nature is best expressed when complying with the dictates of only one virtue Gustice)? Why is there no place for individuals who may not need the opinions of others to affirm their own worth? Though there are no clear answers, this vision of community does not seem to provide any room for isolation, and even less for the democratic sentiments that, say, George Kateb, ascribes to an Emersonian individ ual. It is, then, a puzzle, why liberals who are sympathetic to Rawlsian tenets decry tight and close communities, while holding fast to the not less tight and close vision of society of Rawls's communitarianism. Equally problematic is the realization that in the harmonious realm of a Rawlsian society there is no place for moral conflicts. The individual does not face tragic dilemmas, since he is guided by the principles of rational choice, and his ends cohere with each other ('Dewey Lectures,'
529).
within the individual, and there are no among individuals either, since all plans are part of the social
There are no conflicts
conflicts
plan of society. Rawls's comrnunitarianism as a utopia of harmony, cooperation, and a uniform human nature that best expresses itself by following the dictates of justice is a radical departure from Kant's vision of society. Kant views antagonisms as the cause of human progress, and proposes 40 the famous analogy of the forest. Without competition, he argues, trees would grow feeble and bent, while, with conflicts, they will grow straight. Without antagonisms, Kant goes on, men would live an Arca dian and pastoral life. Rawls's account of community clearly suggests that he prefers the Arcadian life Kant rejects to the antagonisms he praises. The end result is thus a striking surprise: Rawls's communita ri-
Universal History,' in Hans Reiss, ed., Kant's Political Writings (Cambridge: Cambridge University Press 1970), 45-6
40 I. Kant, 'Idea for a
318
Raw/'s Communi/arianism 99 anism and its goal of harmony turns out to be not Kantian, but Platonic. Just as, in Plato, the individual's character has to be in harmony, and this
harmony is reflected in the state,41 so also with Rawls's account: the harmony ruling the individual's plurality of ends finds its way in a social order where individual plans are part of a larger plan, and where
individuals are part of associations that in turn constitute a social union of social unions. Hence, there is harmony both in the individual's char acter and in society. This picture may appear as an exhilarating descrip
tion of present possibilities . I tend to think that it is rather an
impoverished vision of the human condition. For the pluralism Rawls
seeks to protect through his conception of justice and community is
anything but plural. It rests on a uniform personality, and a no less uniform society whose sameness may strike actual individuals living in the real world of a pluralism made up of incommensurable plans and
perspectives, as an ideal that not even for the sake of Rawls's conception of justice is worth pursuing.
Received: June, 199 1 Revised: May, 1992
41 Plato, The
Republic, R.W. Sterling and W.C. Scott,
Book IV, 434d, 435e, 441a, 44lc,d
319
trans.
985), (New York: Norton 1
TifE INDMDUAL, THE STATE, A.I\ID THE COMMON GOOD BY JoHN HALDANE The fell owship of society being natural and necessary to man, it fol _ lows With equal necessity that there must be some principle of gov �rnm ent within the society. For if a great number of people were to live, each intent only upon his own interests, such a community wo uld surelv disintegra te unless there were one of its number to have a care for the comm on good. - St. Thomas Aquinas, De regimine prindpium, I, (1266) 1
If there be not [virtue among usJ we are in wretched situation. No a
theoretical checks, no form of government, can render us secure. To s�ppose any form of government will secure liberty or happiness �I out any virtue in the people, is a chimerical idea. If there be suf aent virtue and intelligence in the community, it will be exercised m th e selection of these men [of virtue and wisdom] .
�
�
- Jam es Madison, Debate on the Federal Constitution (1788)2
I. INTROOI.:CTION let me begin w ith w hat should be a reassuring thought, and one t�at ma� serve as a corrective to presumptions that sometimes ch�actenze _ politic al p hilosophy. The possibility, which Aquinas and �adison �e concern ed \A.ith, of '"ise and virtuous political deliberation res tmg m be neficial and stable dvil order, no more depends upon posse�sion of a Philosop hical theo ry of the state and of the virtues proper to It, t�an does the possib ility of making good paintings depen d upon possessiOn of an aesth etic theory of the nature and value of art. This is not to claim that theory is, or must be, irrelevan� or unhelpful, I�t alo ne that A super philosophical understanding is gratuitous or mert. ficial reading tt and esho Oak I ae 1 of conservative philosophers such as Mich . . rpre mt Roger Scruton tical � is frequentlv taken to support an antitheore own tation of th ors au eir outlook, but in addition to neglectin g these the of on understan ding quesn of their work, this interpretation begs the
�
?oth
·
�
Dawson 1 Se e A qumas: Selected Political Writings, ed. A · p D'Entreves' trans . J. G. · . (0�ord: Blackwell. 1 959), p. 3. See The Debates in the Setrerrzi State Conventions on the Adoptt.on of the Fedmzi Constitution' ed. Jonathan Elliot (Philadelphia, PA: Lippincott, 1907).
© !9% Social Phil osophy and Policy Foundation. Printed in the USA.
321
IOHN HALDA�E
60
nature and role of political theory.3 Regarding the value of philosophy in relation to practice more generally, I agree with G . K. Chesterton "· hen he writes that "[p]hilosophy is merely thought that has been thought out. . . . [M]an has no alternative, except between being influenced by thought that has been thought out and being influenced by though: th�t has not been thought out . " � But to say that reflective understandmg 1s worth ha,·ing, and even that unless one has it one's practice will be con fused and misdirected, is not to say that the possibility of a reasonable political order awaits the articulation and reception of an adequate gen eral theoiY. This m�ch mav seem obvious, but the scale and stvle of contemporary political philosophy assumes an importance for the s�bject v;hich, viewed from the side of political and social life, it simply may not have. More ger mane to what follo,,· s. it is important to bring forward the idea that the fate of political order need not, indeed should not, depend upon the development of a general and generally acceptable theory of the state. the source of its authority, and the scope of its operations. I place that posi tive thought at the outset, because shortly I shall be suggesting that what John Rawls and many others involved in philosophical debates about lib eralism have been trying to do cannot be done. In the circumstances in which we find ourselves there cannot be a full legitimation of the liberal or communitarian state as the arbiter of social justice. This sort of thing is sometimes said by cultural relati\ists and b�· anti realist pragmatists such as Richard Rorty;5 but the standpoint I favor in essentials a neo-Aristotelian-cum-Thomistic one - is certainly opposed to both of these. Indeed, it is because I believe in an objective moral order (a "moral law" even), and in its relevance for the conduct of polit ical life, that I find what many liberals have to say on such issues as abor tion, blasphemy, education, the public celebration of " alternative sexualities," pornography, and the provision of social services, for exam ple, to be troublesome. Yet just as the morally neutral state seems an illu sion, so the extensively morally committed state seems an impossibility. Communitarians are correct, I shall argue, in some of L�e criticisms they make of philosophical liberalism. They are wrong, however, to the extent :; See !-.1ichael Oakeshott. "The Concept of a Philosophy of Politics," in Oakeshort. Reli
gion, Poilt'�. and the M:>...: lifr!,
ed. Timothy Fuller ( �ew Haven : Yale Cniversitv Press, 1993j; and Roger Scruton, Tne M.i4ning o• Conservatism (Harmondsworth: Penguin; :980). "Con servatism may rarely announce itseli in max1ms, formulae or aims. Irs essence is inarticu late, and its expression, when compelled, sceptical. But it is capable of expression . . . " (Scruton, The Mfll nir.g of Consen>atism. p. 11). 4 G . K . Chesterton. "The Revh·al of Philosophy - Why?" in Chesterton, The Common Man ( London: Sheed and \\'axd, 1950}, p. 176. Regrettably, Chesterton's "Titings haw hardly been appreciated by social philosophers . See, for example, Chesterton, What's Wrong U'ith the World (London: Cassell, 1910; San Francisco: Ignatius Press, 1994).
5 See , for exampie. Richard Rorty. "The Contingencv of Communirv," in Rom·, Contin gency. Irony, and Solidarity /Cambridge: Cambridge UniV'ersity Press, 1989); and R�·. "The Priority o! Democracy to Philosophy," in Reading Rorty, ed. Alan !\Ulacho wski {Oxford:
Blackwell,
1990).
322
mE INDIVIDt.:AL, mE STATE, AND THE COM.\10N GOOD
61
that they think that mora l comm unity - in the precise sense that Augusti ne, for exam ple, has in mind when he speaks of "a gathering of _ ration l beings unite d in fellow � ship by their agreement about the objects of therr love" 6 (and, one might add, of their aversion) -can be a general model for the legitimation of the modem nation-state. Here, indeed, I find n:yself in qualified sympathy with Rawls when he writes that "the hope o poiitical commu nity must indeed be abandoned, if by such a commu mty we mean a political society united in affirming the same comprehen sive doctrin e " 7 (the nature of the qualification will become apparent). In the remaining sections, then, I shall be arguing for the following aims . First, the project of liberal political theory, of the neutralist and Individualist sort pursued by Rawls, fails and does so for foundational and structural reasons . Second, an important but still neglected notion in social philosophy is that of the common good - Aquinas 's "bonum com mune." Third, while communitarian conceptions of social life may in dude a nonreducibl e common good, it is certainly questionable whether the con ditions necessa rv for the establishment of communitarian states gen erally exist . In these' circumstances we should be grateful for the pos sibilitie s for moral development offered by various other forms of com munity - for which, following Augustine, I propose the term "fellowship." The acknowledgment that acceptance of a transcendent justification of the political order and its essential operations is not likely to come about (not: that such a justifica tion is altogether impossible) suggests that the appro _ pria te attitude to ward the state is a blend of long-term moral asprratiOn, and short- to middle-term practical participation in ted political goals. Contrary to the position a efense mvolves element latter of Rawls, this of a form of political arrangement that probably is a modus ."1"endr. How ever, th e proportions of this blend, as indeed the need of �t, are matters of sod ohistorical contingency; it is not inconceivable, theretore, t at they may ch ange over time or differ geopolitically, as between the C .S. � nd gl the U. K. for example'. Indeed, I end with the thought that E l angua ge poli 0 related n conditio a tical philosophy suffers from arated by a note d by Oscar Wilde when he wrote of "two nations sep commo n langua ge ."
�
?
�
�
�
� ��;
II.
RAw Ls A N D THE U�AV OIDAB ILITY
oF
CoMP REHD ISIVE
0
DocTRINES
cr a waited · and since Th e ve ar 1993 saw the publicati on o f two I na· twr:s · _namelv m uch -discussed prescnp d 1 ' works on issues of valu es an _ Joh n Rawls's Political Liberalism and pope J.0hn Paul II's Ven tahs ::,p endor. e is cer-� _ Thus far I have not seen these exammed m tandem, though ther .
.
·
·
. . . . XLX. ch. 26. ;\ ugustm e , De cwrtate der ( London : Loeb • l960·I ' Book s . John Rawls, Political Liber.llism (New York: Columbt. 1.: " ·versitv Pre s. !993), p. 146 . c hoik Truth SocietY, 1993) � Jo h n Paul !1, Veritatis Splendor ( Lon don: lncorporat�is , ;ed _ statement oi iundamen ·0 w htch may be added a second much-heralde d - and 4 t t al Ca Church \ mdon : Chapman. 199 ) .
�
� :
� ·
tholic doctnne:
the Cat.:chism of the CaJ/wlzc
323
62
JOHN HALDANE
tainly scope for comparing and contrasting them. That is not my aim on . this occasion; but I shall discuss a difficulty for Rawls's position arismg from the existence of a work such as this encyclical addressed to the
Roman Catholic Bishops - an encyclical directing them, in their teaching
of the nine hundred million faithful, to uphold the unconditional and unlimited character of fundamental moral requirements .
First, however, recall the basic enterprise pursued in Rawls's recent
work. His concern has been to give an account of how political institu· tions governed by principles of justice can be warranted in circumstances
in which they are required to regulate the lives of people who may,
indeed do, pursue different conceptions of their O\\-n good. Early on,
Rawls presents this issue as a question: "[H]ow is it possible for there to
exist over time a just and stable society of free and equal citizens, who
remain profoundly divided by reasonable religious, philosophical, and
moral doctrines? " 9 Although this form suggests a Kantian enquiry into
the a priori conditions of the possibility of justice, the content of the ques·
tion looks to be socioempirical; and certainly Rawls repeatedly empha
sizes the nonmetaphysical character of his investigation and of its
conclusions.
Already, however, that suggests a problem. To the extent that princi
ples of justice carry normative force, they must appeal to considerations that can serve as justifying reasons and not mere psychological motives
for those to whom they are addressed; but in order to do that they must
ha\·e, if not a priori universal validity, at least some element of necessity
or rational inescapability. Otherwise it will be too easv for the claims of justice to be evaded by those who fail to acquire, o� choose to divest themselves of, the relevant desires. One possibility here would be to fol
low Aquinas (at least as I read him) and Aristotle (as he is traditionally read)10 and argue that while prescriptions generated by practical reason
are not categorical in Kant's sense, nonetheless in appealing to an agent's strivings they need not be void on account of the contingency of
desireY For the strivings in question may be ones the agent cannot fail to have inasmuch. as they are partly constitutive of a normal (i.e., norma tive) human nature. Such "assertoric hypotheticals" (to stay with Kantian termino1ogy )12 rooted in an animate essence may be available to those who reject the
pure practical reason of the categorical imperative, but they remain too Politiclll libmzlism, p. 4 . 10 For a recent and influential departure from this tradition, � John McDowell, "A:re Moral Requirements Hypothetical Imperatives?" Proceedings of the Aristotelian Society, sup _ plementary volume ::>2 (1978), pp. U-29; and McDowell, "The Role of Eudaimonia in Aris totle's Ethics," in EsSilys on Aristotle 's Ethics, ed. Amelie Oksenberg Rorty (Berkeley : University of California Press, 1980). 11 For Aquinas, �. for example, his Summa Thtologiae (1265-1273), trans. Thomas Gilby (London: Eyre and Spottiswoode, 1976), la, Ilae, q. 1, a. . 6 12 See The Morul l.Aw: I<Jant's Groundwork of the ML#Qphysics of MoTIIls, ed. and trans. H. J. Paton (London: Hutchinson, 1976), ch. 2, p. 78. 9 Rawls,
324
TifE
INDIV IDUAL , TifE STATE, AND THE COMMON GOOD
63
de�pl stained in the hue of metaphysics � for Rawls's purpose. His opp
tUY
,
from such a doc f
of conflicting 13 ble pluralitYason [Tlhis rrzf . . reasona . See, for example, Rawls, Political Ubethe over time re � of rit and mcom tnS_hc diii TPCt as see n mensurable doctrines is a ligious. re phil· ble n reaso f � rsity diVe under_enduring free institutions"; "w_e also t� ; ot their feature nent osophJCal, and moral doctrines iound in democratiC e�tv of =able religious. phil· public. culture"; and "[a)s always, we assume th�t theS: . ��I a pemtiiN"_I feature of the17� phical and moral doctrines found in d��tJC societJes � :wav"' (pp. 1:15. 136, 216- ; my lie culture and not a mere historical conditiOn soon !,0 p emphases). Whv "characteristic" and "pennanent unJasesss � r skeptical reasons? 14 lbid., p . 9. .
tSm:
"
VIeW
a
0
.
325
JOHN HALDANE
64
all citizens to affirm a comprehensive doctrine to which the political
conception they accept is in some way related. But a distinguishing
feature of a political conception is that it is presented as freestanding
and expounded apart from, or without reference to, any such wider
background . . . . [I]t tries to elaborate a reasonable conception for the basic structure [of political society] alone and inYolves, so far as pos sible, no \\ider commitment to any other doctrine. 15
Additional!\·. \\·e are to consider a threefold distinction among a com
prehensive doCtrine, a political conception of justice, and a modus vivendi.
For these purposes, the last is to be thought of as an agreement or treaty adhered to because the participants regard it as being to their individual
benefit. Such a convergence of interest is contingent, and therefore any
appearance of political unity among the parties is illusory; all that exists
is a precarious arrangement sustained by self-interest. It is against the
background of this tripartite division that the claims of political liberalism
are elaborated. ln answer to the question of how , gh·en a pluralism of
comprehensiw doctrines, there can nevertheless be a just and stable soci
en· and not merely a modus vivendi, Rawls offers the idea of an overlap
p�g consensus o� a political conception of justice - that is, a principled
agreement on \·aiues and norms whose content and justification are inde
pendent of any distinctive comprehensive doctrine, but are compatible with many, most, or all such doctrines. unlike a modus vivendi, such a condition does express and sustain genuine social
unity,
but compatible
with Rawls's requirement that liberalism be neutral between competing conceptions of the good, it is not the expression of one, as against another, comprehensive doctrine .
Such is the claim, but the problems seem resistant to this form of solu
tion. First, the initial separation of practical reasoning into ethical and political spheres is not innocuous - if it were, it would hardlv serve Rawls's argument, which requires a degree of independence of the polit
ical from the moral. There are several traditions, including the Aristotelian
Thomistic and the Kantian ones, which would deny that a political con
ception can be "freestanding," precisely because they assert the unity and
continuity of practical reasoning. Put in terms of Thornism, for example, the counterassertion would be that there can be no account of a polit ical
"right" that does not derive from a theory of the good, and that this lat
ter is the general presupposition of all individual and social action. A
related claim is expressed by John Paul II in a section of Veritatis Splendor where he is considering objections to traditional natural-law moral theology:
!� See ibid. , pp. U-13; see also ibid., Lecture V, "The Prioritv of the Right ove the r · Good."
326
65
11-fE I:\Ti l\1D UAL, THE STATE, AND TifE COMMO� GOOD
The separation which some have posited between the freedom of
individuals and the nature which all have in common, as it emerges from certain philosophical theories which are highly influential in
present-day culture, obscures the perception of the universality of the moral law on the part of reason. But inasmuch as the natural law
expresses the dignity of the human person and lays the foundation
for his fundamental rights and duties, it is universal in its precepts and its authority extends to all mankind .16 It is important to see that concerning the issue of the duality of practi
cal reason and the further critical points that follow, Rawls may be in
some difficulty even if Thomist, Kantian, and other theories are them
selves defective, since it is central to his approach that it not rely on con
tentious philosophical doctrines. Thus, if it is controversial whether the
moral and the po.litical stand in the required relation, this fact alone under mines the p ossibility of advancing the political conception as the �bj e_ct
of an overlapping consensus Certainly one might argue directly tor It, . bu t to do so would be to violate the requirement of political autonomy. Perh aps , however, that requirement is not absolute but a ts of _ degree. Certainh· hiDl quotea I Earlier this. Rawls sometimes suggests writing that "the inde as conception of justice should be, p:nd ent of the doc religiOus l d osophical op!'osing and conflicting phi � _ trines that citizens aifinn " (mv emphasis), and at one pomt he c�nstders directly the possibility of op osition to his p� litical conception advocate of a comprehensive religious doctrme. What he says 15 e revealing and it faces, I believe. a serious objection. In order to show both points I need to cuote at some length:
�
as far pos_si�le,
":0� �
p
•.
[B ]y avo idina comprehensive doctrines we try to by1'ass religion and p h ilo sophy' profoundest controversies so as to have some hope or unco verin" a 'basis of a stable overlapping consensus . . . . . . Nev rth eless ' in affirming a political conception of JUstice \\ e · aspects of our own may eventuallv have to assert at least certatn comp rehensiv� religious or philosophical doctrine (by no means nec es sarily fullv comp rehensi \'e ) . This will happen whenever som eon . . . th a Insis ts, for ex ample, that certain que ti0 s are so fundamen tal � �. . trife A this t · ·: to in s ure their being rightlv settled JUstiftes avil s I\' ;ts _ or o t point we rn a \· h ave no altern ativ e but to en th t mP . y , ·•0 ' l . \\ e had 1 ope d g thm t d m d en ial k' the 0 and hence to maintain avoid. . contend th at To co nsider this, imagine ratwnartst bertevers who reason ( unbv d th e se beliefs are open to and can be fully � tabl'ISh vers sim belie he com m o n thou gh this vie,...· might be). In t IS case
�
.
�
.
·
··
�
�
I&
John Paul
II, \.-ema:zs 5plendor, section 51. P· 80.
327
�
:
�
�
JOHN HALDANE
66
ply deny what we have called "the fact of reasonable pluralism. " So we say of the rationalist believers that they are mistaken in denying that fact; but we need not say that their religious beliefs are not true,
since to deny that religious beliefs can be publicly and fully estab
lished by reason is not to say that they are not true. . . . [W]e do not put forward more of our comprehensive view than we think needed or useful for the political aim of consensusY
First. then, it is conceded that the method of avoidance may fail and
that when it does so it may be necessary, and is permissible, to defend
the political conception against challenges from a comprehensive concep tion by im•oking a rival - one's o•vn - religious or philosophical doctrine.
Second, however, it is supposed that in the imagined example of believ
ers who deny the (purported) fact of "reasonable pluralism," one's doc trinal counter only challenges their epistemological claim and not the
content of their ov>n comprehensive conception. The point of this second
obsen·ation is to emphasize the limited character of the departure from
wU\'ersal toleration. Against this, however, one should observe that any
lapse from strict neutrality undermines the claim that a political concep
tion can be founded on an overlapping consensus only and need not rest
upon a distinctive comprehensive doctrine. Further still , the departure
may not be as limited as Rawls supposes, since it might be part of the rationalist believers' doctrinal commitment that pluralism with regard to fundamental claims is
not
reasonable. In short, epistemological claims
may fall within essential doctrine. Consider, for example, another papal -ument - Pope Pius XII 's encyclical d�
ern
TeJUhing ):
Humllni Generis (False Trends in Mod
�otoriously, the Church makes much o f human reason, i n the fol
lowing connexions: when we establish beyond doubt the existence of one God, who is a personal Being; when we establish irrefutably,
by proofs divinely ·granted to us, the basic facts on which the Chris
tian faith itself rests; when we give just expression to the natural law which the Creator has implanted in men's hearts _ Is
Here Pius is reiterating long-standing Catholic doctrines, the first of
which is the provability of the existence of God - it being contrary t� faith (not merely theological tradition) to deny that there can be such a proof.
Earlier, having made similar claims, Pius asks why there should be dis
agreement over such matters and in response cites "the impact of the 1;- See Rawls, Poliliall Liberalism, pp. 152-53. •� FGise Trrruis i� Modem Tau:hing: Encycliall Letter (Humani GenerisJ of Pius XII Concerning CnWn Flllse Opinwns, trans. Ronald A. Knox ( London: Catholic Truth Society, 1950), Part II, 'The Field of Philosophy," para. 29, section 1, p. 16.
328
67
THE INDMDCAL, THE STATE, AND THE COMMOS GOOD
senses and the imagination, [and] disordered appetites which are the con sequences of the falL" 19 The interpretation of papal encyclicals is a fine that phil� sophers now rarely practice,20 but it is difficult to escape the 1dea that so tar as Pius is concerned, pluralism with regard to the primary precepts of natural law (for example), though explicable, is not reason able. In asserting otherwise, therefore, Rawls would be saying that the rationalist believers' religious beliefs are false. Here my point is not that there can be no argument on behalf of a polit . ICal conception against the claims of those who would pursue their creed to the poin t of unsettling civil peace.21 On the contrary, I believe that one can and shoul d fashion robust defenses of law and social order against, for example, those anti-abortionists who would murder clinic staff-or those who would assassinate blasphemers. As Rawls reluctantly con cedes, how ever, the possibility of doing so depends upon bringing into
�rt
�he political domain a distinctive comprehensive moral doctrine. The to�e
m which he writes of this need suggests a socially regrettable necessity
_ akin to the use of force to expel a drunk and boorish guest from a party; but the problem reveals faults in the very structure of political liberalism. By Rawls's own account, even though the political conception may not be the object of an be overlapping consensus , it should non�theless. . affirmed (an d upheld) because it is implied by a favored philosophical persp ective . In connection with this objection, consider what Rawls has to say �bout public re ason. He e or ra�onal, asks: "How can it be either reasonabl a public conto when basic matters only are at stake' for citizens to appeal . · ?"ll The lt. cephon of jus see they . tice and not to the whole truth as be av� to d se answer elaborates interpretations of ideas that are suppo nd that we live able and acceptable to all, and concludes with the dema ·one · ns · together poli 0 and JUStifi'ca 0 that even ticallv - on the basis of claimS uform ese th er ' can reaso nablv be expected to endorse. As before, howev .even ost Iations alm I , fail to withstand the test of real examp1es. 0 ear � . thIn . g turns on the interpretati. on of , reasonableness " Rawls wntes: •
·
·
·
·
lic reason are The only comprehensive doctrines that run a£ ul 0f pub ues . nce 0f .Political val tho se that can not support a reaso nable bala° hard e b · portance 1t would [And he continues, in a footnote whose un to exaggerat e : ]
19
Ibid. , P: 3. "How C.tn We Leam . Maclntvre. For � distinguished n, however, see Alasd�II' ' exceptio . l discuSS plul· lil-95 · . Wh at Ventatis Splrndor Has to Teach?" The Thomzst, val . 1994), :'.gain: On Vmtahs v e ( h �so p ical aspects o f Veritatis Splrndor in "From Law to. of Sheffield ef teld.· in The Use of the Bible in Ethics, ed. M. DaVIS ( ress, 1995) · . . . s ....,.; �l Issue F't thtsis . n . " 5Yn lerat10 T I ious ''R in offer o elig such an argu me nt • on ,�olerat ion, vol. 9 (1994), p p . 2 -26. 1 - Rawls, Li17erolism, p. 216. :w
ck rsity ; ;nd�a!Jni� ·e Phllosopmca,
:1�
rlendor,"
Political
329
JOHN
68
HALDANE
As an illustration consider the troubled question of abortion. Sup
pose first that the society in question is well-ordered and that we are
dealing with the normal case of mature adult women . . . . Suppose
further that we consider the question in terms of these three impor
tant political ,-alues : the due respect for human life, the ordered
reproduction of political society over time, including the family in some form, and finally the equality of women as equal citizens . . . .
!\ow I believe am· reasonable balance of these three values will give a
woman a duly q�alified right to end her pregnancy during the first tri mester. The reason for this is that at this early stage of pregnancy the political value of the equality of women is overriding and this right
is required
to give it substance and force . . . . [A]ny comprehensive
doctrine that leads to a balance of political values excluding that duly qualified right in the first trimester is to that extent unreasonable . 23
Without entering into the abortion debate, it should be clear that any
notion of reasonableness that renders an opinion contrary to that pre
sented bv Rawls "unreasonable" is almost certain to be (reasonablv) con
tentious -and thus not fitted to occupy a central role in a conception of
justice that purports to apply the principle of toleration to philosophy itself - thereby "to bypass religion and philosophy's profoundest prob
lems. " In this connection consider again a view presented by an author itatiw Roman d ocum ent, this time an "Instruction" on abortion ( Donum
Vitae) issued by the Congregation for the Doctrine of the Faith:
The inalienabie rights of the person must be recognised and respected
by dvil society and the political authority. These human rights
depend neither on single individuals nor on parents; nor do they
represent a concession made by society and the state . . . . Among such fundamental rights one should mention in this regard every
human being's right to life and physical integrity from the moment of conception . . . . As a consequence of the respect and protection
which must be ensured for the unborn child from the moment of
conception, the law must provide appropriate legal sanctions for e,·ery deliberate violation of the child's rights.24
In the face of a dear conflict of views of the sort which this example
makes \i\·id, Rawls's true position reveals itself to be far from neutral: try
for an overlapping consensus, but where it is not available and where important issues are at stake, affirm your own comprehensive doctrine.
The whole raison d'etre of Political
Liberalism,
23 Ibid., pp. 243-44.
3,
however, was to offer a way
24 Donum Vitae (1987), as quoted in the Catechism of the CAtholic Church (supra note 8), Part section 2, paragraph 2273, p. 490.
330
THE l�DIV! D L A L, THE STATE, AND THE COMMO � GOOD
69
forward that possesses the princip les lacked by a mere modus vivendi, yet does not rely upon appeal to distinctiv e doctrines . It is apparent from the foregoing, therefo re, that whatever its other merits this defense of liber alism fails in its own decl ared aim .
III. TH E CoMMON Gooo
One way of presen ting some of the problems that Rawls runs into is by saying, as above, that he tries, unsuccessfully, to secure a political right that of justice withou t deriving i t from any distinctive account of the a good. This is criticism that is increasingly voiced, in one form or another, articu ar y b y l l advoca tes of "perfectionist" Iiberalism.25 It might als o be P noted, tho ugh it rarely is , that although the notion of the right Rawls seeks is a co mmo nly shared one, he conspicuo usly esche\�·s any th7ory of the common good. So far as I am aware, he only mentions th ehId�a . once an d that is in a passage characterizing views that contras t wit his own account of justi ce as fairness : [WJhatev er these religious and philosophical doctrines may be, I . . d that assume the v all contam · h t and the goo a conceptiOn of th e ng . · Iude as m som e t d me ers 00 be d un s a conception of justice that can a w y advancing the common good . 26 ·
This, then , brings me to the issue of the ro1e \\1·thin P0litical. and social . philoso phv o wt1sJan f th e idea of the common good. The body of anti-Ra h · · · · ese ets b ut cntiasm gen erallv dubbed "communitarian " has several ' targ . diV1·dn. eous m caz: be gathered erro together under the general charge 0f " i:h Ualism." Th us, ot e noti n of it has often been allege d that Raw� 's use ti ted original position ons as c implies a view of the identi� ot ers claim �dependentlv of their soda! attachments and inhented meta Is usually . erent m ade in an effort to show that Ra wls h� an mco �on " · a1 · ,,ous St:\. re phv . S the IC . ' an thropology ; but as with the cntlcisms ot p . · bv th e d It 15 i mp ortant to see h ow political liberalism may �e un demun.e ns 'O ne s p tJ o mere fact that it draws upon substantive Philosophical as um are coherent . need not en gage the q uestion of wheth er those s umptJo ... is h defens� !et th Rawls ,.. , alon e t e ru . So, for example, while I a gre e w1 . Io ·c al device and f th e Ori ginal position - holdin it to b� an e � he is mistaken �ot a m t aphysical description ot the soaal wo r � ho u t al ersons is wit tn argu m g that the idea of citizens as free and e�� p ' e hav . If v ottzenshIp !Ihould contravers1. al assumptions.27 Ask yourse wh •
:u'
�
·
�
�alu�- �
·
·
a.:
0
�����
�
J
·
;:s
I'
·' Frttdom
ph Raz: The .\tonz zty o, . . thinking especially of recent wn�gs by Jo 'oxt d : aarendon, 1994). or �ord: Oare ar 1 ndo n , 1986) , and Ethics in the Publu: [)om Raw ls. Politic al Liberalism , p. 109. [neliJninable ;:; f persons: An i!-95, and ·•Idenntv, ohn H aldane, "Political Theory and the Nature 1° J e Se . h p YSlcal , PP · vol. 20 Presuppos·1t1.o n, .. Ph r.losoph''""/ , PP· 199-I ., z.M Papel'5, rnunlty . .., . rs Quarlerlu, vo . ' · , and th. e Limits of A!1"1 Public , " ure \1 u lticult
(0
Here I
:
am
(199 )
331
•1993),
Met;� (�1-1 .
]OH!'\ HALDANE
70
special normative significance ahead of some other aspect of a person's identitv, such as his or ·her religious affiliation, to the extent that it can be appea led to as a trumping factor. The answer has to be that for Rawls
these other identities are in some important sense secondary to that of
our nature as free and equal persons . The exact sense in which the latter
identity is prior and deeper may be held to be normative and not meta physical, but in ei:her event its assertion is philosophically contro.versial.
Having noted this distinction between kinds of criticism, it is unsurpris
ing that those Kho complain of individualist assumptions are concerned to provide a communitarian alternative to liberalism. It is an interesting question what features are essential to such an option. One dominant strand in recent thinking has been the thesis of social constitution: the claim that persons are made to be such - advanced from the status of mere human animals -by being worked into a network of social relations. An analogy here might be \\·ith the process of sculptural assembly, whereby various items, each possessed of a pre-sculptural nature, have a new compositional identity bestowed upon them; for example, what was (and in one respect remains) a piece of windshield wiper is then a grinning mouth. 1 believe that there is something in this idea and that it
shows itself in the fact that most action descriptions presuppose socially constituted patterns of beha\cior. However, here I am more interested in
the claim that liberalism fails inasmuch as it neglects, and cannot accom modate, the fact that some or all of the goods we pursue, and which a system of rights is concerned to protect, are goods possessed in common.
By contrast with Political Liberalism, the term " common good" occurs
frequently in recent Roman Catholic documents, including
Veritatis Splen
The history oi the term's present prominence in Catholic social teach 28 ing goes back to the writings of Jacques Maritain and Yves Simon,
dor.
which in tum look to the moral theology of Thomas Aquinas. In the Prima SeamdJU of �he Summa Theologiae, eight questions (qq. 90-97) are devoted
to aspects of law (indeed, this group is often referred to as "The Treatise on Law" ) . In question
90,
article
2,
Aquinas writes as follows:
[SJince every pa11 is ordered to the whole as the imperfect to the per
fect and one man is part of the perfect society, it is necessary that the law properly regard the order to the happiness of the society . . . .
Hence, since law is most of all ordered to the Common Good, it is necessary that any other precept concerning a particular matter must needs lack the nature of law except insofar as it is ordered to the
Common Good. And therefore every law is ordered to the Common Good.29
2li See , for example, Jacques Maritain, The Person and the Common Good , trans. john Fitz gerald (New York: Scribner, 1941); and Yves Simon, The Trrulilion of Natura/ Law (New York: Fordham U niversity Press, 1965).
29 See Thomas Aquinas, The T� on lAw, trans. R. J. Henle (Notre Dame: University of Kotre Dame Press, 1993), q. 90, a. 2, pp. 132 and 134.
332
�
THE INDM DCA L, THE STATE, AND THE COMMON GOOD
71
It is portant to understand what Aquinas mean s by the common good, s ce present-day writers sometimes speak of general or collectiv � e go� ds m � ays which superficially resemble the Thomist notion, but whtch are m fact quite differe nt from it.30 Notice two elements in the quoted passage: man stands in relation to society as a part to a whole; and every law is directe d toward the establishment, maintenance, and improvem ent of the commo n good. The first is a familiar thesis of conimunitarianism the irreducibility of society as a unified substance that bes tows a form of moral identity on its members. As a corrective to a radical .corpo ratist reading, it is relevant to add that Aquinas also regar individual persons as complete substances. By implication, then, he reJects a dichot omv that bedevils current debates, that which regards ? ersons as either parts of a greater whole - society- or else as preexis t �g in viduals out of which society is formed. A way through lies in the direction of saying that persons are both wholes and parts -wholes as selves, parts as soc ial selves. This may seem evasive and invite the question: Are selves made by �iety o r is society made by selves? Again, however, a w�y through the dilemm a is offered by saying "both"; that is to say, there tS a p�ocess of mutual determination. Think again of an artistic analogy. CortStder, for example, an artist working on a composition by moving around cut-{)ut shapes of diffe rent colors within a rectangular background. The compo sition is made out of parts which (in some, substance-involving, �e�� ) c_an be d es cri bed independently of it, but they also change_ theu � enti ties as aspec ts of a greater substantial whole as they move m relanon to one ano ther matter and the background . A more abstract account of . . · zrught sav . mtght be fashione d in terms of potentiality and actuality One . hat a person which are actualiz has a range of potentialities some so m so cie ty while at the same time the potentialities of a sty!� arrang ment are aJ . tnnSIC powactualized through the exer�e ot natur , � ers of mdiv 5'U:'the person iduals . To adapt a formula of Hila!!, Putna� and the soc iety jointly make up the person �d th s · is the idea t So far as conce rns the common good , what 15 mo ng f society as that every law should have as its proper goal the we_ll-bei . ll s moder a whole . This . d'IV!. dualism tS somet::ile appare ntly radical anti. m · tv as an aggreated by co f mmentators \�ho urge an interpre ati� n no on, equivalent nve gate, an d thereby u treat "common good" as a diStnb
�
�
�
th�
:
��
'
·
;� ·
0�
·
,
:= ;
-
�
�
500�
_
•
· modem :;a in Aqwnas and ·Ll1 Th �or recent accounts of the cominon good as J. t f atures Society." in Glt�· n O d the pe OllUsllc writings. G see Louis Dupre. "The Common dge Uru an ' . Cambridge: Ca"mbri ·:zs� LiberrUism, ed. David Hollenbach and Bruce Dou��mmon Good. Tht Thomtst, \ eftty Press, 1993); Gregorv Froelich, "Ultimate End an on Good': Reflections \' o · 38 (1994), pp. 609-19; Kibujjo Kalumba, "�aritam on ·The Comm 993), . 93-104; and Tho�as the Con cept," Lwal Theologiq�.te et Phi/osophtque, v ctions Econ��c Justice: Refle · Rourke and Clarke E. Cochran. "The Common G 231-;2. pp. , 1992) n _ ;4 th ( o e Though t of Yves R. Simon. " Revitw of Polttzcs, o1 · v
�n
�
ll1ld
�4=��
333
)OHJ'\ HALDANE
., ... ,_
to " the good of each and every member. " At other times it is suggested that while some goods are indeed commonly possessed, they are social means to indi\idual ends - in Aquinas 's supernatural teleology, the beatific \ision. On this account, Ia\\" should promote civil order and public health, tor example, because these are conditions that each may benefit from ( since they and other public goods are objects of convergent interests).
However, neither of these interpretations is plausible exegesis, for nei
ther takes seriously enough the phrase "bonum commune" and Aquinas's
claim that this is a "bonum honcstum," a genuine constituent of perfection.
Consider first a modern attempt to interpret Aquinas 's notion. I quote
from Maritain' s The Person and the Common Good :
[T]hat which constitutes the common good of political society is not only: the collection of public commodities . . . a sound fiscal condition of the state and its military power; the body of iust la,vs, good cus toms and "·ise institutions, which provide the nation with its struc ture; the heritage of the great historical remembrances, its symbols and its glories, its li\ing traditions and cultural treasures. The com mon good includes all these and something more besides . . . the whole sum itself of these; a sum which is quite different from a sim ple collection of juxtaposed units . . . . It includes the sum or sociological integration of all the civic con science, political virtues, and sense of right and li"Qerty, of all the activity, material prosperity and spiritual riches, of unconsciously operative hereditary wisdom, of moral rectitude, justice, friendship, happiness, virtue and heroism in the individual lives of its members. For these things are, in a certain measure, communicable and so revert to each member, helping him to perfect his life of liberty and person.�: The emphasis on the communicability of the integrated sum of social and personal elements contrasts with a notion of commonality as a mere func
tion of convergent interests . The common good is essentially shared: it
is a good-for-many, taken collectively, rather than a good to many, taken
distributively. Aquinas's claim that this good is a bonum honestum, a per
fecting end in itself, might be thought to be incompatible with_ a Chris tian view of human destiny, which is usually treated individualistically;
however, his trinitarian theology grants him the idea that, even eschato
logically, the good of individuals resides in their participation in the life of a community of persons. Theological interests aside, Aquinas's idea of the common good as a participatory end, for the sake of which civil societv exists and must be ralism's "co mmuregulated, has been insufficiently explored even by
libe
3!
Maritain, Tilt Person and the Common Good, pp.
334
52-53.
THE INDI VIDU AL, THE STATE, AND THE COMMON GOO D
73
nitarian critic s . " Yet some thing of this sort seems to be presupposed in . a w1de range of moral-cum-socia l judgments. Concerns about the conduct of nati ns in g ing to war and in prosecuting wars are not easily repre � � . _ sented m mdtvt dualistic terms. To cite two examples from recent historv, t e agonies felt in the U.S. about Vietnam present themselves as colle� tive shame o r guilt - not: a 's plus b's plus c 's, etc., but "ours"; likewise, the British experi ence of the Falklan ds war and subsequent reflection �pan it are most perspicu ously represented in tenns of shared participa tion in honor and tragedy . It is too easy to respond to claims of commu nal good and e vil with charges of romanticism. There is unquestionably something that is recognized, particularly but not exclusively within older traditional societies, as "our" well-being, or "our" corruption, and it is a serious omissi on not to give an accoun of this. t Rawl s's political liberalism cannot find a place for the common good bec aus e of its commitment to neutrality between life-shaping values. At mos t it ca n register and even celebrate convergence in evaluations, see
�
ing in this happy coincidence possibilities for establishing and extending
an overlapping consensus. However, the idea of citizens as free and equal
persons remains incfu;dualistic : the good of citizens that results from par
�
tic pation in an order regulated by the political conception of justice is a pnvate on e . This is sometimes overlooked on account of the regulated order being a p ublic good; but therein lies a lesson: public good does not equal common good. , . . s The argument in favor of the common good is not the tailure . of project. My objections to the latter principally concerned it mabihty to � . meet o ne of its own main criteria namelv, the need to establish a case tor the form and on content of the pol tical c�nception without relia�ce �� g controversial comprehensive doctrines. However, unless one IS w political an d able to argue the in ce differen in the case for greater moral order, the recognition that Rawls's neutrality laps es is likely to prompt the . question of how the what the nature aood should bear upon the nght and · h . .0 lli that the probng ot t e nght m question ca re an d ar, IS . Havmg got th at f ne mav th en o Iem for res 'al life soc! f o olutio n is the normative conditions ' . . . common the of be better placed idea the b . es offered Y to see the poss1bilin . . the expen. tifies J · us goo d - incl t a h w uding, for example, the notio n th at . . . rem peop1 e are sup dIture of societ v's resourc es upon uruversi.tIes whe orted in their thin.k a about these very issues is the fact th the ood in p t e ach e attaine d ther eby are " c mmunicab le , " reverting to � r d al l t�bly , it has � become all too natural t� ask how t e to tax pavers, ot p hilosop hers could possibly be ot mst rume� . per�pec F ssue sm e it s assum . �om the at i ed that this is the only value _ I c merelv ti.ve of A quina ot corrup t10 n not s the question betrays a k'10d . ods ' but go . tal en trum becau se. 1t ov I ' . t non 1ns erlooks th e p oss1'b'l 0 better . a co mu nitv we are all bec a use it neglects the ide a that wit m g the com mu whe n some of us achieve unde rstand mg . Thu ' discu'ssin
�wls
i
·
.
�
•
.
·
·
.
_
�
��
� l �
� : :� :�� �;� � -
·
1�1�
335
�
·
JOHN HALDANE
74
nal division of virtue in connection with the contemplative life, Aquinas writes:
There are things required of the community which one individual
alone cannot meet; this community-duty is discharged when one
does this and another that . . . . [T]he command to be fruitful falls on
the people as whole. They are bound not only to multiply in body
but to grow in spirit. The human family is sufficiently provided for
if some undertake the responsibility of bodily generation, while oth
ers deYote themselves to the study of divine things, for the beauty and health of the whole human race.32
IV. STATES, NATIONS, AND COMMUN ITIES
!\lot ever; state is a nation, nor even· nation a state - as is indicated by
the exampies of the former Soviet U�on and of Scotland respectively.
Both sorts of social entities can be communities, but there is an internal connection between the ideas of nationhood and community that is
absent between those of community and state. Let us say, then, that a
state invoh·es the organization of a collection of people under a system of justice and within a given territory, and that a nation is a people united
by common history, language, customs, traditions, and interests. Given
these features it is unsurprising that nations aspire to be states, or indeed that they regard this condition as their final end.
In a \•.reak sense, a community is any social group whose members
cooperate in pursuit of common interests. So defined, it easy to see how
many states will acquire the character of communities even if they do not originate as expressions of them. Thus, although a collection of previ
ously unassociated individuals may be brought together within a jurisdic
tion, it ·will generally not be long before they begin to interact in ways and under descriptions that indicate membership in a self-identifying social
group. Earlier, however, I quoted Augustine's definition of a people as
" a gathering of rational beings united in fellowship by a common agree ment about the objects of its love ." This clearly specifies a narrower,
moral notion of community, one which I termed "fellowship." Restricted as it is, Augustine arrives at this after having set aside an even more
exclusive definition of a people and of a state as a gathering united under proclaimed obedience to divine law.
Rawls v.-rites that the hope of political community must be abandoned "if by such a community we mean a political society united in affirming
the same comprehensive doctrine." 33 Obviously this would be taken to exclude the possibility of fashioning the state along the lines specified in
32 Aquinas. 33 Rawls,
Summa Theologiu (supra note 11), U, ll, q. 152, Political Liberalism, p. 146.
336
a.
2, ad. 1, p. 173.
73
rnE INDIV1Dt.:AL. THE STATE, AND TilE COMMON GOOD
Augustine's more restrictive account. It is not that the latter requires as much as a theocracy, but that even the assumption of theism is no lon ger tenable as a basis for civil society. In a widely viewed and sub sequently discussed television documentary, the Prince of Wales recently reflected upon his constitutional position as head of the Church of En and and "Defender of the Faith," remarking that, assuming his succes he might prefer to be "defender of faith." From the position within which Rawls develops his political liberalism, however, even this conces pol �i� n to religious pluralism harbors an unacceptable attachment to the should state the Itics of commitment if it presumes that the institutions of ack�owl�dge and give priority to one attitude toward religion (belief) as aga�nst others (agnosticism and atheism). Even Augustine's modified account fails to meet the stated conditions for an adequate political conception of justice, sirlce it violates what Rawls calls the "general facts of the political culture of a democratic society, " namely: the fact of reasonable pluralism (that a diversity of comprehensive on doc trines is a permanent feature); the fact of oppression (that comm state by ed adherence to one comprehensive doctrine can only be maintairl ires oppress ion); and the fact af majo rity suppo rt (that a secure regime requ s").34 citizen the support of "a substantial majority of its politically active of reason Earlier I suggested that Rawls's insistence on the permanenc� n of skep able pluralism is difficult to make sense of save as an expressio as a much so ticism . Equally, "the fact of oppression" is presented not r othe could not be historical datum but as a claim to the effect that things m alis ity of plur wise, the implied explanation for this beirlg the inevitabil . controversial are se The on. as a characteristic result of the exercise of reas t eve� so , want to oppose them - m�ght claims and the rati onalist believer may soaenes it is hard to make plausible the idea that modem Westem e that they alon let s, i com e to exhibit the u,nity of Augustinian fellowsh � egated surfaces. already are so ben eath their visible and visibly van of a comect the coherence My purpose in obsening this is not to rej earunplicati0ns of. the munitarian polity nor to distance mvself from the · so n:e that we az:e 1n lier discu ssion f the common g�od - na mely, . . on se telos is partiapa� respects social bein gs ' a genuine aspect of who ralism 1s c1atm · s 0f plu . share d ends Yet som e acknowledgment of the tana� eration of corrunuru certainly nec ssary' and in consequence a mod aw ' ticisms of R ls s · ..; n the ear1.ler en aspirations is called for However, give o· ....�.ens his account of a at h t ge char r attempts at neu trality and the briefe . from d " . over t gives (unw arranted) riority to liberal identi�es tion la _ g ones e e membe rship in oth er communities - mcludm . here w ' ust ed to cons1der 1 to co mprehensive doctrines - there remains a ne .an fell wshi and Rawls gustim Au of s pole ne sh the be ould stand tween � lan citizen shi p.
� SlO�,
�
�
.
�
·
p
.
34 n.u
'""'·· pp. 37-38.
337
�� � ;x;! 0
P
�
76
JOHN HALDANE
One's political nature is not independent of that "second nature" which results from being born and raised within particular social groups shar ing aesthetic, moral, philosophical, and religious inclinations communi cated to successive generations in part through the cultivation of a complex sensibility. Real-world political personae rest upon these cultural
identities (which they rarely obscure). Accordingly. if the mask is to fit, it must be shaped to the contours of the face, which tells against the attempt to fashion it out of a universal mold. In consequence, the order oi construction in practical political philosophy should be to define the characteristic values of given communities and ret1ect upon how these might be expressed in the political order of a state. Among other things
this approach limits the scope for a priori reasoning about the conditions oi i ustice and political stability and directs attention toward the histori
cal facts of community. In a way, of course, this is what Rawls himself claims to be doing, but in order to maximize the acceptability of the polit ical conception, he then tries, unsuccessfully, to detach the account from distinctive conceptions of the good.
Elsewhere I have developed these reflections in support of the ideal of nationhood and tried to respond to the familiar concerns that this notion is anachronistic, ontologically problematic, and politically coercive.35 I shall not return to these matters now save, first, to emphasize that, as I under stand it, the idea of a nation is a cultural and not a :acial one - conjoining history, tradition, and language, not color and blood-group - and, sec ond, to acknowledge that its deployment involves the sometimes uncom fortable notion that the state may legitimately concern itself with social iormation.* In the present circumstance, howe,-er, I want to suggest that while national identity may be an appropriate reference point for political reflection in some contexts, it may not be so in others; and where it is weakened or absent, there may be no other unifying cultural form that can sen·e as a naturally eligible, prepolitical foundation for civil soci ety. Attempting to establish political divisions along ethnic lines is always possible; but if this is the true condition of things then it may be better to acknowledge that in these circumstances politics is a solution to a prob lem of disunity rather than the expression of prior community. For some, this picture of the political order as no more than a procedural arrange ment for regulating the interactions between " strangers" will be d ispir iting, especially if they are attracted by the many-layered richness of a cultural-cum-nation-state such as Great Britain; but it is often, if not always, possible to combine political thinness with a rich community life; 3:; See Haldane, "Identity, Community, and the Limits of Multiculture," section IV. 3< For an example of what this might validate, see John Haldane, "Religious Educa tion in a Pluralist Society: A Philosophical Perspective," British jounuzl of Eduaztional Studies, vol. 34 (1986), pp. 161-81.
338
THE INDIV IDCA L, THE STATE, AND THE COMMON GOOD
77
and i circumsta nces of radical pluralism, wisdom may caution again � st granting a great er role to the state than the maintenance of civil peace.
V. A Mo o u s VtVE:-.; ot AND PoLITICAL PHILO SOPHY
IN
Co�xr
By now the thoug ht is e merging that in certain circumstances, and some might argue that these are coming to be general within Western societies, the hope of political community having had to be abandoned and the prospec t of political liberalis m having proved illusory, the way forw ard may lie in the developme nt of political order as a modus vivendi. For Rawls , this is an unsatisfa ctory condition that must be sharply distin guish ed fro m an overlapping consensus on a political conception of jus tice. The differenc e is marked by three features: {1) an overlapping c onse nsus focuses on a moral conception of justice; {2) it is affirmed on moral grounds; and (3) because of these facts it enjoys a kind and degree of stability differen t from that of a mere modus vivendi, which can only de pen d on "happenstance and a balance of relative forces." 3; At this stage in the discussion I shall not begin to consider whether
Rawls is entitled to or justified in his claims on behalf of the moral supe
� the su� VIV�1· �or t� . possibility mav seem to force a less critical reassessment ot political lib riority of an overlapping consensus. All I am concerned �ith
gestion that it possesses a greater stability than a modus
eral is m . In fact, Rawls allows that an overlapping consensus may not be necessary for social unitv and stability and goes on to address the con�em that it is u topian to believe that it is sufficient for it. Ho�ever, the tdea remai n s u nquestio n e d that while stability may be achiev ed by other e ans m it is then a verv precarious condition . . Why shoul d this be.thought to be so ? I suspect that part of the reaso.n 15 15 th at alth oucrh he refers back to the religio us wars of Europe. Rawls thinking in te some of American societv and political life, which may times bear the appearance of an arra�gement that could eas�y coi?e apart . power . as the bala n ce of forces changes, and with it the distnbuno n ot . o£ ThI s pros pect seems far less likely in a po 1.I�I· ca1 cultu. re such . as ,that . . . due 15 soae� Gre at Brit ain. Yet it is doubtful that the stability ofBnti.sh s s sense . t an overl Rawl in eption one c a1 · apping consensus on a po 1.ItiC . this civil . tor credit If any perspective of noble principle is entitled to claim order, it is a com munitarian one . . t Of VIeW 01· cu....; u cu Bu t that is too quick and too simple. Fro� the pom terv. mvs lated theory shrouded in a. British political societv is an emgma ' blic inStitutions owes On the on e h pu ts . f anor o I and the intellectual deme d John much to the ci vil' a n d eudaimonistic liberalis ms of John Lo�ke an the On al conscience. Stuart Mill, both of which are respectful of individu
�
0
.
.
•
·
37 Rawls, Political Libera lism. pp. 147-48.
339
·
78
JOH!\: HALDANE
other hand, and notv.ithstanding its lack of a formulated constitution, it
is ruled by a "constitutional" monarch who is the head of an established
church and whose government is conducted in part by an unelected nobility which is most effective when rallying to such causes as the pres
ervation of rural bus services and the maintenance of Christian education in state schools. That education derives nominalh· from the
Church of En
gland (and from the Presbyterian Church in Scotland - of which the sov ereign is also a protector), yet the largest worshipping denomination is Roman Catholic, whose members, by the act of succession, are debarred from ascending to the throne or marrying the sovereign.
lt is difficult to discern a principled justification for this general pattern
of arrangements, and I believe it is a mistake to seek one - both insofar as
there is no coherent set of organizing principles and, more importantly,
because the whole set-up has no obvious political purpose independent of its ovm continuation. Piece by piece, the elements may have been fash ioned to serve particular purposes - some internal, some external - but over time they have become part of a system that remains stable because
it is an embodiment of a social order that is found congenial to a civilized life.
I realize this may sound self-satisfied; however, the point I want to end
on is not the supposed superiority of the British political order but the fact that for good or ill this order exists, and (to the extent that it offers any
j ustification) justifies itself independently of a theory of the right. Even
if it originates in and is maintained by a series of pragmatic resolutions,
these quickly come to be the object of civic allegiance, particularly as they are given the protection of law. If this is a modus vivendi ·writ large, it cer
tainly seems no less stable than an overlapping consensus, and it is a way of going on socially that is compatible with active participation in a range of subordinate moral communities, and with the periodic accomplishment of principled political goals. It would be presumptuous of me to speculate at any length on the extent to which a similar account might be given of the United States (not
to mention the interesting case of Canada). However, if there are real dif ferences, as the emphasis on individual rights and the recurrent preoccu pation with the meaning of the founding constitution su�t, then it would be as well for political philosophers on both sides of the Atlantic to consider the possibility that these cultural and political differences may have found their way into the reflective understanding of such concepts
as community, individual, and state, and that most of the foundational
work done during the renaissance of English-speaking political philoso phy initiated by Rawls has really concerned the political foundation of North American society - an issue formulated not inappropriately, but not altogether accurately either, as "the problem of liberalism." Of course,
it may be that even if this diagnosis were once true, the influence of U.S.
culture in Europe and elsewhere has been such that we are all now in the
340
THE
lNDlVTDt:AL,
THE STATE,
79
AND THE COMMON GOOD
same � oral and political situation. I doubt that this is so, however, and I c:rtamly hope that current uncertainties about the direction of political
philosophy may lead to some cultural differentiation of issues, methods, and doctrines, and to a more extensive investigation of what is available from "old" and "new world" writers of the past, such as those with whom I began.
VI. CoDICIL
In challenging the project of liberal political theory, I have been con cerned principally with Rawls's neutralist and individualist version of it. Directed criticism of this sort raises both a general question of the ade �uacy of other forms of liberalism, and, given the perspective developed
m later sections, a more specific question concerning the compatibility of Catholic social teaching with any kind of liberalism. In an earlier essay, addresse d to the question of whether a Catholic can be a liberal, I described the opposition between liberal theory and Catholicism in tenns
of fundamental disagreements about the place of morality in politics, and t�e status of the community and the common good.38 There as here, the liberalism in question was principally that associated with author� such as R awls and Ronald Dworkin, rather than Locke and �1ill . While the implied distinction behveen "old" and "new " liberalisms may not be as ?I'eat as some like to suppose, there is certainly a question as to whether tdeas of political lib ertv other than those associated with contemporary contractu alism, better. accommodate the value of community and the claims of m orality in the public sphere. In a (loose and popular) sense, \\:e are all lib erals now, or should be; and the ackno wledgment of presents a chall es enge to advocates of general and comprehensive doctnn to show how er Secular. values. � they can endorse commonly held liberal fectio nists such polincal as Joseph Raz do so by arguing that personal of autonfreedom is an aspect of the good life (deriving from the value_ de\·elomy and fro bemg is :-:t . m value pluralism),39 and someth mg swwar and oped by American of Catholic writers influenced by the likes these s m Joh n Co urtenay M succes urrav .w However, while I hope for ventur es, the seem either to beg th e · ume�ts generall Y presented �rg . d unporta nt questions · s. For . now an or to fall short of thetr conduswn . . tbillty sens perh ap s forever, liberal the best hope of liberty lies in a
�his
��an�am .
.
.
·
'
.
'
Moral Ph ilosoph y, Un il.·ersity of St. Andrews 38
pp.
.
vol. 43 (19921, . ral.?" Melita Theologica. See John Haldane, "Can a Catholic Be a Ltbe
44-57 .
39 -10
nd 15 · See Raz, The Moralil·ll .:�f Freedom , esp. chs. 1 4 a
ism ss eds . Cath olic S ee, for example, the essavs in HoUenbach an d Dougla '
Li beralism .
·
341
.
wnd
Acknowle dgmen ts
Deigh , John . "Sha m e a n d Self-E steem : A Critiqu e," Ethics 93 ( 1 98 3) : 225-4 5. Repri nted with the permiss ion of the University of Chicago Press. Sach s, Davi d . "How to Distinguish S elf-Respect from Self-Esteem, " Philosophy and
Public Affairs 1 0 ( 1 9 8 1 ) : 346-60. Copyright 1 98 1 by Princeton University
Press . R eprint e d by permission of Princeton University Press. Th om as, L arry L "Ra w!s ian S elf-Respect and the Black Consciousness Movement," Ph ilosoph ical Forum 9 ( 1 977-7 8 ) : 3 0 3-1 4. Reprinted with the permission of Phil os oph ical For um, Inc. Kohlbe rg, L awr ence. "The Claim to Moral Adequacy of a Highest Stage of Moral Judg m e n t," Journ al of Philosophy 70 ( 1 973 ) : 630-46. Reprinted with the p ermission of the Journal of Philosophy, Inc., Columbia University, and the aut hor.
Bate s, S tan ley. "The Motivat ion to Be Just," Ethics 8 5 ( 1 974): 1 - 1 7. Reprinted with the p e rmi ssion of the University of Chicago Press. . Gib b ard, Alla n . "Hum a n Evolu tion and the S ense of Justice," Social and Political Ph ilosophy, Midw est Studies in Philosophy 7, edited by Peter A. French, Th eod ore E. Uehl ing, Jr., and Howard K. Wettstein ( 1 982): 3 1 -46 .
R e prin ted with the permission of the University of Minnesota Press. Cop yri ght 1 9 82 . Freem a n, S am uel. "Reaso n and Agreement in Social Contract Views, " Philosophy and Public Affairs 1 9 ( 1 990) : 1 22-57. Copyright 1 990 by Princeton University P res s . R eprinted by permissio n of Princeton University Press. . McC len ne n, E d w ard F. "Justice and the Problem of Stability," Philosophy and Publzc
University Press. Affairs 1 8 ( 1 989) : 3 - 30 . Copyrig ht 1 98 9 by Princeton R e pri nted by permission of Princeton University Press. . . . H:Ill, Th o mas E., Jr "The Stabili ty Proble m in Politica l Liberalism," Pacific Phzlosophzcal . · · Qua rterly 75 ( 1 99 4 ) : 3 3 3-52 . Reprinted with the permissiOn °f Basil Bl ackwell. . Position: Philosoph zcal r, S a m uel. "Mora l Independence and the Original . . · of Klu wer Studzes 35 ( 1 9 79) : 397-403. Repn.nted Wit h th e permission Comp Y· n D. Reidel Pu blishi g Acad e mic Publish ers. Copyright 1 979 . Ethzcs ous Pro u dfoo Relzgz f o Social: Journal t , Wayne. "Rawls on the Individual and the S che ffl e
��
344
ACKNOWLEDGMENTS
2 ( 1 974) : 1 07-28. Reprinted with the permission of Blackwell Publishers.
Sandel, Michael J . "The Procedural Republic and the Unencumbered S elf, • Political Theory 1 2 ( 1 984) : 8 1 -9 6 . Reprinted with the permission of Sage
Pu blications, Inc. Delaney, C.F. "Rawls and Individualism," Modern Schoolman 6 0 ( 1 9 8 3 ) : 1 1 2-22. R eprinted with the permission of the St. Louis University. Gutmann, Amy. "Commu nitarian Critics of Liberalism," Ph ilosophy and Public Affairs 1 4 ( 1 98 5 ) : 308-22. Copyright 1 9 8 5 by Princeton University Press.
Reprinted by permission of Princeton University Press. Schwarzenbach, Sibyl A. "Rawls, Hegel, and Communitarianism," Politica l Theory 1 9 ( 1 99 1 ) : 5 3 9-7 1 . Reprinted with the permission of Sage Publications, Inc.
Alejandro, Roberto. "Rawls's Communitarianism," Canadian Journal of Ph ilosophy 23 ( 1 99 3 ) : 7 5-99. R eprinted with the permission of the University of C algary
Press. Haldane, John. "The Individual, the State, and the Common Good," Social Philosophy and Policy ( 1 99 6 ) : 59-79. Reprinted with the permission of Cambridge
University Press ( North American Branch ) .