The Politics of Persons: Individual Autonomy and Socio-historical Selves

THE POLITICS OF PERSONS

It is both an ideal and an assumption of traditional conceptions of justice for liberal democracies that citizens are autonomous, self-governing persons. Yet standard accounts of the self and of self-government at work in such theories are hotly disputed and often roundly criticized in most of their guises. John Christman offers a sustained critical analysis of both the idea of the self and of autonomy as these ideas function in political theory, offering interpretations of these concepts which avoid such disputes and withstand such criticisms. Christman’s model of individual autonomy takes into account the socially constructed nature of persons and their complex cultural and social identities, and he shows how this model can provide a foundation for principles of justice for complex democracies marked by radical difference among citizens. His book will interest a wide range of readers in philosophy, politics, and the social sciences. john christman is Associate Professor of Philosophy, Political Science and Women’s Studies at Pennsylvania State University. His previous publications include Social and Political Philosophy: A Contemporary Introduction (2002), The Myth of Property (1994), and as co-editor Debates in Political Philosophy (2009) (with Thomas Christiano) and Autonomy and the Challenges to Liberalism: New Essays (2005) (with Joel Anderson).

THE POLITICS OF PERSONS Individual Autonomy and Socio-historical Selves

JOHN CHRISTMAN The Pennsylvania State University

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi Cambridge University Press The Edinburgh Building, Cambridge cb2 8ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521760560 © John Christman 2009 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2009 Printed in the United Kingdom at the University Press, Cambridge A catalogue record for this publication is available from the British Library isbn 978-0-521-76056-0 hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party Internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

To the surviving spirit of the people of New Orleans, my hometown, and to the courageous resilience of my family, who continue to recover and flourish with grace: my sisters Jan and Bonnie, my brother Tim, my father John and step-mother Dolores, and all the members of their families.

Contents

Acknowledgments

page ix

1 Introduction

1

part i 2

19

selves

The social conception of the self: a critical taxonomy

21

3 The post-modern subject

48

4

66

The narrative self

5 Memory, agency, and the self part ii

86 107

autonomy

6

Political persons

109

7

The historical conception of autonomy

133

8

Relational autonomy

164

9

The dynamics of social identities

187

part iii

217

justice

10 Justice over time: history, public reason, and political legitimacy

219

Conclusion

245

Bibliography Index

247 270 vii

Acknowledgments

This book evolved over a number of years, so in the process of writing it I benefited from the input and support of many individuals and institutions. Some of the early stages of the project began while I was on a Fellowship for College Teachers and Independent Scholars from the National Endowment for the Humanities in 1998–99, for which I am very grateful. Some of the material in the chapters below is a distillation or expansion of papers presented in various colloquia and conferences. I am grateful to the hosts, commentators, and audiences in those venues who provided comments on earlier versions of these ideas and no doubt improved them greatly. There are too many such people to mention them all, but they include: Linda Alcoff, Joel Anderson, Kim Atkins, Holger Baumann, Kenneth Baynes, Monika Betzler, Philippa Byers, David Copp, Marilyn Friedman, Marlène Jouan, Anton Leist, Catriona MacKenzie, Marina Oshana, James Stacey Taylor, and Lori Watson. I also gained tremendously from the careful reading that two anonymous reviewers for Cambridge University Press gave to the manuscript. Many additions, clarifications, and qualifications were added as a result of their insightful comments, for which I am very grateful. In addition, some of the chapters below include adapted material from previously published articles. Specifically, chapter 4 is a revised version of “Narrative Unity as a Condition of Personhood,” Metaphilosophy 35(5) (October, 2004), 695–713; chapter 5 is a revised version of “Why Search for Lost Time: Memory, Autonomy, and Practical Reason,” in Catriona Mackenzie and Kim Atkins, eds., Practical Identity and Narrative Agency (New York: Routledge, 2008); chapter 7 uses material from “Autonomy, History, and the Subject of Justice,” Social Theory and Practice, 33(1) (January, 2007), 1–26; chapter 8 is a much revised version of “Relational Autonomy, Liberal Individualism, and the Social Constitution of Selves,” Philosophical Studies, 117 (2004): 143–164 (all material used by permission). Also, chapter 5 is based on a presentation I made under the title ix

x

Acknowledgments

“Individual Autonomy, Social Selves” for a conference on Autonomie, Moralité, Affectivité, Université Picardie Jules Verne – Amiens, March 6–8, 2007. Writing this book was made easier by friendships, associations, and professional assistance from a number of individuals and institutions. I am supported in my research by the Pennsylvania State University Departments of Philosophy and Political Science. In particular my colleagues in Philosophy have been an endless source of stimulation and ideas. In addition, I am grateful to Joshua Miller and Evan Seehausen for very helpful copyediting and comments on the manuscript. Friends and associates such as Thomas Christiano, Vincent Colapietro, Matthew Jordan, Elizabeth Lewis, Matthew McAllister, Gerald Nosich, and Shannon Sullivan have been a constant source of help and support. Finally, and as always, I again attempt to thank Mary Beth Oliver for constant friendship, love, and support in terms sufficient to her contributions. And again, I know I come up short.

chapter 1

Introduction

Who are you? Really, how would you answer such a question, coming as it does from an anonymous and context-free page in a book? You might give your name, or some other designator that will pick you out (“I am the one reading this book here …”). But could you give a substantive, descriptive answer? Most likely, whatever comes to mind as a possible answer would depend on the point of asking the question. Different contexts would evoke different, or at least very differently organized, descriptions of yourself. Though the question that is really at issue here – and this will be one of the central themes of this book – is “who should I say you are …?” And I should quickly add, nothing I have said or want to say in what follows should preclude your answer beginning with the words “We are …” Philosophical reflections on the nature of the self are wide ranging of course. One could say, in fact, that concern with the nature of persons, human nature, and the self represents some of the central topics of philosophy, in all its various traditions, throughout its history. Metaphysical accounts of the essential nature of (human) being have abounded in that history. Social and political philosophy, in its way, has taken up such theorizing and proceeded on the basis of assumptions about human nature, the self, and agency to construct or assume models of the subject of political and social institutions.1 At the same time, countercurrents have developed that decry any such attempt to develop a single, metaphysically grounded, account of what it means to be a person. Emphasis on difference, cultural variation, and variegated modes of living (previously ignored in mainstream theory) has motivated this rejection of all-encompassing models of the self. The question for normative social 1

I make many observations in this Introduction that will be documented with references in the chapters below. Here I mean only to set the stage for those discussions.

1

2

The Politics of Persons

theory, then, is what conception of the self can be utilized in articulating general principles while taking into account the radical disagreements that have always existed about the nature of the person. As Michael Sandel has claimed, every political philosophy presupposes a “philosophical anthropology” – a conception of the person assumed as the subject of that philosophy (Sandel 1982). Most such anthropologies are implicit, in that the capabilities, interests, perspectives, and relations of the persons to which political principles apply are never articulated as such. Rather, they are often assumed as background facts, perhaps as aspects of human nature taken to be obvious or universal, and not the controversial subject matter of the particular political philosophy being developed. But political and social institutions guided by principles all presuppose a model of the individual(s) acting within them. What can be called the “subject of justice,” then, names the model of the person that is operative in the principles guiding the workings of political institutions that answer to the dictates of justice. Notoriously, Western political philosophy in the modern age – dominated by what is broadly characterized as liberal theory – has assumed that the model of personhood to be utilized in these contexts is fundamentally individualistic. The subject of principles of justice is first and foremost the individual as such, however many connections and close associations such persons might have in their actual lives. In addition, the picture of the citizen of the just polity includes no specific reference to the marks of social identity, such as race, gender, sexuality, culture, and so on, that many actual individuals might immediately mention when describing themselves. The model person, in the liberal tradition, is characterized without essential connections with past or present others or social factors external to “him.” This view is now deeply resisted in many quarters. In fact, one could say that an uneasy consensus has developed in opposition to the assumption of a thoroughly individualistic conception of the person (as the subject of principles of justice), a conception where no reference is made to relations with other persons, traditions, historical practices, and social forms. The political self is, in some ways at least, a social self, marked in various ways by indicators of a social identity, however this is conceived. At the same time, it is also quite controversial to conclude that the interests and perspective that are represented in normative principles are specified at the level of groups, such as identity groups. It is a matter of much debate whether social relations as such must be represented as the primary object of interest in political theory. The challenge, then, is to construct models of the person as represented in political theory that, on the one

Introduction

3

hand, make room for (if not assume) the deep interpersonal, social, cultural and historical connections that structure the self-concept of many or all of us. This must be done, however, in a way that avoids begging the question about whether the interests of individuals per se should be the fundamental object of social concern. In the present work, I attempt to wade through some of these deep and murky waters by investigating the various ways in which selves are said to be, not only social but, I will stress, historical, in the sense of being diachronically structured and subject to change over time. More specifically, I argue that however we model “selves” in political principles, room must be made for the ways in which we are often defined by our social and temporal relations, in short, that we are socio-historical selves. It will also be claimed, however, that the aspect of persons that must be assumed as well as promoted and respected in normative political principles is their autonomy. Putting these points together, the central focus of this book is the project of working out a conception of individual autonomy for socio-historical selves operative in principles of justice applying to modern pluralistic societies. This work can be seen as a continuation of the multi-vocal discussion that has emerged over standard liberal theories of justice, a discussion that includes the allegation that such theories have systematically occluded the representation of persons as socially identified and historically embedded. The challenge to liberal political philosophy – voiced by feminists and other defenders of identity politics, post-modern theorists, communitarians, multicultural theorists, and others – has been wide ranging and voluminous. The liberal response has also been robust. Both sides, it appears, have agreed on the rejection of the traditional liberal conception of the autonomous agent as the separated, independent self of modernist lore. The social self (or in the post-structuralist idiolect, the “de-centered self”) has replaced the unembedded and disembodied autonomous “man” of traditional theory. What remain to be examined in this exchange, however, are the precise aspects of this socially constituted or de-centered self that must be taken into account in normative theory. Specifically, the question arises whether the idea of autonomy at the center of models of modern democratic theory must be rejected along with the idea of the “lone wolf self.” The way to answer this question is to connect the contours of the socio-historical self with a model of autonomy in order to see if the latter’s requirements preclude the former’s assumptions. Can we see individual autonomy as a property of social selves (specifically socio-historical selves)? I will argue here

4

The Politics of Persons

that we can, and for the purposes of constructing at least one (I hope plausible) framework for principles of justice applying to modern, pluralistic societies, I hope to show how this might be accomplished. i. the context As I mentioned, normative political theories, in particular conceptions of justice, always presuppose or assert a conception of the person taken to be the subject of those principles. There are two chief reasons for this (among others): the legitimacy or justification of such principles will, according to a certain tradition, rest upon their being acceptable to those living under them. Such collective endorsement will be postulated or aimed for in ways that reflect the kinds of people whose acceptance is in question, so specifying the political subject determines the nature and possibility of political legitimacy. In addition, people’s interests will be represented in the structure and aims of those principles. That is, principles guiding the construction and operation of social institutions will assume a profile of basic interests the protection or promotion of which provides the content of those institutional directives. Even if such interests are described formally so as to avoid any substantive conception of the good for human beings, they will still express a broad view of human purposes, needs, and so on and therefore contain a representation of such people whose interests are being so served. Of course people’s interests are expressed in a variety of ways in democratic institutions, notably by their actual representatives in legislative assemblies. But we are talking here about something more basic, the model of what a “person” is, fundamentally, in the design and articulation of the basic principles of the constitution. Traditionally, the task of constructing the conception of the citizen represented in these principles has been left to philosophical (and religious) thinkers whose ruminations on human nature have provided metaphysical accounts of personhood to be utilized in politics. As I mentioned, such traditional metaphysical accounts have also been roundly criticized for ignoring several factors central to many people’s conceptions of themselves: the importance of embodiment in our natures, the way that intimate relations form the core of many people’s selfconcept, and the central function that social markers of identity play in this context. Moreover, reflections on the nature of the self can be notoriously free floating, as if a single conception of the person or self can be fashioned without attention to the geographical, social, and historical locations of both the theorizer and her subject.

Introduction

5

My task here will not be to systematically engage in these critiques of metaphysical accounts of personhood. That is, I proceed on the assumption that we are functioning in a “post-metaphysical” context, as Habermas and other commentators have put it (Habermas 1999, 143; Rawls 1993). We must posit conceptions of the political self that have no necessary pretensions about universality or metaphysical necessity. So we should consider models of selves that will be useful in particular contexts and for particular theoretical and practical purposes, in this case the construction of principles of justice for constitutional democracies in the late modern age. The intellectual tradition in which this discussion occurs emerged out of the seventeenth- and eighteenth-century European political philosophies which gave rise to contemporary liberal theories of justice. This tradition was built on the rejection of hierarchical models of political power in which the legitimacy of regimes was thought to rest on natural inherent powers of hereditary title or military prowess. The social contract theories of that earlier age expressed this opposition and constructed theories of political power that rested on popular will, expressing the fundamental rights of persons defined pre-politically (as natural rights). The contemporary inheritors of that tradition vary in the ways that they interpret the demands of such popular sovereignty. However, all such writers now admit, and in most cases face head-on, the dramatic degree of pluralism found in the populations where such political power is exercised. Differences in identity, embodiment, life histories, cultural associations, language, moral commitment, and so on mark the landscape of contemporary theory, even liberal theory. Opinions differ, of course, about what to do about such differences. But also, theorists of political authority in this tradition have increasingly taken seriously the predominance of violence, domination, and power differentials that shape our history (and in fact track those differences in social identity in many cases). Moreover, and I will discuss this in chapter 10, the context to which the issues discussed here apply are democracies with their own histories of specifically non-democratic practices. Most discussions of political principles assume a less than ideal setting to which such normative provisions apply – the so-called “circumstances of justice,” for example, which specify such things as limited altruism and relative scarcity of goods. However, I want to make the context more realistic in assuming that the histories and to some extent current practices of such societies include systematic inequalities of power and systematic patterns of violence. Relatedly, I assume that such democracies exist and function in a world that also includes many powerful non-democratic regimes. As we will see,

6

The Politics of Persons

the interests and viewpoint of persons represented by principles for such democracies will be different from those in more pristine and less violent settings. The question that we must face, then, is whether conceptions of the person at work in theories applied to such contexts must radically shift away from traditional models because of these factors. I will suggest that they must to some degree, though not in ways that some commentators have insisted. In particular, I will defend the view that individual autonomy, specified in ways that make essential reference to individual history but not necessarily to particular social connections, will be necessary as a lynchpin of pluralist democratic theory and will also survive critiques from those sensitive to social and historical constituents of identity. Also, a word about a word: liberalism. Most of the discussion in what follows will concern “liberal” political philosophy. And in much of my own past writing, some of which is inherited in revised form here, I took myself to be defending a version of liberalism, or at least some of its central ideas. But I have found that in discussions of these topics, the term “liberalism” has taken on such a volatile and variable set of meanings that I largely avoid it in what follows, especially in statements of my overall aims. I prefer to talk about anti-perfectionist, autonomy-based, democratic theories of justice. These are normative approaches to politics and social life that, like traditional liberalism, assume that conflicts among moral worldviews and value systems are a permanent element of political life, and therefore principles of justice should, as much as possible, avoid resting on any one of those controversial views. However, I also think, and this contrasts with much liberal thought, that democratic procedures are constitutive of the generation of principles of justice and not merely complimentary to it. Here I have more in common with some self-described non-liberal thinkers who align themselves with (radical) democratic theory. I cannot defend a broad theory of justice here and won’t try. Instead I merely discuss, in some detail, some of the key conceptual elements at work in such a theory. But I want to make as clear as possible what my theoretical commitments are, and by extension the theoretical frame into which those elements are intended to fit, so as to illuminate those broader theories (in ways that, I hope, will be of interest to both their defenders and detractors). In specific instances, however, such as my rejection of perfectionism in the shaping of principles of justice, I will argue for the view in question, but in others I will simply designate the overall framework as I work within it. Most of these broader reflections are contained in part III of the book (containing only chapter 10). The principal elements of the view developed

Introduction

7

here, however, concern the model of autonomy and the socio-historical self to which such autonomy is meant to attach. I consider these ideas in turn, first discussing the self and then the autonomous self. Let us proceed then with an overview of the topics as they unfold in the chapters below. ii. selves Talk of “selves” and “the self” carries many connotations and ambiguities. Why, for instance, do we refer to “the self” rather than “the person” or “the human being”? What about “the agent,” or “the subject”? Philosophical literature on this matter has ranged over all of these terms and topics. The concentration on the self in this work is not meant to rest on any deep philosophical claim about what is truly central in talking about the human experience such that reference to “persons” or “humanity” or “agents” is somehow wrongheaded. Rather, reference to selves seems to avoid at least some of the connotations that the other terms have but not, I hope, include others of its own that will confuse matters. The self, in the current discussion, is simply the set of elements of human persons that are relevant to normative political principles and social theory of the sort being considered. Such terminology avoids the view that such a self is essentially a rational chooser (as the term “agent” might imply) or a metaphysical entity with identity conditions over time, as suggested by the word “person” perhaps. And as I mentioned, it is also important to avoid the assumption that the self is somehow a specifiable formation at the core of all personalities and lives, faithfulness to which is the prime directive of self-government. Talk of selves is always talk of model conceptions of the wide variety of embodied capacities, commitments, bodily traits, values, and desires that structure the perspective and ground the interests of the subjects of justice. In this way, I join others in rejecting the view that there is a true, core self inside us all that our freedom and autonomy expresses, a self that can be specified psychologically or philosophically and which functions across contexts in our lives (see, e.g., Meyers 2005). The particular “selves” that operate in various contexts vary for all of us, and therefore the model of “the” self should also vary according to the purposes of specification of such a model. What this means, though, is that this fictitious entity, the onetrue-self, is a vanishing point that is always being represented or modeled but which does not, in a sense, exist. There is no single true self guiding my life; elements of the self that are relevant for practical reasons emerge as one acts, speaks, and expresses oneself. Therefore it is implausible to postulate

8

The Politics of Persons

a static set of values, interests, capacities, and the like that make up the settled self prior to such actions. I will attempt to sort out these paradoxicalsounding claims – in particular that models of the self represent an entity that cannot be specified independently of that representation. This may suggest an affinity with the post-modern rejection of any talk of stable selves or settled meanings in discourse about politics or humanity. In a variety of ways, so-called post-modern or post-structuralist thinkers have taken skepticism about foundations generally – of knowledge, morality, politics, and meaning – to extend to conceptions of identity and the self. Not only is there no stable true “self” at the center of our being, they suggest, but all thinking and language utilizes unstable symbolic systems whose “remainder” (elements of meaningfulness not captured sufficiently by those symbols) are operative surreptitiously in reflections about the world and ourselves. Such positions will be explored, specifically in chapter 3, and I will make a variety of crucial distinctions between terms and claims operative in this post-modern landscape. The conclusion will be that we can accept (or be agnostic about) much of the challenge raised by these thinkers, in particular, their healthy skepticism that selves are fixed and transparent structures that can be accessed through introspection. I want to maintain that “self”reflection (I’ll avoid the quotes from now on) should still play a role in a person’s representation of themselves in social discourse, even if this representation goes “all the way down,” as it were, without there being an entity or set of functions or beliefs that exists independently of that representation and that provides its ground. Moreover, we should be able to distinguish adequate or valid reflection from manipulated or distorted reflection without presupposing a stable object of self-understanding of the sort those critics decry. Also, on the view to be developed here, selves are not entities that can be fully grasped at a single instant. They are diachronically structured in that their elements exist in and over time. We have pasts and futures, life spans which include childhood and, we hope, old age. Our memories and future aspirations are as much a part of us as our current array of capacities and desires. And, in order to emphasize the elements of our selves that have been most neglected in standard models, I will call the view the sociohistorical concept of the self. I will also discuss a version of this approach that views selves and their lives in narrative terms. For quite some time now, reference to narrativity has been prominent in certain corners of psychology, social theory, political philosophy, and theories of persons more generally. I will adopt a version of

Introduction

9

the narrative conception of the self, but as I will argue in detail in chapter 4, narrativity in my sense does not refer to a set structure resembling a story or tale but rather more generally a set of experiences, actions, events, and traits that are structured according to standards of coherence provided by the subject/person herself, mediated by socially operative meanings. The historical self is narratively structured only in this broad sense of being diachronically comprehensible. But I will argue that rather than seeing that there is a core of such narratives – the self as author – we should look at the ways in which we are always reinterpreting the events, character traits, relations, and memories that make up our lives. In this way we are less authors of this narrative than, as it were, its literary critics! The self modeled here is also social, in a deep sense. As most theorists these days accept, the self is socially constituted by virtue of the language we must use to understand and express ourselves, the values and commitments we structure our lives by, our emotions and memories, our bodies, and much else, so much else in fact that it will take a full chapter to sort it out. Chapter 2 includes a discussion of the many ways in which selves can be thought to be socially constituted. I will reject some proposals in this vein, but my major aim is not to settle on a specific social conception but rather to argue that even if we accept the social thesis, the possibility of individual self-reflection and self-evaluation central to notions of autonomy (of the sort I develop here) remains a live option, psychologically, metaphysically, and most importantly, politically. In this way, I attempt to maintain a robust ecumenicalism about conceptions of the self, a stance in keeping with the pluralism in politics that I go on to embrace. However, seeing the self as diachronic, indeed as historical in the sense I suggest, requires that we take a keen look at the capacity we (most of us) have to grasp ourselves over time, and in particular to understand our past. For that reason, I consider the nature and importance of memory in understanding the self (chapter 5). Specifically for the purposes of theorizing about agency and active selves in society – the purposes that are engaged in the context of normative social theory – I suggest that memory is important because it both structures and presupposes a temporally extended self-concept which, in turn, is needed to reflect on ourselves in the way (I will argue) autonomy requires. Therefore, part I tries to establish the conclusion that selves can be plausibly understood as socially constituted and irreducibly diachronic. That is to say, they should be understood as narratively structured but only in a broad sense. As such, the model of the self that will play the most useful role in social theory will have a capacity for autobiographical,

10

The Politics of Persons

narrative memory, the function of which will be necessary for the development of that very diachronic self-concept. In the subsequent chapters I interrogate the idea that citizens in democratic societies should be (individually and collectively) self-governing, but the self that does the governing, so to speak, is the socio-historical self described in these chapters. One last portion of this picture is needed, however, and it is set out for emphasis rather than as a profound discovery. That is, selves should be seen as to a large extent formed by factors not under the control of those reflective agents themselves. Most of the central elements of our existence are things that were not (and in many cases could not be) chosen by us. Our parents and childhood conditions, for example, could not have been chosen by us as adults, yet who our parents are, what they did, and the kind of lives we enjoyed as children have a tremendous impact on what our values, options, and perspectives are now. Moreover, we are embodied creatures, and our bodies grow old and are subject to sometimes radical changes, such as from disease, injury, violence, pregnancy and childbirth, growth, and so on. Moreover, many of our social relations and the social context in which we define ourselves were not chosen and in some ways cannot be escaped. We can emigrate (though not always, or only at great personal cost), but we cannot choose to have been born somewhere else. Insofar as we want to define ourselves by our geo-political location and its social inheritance – and many of us do, either positively or negatively – we often have only one such legacy to choose from (albeit one with multiform elements). I therefore want to take stock of the myriad ways in which selves are not self-created. This will help accomplish two things: to provide grounds for the rejection of models of agency and citizenship that assume Herculean abilities to fashion ourselves out of whole cloth; and to force us to focus more carefully on what powers of self-shaping we therefore are left with. This discussion, carried out in chapter 6, is not about metaphysical or even social determinism, for my observations there will be much more meager (and perhaps less interesting). I will talk in terms of degrees, so that the conclusion that we are not wholly self-creating can be established in ways that do not disturb the beasts of the free-will or agencystructure debates.2 The point must be that the role of the self’s control of the self (and the attendant social elements of both “selves”) will be circumscribed by the ways in which our lives are shaped for us and not by us. 2

For recent discussions of these issues that connect to our current concerns see Appiah 2005, 51–61, Giddens 1991, and Habermas 1999.

Introduction

11

In particular, it will be necessary to take stock at the various conclusions of part I – that elements of selfhood that must be accounted for in such models include social elements, historical factors, and embodiment – and thereby construct an overall template for a political conception of the person. As we will see, the malleability and contingency of selves and hence self-concepts will force us to adopt an abstract and flexible political conception. But this does not prevent us from claiming that, despite the social and temporal elements in the self-concept we have identified, the idea of individual autonomy – self-government – can still be defended as a fundamental normative notion. iii. autonomy The tradition of which this book is a part understands autonomy as selfgovernment, guided by the etymology of the word (auto-nomos) as well as the analogy with political autonomy that the ancient use of the term implied. In part II, I want to explore the concept of autonomy as it might function in the political context of democratic theory. While I directly confront the claim (in chapter 6 and part III) that it is the autonomous person who should be the subject of democratic justice, in this section I want to focus simply on the concept of autonomy itself. In this context, the theme that selves must be seen as both socially constituted and historically structured will be emphasized further. As a result, we will look at autonomy as it applies to such selves. For reasons that will be set out, I will suggest that autonomy should be seen as a historically structured aspect of persons, or more precisely, a dimension of persons conceived as socio-historical entities. What this means is that to be self-governing or self-determining (or as I will prefer self-managing), it will be necessary to inquire into the diachronic processes that gave rise to our current condition. Autonomy, then, will involve selfreflection as temporally extended, so that the broad range of characteristics that make up the self will come into focus. In chapter 7, I will work out and try to defend a historical conception of autonomy along these lines. But as I said, people enter and act in social life with identities, ones which connect them with others both in their own self-conceptions and the eyes of their fellow citizens. In addition, certain claims made on social institutions are not merely desires to advance one’s interests but flow out of, and are necessary to maintain, those very identities. Such identity-based claims often do and should get greater weight in the calculation of interests by democratic procedures, and philosophical room must be made for the

12

The Politics of Persons

extra heft of such claims. Doing so, however, seems to run counter to seeing autonomy, which traditionally is understood without mention of such selfdefining social ties, as the center of interpersonal respect required by democratic justice. But I will argue in chapter 9 that seeing autonomy in the way I suggest does not run afoul of these powerful considerations. Recognition of identity-based needs is consistent with respect for autonomy as long as we see autonomy as a characteristic of social selves, whose connections with others and place in a socio-historical network of similarly marked others are deeply part of their self-conceptions. Autonomy should not be seen as operating in resistance to the acceptance and celebration of those connections. These reflections should be set against the backdrop of not only voluminous discussions of the concept of autonomy in recent years, but also broad skepticism and critique concerning the value of autonomous agency in Western political thought. The idea that the rational, self-aware, independent individual agent is the seat of social and political concern has been challenged from any number of directions, three of which it will be useful to mention here: Marxists who think that attention to the agency of individual actors in political contexts blinds us to the more fundamental issue of how material conditions and our place in the dynamic of class relations shape social structures and hence define the possibilities for social change; communitarians who argue that the kind of individual agency that is lionized in liberal theories of justice blinds us to the fundamentally social nature of thought and value; and feminists, critical race theorists, and others associated with a politics of identity who have stressed the overly narrow conception of the person and his interests that is operative in autonomy-based conceptions of social justice, a view that downplays the importance of social connection, intimate relations, the affective and emotional sides of our personalities, and other elements of social existence that have traditionally been associated with marginalized groups. I am profoundly aware of, and largely sympathetic with, a large part of these critiques, and indeed I hope to add to their force in what follows. Only if the interests and values that are highlighted in autonomycentric theories of politics reflect attention to these various concerns will they be worth defending. As I noted earlier, the idea of the social self that communitarians and others have worked to develop will be accepted and in fact fleshed out in these pages. The relevance of other markers of social identity such as gender or race will also be discussed relative to conceptions of autonomy, specifically in chapters 8 and 9. And as I mentioned, I emphasize that the context in which principles of justice must be

Introduction

13

articulated should be seen as rife with victimization and oppression operating specifically along lines of identity (as well as other dimensions of social existence). So what emerges here should surely be seen as the work of a fellow traveler in the furtherance of these critiques. One particular, and particularly fine-grained, line of resistance to autonomy-based models, however, bears special mention. This is the view that traditional conceptions of the free, autonomous individual put undue emphasis on the ideal/assumption of rational self-awareness and cognitive mental operations and have ignored the importance of embodiment, affect, and instinctive (and socially embedded) action. I have already mentioned how I will take on this critique by discussing both the social constitution of selves and the unchosen and unchangeable aspects of our social existence. But it should be emphasized that the model of autonomy worked out here does put self-reflection at the center of its conditions. As I argue later in the book, requiring self-reflection, even the hypothetical reflection demanded in the model I defend, is mandated by the social dynamics of interpersonal communication in which autonomous agency plays a key role. Only when we view autonomous persons as able to reflectively appraise their commitments and conditions can their value commitments (connected to those factors) count as a reason for the rest of us in our interactions with them. As I will argue, interpersonal respect for the autonomy of the other – something required by democratic social forms – assumes that autonomy means reflective self-acceptance of some form. So despite the trenchant criticisms of traditional views in the “rationalist” tradition of political thought, I want to hold on to the requirement of self-reflection in the model of the autonomous self worked out here. I also argue that self-acceptance required for autonomy be read as nonalienation, in a special sense of alienation I describe. If to be autonomous required that we positively value each aspect of our motivational structure and conditions of action, only the supremely lucky and fulfilled among us would count as autonomous. The rest of us sorry sorts would have to make do with imperfections, ambivalences, and the acceptance of tragedy in our lives. In order to keep “autonomy” from simply being a label for the ideal life (which for reasons I talk about, we should resist), we should require of the autonomous person only that she accept herself in the minimal sense of not being acutely alienated from the basic elements in her motivational structure and life conditions. The emphasis, however, will be on the diachronic aspects of autonomous agency, how such a trait is developed and expressed over time and contingent on the particulars of personal history. As I argue, historical conditions for

14

The Politics of Persons

autonomy should also not instantiate narrow ideals of the good life, such as requiring that all of our desires actually result from reasoned reflection to be authentic (since most did not for most of us). Rather, the agent’s own perspective on adequate historical processes by which she came to be how she is should remain paramount in the determination of autonomy for that person, though that prominence will be circumscribed in a number of ways. The claim that the agent’s perspective should remain central in considerations of respect and justice in political contexts should be stressed. This is motivated by the idea that principles of justice built on considerations of autonomy must recognize the ineradicable pluralism of values, attitudes, identities, and philosophical perspectives that are found in complex contemporary societies. Acceptance of pluralism and difference of this sort will be contrasted with specific forms of “perfectionism,” the view that there are ideals of life and human flourishing which can be formulated and defended philosophically and which apply to individuals generally, whether or not they have chosen or accepted such values. Political perfectionism, then, is the claim that such objective values, whose validity does not turn on individual acceptance of them, should be used to formulate political principles for a just regime. The type of democratic justice to be explored here will contrast with this form of perfectionism. For these reasons, autonomy should be seen as a characteristic attainable by a wide array of types of people, whose style of living and thinking, as well as whose moral commitments and values, vary inordinately. Considerations of this sort have led many to build resistance to the outdated conception of the independent man (all gender pronouns intentional here) into the conception of autonomy itself, hence making autonomy “relational” in a special way. Specifically, many have rolled a critique of hierarchical and oppressive social relations into the conceptual conditions of autonomy itself, claiming that autonomy means, among other things, having social relations of a certain acceptable sort. However, in chapter 8 I will argue that, while I applaud the motivations behind such a move, and indeed am similarly guided by them in my earlier arguments about the social self, I want to resist the temptation of making autonomy a relational property itself. Doing so, I argue, often drifts over into an overly idealized conception of the autonomous person which, for reasons connected to recognition of difference and pluralism, I suggest we should avoid. So autonomy is described here as a characteristic of socio-historical selves whose identities and life conditions are thoroughly entwined in social networks and connections. Yet, the ability to reflectively reject aspects of

Introduction

15

the self related to or defined by those networks, when we are alienated from them, will remain central to autonomy. Seen in this way, autonomy can function as the locus of interpersonal respect and reciprocity that principles of justice for pluralistic democracies are built upon. But those principles should themselves be seen, I will urge, as diachronic and historical, both in terms of the social context to which they apply and in their own structure. iv. justice We have, then, arrived at the point where autonomy has been characterized and applied to socio-historical selves. This was done in order that autonomy might be utilized in normative political principles for modern democracies. But why autonomy? Why should the conception of the self modeled in theories of democratic justice see autonomy as the chief characteristic to be recognized and promoted in citizens? The argument for this will be complex but generally follow these lines: rejecting perfectionism means, in part, viewing social values that guide our lives as valid only if acceptable to those governed by them. That is, since we don’t accept the view that there are objective ideals that should guide citizens’ lives independent of their acceptance of them, we must focus on the process of acceptance (or non-rejection) that grounds that validity. Only if that acceptance arises out of authentic, competent judgments about our social and personal values can the outcomes which flow from those judgments be properly binding on others. Let me put what is essentially the same point a different way: since selves are socially constituted and values are defined in terms of interaction with others, our abilities to pursue valued ends are inflected by the social dynamics in which we engage in those pursuits. Political structures and other institutions of power shape and codify those dynamics in broad and powerful ways. That power structure is justified only if it can be seen as flowing from (or at least in tune with) our own judgments, our perspectives about what is valuable to pursue given the fact that we live among people with contrasting values and who (like us) are products of the contingencies of history, both their own and society’s. Further, only if the structures that shape our lives are seen as endorsable by the independent, authentic judgments of ourselves and our fellow citizens can we live under them (and define our values by them) and remain free and equal citizens. Therefore, only if the principles that guide and shape (and justify) those power structures rest on a model of the citizen as autonomous – able to competently, independently, and authentically judge values for ourselves – will they gain the legitimacy that just institutions require.

16

The Politics of Persons

This is a complex argument, and it will be worked out further in chapter 10, but it also should be familiar. It echoes Rawls’s political liberalism, for example. But I hope to add key elements to this picture that might make it more attractive (though as I said, I cannot hope to defend a full theory of justice in these pages). To see this, let us take the discussion further into issues of democratic theory. Self-government expresses an individual capacity or ideal as well as a political one. Democracy, under some conceptions, expresses the idea of collective self-government analogous to the way autonomy expresses this idea at the individual level. It makes sense, then, to include reflections on the way that seeing the person as represented in political principles as an autonomous person relates to democratic principles generally. As I said earlier, philosophical reflections about justice in the past few decades, in the Anglo-American tradition at least, have centered on the project of liberalism. With the development of Rawls’s political liberalism, a view which evolved in the last 15 years or so of Rawls’s life, a protracted debate has gone on about the degree to which justice involves dependence upon objectively determined moral values or purely proceduralist (and in that way purportedly “neutral”) political ideas. That debate can be seen as involving positions that range from full-blown perfectionist defenses of liberal justice (Raz 1986; Wall 1998) to political liberalism (Rawls) to more thoroughly proceduralist defenses (Gray 1993; Habermas 1996a; Larmore 1996) to radical democratic (sometimes called “agonistic”) theory (Connolly 1991; Laclau and Mouffe 1985). These discussions raise complex and fundamental questions of political philosophy in that they involve asking about the possibility of justifying moral values that apply to radically pluralistic populations, with not only deeply different life experiences and histories but also passionately held political positions. I cannot sort through, much less take a responsible position on, these complex fundamental questions. My more meager enterprise here is to spell out some themes that are crucial for a political, proceduralist (and hence anti-perfectionist) democratic view. More specifically, I will try to show how the conception of the autonomous person specified in part II can provide a model for a proceduralist, democratic conception of justice that rests on a healthy skepticism about the possibility of establishing objective, philosophical grounds for substantive moral values that could be used to justify political power over modern pluralistic populations. I will suggest, on the one hand, that such a conception of democratic justice must be seen as an ongoing project rather than a settled set of institutions, and that such institutions must be understood to function in a far from perfect

Introduction

17

world. In these ways the attention to history and social embeddedness stressed earlier will reappear. A frequent criticism of theories of democracy that rest on the model of collective self-government is that large modern states cannot accommodate, nor do they ever actually involve, direct citizen control over government action. The idea that state policy is an extension of the wills of citizens aggregated and collectivized assumes a causal relation between citizens’ will and government policy, but no such causal connection can realistically be established for complex societies. Most members of such societies make no effort and have no real opportunity to directly affect state policies, so the idea of self-government in this causal sense is at best a metaphor and at worst a dangerous illusion. But in the model of individual autonomy worked out here, it is not selfgeneration that is assumed to operate at the center of self-government. Rather the core idea is that reflective persons can reject commitments, values, or conditions when they are alienated from them; this by no means assumes or implies that they, or their reflective selves, are the causal source of those factors. In the same way, collective self-government need not imply citizen generation of social policy, only opportunity for citizen correction. In this way, this objection to autonomy-based conceptions of democracy can be sidestepped, or so I shall argue. In a related manner, overall political legitimacy need not be seen as complete collective acceptance of the constitutional fundamentals of the political structure, as some have assumed Rawls’s political liberalism demands. In response to criticisms of that view I discuss, in chapter 10, an alternative model of political legitimacy. Such criticisms are raised by those who argue that citizens who are committed to comprehensive moral conceptions cannot be expected to compromise those commitments in favor of political loyalties and ideals of reciprocity at the center of liberal justice. I suggest, however, that legitimacy need only be seen to demand the reasonable expectation that ongoing democratic processes allow comprehensive doctrines to get a proper hearing, not that they prevail in the current constitution. This proceduralist response to the challenge to Rawlsian political liberalism mirrors the historicist approach to autonomy worked out earlier. In both cases, the fundamental commitment is to an ongoing process of reflection and review, not to particular values and commitments functioning at the foundation of principles of justice. In traditional approaches to the concept of justice in the liberal tradition, certain background assumptions about the social conditions to which principles are meant to apply are consistently made. The so-called

18

The Politics of Persons

“circumstances of justice” are laid out in this tradition, including moderate scarcity of goods and limited benevolence of citizens. Critics of this approach, most notably from the tradition of Marxism, critical theory, and, in a different way, radical democracy, have argued that such a view is insensitive to the historical and ongoing injustice that characterizes the current state of societies, and thereby effectively supports the power hierarchies at the root of that injustice. A set of rules for a race that assumes an even starting line is inappropriate when applied to one with staggered blocks. Feminists in particular have made a powerful case that traditional principles of justice – such as the view that “equality” means treating like cases alike – are distorting and inadequate when applied to current conditions of patriarchy and domination (see, e.g., MacKinnon 1989). When women, for example, are consistently the victims of violence, domination, and discrimination because of their sex, principles that are blind to those entrenched power inequalities will be inadequate. Generalizing from such observations, I want to press the conception of democratic justice as extended in time and space in that principles must be understood to operate over time (and hence respond to changing conditions as well as express sensitivity to historical facts) and in a context of ongoing violence and domination. Any work of this sort is more of a patchwork than a fine cloth. I have no pretensions about spinning out a broad and seamless fabric covering the topics of self, autonomy, and justice. I hope merely to say some fruitful things while making my way through this vast and multi-faceted terrain. The final hope is only that what is said will aid in the collective construction of socially negotiated visions of ourselves that we collectively constitute but also constantly revise and reconsider. There are stories woven in the fabric of our lives, but they are stitched by several hands and in constant need of mending. Ideals of justice and democracy might, if we are lucky, help us to organize that process.

part i

Selves

chapter 2

The social conception of the self: a critical taxonomy

After several decades of discussion of identity politics, multiculturalism, and communitarian challenges to liberalism, it has become a commonplace to acknowledge that citizens harbor identities and formulate values based in part on their membership in larger social groups. Whatever particular account is given of the relationship between social identities and principles of justice, it is everywhere admitted that a conception of the citizen as entirely unconnected to social practices and categories, cultural traditions, and other marks of identity will be insufficient to gain legitimacy in the contemporary context. The age of classic individualism, at least in its unwashed form, is behind us. The communitarian critique of liberalism discussed so protractedly in the 1980s and 1990s has now morphed into the challenge posed by identity-based movements and cultural claims, which has caused liberal models of autonomy to become more sensitive to this social conception of the self.1 So, this is admittedly well-trod ground. But the conception of autonomy defined and defended in the chapters to follow will need to be set against the backdrop of the many dimensions of the social and historical “self” that can plausibly be postulated. The standard liberal responses to the challenges built on such postulations have not adequately surveyed the many different ways that selves can be considered “social.” This has caused more than a little cross-talk among liberal apologists and their anti-liberal critics. But moreover, in chapters to come we will see how very much richer this socio-historical perspective about persons can be made while still defending a basically individualist account of autonomy, one that will be of use to a conception of justice as it applies to the legitimating principles for democratic politics. 1

Will Kymlicka laid out the most detailed version of the liberal response to the challenge of a “social self thesis” (Kymlicka 1989, ch. 5). The social self thesis he examined was defended in this form by Charles Taylor (see Taylor 1979, 1989b, ch. 2, 1991).

21

22

The Politics of Persons

We will proceed with our examination of the claim that selves are socially constituted by looking, first, at this view as a metaphysical claim about selves as such; we will then consider it as a contingent proposition, specifically psychological views about cognition, emotion, and motivation; and finally we will turn to the view that selves are social in relation to their value commitments. Along the way we will also examine arguments about the connection between social relations and emotions, as well as the social elements connected to modes of embodiment. The major goal of this chapter is to distinguish in new ways how the claim that we are social beings can be understood, and to do so in a way that clarifies the challenge such a thesis mounts for an individualized conception of autonomy. Of signal importance will be the manner in which autonomy requires that individual persons critically reflect on central aspects of their selves and conditions from a position somehow independent of those factors. As we will discover in later chapters, questions about that process are more complex than is usually thought. The main focus here, though, is the ways that “selves” can be said to be constituted by relations external to the individual person and whether any plausible versions of that claim rule out or complicate those individualized conceptions of autonomy that require critical self-reflection. My conclusion will be that they do not force revisions or complications that a plausible conception of autonomy along the lines worked out in part II cannot accommodate.

i. the social self as a metaphysical thesis The first version of the social self thesis is, in many ways, the most literal. The claim is that, from an ontological or metaphysical point of view, persons are constituted by external, interpersonal, or social factors.2 The person is essentially who or what she is in virtue of those external relations. To understand this claim, however, we must first distinguish between a diachronic and a synchronic sense in which selves are said to be socially constituted. We must be sensitive, that is, to whether we are talking about the relational aspects of human development or relational aspects of a person’s current ontological status. Even operating on this ontological level, it is important that we separate the question of who a person is from that of what she is. To say a person as such is relationally constituted (the latter question) is to make claims 2

Here “factors” is a severe truncation for any of the following: historical traditions, socializing influences, socially defined habits or bodily dispositions, practices, etc.

The social conception of the self

23

about persons as entities. To claim that a particular person is constituted by social relations (answering the “who” question) is to advance a view about her identity. In both these cases the plausibility of the position will turn on the specific relations that are alleged to constitute the person, the varieties of which we will discuss as we proceed. In either case, to say that something or someone is essentially constituted by its relation to something else is to claim that without it, the thing loses its identity – the person in question would cease to be.3 The test of the plausibility of any such claim, then, will be expressed in counterfactual conditionals of the form: “if factor X were absent or changed, person P would cease to be the same entity” (ceasing either to be the same person or to be a person at all). The challenge this might pose to individualist theories of autonomy and justice will turn on the relation between such constitutive factors and the possibility of reflection and self-appraisal. As we mentioned, classic liberal accounts of the self can accept that the self is social in these ways as long as such constitutive factors can be brought to consciousness, critically evaluated, and be accepted or rejected by the person herself (Kymlicka 1989, 51–52). But if crucial components of the self are external relative to the reflective scope of self-appraisal, then introspection will serve as a misleading or distorting anchor for the validity of social principles. One way for this challenge to be expressed is that whatever the constitutive elements of selves turn out to be, whether they are relational and social for example, they are not graspable by internal reflection or cognitive appraisal. Factors of this sort will not amount to mere beliefs that can be given propositional form and representational content. They are ways of being, whose character is picked out by how they are enacted rather than how they are represented to ourselves.4 They are things such as skills, 3

4

The metaphysical claim that personal identity is fundamentally social can be further divided between a claim of identification per se and mere characterization. For discussion of this distinction, see Schechtman 1996. Criteria of metaphysical identity over time refer to what Schechtman calls the “reidentification” question. The “characterization” question, however, refers to criteria for identifying those beliefs, desires, values (and perhaps relations with others) that, in general, make the person who she is at a given time, without which she would suffer an “identity crisis” but would still, metaphysically, be the entity she was before. The characterization sense of the social self thesis will be considered below, however, as an example of a contingent identity claim, since it does not presuppose that the conditions specified are metaphysically necessary for the continued existence of the person. These claims are echoed in Heidegger’s view that individual persons must be seen as fully contextualized entities who are intertwined in their own socialization, surrounded by social practices that form the background of their activity and understanding, and marked by incomplete selfinterpretations. This view is developed in Being and Time; for discussion, see Dreyfuss 1991, and Bell 1993, 27–54.

24

The Politics of Persons

habits (or “habitus” in Bourdieu’s technical sense), propriocentric factors that shape experience, and the like – in short, ways of being in the world rather than beliefs about it. If this is plausible, then the piecemeal re-evaluation envisaged in the liberal reply would not readily be possible. (In the next chapter we will consider the non-propositional nature of these factors in more detail.) The crux of this position as it affects our concerns here is that essential relational elements of selves are either impervious to critical introspection, or even if we can consider them from some point of view not defined by them, we will misrepresent them in some way and/or be unable to alter them. For example, if the language in which we express our selfconceptions is seen as somehow constitutive of the person or kind of person we are, and introspection always utilizes, but can never fully consider, language as something external to us, then self-constituting relations of this sort will undercut the standard view of critical introspection. I will return to this shortly. As I mentioned, one way to press this view is to stress that many relational elements of identity are not amenable to expression in propositional form, as with beliefs or explicit value commitments. Consider the manner in which elements of our identity might be captured only by poetic or metaphoric expressions, or ones that are fundamentally emotional in character, or involve relationships whose force is difficult to measure as well as to express. One of the most famous examples of this is Marcel Proust’s powerful evocation of the rush of memory (in Remembrance of Things Past) with a description of the taste of madeleine cake or the feeling of stepping on a paving stone. What was evoked could partially be captured in a truth-functional set of propositions but it was clearly much more, such as the sense of an association between being in love and smells and visions of one’s past. Or consider the sense one might have of the power of one’s devotion to a family member or the connection one has to the place where one grew up (picture re-entering the house of one’s childhood): what is the force one feels in the presence of these memories? How does one evaluate the commitment one feels to one’s mother or father, say, while weighing that commitment against some competing considerations? In considering whether to embrace or resist the force of values associated with such connections, the question one asks is not whether the values are “valid” or “true” but rather whether they make sense, whether they cohere into an acceptable autobiographical narrative. The position at issue, then, is that certain external relations and elements that are central to the self cannot be adequately evaluated by internal

The social conception of the self

25

appraisals because they escape the cognitive grasp of our acts of introspection; and this is more than a psychological contingency but part of what makes persons what/who they are. However, only if the theorist defending such a position can claim with confidence that there is an ontologically “true” component of the self can she proffer the view that a subject’s own reflection on that component fails to capture it “accurately.” And we will see below that there are few reasons to believe that such a true self can be identified and few theorists who want to commit themselves to its existence. The other strategy for claiming that self-reflections will be systematically distorting rests on the view that such reflections are themselves mere outgrowths of the very background forces and relational elements that are being evaluated in the first place. What Charles Taylor describes as the horizons of significance, the “given” which structures all substantive judgments and thought, are not merely objects of our consideration but mold and structure the acts of consideration themselves.5 This amounts to the worry that self-reflection provides no reliable guide to the authentic self since such acts are just as embedded in the social milieu they are analyzing as are the first-order commitments being evaluated (and so further levels of reflection relating to those second-order appraisals are needed to appraise those connections, etc.). However, defenders of individualism can clearly accept the general ontological thesis that human beings exist in a world, either as naturalistic, biological organisms or some more abstractly conceived being-in-the-world. That is distinct from saying that individuals are identified as related to any particular and fixed aspect of the world or of history, such as their nation, the traditions of their culture, or their community. The corresponding counterfactuals that test the plausibility of these latter kinds of claims have notably more bite: “if she left her country she would cease to be the same person really,” “since her religious conversion I don’t recognize her anymore,” and the like. But, obviously, the more localized conception of the social self must also account for the fact of personal change, even radical change. The very charge of insensitivity to the varieties of human experience made against liberalism comes back to haunt its critics at this juncture. For just as it is overly narrow to assume a univocal profile of the rational, reflective self in political theories, by the same token, we cannot ignore or forget the very real mobility, in the deepest sense, of which many are capable. We need not regard such change as desirable, frequent, or characteristic of many or 5

See Taylor 1979, 157; for discussion of this issue see Kymlicka 1989, 51.

26

The Politics of Persons

most lives. All that need be the case is that some individuals move from one social milieu to another, including movement which involves fundamental alteration of habits, values, and patterns of thought, in ways that do not destroy their identity as persons. One response to such cases is to revert to the diachronic sense of social embeddedness, and to claim that although many leave their pasts behind, they nevertheless bear eternally the marks of that heritage; they will forever be “former-” whatever (a lapsed Catholic is still fundamentally different from a person who has never been a Catholic). Either this can be taken in a psychological sense, that we are never able to shed the influences of earlier commitments, or in a stronger, ontological sense, that we must always be identified as the person who once was … this or that (see Bell 1993). In the latter sense alone, however, this cannot be raised as a challenge to liberal individualism, since rational reflection and self-revision is in no way ruled out by it. Only if some version of the psychological claim is true, indeed a robust version of it according to which rational reflection will be inescapably shaped by those earlier commitments, will individualist autonomy be ruled out as a meaningful possibility.6 Moreover, for every element that the (ontological) social self theorist lists as constitutive of the person’s identity, that element’s very articulation on such a list makes it a candidate for critical reflection (“oh I see now, I am the way I am because of my Catholicism”). But once such an element is articulated, it is at least possible (metaphysically possible) to ask the further question: what if I wanted to question or change that relation? But once I raise the question I am leaving open the possibility that I (and not some different ancestor of my present self) would survive that act of questioning and that change, should it occur. Of course this is not to say that such questioning or change is psychologically possible or desirable for any given individual, but merely that it is metaphysically possible, for that implies that further argument is needed (of a psychological or normative sort) to conclude that the ability to engage in individualized self-reflection and possible self-alteration is a problematic assumption about persons. The apparently seamless move from a claim about the overall and general embeddedness of selves to the more “local” specification of social identities is often made through an analysis of language. Since language is a social 6

In Chapter 4 we will continue this discussion by looking at narrative views of the self, where internal self-interpretation is still required for the disparate elements that make up the self to cohere. For discussion, see Flanagan 1996, 142–170.

The social conception of the self

27

construction which shapes cognition and thought, and particular natural languages play this role in unique ways, then selves are structured by such local social factors. This approach has been put in general form by Charles Taylor: The general feature of human life that I want to evoke is its fundamentally dialogical character. We become full human agents, capable of understanding ourselves, and hence of defining an identity through our acquisition of rich human languages of expression. For purposes of this discussion, I want to take “language” in the broad sense, covering not only the words we speak but also other modes of expression whereby we define ourselves, including the “languages” of art, of gesture, of love, and the like. But we are inducted into these exchanges with others. No one acquires the languages needed for self-definition on their own. We are introduced to them through exchanges with others who matter to us – what George Herbert Mead called “significant others.” (Taylor 1991, 32–33)

Taylor mentions Mead, but the pedigree for this way of thinking includes a host of major philosophers of the nineteenth and twentieth centuries from Hegel to Heidegger and Gadamer (and more recently Bourdieu). At its most general level the thesis connecting a social conception of the person with language can be stated as follows: thought is shaped by language; language is a social artifact, both in the sense that meaning is fixed only through social patterns and the history of use but also in the sense that meaning is a function of ongoing interaction with others; therefore thought, and hence self-reflection and judgment, is fundamentally social in its constitution. I will return in a moment to the first premise. Concerning the social nature of meaning, that view emerges most clearly (though certainly not originally) with Wittgenstein, who argues that pure ostension using atomistic semantic elements could not succeed in fixing reference without a host of shared conventions among language users. Indeed, words alone cannot bear meanings, but must function within sentences which themselves gain their meaning through a rich, and socially anchored, language game (Wittgenstein 1958; for discussion see Cuypers 2001, 138–139 and Taylor 1991, 74–78). The argument, in much the same form, has been restated by Richard Rorty.7 But there is no real controversy at this level, for no one believes that agents individually invent, out of whole cloth, a set of linguistic structures with which they can project meanings onto the world. Elements of our 7

See Rorty 1989. For a discussion of this position, see Hekman 1995, chs. 3–4. On the linguistic components of the self-concept, see also Vygotsky 1930/1978; cf. also Bruner 1983a, 1986.

28

The Politics of Persons

world enter our consciousness with tags on them, so to speak, or at least can gain their name tags because of an elaborate social practice that fixes the procedures of naming. No, the controversial aspects of the above argument will lie, first, with the first premise, then with the correct interpretation of the conclusion. The controversial idea driving this argument is the view that thought is nothing but a system of semantic elements, a system of signs. Consciousness amounts to the operation of this grammatically structured sign system and is given shape and structure by linguistic matrices.8 Such a claim about the language of thought is controversial in that it assumes that thinking cannot take place except in the form given by semantically structured systems. Views of this sort are complex and subtle and cannot be fully explicated here. But we must again distinguish global, abstract relational models from more local mechanisms connecting self-definition with particular language structures. If the tack we are taking is the general Wittgensteinian one that thought reflects a publicly constructed semantics, then no model of liberal autonomy presently on offer will be in jeopardy. The reasons will parallel the liberal response to the general metaphysical claim discussed above concerning being-in-the-world. But if we are making the more localized claim that cognitive mechanisms reflect particular languages, even specific culturally localized dialect patterns, we then have a more challenging claim against those models of autonomy that assume abilities to abstract from such concrete connections during self-reflection. However, it is also important not to reify particular languages as if they constitute, in a hard-wired fashion, thought processes themselves. Remember, language patterns are malleable and subject to negotiation and, to a large degree, translation. We should not forget the linguist’s lesson that “a language is just a dialect with an army.” We should look, for example, to the social dynamics of multi-linguistic communities where maintaining bi-lingual abilities involves holding on to specific linguistically grounded concepts and values while navigating among the complexities of alternative semantic structures.9 The fact that some concepts are only or best expressed in one language does not forestall the possibility of considering that concept from a location within that language or from the perspective of another closely related one. That is, even if all second-order 8 9

The starkest version of this approach can be found in the semiotics of Peirce (at least as some interpret him). See, e.g., Colapietro 1989. For discussion of a similar point, see Bell 1993, 158–167. For discussion, see Kymlicka 1995, 89–93. Also, see the description of “mestizo conciousness” where multi-lateral and conflicting sources of identity and value must be negotiated for those living in “border” communities (in every sense of that word), found in the writings of Gloria Anzaldúa (1987).

The social conception of the self

29

reflection was somehow structured and/or constituted by (socially structured) language, it is nevertheless possible to reflect on any one instance of such structures in a piecemeal manner. Here again we find a limited sensitivity to the varieties of human experience and styles of self-identification. For while instances of linguistically determined thought can be located, a general thesis about how socially structured linguistic conventions deterministically shape all thought ignores those life histories that have successfully negotiated across linguistic boundaries and maintain an identity that trades one kind of language game for another. The ability to identify and resist those styles of judgment embedded in a language need not be universal or even ubiquitous to be worth protecting when needed. Liberation often takes the form of uncovering the hidden constraints of linguistic practices – seeing, for example, how use of the generic “he” masks any number of power hierarchies – in a way that presupposes the ability to turn an eye inward to those linguistic practices themselves. What defenders of autonomy-based theories must maintain, however, is the position that this ability can be described generally enough so that its promotion or protection does not marginalize populations who do not organize their sense of self in this manner. (This will be the specific subject of chapter 9 below.) But saying all this is not to imply that self-alteration in these ways is always possible. Stepping outside a linguistically structured matrix of value and meaning in order to consider critically one’s commitment to that matrix, might be possible for some in some ways. And for this reason, the strong social self thesis that implies this is never possible should be rejected. But this does not imply the corresponding view – defended by many liberals – that such reflective separation is always possible and hence should be venerated in the design and justification of social institutions. The middle ground between these two positions is the theoretical location I want to defend in what follows. This means, though, that we have now moved some distance away from the pure metaphysics of the social self view and have proceeded to the area of pure contingency. In the next section, then, we will consider those versions of the social self thesis that have an unapologetically contingent character, based on theories from psychology and the other social sciences.10 The views to be considered now do not boast of ontological necessity – people surely could have been otherwise – but rather trade in the currency of actual fact. 10

For an excellent discussion of this issue using social scientific material (especially regarding developmental psychology) see Crittenden 1992.

30

The Politics of Persons ii. the social self thesis and human psychology

Leaving aside the view that human selves are essentially social in their defining conditions, we can now turn to the related but different idea that, as a matter of fact (many, most, or all) people are, to varying degrees, deeply connected to external factors in their self-identifications, cognitive structures, values, and the like. That is, as a matter of contingent fact, selves are social. Now, as the list in the earlier sentence shows, there are several possibilities to be considered here. Indeed, psychologists generally would blanch at the idea that there is a particular self for every person that may or may not be “social” to some degree. There are, most would say, various self-systems that predominate in various contexts, such as the knowing self, the remembered (and remembering) self, the feeling self, the social self, and so on.11 To consider the ways in which selves are to some degree “social,” then, we have to take things a step at a time. First, we could say that human beings experience themselves only in relation to certain others (symbolically or actually); second the cognitive processes (including the use of language) by which people deliberate and choose are definable only interpersonally; third, the structure of human emotion and affect contains a crucial interpersonal component; finally, our bodies, which are part of our selves, are defined in irreducibly social ways. All have received attention in the literature, and as we will see, each of these possibilities has different ramifications for the prospects of autonomy-based approaches to justice. Self-understanding and self-concept To have a self-concept is to have an experience of oneself as the object of reflection as well as to have/be a mechanism of agency and thought. Approaches to self-conception in the psychological literature of relevance here can be usefully traced to the work of William James, who distinguished between the “I-self” (the mechanism of subjectivity and agency) and the “Me-self” (the set of characteristics that are available to self-reflection and external observation by others); he referred to these as the “self-as-knower” and the “self-as-known.”12 11 12

See, for example, Neisser 1988 and the essays in Ashmore and Jussim 1997. The Me-self is composed further of the material self (one’s body and possessions), the social self (characteristics recognized by others), and the spiritual self (one’s mental life). These are arranged, James claimed, in a hierarchical matrix, with the spiritual self at the highest level. See James 1890. For

The social conception of the self

31

Theories of the self that emphasized the social constitution of persons gained further prominence in the work of George Herbert Mead, C. H. Cooley and the “symbolic interactionist” school of sociology and psychology.13 On these views, the self develops through a process of reaction to and internalization of others’ actions toward oneself, giving rise to what Cooley called the “looking glass self.” Mead summarized the view this way: “We appear as selves in our conduct insofar as we ourselves take the attitude that others take toward us. We take the role of what may be called the ‘generalized’ other. And in doing this we appear as social objects, as selves.”14 These processes of internalization are mediated in a crucial way, on these models, by language and the process of developing linguistic skills. Much work on the psychology of the self, however, has centered on the “Me-self” – the self as known. Also, much of this work in social psychology equates self-conception with “self-appraisal” and measures the ways in which individuals rate themselves and come to conclusions about their own relative abilities in different contexts. These investigations bear on questions of self-esteem, which might be tied to questions of autonomy but in the present context can be sidelined.15 Of particular relevance here is work done in connection with people’s “self-schemata,” which are cognitive generalizations about the self, derived from past experience, that organize and guide the processing of self-related information contained in the individual’s social experience.16 The social and interpersonal elements in self-schemata are particularly underscored in work investigating the comparative degree to which members of various groups utilize relational concepts in their schema for self. It has been suggested, for example, that (some) women and members of some east Asian cultures have a more highly relational sense of self as measured in this manner. That is, subjects from such groups pick out

13 14 15

16

discussion of the historical development of theories of the self see Susan Harter (“Historical Roots of Contemporary Issues Involving Self-Concept,” in Bracken 1996, 1–38; see also Gergen 1971, 1–12. For a parallel discussion that refers to Wittgenstein’s ideas of the self, see Cuypers 2001, 135–137. Also, I will discuss later (in Chapters 4 and 7) David Velleman’s view of the self that relates to this distinction: see Velleman 2004. See, e.g., Mead 1934. For discussion of Mead and Cooley, see Burkitt 1991, 28–54. Quoted in Susan Harter, “Historical Roots of Contemporary Issues Involving Self-Concept,” in Bracken 1996, 4. For an overview of recent theories of self-concept, see Ross 1992. Another line of inquiry examines the “social self-concept” in which self-concept and self-esteem is measured in social settings (or in reaction to interpersonal interaction). See, e.g., Thomas J. Berndt and Leah Burgy, “Social Self-Concept” in Bracken 1996, 171–209. Hazel Markus, “The Self in Thought and Memory,” in Markus and Wegner 1980, 64. Various models of the self-schema have been developed, from multi-dimensional models to hierarchical ones. For an overview, see Ross 1992, 1–49.

32

The Politics of Persons

terms which refer to relationships (mother, friend, business manager) to a greater degree than alternative subject pools (Western males, for example), indicating that such relational concepts occupy a place in the self-schema which is closer to the “core” for such people.17 Such work is varied and multi-faceted. However, it may be premature to conclude from it that members of particular groups have a sense of self which is socially constituted in a sense which is incompatible with a capacity for autonomy. For granting that subjects often identify themselves as occupying a certain interpersonal role or as holding a place in a relationship, it is nevertheless possible that such subjects may often retain the capacity to reflect on those very relationships when called upon to do so. More importantly, the methods used to construct a subject’s self-schema, in the psychological context, are not sufficiently sensitive to the difference between essential, identity-creating, aspects of the person and merely deep and important personal characteristics. We must be aware of the distinction between, in Gordon Allport’s words, “what are matters of importance to the individual and what are … merely matters of fact to him [or her]” (Allport 1968a, 26). Also, such investigations must be sensitive to the diachronic and dynamic nature of self-identity. The idea of an internalized “other” comprising a component of the self-concept leaves quite indeterminate where the focus of such an other lies: in the caregiver, one’s childhood playmates, one’s current intimates, the members of one’s gender, race, class, etc.18 Such indeterminacy leaves open the (psychological) possibility of shifting identifications, where reflective questioning of particular connections comes about while moving on to other relations. This is not to say that reflective self-appraisal will be a universal capacity or a universal value. I am suggesting only the cautious conclusion that even for the thoroughly social self-schemata discussed in this literature, the capacity for piecemeal self-reflection has not been ruled out. That is, while models of the self-concept show how social relations and factors figure importantly in the construction and ongoing operation of such structures, it remains unclear whether these factors are such that they cannot be brought under reflective scrutiny by the agents themselves. We will return to this point. For now, let us press on in our investigation of other social components of selves. 17

18

See Markus and Kitayama 1991, 224–253. These experimental results correspond roughly to conclusions drawn by anthropologists such as Clifford Geertz who emphasize the manner in which various nonWestern cultures utilize essentially non-individualized concepts for self (Geertz 1973, ch. 4). For example, Neera Badhwar questions the common assumption among communitarian thinkers that the principal object of identification will be with the political aspects of one’s community (as communitarian politics assumes). See Badhwar 1996.

The social conception of the self

33

The social conception of cognition As I mentioned, the research just discussed on self-conception deals specifically with the self as the object of internal reflection – the “Me-self.” However, defenders of the social self thesis may well insist that the notion that we are constituted by our external relations has more to do with our subjectivity and agency, the “I-self.”19 Indeed, many anti-individualist critiques of autonomy posit that human thought – cognition and affect as well as memory, judgment and reflection – has a fundamentally interpersonal structure, for many or all people.20 One influential approach to modeling the mental posits that meaning and thought should not be understood as undertaken in or by a single mind (or embodied mind). Rather meanings are determined by way of constitutive relations to external phenomena, external to the functioning of a single thinking being. On this view, thoughts per se are possible only when a thinker is embedded in the spatio-temporal world in the right way. The concept “chair” is not simply inside the head of the person using and understanding the word; rather it has meaning only in relation to other aspects of the world. (The classic statement of this view can be found in Putnam, 1975; for an overview of the issues, see Lau 2002.) A “social” version of this thesis claims that meanings function only in relation to a system of language use which is socially ordered and dependent on patterns of activity and relationships in the person’s environment. Relatedly, some philosophers stress the way thought itself, the activity of reflection, cogitation, and decision, takes place in a social network where other people, institutions, spaces, and practices not only (instrumentally) 19

20

This distinction raises the question of whether the “self” is fundamentally a “thing” (whatever its metaphysical status) or a process. For an analysis of that question (which claims the question can be avoided rather than answered), see Gergen 1971, 1–12. One research program in psychology that directly posits a social model of cognition is Social Learning Theory (Bandura 1986). According to this view, human cognition is mediated through processes of symbolization and vicarious learning – called “modeling” – which operate by virtue of close interaction with certain others creating cognitive structures which embody those interactions and which form and guide thought (Bandura 1986, 48). Cognitive structures are molded, through modeling and vicarious learning, in a manner which then shapes and structures later cognitive events and hence behavior (Bandura 1982). However, social learning theorists also describe the agent’s capacity to utilize various cognitive and affective tools in order to reflect upon, regulate, and direct her own behavior. As Bandura puts it, “making choices is aided by reflective cognitive activity, through which self-influence is largely exercised. People exert some influence over what they do by the alternatives they consider, how they foresee and weigh the consequences, and how they appraise their capabilities to execute successfully the possibilities they are entertaining” (Bandura 1986, 39). Therefore, although a person’s perspective and values are shaped in a manner that involves ongoing interpersonal dynamics, the agent maintains the ability to evaluate and analyze the thoughts, values, dispositions, and the like that are the result of those connections, according to these theories.

34

The Politics of Persons

affect thought but help constitute it. David Zimmerman argues, in fact, that externalism about mental content of the sort described has direct implications for theories of autonomy, in that one cannot be a mental externalist and hold a thoroughly internalist conception of autonomy. If mental content requires that one be related in some way or other to objects and phenomena external to the self (in the present and the past) then it is conceptually impossible for a person without any past at all to be autonomous; and those who think historical relations are irrelevant to autonomy must, then, reject this kind of externalism.21 This is a complex issue that clearly bears on some issues outside of our concerns, but for now we need only say that externalism about mental content implies only that relations in general must hold to phenomena outside the person (currently or in the past), relations for example with the language-independent world or causal relations with the initial act of naming that fixes the reference of terms, and so on.22 Mental externalism merely states the general proposition that mental states (beliefs, for example) must be related to the world in certain ways for them to have the content they have (along with some very fine-grained specifications about the kinds of relations that must hold for various types of language use). It is neutral about whether having those thoughts (or being in whatever states those external relations are necessary for) is of any interest to the agent. As Zimmerman concludes in his analysis of this issue, if one holds to robust externalism about the mental, where having thoughts at all implies having certain relations to the world outside the individual’s head, one must reject wholesale internalism about practical reason where being a fully capable agent requires no particular relations with the external past or present at all. The view of autonomy we will sketch in later sections deviates significantly from such a view. Our only conclusion here, though, is that mental externalism of the sort that seems so influential in the philosophy of mind and language is acceptable for the kind of ecumenicalism about the self we are proposing in general and the historical conception of autonomy I will sketch in particular. Affect and emotion Another way in which it has been claimed that humans are fundamentally social concerns the structure of emotion or affect. Naomi Scheman claims, 21 22

See Zimmerman 1999. Mele claims the two questions are completely independent: Mele 1995, 147. These vaguely described examples are meant to point to various positions in the philosophy of language such as the causal theory of naming. For discussion, see Evans 1982 and Lau 2002.

The social conception of the self

35

for example, that certain affective states, and many mental states in general that partially define individuals, can only be specified in social terms. To be “in love” is to be in love “with someone,” she writes (Scheman 1983, 232).23 Since such mental states have as their object external phenomena, selves are essentially social. It is important to distinguish, however, the relational aspects of the states of our selves – our emotions and desires, for example – and the relational aspects of the self per se. Jean Grimshaw, for example, has criticized Scheman by pointing out that while emotions are intensional states and have others as their objects, a personality who loses the ability to distinguish the loved one’s personality from her own is suffering from a pathology (Grimshaw 1986, 162–186). It also must be remembered that emotions are sometimes transitory, especially when viewed from the point of view of an entire life. Although many people cherish their emotional connections with others to such an extent that they experience their identities as interconnected with them – their children or parents for example – such relationships are seldom life-long (our parents may die before we do and we exist prior to our children coming to be). And for many people, such connections shift at various points in the life cycle. This is not to argue that a life of shifting emotional connections, and hence, perhaps, shifting identities, is a superior or even minimally flourishing life; the claim here is that for many people, emotional connections do not form the essential components of their identity for the entirety of their lives. So while at some given time for some people emotional connections bespeak of relational identity, such interconnections are not an invariable fact about humans generally. (This issue is taken up again in chapter 8 under the rubric of relational views of autonomy.) Satya Mohanty comments on Scheman’s work as part of the development of a cognitive account of emotions.24 Mohanty’s claim relevant here is that [a] necessary part of [the emotion’s] form and shape is determined by the nonindividual social meanings that the theories and accounts supply. It would be false to say that this emotion is the individual’s own “inner” possession and that she alone has “privileged access” to its meaning or significance. Rather, our emotions provide evidence of the extent to which even our deepest personal experiences are socially constructed, mediated by visions and values that are “political” in nature, that refer outward to the world beyond the individual.25 23 25

Quoted in Grimshaw 1986, 167. See also Scheman 1993. Mohanty 2004, 394, quoting from Scheman 1983.

24

Cf. also Nussbaum 2003.

36

The Politics of Persons

That is to say, our emotions are what they are because of the meanings operative in the beliefs that constitute them, and such beliefs are socially structured. This combines a cognitive theory of emotions with an externalist theory of meaning to support the view that emotions of many sorts have the structure they have only as a function of the social structure in which they operate and hence cannot function simply at an individual level. Nevertheless, as we said, it is clear that having an emotion, like winning an award, might have validity only in relation to other social factors. But this does not gainsay that it is my (individual) emotion and my award. In later discussions we will consider and indeed accept the idea that the value and meaning of lives require ineliminable reference to significant others. By extension, it will be necessary to reject individualist models of the self that label close emotional ties as separable aspects of the personality that can simply be shed upon reflection, as if a parent’s connection to her children is something she could just abandon without great cost. What I have concluded here, however, is that general models of the self that make particular emotional ties to others an essential, constitutive element of the self go too far and are left unable to differentiate pathological connections between self and other and such close connections as just described. To continue our investigation, however, it will be necessary to examine views of the self that place our embodiment at the center of the analysis. In our overview so far, we have mirrored the tradition of thinking about the self in the major currents of Western philosophy by downplaying the role that our embodiment and physicality play in molding our selfconcept. Indeed, according to some theorists, seeing ourselves as fundamentally embodied creatures makes more apparent the ways in which we are socially constituted. The social body It may sound initially paradoxical to say that our embodiment marks us as social beings in a deep and significant way. After all, our physical individuation seems to manifest the separateness of our existence in a rather obvious manner. And in much analysis of the question of liberal autonomy, very little mention is made of our bodies, especially regarding the social nature of the self. Drawing particularly on the work of Merleau-Ponty, several writers have emphasized the way that the structures of human bodies – the relation between head, arms, legs, torso, front and back, and so on – fundamentally orients thought and mental life. Merleau-Ponty drew the connection

The social conception of the self

37

between the body’s movement forward in space and the constant sense of “I can” that structures practical reflection.26 The sense that we are always poised to act is part of our reflective experience of ourselves and as such shows how being embodied (and so able to move in space) structures our sense of self. Feminist writers such as Iris Young show how bodily comportment and movement reflect not only thought generally but particular modes of moving, sitting, and standing that reflect social expectations and relations (Young 1990c). In particular, Young examines modes of female bodily forms (in the context of Western society) that show the internalization of the terms of social self-conception. Young traces certain stereotypical feminine movements such as relatively closed posture, delicate arm movements, and hesitant motion that indicate a mode of self-presentation that mirrors a sense of self-protection and vulnerability. Such a sense of self, she argues, reflects an internalization of feminine roles and expectations. While some of her examples may be dated – even her title, “Throwing Like a Girl,” rankles when one considers female athletes – the general point remains: modes of thought and feeling are structured by bodily comportment, structure, and movement. Such movements are internalized habits reflective of social relations and expectations, such as men’s more open posture and women’s more closed and protective bearing; selves are, in part at least, embodied modes of being in the world, so selves are social in reflecting these relations and expectations. (An alternative conclusion is this: the self-concept reflects manners of thinking and feeling that are in turn shaped by our embodiment; insofar as the latter manifests internalized social factors, the self-concept will reflect the effects of those factors.) In these ways our bodies are marked as exemplifications of social categories that, in turn, shape our thought and reflection. What is important for our project here is not so much whether these accounts of the social nature of bodies are defensible in general or in detail – though I am certainly sympathetic to their broad outlines – but whether accepting them would force us to radically shift our understanding of the person from a political point of view. In particular, the question will be whether seeing embodied persons as socially constituted in these ways induces a change in the way we must think about the respect owed to persons, the 26

Merleau-Ponty 2002. More recently, Lakoff and Johnson have analyzed the metaphors that structure thought and their conceptual and phenomenological ties to the structure of bodies (“up” is used to mean “increase” for example) (Lakoff and Johnson 1999).

38

The Politics of Persons

protections afforded them, or the interests that political principles and institutions aim to protect on their behalf. I will delay answering those questions directly for now. But I do want to make some observations concerning the upshot of these analyses here. One can accept the view that both our self-concept and our mode of selfpresentation are socially formed in largely the ways alluded to in these views, but also emphasize the extreme variability of the effect and shape of the social dynamics in question. Indeed, there are several levels of variability that ought to be distinguished here. First, there is the question of whether all, most, or some people can be described in these specific ways at all, whether the models of bodily presentation or phenomenology can be applied consistently even to members of the identity groups being considered. The point of these theorists may well only be that since some (perhaps many or most) do reflect these models of the embodied social self and that traditional models of the political person neglect the importance of these factors, then traditional models are defective. That point is well taken, but another manner in which such models of the person admit of variation has other implications. That is, people can vary in the way they themselves want the social elements of their identity interpreted. People may well manifest, in their bodily type, comportment, movements, and modes of experience, the social dynamics of their world. But the question nevertheless arises whether the descriptions of them that attempt to capture those social elements can be given as general accounts of the self apart from their own expressions of that way of being. It is one thing to give a complex and subtle account of the way (say) my body reflects social expectations and classifications concerning being a male; but it is another that this is the way I will be represented in political principles (and policies arising from them) independent of my own articulation or interpretation of that social self. Moreover, the description of the social elements at work in the function and construction of the self leaves open the question of what must be done to protect, secure, or resist the operation of those social dynamics. As we will discuss in chapter 8, claiming that certain social relations, say, are constitutive of the self (or autonomy) and then concluding from that that such social relations must be protected in order to support a person and/or her autonomy, may well lead to unacceptable conclusions. In this context we can say this: analysis of the ways that social relations (such as those involved in speaking a certain language, participating in cultural practices, and so on) construct the self leaves open the question of whether support of those relations are necessary to respect or advance the interests of those selves.

The social conception of the self

39

Young’s analysis, for example, suggests specifically that we should not: understanding the source of women and men’s bodily comportment (and the role this plays in the self-concept) as the patriarchal social environment they live in may well imply that we should resist and reform that very environment. This completes a brief survey of the ways one could claim that a person’s psychological make-up is fundamentally interpersonal. I make no claims of comprehensiveness here, but I hope that even this selective overview has been sensitive to the wide array of contingent social self claims. However, I turn now to the view that it is at the level of values that the social self truly emerges.

iii. the social self and value commitments In chapter 6 we will engage in a detailed discussion of the response by liberal thinkers to the challenge of the social self thesis, but here let us set the stage by laying out some of the components and implications of that thesis as it concerns values and moral commitments. There are two importantly distinct lines of critique centering on liberalism and values. The first concerns the social constitution of the person and the individualistic values presupposed by autonomy-based conceptions of justice. The second concerns actual social consequences of liberal regimes, where it is claimed that liberal institutions have caused, and will continue to cause, social atomism, separation, and general malaise, due to liberalism’s individualist value structure.27 These two lines of thought are certainly related, as we will see, but conceptually they are quite different, and the kinds of response they call for will reflect those differences. Let me first comment on the second of these two challenges. According to sociological analyses of (more or less) liberal societies like the United States, the individualistic values that tend to dominate in the culture – the importance of individual rights to associate with whomever one wishes, including freedom to sever associations, the right to property, and freedom from general social and political interference with personal decisions – have had the effect of producing a highly mobile society, where communities have dwindled, family connections are less stable, and general social connectedness has become fluid (Putnam 2000). Most pointedly, the very conditions of meaningfulness that people would list to capture the “moral 27

For discussion see Kymlicka 1989, 61–62; see also Bellah, Madsen et al. 1985, Barber 1995, Beiner 1997, ch. 1, and Putnam 2000.

40

The Politics of Persons

coherence” of their lives – enjoying and cultivating relationships, engaging in long-standing and fulfilling familial and social roles, and the like – are diminished by the struggle to conform to standards of success set by the reigning public values of the liberal, individualist, society in which they live and work. As Bellah and his co-authors put it, “the freedom to be left alone is a freedom that implies being alone” (Bellah, Madsen et al. 1985, 23). If these observations are borne out (and I leave that question open here), for them to be parlayed into a critique of autonomy-based conceptions of justice, a connection must be established between the model of autonomy lying behind the protection of individual rights in the liberal society in question and the social effects enumerated. My own view is that what is more directly connected to the instability of communities, neighborhoods, marriages (or romantic partnerships), families, and the like are the economic pressures people face as they pursue professional or employment opportunities. The fact that moving 500 miles from my home is expected of me when I take a professional position, and the possibility of losing that position and having to repeat that move to even greater distances from my “home,” have the most direct impact on my ability to cultivate and protect communal relations which constitute what is valuable for my life. And this set of opportunities and pressures is related to the property system operative not only in my direct vicinity but in the global marketplace that (through governmental policies) sets the stage for those local factors to fall into place. But even if we accept these sociological speculations (and surely some may object to them), the connection between autonomy-based liberalism and the kinds of property systems that give rise to this pattern is, at best, tenuous. Indeed, all of the “left-liberals” writing today wholeheartedly decry those very property systems on the basis of liberal principles.28 But surely the charge that a public understanding of the value of individual autonomy may have a deleterious effect on valued relations with others must be taken seriously, even if it cannot be adequately adjudicated in a philosophical discussion. (It is, after all, a complex sociological issue.) What is a more troubling challenge, however, is the view that there is a direct connection between the protection of individual autonomy itself, independent of property rights connected to it, and the kinds of value commitments that are supported or suppressed in society. To establish such a connection, a conceptual bridge must be constructed between the metaphysical or psychological profile of persons presupposed 28

My own attempt to spell out such an egalitarian view can be found in Christman 1994.

The social conception of the self

41

in conceptions of autonomy operative in political principles and the kinds of values that are protected or, more importantly, ignored, in a regime living under them. That bridge is often built on the claim that since individual identities are interwoven in social relations, and value commitments form a central element in one’s identity, then a person’s value commitments are themselves so interwoven. Correspondingly, insofar as a political order protects only, or chiefly, values held by individuals in separation from social ties, or if that order encourages such separation, then that regime is subject to the charge of assuming narrow personality and value profiles in its citizens. But several crucial distinctions have to be made in order to wrestle with this challenge. First, we must separate claims about the nature of people’s values from those concerning their motivations. It is one thing to say that our deepest ideals concern, irreducibly, social relations, yet another to say that only if direct reference to those relations are made in the principles protecting or promoting our interests will we have a reason to see those principles as legitimate, or to obey them when they conflict with our other interests, or the like. Let us consider these separate issues in turn. Motivation An interesting line of argument relating to this issue, and one not often directly addressed, is that which relates a claim about constitutive social identities and motivations required to support and conform to the dictates of distributive justice. According to Sandel, for example, in distributive principles that demand that citizens sacrifice some of their goods either to maximize the public good in accordance with a utilitarian principle or to satisfy a more egalitarian requirement such as Rawls’s difference principle, the model of the citizen cannot assume a total separation between self and society. Persons who lack entirely any identification with the community of co-citizens feel used by state policies that ask them to sacrifice their own (separately identified) pursuits in the service of a social cause. If any such principle is to avoid simply using such persons as means, he writes, such redistribution “can only be possible under circumstances where the subject of possession is a ‘we’ rather than an ‘I’, which circumstances imply in turn the existence of a community in the constitutive sense” (Sandel 1982, 80). The claim such an argument depends upon is that identity with one’s co-citizens plays a fundamental motivational role in participating in social policies that demand some degree of sacrifice of one’s own individually defined good for the society as a whole.

42

The Politics of Persons

Now one mode of response to these sorts of claims is to analyze more deeply than Sandel does the structure of ownership to show that redistributive policies that require transfers of wealth from some to others (or to other social causes) need not violate what citizens see as “their own” in a deep, identity-related sense. The force of Sandel’s argument trades, for example, on the blurring of a distinction that I have argued elsewhere is central to the understanding of ownership, between “possessing” (and controlling) an asset and receiving income from it as a result of a marketoriented exchange. The latter is sometimes referred to as receiving the “fruits” of the asset, as if it is a natural extension of what I have in my possession, like the fruit on my tree whose possession and enjoyment is within my grasp. But this is to misunderstand the complex nature of both ownership and market exchanges (which themselves are only possible from within an established set of property rules); for these “fruits” are the result not only of whatever effort I might put forward in generating them (if effort on my part is involved at all) but the choices and opportunities faced by others in my surrounding area and the various rules of interaction and trade under which we live that together produce the price structure that generates my “income.” Rules that attempt to counteract such a phenomenon need not be described as taking “my” assets from me (see Christman 1993). But the more striking part of Sandel’s analysis is that constitutive identification with a community is required motivationally for the purpose of generating social policies that involve citizens’ sacrifice, which virtually all such policies do. That is, the relation between individual citizens and the community must be a “constitutive” one if distributive schemes of any sort are to have legitimacy, if they are to connect with actual people’s motivations. Now one might question whether a motivation to cooperate in social policies of this sort requires identification in the constitutive sense with one’s community. Motivation can be stable and general if participants share a common moral view, and that view has been “internalized” into their motivational system – if they are convinced (by tradition, upbringing, or simply good argument at the political level) of its justification. Having one’s identity constituted by the value system in question is one way for motivations to be established, but it is not the only way. One can recognize no deep identity with the overall value system that structures one’s community, in the deep self-constituting sense of identity, but nevertheless be consistently moved to respect its dictates. Nevertheless, this line of argument lays down a challenge to autonomybased conceptions of social justice: what is assumed by such conceptions about the ways that citizens relate to their society, identify with it, or

The social conception of the self

43

share its dominant values, and what does this assumption imply about the social elements of the identities of those citizens? More pointedly, can a conception of individual autonomy at the heart of political principles of this sort assume no social elements of identity if the regimes governed by those principles are to establish legitimacy? What we will see, especially in chapters 6, 9, and 10, is that conceptions of autonomy that apply to individuals without (conceptual) requirements for particular social relations can nevertheless accept that many or most citizens do identify with social elements of their life in a deep constitutive manner. Now, possibly a more powerful version of the motivation question arises from considering the relation between the conception of the person (and its alleged social constituents) and the value of autonomy itself. If selfdetermination is merely the abstract ability to reflect upon and revise any and all of our commitments, with no direct reference to any selfconstituting connections in the explication of the ability to so reflect, then it will be unclear how self-determination in this sense is a value that anyone would have reason to promote (see Kymlicka 1989, 47–52). What might be claimed (as liberal theorists such as Raz, Kymlicka, Wall, and others have done)29 is that enjoying autonomy as part of one’s pursuit of the good is itself an objective, fundamental value, an ideal forming part of a perfectionist conception of justice (one which is not meant to be neutral among rival conceptions). However, as we will discuss in greater detail below, autonomy as the ability to question and revise one’s commitments in general is simply not a universal value for all people. The special challenge for a value-neutral, proceduralist conception of autonomy (of the sort developed here) will be to make a connection between being autonomous in this sense and valuing autonomy to an extent required by the strictures of social justice. My view will be that participation in democratic processes itself presupposes and promotes valuing autonomy in this (procedural sense) without assuming fixed connections between selves and aspects of their social surroundings (but certainly allowing such possible connections). It is now clear, however, that we have slid from the issue of motivations to the question of values. Let us now continue the discussion in that language. Values A powerful way to connect more or less metaphysical claims about the embeddedness of existence and claims about value commitments is to 29

Kymlicka 1989; Raz 1986; Wall, 1998.

44

The Politics of Persons

argue that such contextualized persons would suffer a special sort of psychological damage were they unable to pursue the values to which their social identities give rise (or if their society’s principles did not recognize the contextualized nature of those values) (Bell 1993). This gives rise, then, to the more general form of the claim that social structures that stress only individualized commitment, and put great weight on the opportunities to sever such commitments (while giving relatively little weight to maintaining them) tend to produce the damaging alterations in life commitments just described. The stress on multiplying opportunities for change, which principles of individual rights and other autonomy-based norms provide, threatens this psychic damage even without actual changes of this sort. Sandel, for example, stresses the idea that identity-forming values always take root in the historical and cultural setting within which people develop. The social embeddedness of these values commitments make the self who is their subject essentially social, and implies that any conception of the self as an independent chooser of values is mistaken. He illustrates this claim with a description of the dilemma that Robert E. Lee faced just before the start of the US Civil War. Although Lee opposed secession, he realized that his (unchosen) obligation to his native state of Virginia outweighed his convictions about loyalty to the Union and the military. Sandel claims that we cannot make sense of Lee’s dilemma, nor of our sympathy with his conundrum, if we consider obligations as binding only insofar as they are chosen by an unencumbered self (Sandel 1996, 15–16). But an important thing to note in this story is that although Lee found various aspects of his situation and character to be binding, unchosen though they were, it was nevertheless the case that he felt compelled to reflect upon these factors and relations in order to come to a settled position. That he arrived at the view that his unchosen connections to his home state were of greater importance than his chosen obligations to the military (for example) only after due reflection on those ties indicates that competent reflection was a necessary condition for the authenticity of his decision. This is not to presuppose that this process of inner endorsement must occur according to universal and foundational canons of reason or the moral law. One can certainly assume that the reflective process, while perhaps structured by the minimal demands of rational cognition, operates by way of concepts and values that themselves are historically situated. As we will see, autonomy is not a transcendent value, so the reflection it presupposes need not be trans-historically grounded. But reflective consideration of

The social conception of the self

45

values is quite consistent with the view that such values are unchosen background conditions of our general ideals and perspectives.30 A crucial element in this picture is that the values that form the motivational foundation for those whose identities they help to form are abstract, fluid, and subject to evolving understandings concerning their particulars. In short, even if shared values constitute identities, those values are continually subject to interpretation. And two things are pivotal concerning those interpretations: they are always subject to review, and they are formed on the basis of discussion by participants who are themselves rational, authentic selves. As Daniel Bell puts it, the shared values that form the identities of co-citizens (of what Robert Bellah has labeled “communities of memory”) emerge from a “tradition.” A “tradition” as Bell defines it, “is based on the understanding that moral principles and virtuous exemplars from history must be interpreted to be applied, to be useful in particular situations with novel features … a historically extended, socially embedded argument about the good of the community whose identity it seeks to define” (Bell 1993, 126). But Bell and other communitarians must place conditions on both the participants of this “argument” and the standards with which it is conducted. Foremost among those conditions will be the authenticity of the voices in the conversation – no one is lying, manipulated by underhanded interests, incompetent as a deliberator, or obviously mistaken – a kind of authenticity that will very much mirror the conditions of autonomy that will be defended here. The story of Lee reminds us that even when choice occurs against a background of inherited values, such choices involve a reflective interpretation of the meanings of those values and their implications for the present decision. And such interpretation can be distorted in any number of ways – the person could be mistaken about the facts, thinking nervously and without concentration, rationalizing a non-cognitive drive, or acting on any number of reflection-inhibiting mechanisms. Once we see this, we 30

We could contrast Lee, for example, with another confederate – call him “Davis” – who unreflectively charged forth in the struggle against “northern aggression” without ever turning a reflective eye toward the value foundations and indeed the internal contradictions inherent in the tradition for which he was fighting. There is an air of authenticity to Lee’s value commitment that seems lacking in Davis, stemming specifically from the former’s acts of reflection and internal endorsement. Though this claim needs much discussion and further support, it effectively resists the conclusions of, for example, Alasdair MacIntyre who claims that canons of rationality are too intertwined with substantive ideals of justice in any historical period to consider the former as a tool to evaluate the latter. See MacIntyre 1988. For discussion of the conflict between this view and Habermas’s claim for universal norms of rational discourse, see Kelly 1990.

46

The Politics of Persons

must then realize that only under conditions where such factors are absent, when the self-interpretation that a value commitment involves is conducted in a non-distorted fashion, will the resulting decisions be authentic reflections of the person. And the same holds for interpersonal, community-wide discussions of shared values: only if all of the voices are sincere and minimally rational (in some sense to be worked out) will the results of the discussion be a true reflection of the community’s beliefs. Indeed, it is surprising to notice that the same theories which assume the impossibility of detached self-understanding ignore the inevitable limitations that infect any shared understandings of public values. Shared understandings, like all other phenomena, exist in time, and are subject to the fluid reflections of changing interpreters. And those interpretations, like the self-interpretations discussed above, are often expressed in metaphorical, impressionistic, and generally abstract terms (a point we will discuss in the next chapter). Hence, the process of understanding the shared values of a community will be ongoing, incomplete, and necessitate the participation of independent, authentic voices. The autonomy of those participants, then, will continue to be of pressing importance. What is needed, however, to make sense of the practice of social selfinterpretation just described is an account of reflection that on the one hand engages in an interpretive explication of one’s commitments and values but which, on the other, does not require the person to completely shed her skin, as it were, and ask whether the system of meaning within which she would engage in that very reflection is worthwhile. The claim that the latter, in its starkest form, is an unacceptable demand to place on all people as a fundamental value in a just society does not foreclose the possibility of reflection altogether, and indeed the kind of reflection required in the social dialogue that must take place for people’s fundamental commitments to gain purchase in their own communal life. In various places in what follows I will attempt to sketch such a view of reflection. iv. general conclusions The purpose of this chapter was in general to place on the table the multiple claims to the effect that selves are fundamentally social, to distinguish among them, and to identify the particular challenges they pose to individualist conceptions of autonomy and conceptions of justice built upon such a conception. As the discussion proceeds, it will be necessary to take a more critical look at these positions, but for now we are only concerned to identify the particular challenges to the traditional individualism of liberal theory.

The social conception of the self

47

I have been referring to “liberalism” and liberal theories, but, as I mentioned, the approach to justice to be sketched in what follows may not reflect all of the basic commitments of traditional liberalism, even in its recent guises. But I have tried to evaluate the various approaches to the social model of the self in light of the challenge they might pose for autonomy-based conceptions of justice and for viewing the autonomous person, in some form, as the basic subject of justice. Specifically, the capacity to reflect on oneself and one’s fundamental commitments, consideration and public discussion of which is the stuff of politics in a democratic regime, was the primary focus of the present examination, as that is the element of autonomy most at risk in confrontations with social models of the person. I have tried to suggest in a preliminary manner that some sort of critical reflection on the self and its commitments is not entirely ruled impossible or problematic in light of these social approaches to selves, though admittedly much relied on promissory notes that will be honored more fully only in later chapters. For now, we should note that many of the social conceptions of self, self-concept, personality, embodiment, and values that we considered will be taken on board here. I think it is a clear mistake to view the subject of justice to be persons defined without reference to their connections to other individuals, groups, traditions, and histories. Indeed, the chief danger of autonomy-based conceptions of democracy is to denigrate such relational elements of selves. I hope we have at least identified some conceptual material that can be put to use as we discuss the nature and importance of those relational elements.

chapter 3

The post-modern subject

We have now seen how those who see the self as essentially social want to put aside talk of a rational autonomous person in considerations of justice and politics. In other circles it is also strongly suggested that the conception of the autonomous person must be abandoned, but for reasons which are in sharp contrast to the perspectives just discussed. In so-called post-modern or post-structuralist approaches to critique and theory – which are hardly a single approach but rather a family with common elements – the idea of a stable subject who can critically appraise elements of herself and her values and whose judgment fixes the justification of principles of justice is thought to be an outmoded and ultimately unacceptable picture of both persons and theories. In some ways, this approach is at the farthest remove from the communitarian views just sketched, if such continua are judged by virtue of how “thick” or “thin” a conception of persons is allowed. For while the social conceptions of the self surveyed urged a “thicker” conception, according to which human thinking is fundamentally social in structure and substantive moral commitments form an irreducible component of those structures, post-modern views urge us to move to the opposite pole, where the notion of a singular unified subject of any sort, however thin the conception, is abandoned. The point of this chapter is to very briefly survey some prominent themes in what is called post-modern thought, particularly about the self and the possibilities of self-understanding, which will be of particular relevance to models of autonomy. The conclusion we will reach, in keeping with the flexibility about conceptions of the self that we want to maintain here, is that many views developed under the banner of post-modernism about the self – in particular the “decentering” of the self stressed by many – will not disturb the model of selfhood developed in the conception of autonomy to follow. Before getting to those tasks, however, it will be necessary to orient ourselves toward the general ideas that make up a post-modern approach 48

The post-modern subject

49

1

to meaning, language, and theory. Following this, we will examine the claim that the self “vanishes” when certain modernist articles of faith are questioned, but I will argue in section III that even if such a claim is acceptable (and there are versions of it that are not), the role that reflective deliberation plays in social relations of the sort advanced by some theorists of this ilk is nevertheless important. And that conclusion is all I want to retain in order to construct the ecumenical conception of self and selfgovernment I will subsequently develop.

i. the post-modern challenge The post-modern turn can be seen as having two separable strands, both corresponding to a rejection of corresponding modernist articles of faith.2 In one, the modernist idea that there are over-arching, if not universal, principles of reason applying to all rational agents as such (independent of particulars concerning those subjects) is rejected. What is discredited, on such views, is the idea that morality is universal and can be specified without reference to contingencies of personality, historical position, or operative dynamics of power. This view is “post-modern” in the sense laid out by Jean-François Lyotard in that it proceeds based on the rejection of the “metanarratives” of universal principles, whether these be the historico-teleological theories of Hegel and Marx or the categorical imperative of Kant (Lyotard 1997). On the other hand, post-modern theory casts suspicion on the idea of a stable subject or self operating behind action, thought, and judgment.3 Accordingly, a close examination of the dynamics of personal reflection, in some cases informed by psychoanalytic accounts of human motivation and thought, reveals in the subject a disjointed and unstable motley of motives, identifications, ways of thinking, and modes of behavior. Indeed, given this disjointed picture of personal reflection, attention is turned to the power dynamics of social institutions in which people’s self-images – their conception of sexual desire, conformity to the law, or reason and sanity – are ultimately constructed. The entire architecture of Cartesian subjectivity,

1 2 3

I cover some of this same ground in Christman 2002b, ch. 7. For example, see Iris Marion Young’s discussion of the various lines of critique of “the metaphysics of presence” in Young 1990b, 303–305. For an overview of post-modern themes, see White 1991. This line of thought draws inspiration from Nietszche’s claim that there is no “doer behind the deed;” see Nietzsche 1887/2003.

50

The Politics of Persons

where the cognitive reflections of the mental self can be understood as operating apart from the external physicality of body and world, is thereby abandoned.4 As we have noted, the traditional conception of the autonomous person is one of an agent able to rationally subject aspects of herself to reflection, yielding a reliable view of herself, and then alter herself in light of principles and values. The post-modern rejection of this picture of the person lying behind such a model is a wholesale one. The assumption of the essential rationality of the thinking subject, for example, is undercut by an emphasis on unconscious (and non-rational) drives and relations. The assumption of the centrality of the mental and the relative unimportance of the physical to conceptions of the self, is rejected by theorists who insist on the fundamentally physical, embodied, nature of experience and reflection.5 The traditional picture of the capacity for pure reason and unmediated self-reflection is rejected for problematically conceiving of thought apart from the play of meanings and symbolic systems which shape our experience of it, systems whose “mental” component (the understanding of these meanings) cannot be separated from the “external” physical component. Thought is language, and linguistic meaning is structured in ways dependent on reflective as well as physical, social, and political activity.6 A similar point is made in the work of Julia Kristeva who emphasizes the fluidity of meaning and the complexity of linguistic usage. For Kristeva, language is not simply structured by a set of transparent rules of meaning and use; it is a “process” involving an interplay of meaning, psychological experience, and communication. It is produced through the play of the literal and figurative, representational and musical, aspects of speech.7 Linguistic elements such as metaphor, figurative tropes, suggestive associations, and terms with shifting connotation all add to the interchange of ideas among language users. The conventions governing the meaning of such elements are open-ended, subject to negotiation and change, and, most importantly, are a function of the complexities of social dynamics inherent in the dynamics of actual communication and the institutional and social settings structuring that interchange. In more specific terms, key categories in the reflective understanding of the self – sanity, reason, guilt, 4 5 6

For discussion of the development of this Cartesian legacy (and parallel strains), see Taylor 1991 and Schneewind 1998. See especially Merleau-Ponty 2002; for post-modern versions of this emphasis, see, e.g., Butler 1993 and Grosz 1994. For discussion see Eagleton 1983, 133ff. 7 Quoted in Young 1990b, 304; see also Kristeva 1986.

The post-modern subject

51

responsibility, desire, sexuality – must be understood as operating inside this historically unfolding dynamic of social relations. And where there are social relations there is power at work in the shape and direction of those relations. So the self is constructed out of the power dynamics of present and past social structures, in particular institutional structures whose conceptual legitimation arises from and in turn reshapes that dynamic of power and meaning (knowledge/power).8 Another strand of post-modern thought that will be of relevance here is that which takes psychoanalysis as fundamental in understanding personality and meaning. Jacques Lacan has done more than any theorist to combine the developmental approach to personality offered by Freudian psychoanalysis with the post-structuralist view of language (as endless play of signifiers) to construct a theory of the desiring and reflecting self. For Lacan, the self emerges simultaneously with the acquisition of language which allows the child’s accession to what he calls the “Symbolic Order.” This process is mediated at crucial points by recognition of others and, in the important central step, of oneself through perception in a mirror by which one sees the external (and incomplete) reality of one’s physicality and simultaneously grasps the perspectival nature of all (including self-) perception. Moreover, this process is intertwined with the relational dynamics of mother and child, where the child experiences the primal “lack” resulting from severance with the mother (an inevitable and traumatic occurrence) which can only be resolved through gaining a place in the symbolic order and accepting the (linguistic equivalent of) the place of the father. This replaying of the Oedipal conflict (for Lacan, a dynamic that is the same for boys and girls) manifests a psychic process that provides the architecture of the self, whose components never shed the traces of this dynamic.9 So language constitutes the self, not merely in structuring thought and organizing reflection but in formulating the very subjectivity that engages these operations. This process of constitution, of adopting linguistic forms for oneself and as oneself, results from the interpersonal dynamics of parents and child. These dynamics, in turn, replicate the power structure of the Symbolic Order manifested by the phallus, which for Lacan (in a famous 8

9

See Foucault 1965, 1990a, 1990b. It can be noted that there is a certain vacillation in this literature concerning the epistemological status of the claims being made themselves. For post-modern writers are famous for rejecting any metaphysical basis for their own claims, so that, taken this way, it may appear there is much that is “more or less empirical” in these views. (For comment, see White 1991, 28.) But the empiricism that may be present is at a far remove from the scientific methodology of modernist epistemology, for it is an “empiricism” (better referred to as a philosophy of contingency perhaps) that is infused at every stage with a process of thick interpretation. See generally Lacan 1977, esp. ch. 1.

52

The Politics of Persons

ambiguity) seems to represent both the father’s penis and the amorphous symbolization of all things authoritative. Not all post-modern thinkers import elements of Freudian theory. Indeed, one prominent example is Lyotard’s forthright rejection of any such metanarrative.10 All such over-arching theories of history, the self, society, or normativity manifest the hypostatization of meanings as a fixed relation of word to world or as rule-bound patterns of language use. Theoretical claims of any sort which purport to universal significance belie the ultimately local and bounded nature of language. These boundaries, suppressed in the falsely general structure of metanarratives, are both linguistic and political (indeed the two cannot be separated).

ii. the vanishing subject Returning to the postulate of a decentered self, one can discern two separable lines of analysis used to articulate this view, one motivated by psychoanalytic approaches to development and the other focusing on the dynamics of power within which self-images are formed. The first emerges in the work of Lacan and post-modern feminists such as Kristeva and Butler, while the latter is seen most clearly in Foucault. We discussed Lacan briefly above. Julia Kristeva revises Lacan’s views on psychological development, offering, among other things, an alternative account of the pre-Oedipal stages of such development where representation of the “other” plays a role in the primary demarcation of the body and self. In discussing the developmental process in the formation of identity, Kristeva replaces Freud’s vision of identification with the mother and intrusion by the father (in Lacan, the intrusion of the phallic “Symbolic Order” at the point of the development of language) with an identification with the mother–father conglomerate, a sexless, genderless unit into which a “third party” intrudes, who is perhaps the father, perhaps an outside figure. While for Kristeva the Oedipal phase of development is crucial to explaining the internalization of patriarchal rules and discourse, it is to the pre-Oedipal processes of development that she redirects attention, in particular as a means of understanding the abiding disunity of the subject.11 10

11

This is explicated in Lyotard 1997. Another line of analysis in this work, however, stresses the destabilization of epistemic foundations brought on by new information technologies such as computer-aided communication. This theme is also emphasized in Gergen 1991. A contrast can be noted between Freudian/Lacanian psychoanalysis and object-relations theory concerning the centrality of the Oedipal phase in the construction of identity. For discussion, see Leland 1989.

The post-modern subject

53

As noted earlier, Kristeva sees the subject as the embodiment of a “questionable subject-in-process,” never fully defined and containing elements not reducible to strict semantic specification. She distinguishes between the realm of the “semiotic” and that of the “symbolic.” The latter corresponds to Lacan’s “Symbolic Order” and manifests the phallic, rulebound, order of society and law, internalized by the subject and forming a constitutive element of conscience and personality. Set against this is the semiotic: the instinctual realm of the maternal connection that exists prior to and independent of structured representational or linguistic systems. The semiotic is the unconscious element of subjectivity that circles around and underneath the operation of structured semantics and which corresponds to the feminine (and hence repressed) components of subjectivity. This element of the subject often emerges through artistic practices and alternative modes of expression. Those aesthetic practices that contrast with, and hence bring into question, the linear, closed, rule-bound structure of language embody, for Kristeva, the semiotic element of the unconscious that lurks within every subject-in-process. Avant-garde art in particular (especially that which simultaneously uses and undercuts structured representational and linguistic forms) allows the non-symbolic realm, the realm of the preOedipal and hence non-rule-structured elements of subjectivity, to shine along the edges of structured discourse. The “feminine” for Kristeva names that elliptical form of negative expression which emerges in the interstices of formal speech. This alternative is sometimes labeled “heretical ethics” or “herethics.”12 This view leads to a vision of the self as fundamentally disunified, irreducibly complex and hence “decentered”: both reflective agency and stable centers of action are placed in question. Avant-garde art and the utilization of extra-linguistic modes of expression, modes which find their source in the non-conscious semiotic order, manifest the way in which the assumption of an “author” of actions and representations is an illusion, albeit a useful one. While it may be indispensable to assume a singular agency in both self-understanding and interpersonal interaction, that conception of agency as unified and identifiable can only be represented in the discursive realm of the symbolic – the realm of rational thought and linguistic precision – and hence is undercut by the recognition of a nonrational and fundamentally ineffable realm of experience and expression. So since self-understanding must be formulated in linguistic forms, and 12

See Kristeva 1986, 185. This is what Diana Meyers terms “dissident speech”: see Meyers 1994.

54

The Politics of Persons

systematic linguistic structures are always incomplete expressions of mental life and bodily experience – that is, the symbolic order is always intertwined with but cannot fully capture the semiotic – then self-understanding is fundamentally incomplete. Hence, the subject (the object of such understanding) is disunified and itself incomplete. Other writers have arrived at a picture of the self as disunified via routes that do not rely so heavily on psychoanalytic explanations of mental development. For instance, several feminist writers who reject humanist conceptions of identity are motivated by a deep wariness of any categorizations per se (of which the unified subject, or the universal picture of a settled human nature, are prime examples). Such categories, as we have noted earlier, manifest a constraint on the fluidity and play of language and ultimately rest on a discredited metaphysics.13 Judith Butler, for example, famously argues that all categories (“subject,” “woman”) should be seen as inherently unstable and subject to deconstruction. Therefore, insofar as women must be delineated apart from the illusory neutral self of liberal modernism, the category of gender must be understood provisionally. For Butler, gender is performative, the manifestation of an embodied, repeated, and creative act of expressivity within contexts of fluid and contestable meanings (see Butler 1990). (I will return to Butler’s views below.) Still further from the rubric of Freudian theoretical foundations are those post-modern thinkers that focus on the social structure of meanings as the source of the discrediting of the humanistic ideal. One powerful theme in Foucault’s work, for example, is the treatment of the subject as constituted by the confluence of vectors of power within which it develops. The self cannot be seen as a stable set of capacities but rather as a mere projection whose perspective, agency, and values are the result of complex dynamics of normalizing power which, through language, institutions, juridical practices, and patterns of interpersonal interactions, operate to shape the person at every turn. Power not only constrains a self, it constitutes it. And since the dynamics of power are fluid and changing, the self is fluid and changing and not able (reflectively or otherwise) to detach from those dynamics and react independently of them (Foucault 1990a, 1990b). Foucault often insists, however, that it is not the disintegration of the self, per se, that he is describing, as if there was once (or could have been) a stable entity that operated in an independent and rational manner but which has 13

See for example Flax 1993 and Fuss 1989.

The post-modern subject

55

been distorted by recent power dynamics.14 Rather, it is the discourses that feature the unified subject as an independent and rational mechanism that are the object of his analysis. More importantly, the dynamics of power that surround the person in any given age should not be seen as a “repressive” force that stultifies an otherwise liberated consciousness, but rather as a pattern which constitutes that subjectivity.15 For Foucault, various processes evolved that manifest modes of “objectifying” the subject – constructing discourses that created the modern understanding of the person as subject. One of these modes is what has been called “subjectification” (sujetissement) and concerns “the way a human being turns him- or herself into a subject” (Foucault 1984, 11). In these relatively active forms of self-formation, social dynamics and relations of power provide the elements with which persons come to operate and understand themselves. It is unclear, however, exactly how the detection of past dynamics as the source of current subjectivity results in the “death” of the subject, that is, in the disappearance of a stable agency accountable for actions and judgments. Of course part of the Foucauldian project is to subject any such stability to a constant interrogation concerning the genealogy of its structure, to repeatedly ask how institutional and social practices normalize its functioning in a way undetected in its routine operation. What this view comes down to, then, is another version of the claim that persons are not transparent to themselves – that no person’s psyche can be considered already fully laundered by rational self-appraisal – and hence the structure of a self is subject to constant reconsideration and interpretation. From these two angles then – from that of individual development through psychological processes and from that of the dynamics of social power – these theorists arrive at the view that selves cannot be seen as stable, unified, and self-transparent structures, but must be understood as constantly subject to deconstructive interrogation. We will discuss below an important implication of the view that conceptions of desire and 14

15

For Foucault, a shift has occurred from searching for a (metaphysical) model of the self according to which questions of self-determination, freedom and, more particularly, the ability of selves to “penetrate the substance of things and give it meaning” are raised, to asking about what function talk of a self serves in contemporary discourse. He writes, “[h]ow, under what conditions, and in what forms can something like a subject appear in the order of discourse? What place can it occupy in each type of discourse, what functions can it assume, and by obeying what rules? In short, it is a matter of depriving the subject (or its substitute) of its role as originator, and of analyzing the subject as a variable and complex function of discourse” (Foucault 1984, 118). For an example of how to wrestle with the paradoxical question of how power both limits and constitutes the subject, see Butler 1997.

56

The Politics of Persons

subjectivity must be the object of recurrent reconsideration. First, however, we should make clear exactly what these claims about the instability of subjectivity, the decentering of agency, and so on really amount to.

Self-reflection without a self? Clearly, then, these lines of post-modern thought cast doubt on the idea of a core self, a fixed and unified set of beliefs, values, capacities, or ways of thinking that define the person (in self-transparent ways) apart from the social settings she encounters and which she uses in those encounters. But to see whether this implies that a conception of autonomy that sees selfreflection as a crucial element must also be abandoned, let us look more closely at the development of the recent views of Judith Butler on the question of the self. Doing so will provide some interesting lessons about the implications of these ideas, some of which lead circuitously back to a position we want to maintain. As I mentioned earlier, Judith Butler has notably developed a line of analysis that deconstructs the conception of a stable, transparent self. However, in recent work, she has also attempted to formulate a conception of the subject which at the same time incorporates the insights of Foucault and others concerning the way in which power dynamics construct the subject but also finds a way of understanding how we are agents nevertheless, that the power dynamics in question, while exerted “from the outside” as it were, nevertheless structure us as agents and as potentially responsible for events and actions (Butler 1997, 2005).16 Taking into account Althusser’s view of the way social authority is exercised in the formulation of the subject’s sense of herself (though without his assumption of unified state power as the source of this authority), Butler constructs a speculative model of the development of the subject. Her construction also makes use of psychoanalytic accounts of the manner in which traumatic separation from the source of love (which is also the source of dominating power) creates a subject through a kind of subjection (to the caregivers): 16

It is important to point out that “the subject” for Butler is not the same as the person or the individual. The subject refers to that which is “designated as a linguistic category” (1997, 10). That is, it is the psychic structure operating in an individual which is formed by discourse. As I would put it (perhaps in a way Butler might not embrace): the subject is the embodied psychological schema whose operation involves concepts that operate and gain meaning from the conceptual dynamics of the social milieu in which the person develops and lives.

The post-modern subject

57

Through that neurotic repetition the subject pursues its own dissolution, its own unraveling, a pursuit that marks an agency, but not the subject’s agency – rather the agency of a desire that aims at the dissolution of the subject, where the subject stands as a bar to that desire. (Butler 1997, 9)

The paradoxical dynamic at work here produces, on Butler’s account, a deep ambivalence: “How can it be that the subject, taken to be the condition for and the instrument of agency, is at the same time the effect of subordination, understood as the deprivation of agency?” (1997, 10). Insofar as a subject is formed by her subordination – her place in a dynamic of power that is imposed externally by unnamed and uncontrollable forces – the subject forms a passionate attachment (by way of her conception of herself) to that subordination. This is the paradoxical nature of subject formation for Butler.17 In this model, Butler is trying to avoid the account of the self in the liberal humanist tradition where it is assumed that agency is always fully formed and stands in opposition to power. Yet, the supposed alternative, that power operates at all points to construct and structure (and hence “control”) agency, implies a “fatalism” about the possibility of resistance that she also wants to avoid (see Butler 1997, 17). As a result, Butler constructs a picture of the agent which has capacities for reflection and self-awareness, but only to a limited extent, as forces that are both social and unconscious operate to form the elements of that self and hence remain somewhat obscured from (introspective) view. Selves (subjects) develop by way of power dynamics modeled on Hegel’s master–slave dialectic, on Butler’s view. What emerges is a reflective psychic function that can take itself (or an aspect of itself) as its object. The reflecting self is as much a product of the power interplay as the object of reflection (desire, say) but that dynamic is hidden from view, as it were: [C]onsciousness is now divided into two parts, the “essential” and the “unchangeable,” on the one hand, and the “inessential” and “changeable,” on the other. The watching self, defined as a kind of witnessing and scorning, differentiates itself from 17

Butler 1997. Although I am not overtly pursuing critical lines of response here, I might mention that part of the “paradox” that emerges in this account stems from the rather loose use of key terms, such as “power” and “subordination.” To say that a subject is formed by the dynamics of power implies that there is an entity upon which that power exerts itself. But it is assumed here that the psychic schema that forms the subject is constructed by that power. This involves a paradox only if we assume there is something like a self or subject already in place upon which power is exerted. But if we drop that language, and merely talk about the social and psychological influences that bear on a person’s development, no paradox arises.

58

The Politics of Persons

the self witnessed as perpetually falling into contradiction. This watching becomes a way of reestablishing the visual distance between a subject aloof from the scene and the subject in contradiction. In this case, however, the witnessing and scorning self cannot deny that the contradictory self is its own self; it knows that the contradictory self is itself, but in order to shore up an identity over and against it, it renders this contradictory self in an inessential part of itself. It thus parts with itself in order to purify itself of contradiction. (Butler 1997, 46)

The attachment to this self (as the object of reflection) is “passionate,” for Butler and amounts to the “will” of the subject.18 For Butler the subject does not form by way of intentional processes that the person enacts upon herself, as a form of mental activity. These processes occur through repetition, performance of bodily and psychic operations that take on a certain form, understood as discursive patterns, patterns that Foucault calls a “regime of truth” (Butler 2005, 22–23). This implies that the formation of the subject involves an “injury” in that one’s identity is constructed by way of an imposition of power (repeated and reiterated by one’s own activity and interaction with those power dynamics). But Butler does not claim that identity then must re-enact that injury perpetually but rather that strategies of resistance to dominant social norms, ones whose discursive structure formulates the terms of the self-concept, will often involve an unsettling of the passionate attachment to subjection without which subject formation cannot succeed (1997, 105) In later work, Butler builds on this understanding of the construction of the self/subject to confront the issue of how an agent could be thought to be, or think of herself as, a responsible being, especially given the external dynamics (grounded in the social complexities of power relations) that structure thought and reflection (Butler 2005). She notes, following Adorno, that moral judgments are conditioned by the historical conditions that give rise to them, and she therefore warns against the imposition of moral universality of the form that defines membership in deliberative practices in ways that are exclusionary (Butler 2005, 7). But even the kind of moral judgment that is sensitive to its own historical conditions will presuppose an “I” which is the seat of such judgment, albeit one that cannot stand apart from the social conditions of its emergence (as sketched earlier). The question becomes, then, what view of such an “I” can be given – a view of the subject in moral and hence political discourse – that can be responsible while understanding itself as structured by a socio-historical context. 18

She writes: “self-consciousness is the form the will takes when it is prevented from simple expression as a deed” (1997, 76)

The post-modern subject

59

This involves, she claims, having to give an account of ourselves while at the same time acknowledging (or at least being sensitive to) the ways that social conditions structure the reflections that underlie such an account. Of course, this act or process of accounting for myself is always enmeshed in power relations; it is in response to an address, recalling Althusser’s analysis of the function of interpellation. So the paradox identified earlier, where the concept of myself is something I discover and present, is the result of power exerted on me from external sources. Moreover, to be “responsible” (“response-able” or “able to respond”) is to react to the very forces which exert this power on my self from the outside. This account, then, is built in the service of a view of responsibility and agency that rejects the possibility of self-transparency and ontologically fixed centers of agency:19 “the very terms by which we give an account, by which we make ourselves intelligible to ourselves and to others, are not of our making. They are social in character, and they establish social norms, a domain of unfreedom and substitutability within which our ‘singular’ stories are told” (Butler 2005, 21). But this contrasts with the social constitution of the self described in the previous chapter, for these normative discourses are not “a structuralist totality nor a transcendental or quasi-transcendental invariability” (Butler 2005, 24), by which is meant (as I read her) a closed system with definitive, public, and transparent meanings. The normativity at work in these discursive power dynamics is fluid and amorphous (yet powerful), as they work, for Butler, at the level of the unconscious and in relation to desire, not merely cognitive propositions that can be grasped and evaluated in their entirety (Butler 2005, 35ff.). Giving an account of oneself, for Butler – something at the heart of ethical responsibility on her view – involves constructing a narrative of oneself in response to (and in dynamic relations with) interrogating others and enmeshed in a normative discourse not of one’s own making. Such narratives, moreover, can never fully produce a complete and canonical representation and hence involve several “vexations,” which include the presence of a non-narrativizable exposure that establishes one’s singularity (uniqueness that cannot be captured by the general terms of a story), and a history that establishes one’s partial opacity to oneself (Butler 2005, 39).20 19

20

Another critical note could be inserted here: suggesting that there is an “opacity” in our self-reflections or that self-understanding is incomplete or inaccurate implies that there is a true self after all, namely that which is not adequately captured by our internal reflections (as pointed out above). Echoing a theme we will take up in chapter 4, Butler suggests how narrating ourselves always comes in response to the address of another, whether the valence of that encounter is positive or negative

60

The Politics of Persons

Indeed, she argues that this opacity to ourselves can actually ground an ethical commitment to others. The limits to self-knowledge can, she suggests, motivate recognition in others of that very limit, spurring one’s humility and “forgiveness” of others for not fully knowing what one cannot know oneself: the complete details of the normative sources of one’s own identity (Butler 2005, 42–43).21 There is always a “pre-history” to the self which autobiographical narration attempts, but always unsuccessfully, to capture. Self-constituting narration of this sort is always a kind of fiction, but it is one that constitutes our “selves” all the same. Agency and responsibility arises from this socially and relationally facilitated self-construction, albeit an agency that is shorn of the presumptions of self-transparency and (ontological) stability. Now I am not concerned here with the details of the positive account of ethical responsibility Butler constructs; nor am I engaging in critical appraisal of her views.22 Rather, I want to note that such a view of the subject can be accepted by an account of autonomy that requires selfreflection, as long as the description of such reflection does not presuppose a pre-existing “me” as the object of such introspection which is fully and accurately captured by it. There are differences, however, between the account of autonomous agency given in what follows and other claims and commitments a theorist like Butler would hold, for example concerning the role of rationality (and reason-giving generally) in the reflection that

21

22

(Butler 2005, 11). There is always an implied (or actual) audience for such a narration and the goal will often be one of persuasion, if only to persuade the other that the voice of this “I” is genuine. This process of self-narration is not based on a pre-existing causal structure we can call a “self” but is a precondition for responsibility for actions and outcomes (even if this self-narration is in the form of gestures, bodily movements – “a nod” – or speech) (Butler 2005, 12–13). A theme similar to this will be echoed in chapter 9 when it is argued that recognition of social identities must be based on an openness to the constant possibility of revision for the categories expressing those identities. One additional critical note: one might point out the ways in which Butler refers to the “violence” and “trauma” of the limits of our ability to narrate ourselves and the imposition of the other in that process of narration. These terms are meaningful, one might claim, only if there is a victim to that violence, trauma, or imposition already in place. But the claim here is that the process of socially mediated narration described here constructs the self who would be a candidate for such victimization. This relates to a second point: Butler’s account of responsibility seems to make no clear distinction (and may not allow such a distinction) between very different kinds of imposition by external social powers. Real violence is different from amorphous imposition by the power dynamics of social discourse in general, and this difference is precisely based on the fact that my identity is crushed and injured in the first case but structured and formed in the second. Her only gesture toward this issue is to call for us to “devise norms to adjudicate among forms of impingement, distinguishing between its inevitable and insuperable dimension … and its socially contingent and reversible conditions” (Butler 2005, 107). But the distinctions between kinds of impingement and injury can be drawn only by postulating the elements of self and personhood which are implicated in them; with such a fluid model of the latter such distinctions may not be available.

The post-modern subject

61

autonomy assumes. The more limited conclusion I want to draw here is that this post-Enlightenment conception of the self is one that the theory of autonomy sketched in what follows here can at least be agnostic about. Moreover, Butler is quite sensitive to the social need to present oneself as a singular entity, a stable personality with firm commitments and motives, even though the reflection that gathers together the sense one has of those commitments and motives is rife with occlusions, distortions, and misplaced concreteness. But social intercourse demands the construction of a presented self nonetheless which, as a result, becomes the “me” that is held responsible for what is presented: “When I tell the truth about myself, I consult not only my ‘self,’ but the way in which that self is produced and producible, the position from which the demand to tell the truth proceeds, the effects that telling the truth will have in consequence, as well as the price that must be paid” (Butler 2005, 132). What I will claim later is that the social dynamics required by democratic culture imposes on us a need to reflect on our commitments and values and to give reasons for them in social deliberation. That reflectiveness may construct the self which is at the same time the subject of the reflection, but autonomy can be fashioned in a way that accepts that picture (or at least is open to it). Indeed, as Butler herself says: “Although self-knowledge is surely limited, that is not a reason to turn against it as a project” (Butler 2005, 46)

iii. a re-emergence of deliberation? As I have stressed, a common theme of post-modern thought is the rejection of discursive speech as the sole medium of expression, reflection, or knowledge. A strongly suggestive case for this is made with reference to poetical expression which, immune to paraphrase into propositional form, nevertheless conveys powerful content. What Kristeva refers to as the symbiotic (whether or not one accepts the place she accords this in the unconscious) refers to that aspect of language which conveys feeling and thought but which is not analyzable via specific rules of meaning and reference. The form of poetical expression as well – the rhythm, sound, musicality, and sensuality of symbolic forms – carries meaning that is not well disposed to reduction or paraphrase. Consider almost any powerful poetic passage and ask how the sounds and rhythms of the language irreducibly convey much of the meaning. One must either subtract all that is metaphorical, figurative, and “sensual” in the linguistic universe from the realm of the

62

The Politics of Persons

meaningful, or one must admit the importance of those non-discursive modes of expression.23 The point is more general when one considers metaphor; for non-literal expressiveness is pervasive in all areas of thought and expression (see Lakoff and Johnson 1980). Now there is a complex and healthy debate about the semantic function of metaphor, with Richard Rorty, following Davidson, arguing that metaphors carry no semantic value but function merely as gestures, ostensive acts that point to the meaningfulness of some phenomenon but do not embody stable meanings themselves (Rorty 1989, 162–172). Others reject this view, arguing that sentences containing metaphorical expression can (though need not be) truthfunctional and subject to debate, thereby indicating the semantic values inherent in such language. However that debate turns out, there is left the problem of determining the difference between so-called literal speech and metaphorical expression. How could semantic or meta-linguistic rules be formulated that distinguish these two forms of expression without, in an ad hoc or question-begging manner, presupposing one or another theory of metaphor itself? Considering the pervasiveness of expressions that arguably do not “literally” express what their words indicate, it will be very difficult to systematically determine what sets some language aside as not truth-functional, semantically stable, or discursive, and hence not meaningful. But indeterminacy in language can only be tolerated to certain degrees and in certain contexts. Modes of self-reflection might indeed allow inclusion of semantically variable symbolization: I may say to, and of, myself that my attachment to a mode of life, say, partly derives from a certain je ne sais quoi. But that indeterminacy becomes problematic when I must convey this observation, or any information relevant to it, to others in contexts where mutual understanding and communication are important. As Habermas has said about post-structuralist accounts of language, it is unclear on those models how the reproduction of social life in general, and learning practices in particular, could occur if such accounts accurately captured the nature of meaning (Habermas 1995).24 In modes of interpersonal communication, and this includes all social life and political institutions as well as intra- and inter-cultural communications, certain presuppositions about both the semantic stability of the language used and the sincerity, rationality, and 23 24

As we have mentioned, this kind of approach is often part of views inspired by Heidegger. For discussion, see Dreyfuss 1991 and Haugeland 1982. For discussion see White 1991.

The post-modern subject

63

authenticity of the speakers in question, must underlie those interactions for any meaningful responses to be made. A similar point emerges also from the post-modern insistence on continual reinterrogation of dominant meanings, including conceptions of self and other. If such constant reconsideration must take place among persons, then the participants in those discussions must assume, at least provisionally, the sincerity, authenticity, and basic rationality of their interlocutors, as well as themselves. This must be so if the outcomes of such discussions are to have any normative purchase for the participants. For the most part, for example, post-modern theorists write as if their audience does not assume a complete lack of stability and agency on their part. Recall how Kristeva described the attribution of agency as a useful illusion: it is illusory (perhaps) at one level, but essential – more than merely useful – in the interpersonal interplay among modern and postmodern fellow travelers. The implications for the interpersonal dynamics of agency will be explored in later chapters. A point to emphasize here is that insofar as meanings as well as agency itself are constantly subject to interrogation and revision in the post-modern picture, such review must be engaged in by individuals whose identity and minimal rationality will be assumed to be stable during that process of engagement. While certainly the selfreflexive structure of post-modern theory has been well noted – where not only the theories but the theorists in question are subject to the deflationary account of meaning they themselves endorse – what has not been noted sufficiently is the degree to which discussion, critique, and even the development of genealogies concerning dominant meanings, is a social, interpersonal process. This must be a process whose participants are able to place the disintegrated nature of their own identities somehow at bay, perhaps through the ironic distancing discussed by Rorty (1989), or through the purely strategic use of the tropes of humanism for the purposes of political struggle and value articulation, as others have suggested (see, e.g., Butler 1995). I would press the point further, namely that only rarely do the authors of texts defending the fragility of identity and the fluidity of meaning themselves display anything but a forthrightly rational, discursive, authentic, and sincere persona in the presentations of their views. And apart from academic writing, actual political and social engagement must be part of the negotiation of identities that the rejection of modernism itself calls forth: in rejecting an unchanging foundationalist metaphysics and all that rests upon it, we nevertheless must continue discussions and

64

The Politics of Persons

interactions with actual others whose viewpoint and self-reports we will (indeed must) take seriously, and as something other than systematically deluded plays of difference. To view our companions and comrades in any other way – to consider them as merely the conflicted dupes of unconscious power dynamics, without a center of agency or a reliable reflective capacity – is to treat them (and ourselves) as actors in a play, to be dealt with as obstacles and barriers rather than persons whose viewpoints matter crucially to us.25 This, then, points to the relationship between interpersonal dynamics and the solidification of a locus of self. That is, practices of giving reasons and holding agents accountable, both for their actions and their words, secure a location for a committed “self” in some way or another (as we saw in our examination of Butler). Even if we admit that the meanings by which the contents of that self are expressed may be fluid, shot through with ambiguity, immune to truth-functional analysis, and ridden with the effects of social power, the interpersonal exchanges involved in sincere communication fix a public meaning onto the participant’s expressions of herself. A “self” is created, as it were, out of such exchanges, even if the beliefs or commitments or values that are so created did not exist in a pure form (fully transparent to the person herself) prior to that expression (because aspects of them were unconscious for example). We can, for argument’s sake, then, accept or be agnostic about much of the lines of thought brought out here but still argue, as Habermas does, that the dynamics of communication and deliberation fix the language and conditions of validity of self-expressions. But I should add, just as this argument focuses on the interpersonal element of self-construction, there is an inter-temporal element as well: just as interpersonal interactions serve to secure a set of commitments and values as part of a person’s self-concept, memory and planning do so also. Of course, as reflections on personal identity have shown, memory alone cannot determine the existence of a self over time, since determining whether memories are valid requires determining identity, but memories do provide a sense of oneself over time, even if strict metaphysical identity cannot be so determined. In chapter 5 below I will examine this aspect of selfhood in much more detail. 25

This is not to deny Iris Young’s claim that alternative, non-discursive modes of expression must be allowed in some form in public discourse in order for differences in communicative styles to be accommodated (Young 1990b). My claim is not, at this point at least, that expression must take on any particular form, only that it be sincere, authentic, and minimally rational to function adequately in interpersonal exchange.

The post-modern subject

65

My overall claim in this discussion has been that we can, for the sake of argument, accept many of the general claims discussed in this chapter. For example, the view of the (model of the) self that functions in the conception of autonomy developed here can accept that meaning and the linguistic expressions that structure and convey it may be more variable and unstable than traditional models of discursive (truth-functional) propositional language assume. Poetical speech, rhythms, and sensual qualities of words, and the gestures and bodily aspects of expression convey meaning but are not themselves subject to logical analysis within truth-functional semantics. However, while metaphorical expression figures irreducibly in communication, it cannot be completely immune to all interpersonally verifiable rules of meaning, since communication with others and the verification (in some sense) of claims necessitate such public rules. Second, we can accept that many aspects of a person’s motivations, beliefs, and the meanings of her conscious thoughts are inaccessible to her conscious reflections; and elements of those meanings are fixed by the unarticulated undersides of the terminology she uses (that is, the “other” of that terminology whose difference determines its meaning). Key conceptualizations involved in her reflections may well be constituted by social processes that are, in turn, structured by relations of power. Yet, accepting all that, we can still claim that a “self” arises out of the public, interpersonal interactions that sincere communication involves. We commit ourselves to fixed meanings even if the meanings of our internal reflections are rife with fluidity and ambiguity, both conscious and unconscious. Despite the fact that post-modernism of this sort is often used to jettison talk of self-government (and hence autonomy) as a normatively coherent model of the political subject, it will be my claim here that none of these lines of thought poses a serious threat to a conception of autonomy that allows that the self of self-government may be as fluid, opaque, and amorphously structured as these lines of thought suggest it is. The proof of that claim, however, lies in pudding yet to be made, a task taken up in part II below.

chapter 4

The narrative self

In chapter 2 we discussed visions of the self that emphasized the social elements relevant to its constitution. Here we take up a theme that connects orthogonally to that vision, the conception of selves as nothing but a sequence of experiences, acts, traits, and judgments that have a particular form. This view accepts the possibility that the beliefs, attitudes, or dispositions that are part of the self are socially constituted, but concludes that what makes a collection of beliefs, attitudes, or dispositions (located in a single biological body) a self is that the set is structured in certain way, in particular in a narrative form. I want to examine this view in order to assess its usefulness in the context of a political conception of the self. What we will see is that its usefulness arises not so much from the nature of narrative – the particular structure that defines narrativity and hence purportedly distinguishes so-called selves from non-selves – but from the ability to form narratives (whatever those are). That is, the reflective capacity to interpret the elements of one’s life (experiences, beliefs, acts) into a unique and coherent structure is what makes the person who she is, not necessarily the fact that those contents have the shape of a good story.

i. background In a wide array of fields in recent years, the claim that selves (or personalities, or persons) count as a single entity only if the elements that constitute them are shaped in the form of a narrative has gained enormous popularity. In philosophical and social scientific research – on theories of the self, the person, the conditions that make for a single (as opposed to multiple) personality – the claim is made that to be a single such entity (self, person, personality) the particular “contents” of that entity must be shaped in the form of a narrative. The idea is that selves, persons, personalities, and so on 66

The narrative self

67

contain elements such as experiences, acts, and bodily characteristics. Such collections are sometimes so disconnected that we would not say that a single person or personality occupies a particular body, but rather several such entities do so. That being so, we need criteria for determining a level of unity or continuity that marks a single such entity. The criteria on offer here is that of narrative unity: the experiences and actions of a body count as that of a single self only when they can be grouped into narrative form. Otherwise, there are several selves (or successive selves) occupying a body. This requirement functions in accounts of personal identity over time, in views of what differentiates one personality from another (in cases of so-called multiple-personality disorder, for example), and in psychological models of the self-concept, among other areas. We are to imagine here that the “elements” I am describing (experiences, actions, bodily traits) can be considered reflectively by a consciousness occupying the body in question.1 The implication of these theories we are discussing is that a subject reflecting on memories and experiences in a way that does not display narrativity would fail the test of being a single, unique, personality. But this raises the question that will be the focus of most of our discussion in this chapter, namely what this requirement of narrativity amounts to as a separate characteristic of being a subject of experiences of any sort, and whether it adds anything substantive to the conditions of personhood over and above the requirement that the subject reflect and interpret “her own” experiences. I will argue here in fact that the concept of narrativity in these contexts must be unpacked, and that in many of the ways in which this idea can be understood (ways often not specifically ruled out by the theorists in question), it is implausible to say that the subjective structure of the experiences of a person (self, personality, and so forth) must be of narrative form for those experiences to coalesce into a unity. Before beginning, we should take note of the ways in which the narrative conception of the self functions in various theoretical arenas. In philosophy, reference to the narrative structure of lives or personalities appears in a variety of settings: theories of personal identity, views about the nature of selves or the unity of consciousness, and social and political 1

This strange talk of consciousnesses occupying bodies is unfortunate, and is not meant to carry metaphysical implications (that mental entities are “in” physical ones for example). But it is forced on us for the moment to avoid very clunky disjunctions, such as “reflection on experiences, actions, and bodily traits takes place at a time by an embodied entity who may or may not be a single self, person, or personality.”

68

The Politics of Persons

theories specifying the communitarian or socially embedded conditions of personhood.2 In these contexts, and in a variety of ways, the view is proffered that something called “narrativity” is an identifiable characteristic of the sequence of memories, reflections, actions, mental events, or other such factors that marks them out as unified and individualized. Narrativity is meant to help explain what it means to be a unique, individualized subject of experiences, as opposed to a dissociated, disconnected series of (perhaps partial) selves. A similar idea can be found in accounts of the self or self-concept as presented in psychology and the social sciences. For example, some theories of self-concept use narrativity as a model for the factors that determine whether memories, self-reports, and experiences lie within the core of the person’s sense of self.3 In dealing with so-called multiple-personality disorder, psychologists have claimed that the impressions, memories, personality traits, and the like that a subject experiences count as a single personality (within the person) when these experiences fit into an identifiable narrative structure (see Flanagan 1996 and Hacking 1995; for general discussion see Radden 1996). Psychoanalytic theorists have also used the model of narrative unity in therapeutic settings as a template through which to construct a working conception of a person’s (perhaps suppressed) core ego (Freeman 1993; Schafer 1981; and Spence 1984). In many of these contexts, however, insufficient attention is given to the complexity and variability of the idea of a narrative. And without careful analysis of this concept, narrative theories will be at best incomplete and at worst implausible on their face. As I will point out in what follows, the idea of narrativity gives rise to many interpretations, several of which would be unwelcome in the theoretical contexts in question. Indeed, I want to argue that under most interpretations of the idea of narrativity culled from linguistics, literary theory, and philosophy, the condition of narrativity for the unity of selves, persons, and personalities is either implausible or otiose.4 My purpose will be to unpack the concept 2

3 4

Atkins 2008; Dennett 1988, 1989; Flanagan 1996; MacIntyre 1981; Schechtman 1996; Taylor 1991; and Velleman 2005. For a treatment of narrativity and the self with which my treatment has much in common, see Goldie 2003. See, for example, Bruner 1983b; Cohler 1982; and Kagan 1989; for an overview of literature on the selfconcept generally, see Ross 1992. These questions are not intended as a full-fledged criticism of any of the views I describe; indeed, I do not explicate any of these views in enough detail to criticize them legitimately. Rather, my intentions are more general and friendlier, in that I want to insist, as a necessary extension of those views, that the condition of narrativity must be unpacked and that many of its typical meanings are inadequate to the theoretical projects in question.

The narrative self

69

of narrativity that will make sense of the unity of selves but will, more importantly, aid us in the view of autonomy developed below.

ii. the narrativity condition What does it mean to say that narrative unity is something over and above reflective (self-)interpretation and that this unity plausibly picks out a unique entity (a self, a person, a personality)? To help answer that question let us look at the way in which the claim that narrativity defines personhood in some way occurs in numerous forms in various settings.5 Marya Schechtman, for example, has developed a theory of personal identity in which she argues that the unifying element of a (single) person’s life is the narrative form of experience (Schechtman 1996). She uses this observation as a way of constructing a theory of identity that answers the “characterization question” – the question of what characterizes a person over time – rather than the “reidentification question” – the question of how to reidentify a person over time (arguing that the former question speaks more directly to the concerns that we expect personal identity to answer, such as survival, moral responsibility, self-interested concern, and compensation). Persons can be distinguished from merely conscious, sentient individuals because of the narrative structure of their experiences. Schechtman writes that “individuals constitute themselves as persons by coming to think of themselves as persisting subjects who have had experience in the past and will continue to have experience in the future, taking certain experiences as theirs … [Persons] weave stories of their lives” (Schechtman 1996, 94). By way of example, she contrasts a “person” in her sense with a Buddhist who has achieved a satori-like dissolution of the self, where each person time slice recognizes itself as a separate existence, without regard to future or past. What is lacking in such a case is a linear connection among the events experienced by the subject, a connection that conforms to the structure of familiar narratives.6 5

6

The claim I am discussing comes in a variety of forms, at times concerning the nature of “persons,” “selves,” “individuals,” “lives,” or “personalities.” Since I want to paint with this broad critical brush, I will use these terms interchangeably to refer to the array of theories in question. Schechtman 1996, 100–101, citing Derek Parfit as a source of the Buddhist example. It may not be clear how that argument is meant to work, however, since we are to imagine an individual who no longer maintains a continuous consciousness in order to show that narrativity is necessary for personhood. But this argument does not establish that narrativity is anything over and above continuous consciousness.

70

The Politics of Persons

For Schechtman, a narrative has the “form and the logic of a story … where ‘story’ is understood as a conventional, linear narrative” (Schechtman 1996, 96). She quotes Jerome Bruner to elaborate: “A narrative is composed of a unique sequence of events, mental states, happenings involving human beings as characters or actors. These are its constituents … Their meaning is given by their place in the overall configuration of the sequence as a whole – its plot or fabula” (Bruner 1990, 43–44). In spelling out what she means by narrativity in her theory, Schechtman clearly relies on the idea of classical, linear, stories, though she points out that such narratives can take many forms. Nevertheless, the form that such narratives take must be socially recognizable in a culture for these sequences to count as a unity: “To be identity-defining an individual’s self-narrative must conform in certain crucial respects to the narrative others tell of his life” (Schechtman 1996, 96). She claims, though, that to the extent that nonstandard narratives organize a person’s life, that individual’s “personhood” is therefore non-standard (relative to a culture): A family of mostly overlapping narrative forms and the practices that go with them count, for our purposes, as standard life stories, and so as the kind of narratives that unproblematically constitute a person in our sense of the word. Narrative styles outside this family or group which retain certain of its most basic features also constitute persons, but persons unlike us. When a self-conception becomes wildly different in form from those standard in our culture – for example, a self-conception that is not even in narrative form – the narrative self-constitution view does not consider it identity constituting at all, nor those who organize their experience this way as persons. (194–195)

Schechtman also adds constraints on the kind of narratives that count as person-constituting: an articulation constraint – a person must be able to give an explanation of her acts by virtue of past or present actions or traits – and a reality constraint, requiring that the narrative not contain egregious factual or interpretive errors. So for Schechtman a person can be characterized as existing uniquely over time only when (among other things) the experiences and memories of the subject are narrative in form. This means that the discrete experiences have a linear, culturally recognizable form in which each “event” in the sequence gains its meaning in reference to the other events, and such sequences could be articulated by the subject without undue errors of fact.7 7

I use Schechtman here (as with the other thinkers I discuss) as a jumping-off point to crystallize how the condition of narrativity functions in theories of the person, and I couch my response in critical tones, but the view I end up with here does not depart significantly from what Schechtman goes on to say about narrativity in the end. So my treatment of her work here should not be taken as a criticism in any direct sense.

The narrative self

71

Psychologists Kenneth and Mary Gergen construct a theory of the selfconcept that also crucially involves narrative structures (Gergen and Gergen 1997, 161–184). They reject mechanistic and synchronic accounts of the self-concept, since these ignore (in the former case) the individual’s ability to shape her own ongoing self-conception and (in the latter) the individual’s self-understanding as a historically emergent being. As a replacement, they adopt an account utilizing the idea of a “self-narrative,” which they define as “the individual’s account of the relationships among self-relevant events across time. In developing a self-narrative the individual attempts to establish coherent connections among life events” (162). The essential aspect of narrative form, for Gergen and Gergen, is “directionality” among events, in which they can be seen to move over time in an orderly way toward certain ends.8 Following Northrop Frye, they describe the four basic forms of narrative: comedy, romance, tragedy, and irony/ satire (Frye 1957). At a more abstract level, these forms (except satire) involve “shifts in the evaluative character of events over time” (Gergen and Gergen 1997, 165). Combining this observation with subdivisions Frye supplies, they claim that the basic forms of narrative sequences are the “stability narrative” (in which the subject remains evaluatively unchanged) and “progressive/regressive” narratives (involving evaluative shifts in positive or negative directions). Narrative structures can be mapped by using these dimensions; for instance, “tragic” structures involve a severe regressive evaluative shift. For Gergen and Gergen, narratives are not objectively defined sequences that structure a person’s life; rather, they are wholly constructed by the person herself. That is, they are not pre-constituted forms taken from the world of art, though it is clear that people are affected by narrative forms present in their culture (Gergen and Gergen 1997, 168). A kind of “dramatic engagement” is involved in constructing a self-narrative, on their view, which involves the capacity to create feelings of drama or emotion that in turn alter or accelerate the evaluative “slope” of the event sequence making up the dramatic structure. Such engagement and the narrative forms or selfconceptions it produces also operate in a person’s daily life by helping to facilitate social interaction. Others know what to expect of us as they refer to the narrative forms our actions and characters suggest to them. These social functions, however, delimit the range of narrative structures available to the 8

In footnote 10 they claim further that mature narrative requires the elaboration of a center or core situation; shifts in sequential events thus clarify, extend, or modify new aspects of the central theme. Cf. Applebee 1978.

72

The Politics of Persons

person. The necessities of repeated social interaction, for example, require a stability narrative, while progressive narratives aid one’s sense (and the predictability for others) of one’s self-improvement. Further, the social element implies that self-construction is, in part, a dialogic rather than a monologic process. On this account of the self, then, a self is a human being whose “selfrelevant events” take on a narrative structure; this means that they are ordered in a way that is coherent to the subject and surrounding others and that manifests identifiable evaluative shifts from good to bad. This counts as a narrative when such sequences can function in social settings to allow communication and comprehensible interaction. A common idea in these accounts is that a narrative is a sequence of events or experiences ordered in such a way that they are socially recognized as such, and each element of the sequence gains its meaning in relation to the others. But the authors also rely a great deal on the pretheoretic understanding we all have of what a “story” is. They often use the concept of “narrative” to refer to the standard linear forms, while noting in passing that there are certainly non-standard examples of narrative. So for these and the several other related views that utilize a condition of narrativity in this way, a fuller account is needed, of what is meant by claiming that, as a necessary condition of the unity of the self, a reflecting, self-interpreting, psychologically continuous individual must understand her experiences (and be understood by others) as having a narrative structure. In pursuing this project here, I will first look briefly at some ways that a sequence of events (or more generally “elements”) might be connected to make it a “narrative” in order to suggest that in most such accounts of that notion the claim just noted, that the elements of a single self’s life must be in narrative form, is not plausible. After this ground is cleared, we will then look for a non-circular, non-empty account of “narrative” that will distinguish plausibly those lives or personalities that meet the condition from those that do not.

iii. what makes an event sequence a narrative? There is voluminous work in literary theory, linguistics, and anthropology setting out the structural conditions of narratives. Such analyses are especially prominent in structuralist linguistics and narrative semiotics.9 Much 9

See Frye 1957; Greimas 1987; Todorov 1977; cf. also Barthes 1982 and Ricoeur 1986, vol. II; for general discussion see Prince 1995.

The narrative self

73

of this work is relevant here, though in large part it provides theoretical analysis of what are already taken to be narratives – myths, novels, legends, folktales – rather than providing necessary conditions of narrativity in a way that would distinguish such structures from non-narrative sequences (see Leitch 1986). I will try in what follows to cut across the various theories of narrative at a level of abstraction that conveys the full range of possibilities for such models. What is needed for the condition of narrativity to pick out unique structures are conditions that distinguish random-event sequences (or whatever non-narrative structure we imagine) from stories. To clarify the issue, let us first consider three kinds of relations holding among events (or descriptions of events)10 that would characterize an event combination as a narrative: relations of causality, function or teleology, and theme. Causal narrative connections exemplify the classical, linear stories with which we are all familiar. Sequences of events unfold in a causal ordering in which earlier events explain later ones. Aristotle’s idea that events in a plot (muthos) follow one another “by necessity or probability” exemplifies this idea.11 But causal connectivity is not a plausible general condition of all narratives. Even leaving aside recent modern (not to mention post-modern) fiction, many classical and traditional stories include events that are separated in time and place. In many stories and novels, for example, event sequences are told in parallel form, related only very distantly, if at all, by causal connections.12 In other stories, major events occur accidentally or randomly; although they begin causal sequences of significance to the story, they do not themselves occur as a result of earlier narrated events, even if we consider the first event reported in the story as the plausible beginning of that causal chain.13 Clearly, not all events in a narrative form a complete causal chain. But even if causal connectedness were a plausible condition of narrative structure, it would not adequately characterize a necessary condition for the unity of the self. For the experiences of a life do not fall into neat causal 10

11 12 13

For simplicity of exposition, I do not distinguish between reference to factors or events themselves and to the representations – texts, discourses, significations, mental representations – of those events. We will see, however, that the assumed distinction between narrativity applied to events and narrativity applied to discourse (about events) is hardly an innocent one. For discussion of Aristotle’s view in this context, see Velleman 2003, 1–2. In William Faulkner’s Absalom, Absalom, for example, events take place in the (relative) present that parallel, but do not all causally interact with, events that took place in the distant past. Consider Proust’s narrator Marcel in Remembrance of Things Past (À la Recherche du Temps Perdu), whose memories and reflections make up the entire novel but are themselves caused by the accidental occurrence of eating a piece of madeleine cake or stepping on a paving stone. For discussion of time and perspective in Proust, see Ricoeur 1986, vol. II, 130–152.

74

The Politics of Persons

chains (or do so only trivially – see below); and many of the most central elements of a person’s identity are not caused by other elements of the self or by previously experienced events. Many sequences of a person’s life proceed quite independently of each other; plans and projects in one area of experience remain quite separate (in time as well as space) from other plans and projects. Of course, it is true in a trivial sense that all events of an individual’s life are causally connected: the individual herself, as a body, is physically connected to them all. But this fact is captured by the requirement that person-constituting events all be experienced by the same physical subject, making otiose the claim that those events must be connected in a narrative fashion. Moreover, the elements of a person’s life comprise only a subset of the causally linked events in which she (her body) is physically a part. They do not include the trivial (but causally connected) events like the movement of a hair on her head, nor the infinite number of occurrences that pre-date and post-date her physical existence. Self-interpretive activity, which forms the core of the self in narrative theories, is selective and partial, leaving out of account most of the causal sequences in which we figure.14 Also, as indicated above, many of the events that become crucial parts of our life histories are not part of a causal chain of which we are antecedently aware. We experience tragedies or bursts of good fortune that could not be foreseen and which were not planned. After the fact we may come to internalize these events and make them a part of the components of our life, but their occurrence itself is, from our point of view, random. This means that not all events in a life are part of a causal pattern made up of other events in a life. Hence, the narrativity that is meant to comprise life histories cannot be comprised of a single, causally connected, chain of events.15 The second possible connection among events that makes them narratives is a “functional” or “teleological” connection, where events are explained by their contribution to later culminating experiences or events 14

15

An interesting question arises here whether the events that truly do make up a person’s life story must all occur while she is alive. It may be suggested, quite plausibly, that being a famous and published poet is essential to the life story of Emily Dickenson for example, but these events happened after her death. In discussing a short story by O. Henry where three unconnected lines of plot action are described independently of each other, Thomas Leitch claims that “since the connection among the three parts [of the story] is thematic rather than causal, and the stories in their composite form display thematic points about fate and social climbing rather than the teleology of a single action, none of the individual sections is truly independent … only the conjunction of the three episodes makes them retrospectively tellable by endowing them with thematic unity” (Leitch 1986, 47–48).

The narrative self

75

(see, e.g., Greimas 1983). This is picked out by the idea of the “quest” sometimes mentioned in certain theories of the self (see, e.g., MacIntyre 1997). Fictional narratives tend generally to meet this condition (though the novels of James Joyce, Alain Robbe-Grillet, and Gabriel Garcia Marquez, among others, might provide counter-examples). For the most part, all such stories have a “point,” a denouement to which all described events lead. However, even as a requirement for narrativity itself, this condition must be questioned. As Thomas Leitch puts it, “The teleological principle will give only limited help in answering this question [of what a story is], in part because it cannot always distinguish between tellable stories and alternative versions of those stories which are inadequate even though they project the same teleology, in part because it cannot account for the specifically antiteleological tendency of narratives like Pickwick Papers” (Leitch 1986, 63). Also, some stories do not have endings at all: consider soap operas. But even if we accept the view that events in stories are unified by a telos toward which they aim, do lives have such a telos that explains all of their fitful meanderings? Perhaps some do – exemplified by the person who says, “This is what I was born to do.” But clearly many do not have such a telos, at least not any single culminating purpose. In creative reinterpretations of our own existence or that of others, one can certainly postulate a variety of raisons d’être, or final purposes to which all our training and experiences inexorably lead. But these are (mostly) the stuff of fanciful biographies rather than plausible structural accounts of everyday lives. Also, people generally undertake entirely separable projects and goals, which, though each has an internal organizing aim, fail to interconnect in a grand scheme. The events that culminated in my teaching at a university, for instance, are in a functional sequence different from those which lead to my working in my garden (indeed, the latter are experienced as a distinct alternative to the former). Separable purposive strands must clearly be consistent with each other for our lives to be tolerable and coherent, though that condition is so weak as to be satisfied by thoroughly dissociated persons (or groups of persons). Of course, it might be said that a person’s experiences must lead to any one of a variety of ends, so long as each serves some functional role or other, but unless it is specified how many such goals can be pursued, the condition of narrativity, so construed, will be trivially met by all individuals, no matter what level of unity or coherence their lives manifest. Certainly, teleology can be mapped onto many sequences of events as a framework within which to evaluate the sequence, motivating the question of whether the actions have led to anything meaningful in the end. But this use of teleology is plausible in picking out at best only flourishing lives, ones

76

The Politics of Persons

that have succeeded according to this normative overlay.16 But as a general characterization of a life’s experiences or events as such, one that is meant to separate the unified sequences from the dissociative ones (selves from nonselves), this condition will not succeed. Finally, we turn to the idea that “thematic” unity among described events determines their narrativity. In many ways, this is the most plausible as a characterization of narratives themselves, if only because it is the most flexible. Narratives bring together descriptions and representations that are connected under some, perhaps quite general and suggestive, idea. Novels and stories portraying events that neither lead to some single purpose nor causally interact in a linear fashion come to mind. At some level, however, they all present those events in a way that is suggestive of a general idea, value, or moral. So, while many narratives may lack a linear, causal, story line, they nevertheless are meaningful relative to an overarching thematic structure.17 This can be understood as a hermeneutic account of narrative, in that each event in the sequence gains its meaning in reference to its (thematic) connection to the others, and it is this account that comes close to some of the views of the narrative theorists discussed earlier (Schechtman, for example). An essential element of narrative is that the ordering of described events matters crucially to the meaning of any of those events. This need not be a causal ordering, as we have seen, but it can be claimed that in a narrative sequence each member of the sequence is meaningful only because of its place in relation to the others. But it is also important to see how broadly this condition can be construed. For any idea can provide a meaningful schema within which event sequences can be understood. Thematic unity can be established for just about any set of distant and unconnected events.18 David Velleman has examined narrativity as an explanatory tool in a way that is relevant here. On his view, what gives a narrative its particular 16

17

18

Kim Atkins develops a complex theory of the self that relies heavily on narrative (Atkins 2008). In the end, the view I develop here is in general accord with Atkins’s, but as the point here illustrates, I want to resist her view that “narrative identity presupposes an ethical aim: the desire to live well, with, and for others, in just institutions” (ibid., 3). Mark Turner discusses the general use of “parable” to portray events or understand experiences. See Turner 1996. Parables are understandable, however, only as representations of events whose meaning is given in the parable’s theme or moral. A number of novels and films have structures which suggest thematic connections among causally unconnected plot sequences. The Polish director Krzysztof Kieslowski, for example, in films such as The Double Life of Veronique, Blue, White, and Red, makes much use of this technique, where characters’ lives cross incidentally and in ways that the characters are not even aware of, but in a manner strongly suggestive of a variety of themes.

The narrative self

77

structure is that the connection between elements in a narrative corresponds to the natural ebb and flow of emotional expectation, tension, and resolution, much of which may be hard wired (or naturally selected) in us (Velleman 2003). In contrast to those theorists who have suggested that narrative organization of elements provides a cognitive script that focuses attention and taps into our expectations about related events (Shank 1990; discussed in Velleman 2003, 7–8), Velleman claims that narrative structures don’t just tell us what to think, they tell us what to feel. He claims that “a description of events qualifies as a story in virtue of its power to initiate and resolve an emotional cadence in the audience” (Velleman 2003, 18) Moreover, on Velleman’s view narratives must always have or imply an ending: “[a] narrative must move forward not only in the sense of telling one event after another but also in the sense of approaching or at least seeming to approach some conclusion to those events, some terminus, finish, or closure” (Velleman 2003, 10). As we will elaborate upon below, whether or not this is an adequate account of narratives per se,19 it is not clear that lives can adequately be defined as a sequence that is leading toward an ending. If what is meant here is that there must be some single idea that provides the lens through which the experiences of a life can be understood, then this will fail as a necessary condition of the unity of (many) lives. For clearly most lives do not have “themes” in this sense, or an end toward which actions and experiences are leading, just as most of us do not aim at a single telos by virtue of which all our actions are understood. Lives lived “in quiet desperation” might be structured by an open searching or merely quotidian coping rather than by endeavors that aim toward a conclusion. Projects and plans have this thematic structure, but it is unclear that the lives of selves do. Indeed, as I mentioned, many factors that crucially define the path of our lives were unplanned and random, making no sense at all except in that we responded to them and built our subsequent life around them. Lives may have themes (plural) – organizing ideas through which certain projects and periods can be understood – but few lives have one single theme. Consider, for example, those people who have suffered severe injury due to a sudden accident. The injury alters them physically and permanently. The response they must make is “I must somehow come to grips with this” but not (typically) “this makes sense as part of my life.” Yet, there are numerous 19

One worry that might be raised about whether it is concerns stories without ends, such as serials or soap operas. Velleman’s response is that we must distinguish between narratives (which do have or imply endings) and genres which utilize them (Velleman 2003, 18–19).

78

The Politics of Persons

cases where such an event and its consequences become defining conditions of the person’s life, especially when the physical change that results is extreme and permanent. Moreover, in the case of a single subject of experiences, there is always an “interpreter” available to make as much sense as possible of the events she experiences. That interpreter is, of course, the person herself. If one grants that the individual in question is a conscious reflecting interpreter of experiences, then thematic unity of this sort will be achieved whenever the interpreting subject can make minimal sense of her experiences, where “sense” is not specified in advance. The further insistence that the experiences of which she is a subject be narrative in form adds nothing to the analysis. Although associated with the narrative camp, Daniel Dennett comes close to this very point when he talks about how “selves” are created by the interpretive operations of our own self-reflections, whether or not what we are interpreting is structured sensibly or not: “It does seem that we are all virtuoso novelists, who find ourselves engaged in all sorts of behavior, more or less unified, but sometimes disunified, and we always put the best ‘faces’ on if we can. We try to make all of our material cohere into a single good story. And that story is our autobiography” (Dennett 1988, 1029). (For discussion, see Flanagan 1996, 72–74.) It is not that people do not think in narrative form or that the ability to understand stories is not an important skill in the development of a person (see Applebee 1978). But when the idea of narrativity is unpacked, we see the deeper condition lying beneath it. What is truly necessary for a unified life in these theories is the capacity for reflection on events (one’s own and those of others) in a spirit that attempts to render the events coherent within the categories of meaning available to the subject. Now others have been critical of the narrative account of persons in ways that parallel the points I am making here, and one instance goes further than I want to go. Specifically, Galen Strawson has argued that we should reject not only the narrative conception of personhood (and flourishing) but also any diachronic view according to which persons or selves exist over time. I will respond to Strawson more fully in the next chapter, but let me note here the main thrust of his claim and comment upon it in the context of our current discussion. The gist of Strawson’s argument is this: When considering ourselves in introspection, we can take the position of what he calls the “Diachronic” – one who figures herself, considered as a self, “as something that was there in

The narrative self

79

the (further) past and will be there in the (further) future.” Alternatively, there are “Episodics” who do not consider themselves as “something that was there in the (further) past and will be there in the (further) future” (Strawson 2004, 430). Strawson considers both the psychological claim – that all human beings are in fact Diachronics (indeed, in the literature he is attacking, they are more than that, they live their lives as a narrative) – as well as the ethical idea that they (we) should do so. Most of his analysis is aimed at the ethical view, which he denies (he writes: “it would be a great mistake to think that the Episodic life is bound to be less vital or in some way less engaged, or less humane, or less humanly fulfilled” (Strawson 2004, 431)). His overall claim, though, is that the mere possibility of there being “Episodics” in the world – and he counts himself among them – shows not only that diachronic conceptions of the self (and hence narrative conceptions) are not psychologically necessary but also that they are not definitive of human flourishing. There are perfectly good lives lived without attention to their diachronic structure, he argues. Now I agree with Strawson that the narrative view of selves, in the narrow sense that we all do or should live our lives as canonically structured stories, is untenable. My reasons are different, though complimentary, to his in that I claim that standard accounts of stories are such that it is implausible or otiose to claim that all lives have that form, and insofar as narrativity accounts imply that lives are all stories then I reject the narrativity view. But I resist Strawson’s stronger claim that selves need not be understood as temporally extended, “diachronic” in his terminology. My argument for this will turn on the role of memory in practical reason so must be put off until the next chapter, but the position I stake out here is simply that he is correct to reject the (narrow) narrative view I have been discussing but not because of the radical, atemporal view of the self he substitutes for it. This brief discussion of the possible conditions of narrativity was meant to show how, in the standard ways that narrativity can be understood, problems emerge in applying that concept to the idea of a unified self as a necessary condition of that unity. What we are left with is that selves are individuals who reflect on experiences and events in a way that gives meaning to them. But more must be said, for clearly many of the things that I reflect upon and can make sense of include the acts and events of others that are external to my “self.” Although narrative theorists have pointed us in a certain direction, we must go further in specifying what is meant by this condition.

80

The Politics of Persons iv. fleshing out (or moving beyond) the narrativity condition

As touched upon earlier, Daniel Dennett has developed a position on the nature of the self that may be helpful here. On his view, the self is not an actual entity, such as a central controlling mechanism in our brain or a non-physical soul doing our thinking for us. It is also not a core set of beliefs, commitments, and ideas (though he might be willing to say this comprises our self-concept). Like the idea of the center of gravity, the self is a kind of fiction, albeit a quite useful one, indeed perhaps an indispensable one due to the nature of thought and social interaction. Specifically, the self is a fictional center of the process of constructing for ourselves the story of our existence, our autobiography. He writes: Our fundamental tactic of self-protection, self-control, and self-definition is not spinning webs or building dams, but telling stories, and more particularly concocting and controlling the story we tell others – and ourselves – about who we are … These strings or streams of narrative issue forth as if from a single source – not just in the obvious physical sense of flowing from just one mouth or one pencil or pen, but in a more subtle sense: their effect on any audience is to encourage them to (try to) posit a unified agent whose words they are, about whom they are: in short, to posit a center of narrative gravity. (Dennett 1992, 418)

The object of this narrative weaving, for Dennett, is the actions we take and our various characteristics, physical, psychological, and so on, as well as our associations and histories. The self, fictional though it is when considered ontologically, is the center of the process of making sense of those elements. “We try to make all our material cohere,” he writes, “into a single good story” (Dennett 1992, 114).20 David Velleman amends this account by claiming that Dennett is wrong to think that a system of interpretation of this sort is not real. Velleman argues that a central controlling subsystem that weaves stories about us, comprised both of what we have done (and what we are) and what we will do, is the locus of our agency. Psychologically, having beliefs that we will do something can actually bring it about that we do it, as a kind of effective wishful thinking.21 Similarly, the central narrative-generating function in us constructing our autobiographies – the self on this account – generates 20

21

In cases of so-called multiple personality disorder (or “dissociative identity disorder”) there is more than one such process weaving stories and hence comprising selves. Dennett’s account of this phenomenon is developed in Dennett 1992. Velleman works out this idea as a central component in his account of autonomy as well as free will. See Velleman 1992.

The narrative self

81

both the account of our actions as well as the actions themselves, as we (as thinking and acting mechanisms) make our future actions cohere in our autobiographical narratives by bringing it about that they occur. The force of narrative unity amounts to agency, for Velleman. As we saw above, however, if by “narrativity” here we mean that account of emotional expectation and resolution that describes the unifying principle of narrative explanations, this will not quite do as an account of selves, at least not without further amendment and elaboration. Let us say more, then, about what narrativity amounts to if, as we have done, we put aside standard accounts of stories in the canonical sense, but keep alive the idea that selves have the function of constructing accounts of our lives and persons. As we will see, however, this function is as akin to literary criticism as it is to authorship. A general complication in utilizing narrativity as a structural condition of selves is that narratives cannot be specified independently of a context of discourse, where a settled semantics is operative and there exists an audience interpreting the events in question. Leitch, for example, surveys the variety of ways that narratives should be characterized and concludes that the essence of stories are their “tellability,” the degree to which they will make sense to a certain kind of audience. Leitch claims that in narrative theory “the whole concept of story depends on a context which involves not only a particular discursive mode and selection of states of affairs but also the particular circumstances governing the storytelling transaction” (Leitch 1986, 25). Moreover, narrativity involves an interplay between discourse and active interpretation by an audience: “The audience’s narrativity should be active enough to recover a given story according to the cues the discourse provides but not so active that it transgresses that discourse’s guidance” (Leitch 1986, 40). Narrative is a transaction between discursive sequences and the constructive activities of audiences. This transaction involves revelation of contingencies, trading on expectation and satisfaction, and interpretive projections (filling in missing information, since no discourse gives a complete account of events). So narrativity is both constructed and grasped by way of public languages and shared social meanings. An idea similar to this is echoed in the work of Paul Ricoeur (Ricoeur 1986). For Ricoeur, narrative unity is the structure that provides a bridge between the purely descriptive conceptions of the self found in many (analytic) theories of mind, identity, and action, he claims, and prescriptive accounts of agency and character. However, a purely structural (atemporal and ahistorical) account of narrative that focuses solely on internal syntactic relations – what Ricoeur calls the “logicization” of narrative – fails in this

82

The Politics of Persons

context, for narrative inevitably depends on readers’ understandings of narrative traditions. Narrativity, on Ricoeur’s view, is not a fixed semantic structure but an ongoing process of construction of interpretation reflexively engaged in by a person in communication and interaction with others and embedded in a particular historical context: “Self-understanding is an interpretation; interpretation of the self, in turn, finds in the narrative, among other signs and symbols, a privileged form of mediation; the latter borrows from history as well as from fiction, making a life story a fictional history or, if one prefers, a historical fiction, interweaving the historiographic style of biographies with the novelistic style of imaginary autobiographies.”22 Ricoeur utilizes these observations in building a theory of personal identity, one that distinguishes between identity as sameness (idem) and identity as selfhood (ipse) (in a way that echoes Schectman’s views). What a theory of personal identity must do, he claims, is to establish a basis for the permanence of the person over time. One of the central elements establishing such a basis is character, a set of lasting dispositions by which a person is recognized. But the specification of character must make reference to habits: such habits cannot be seen as fixed rules of behavior mechanically reacting to environments but must be a set of dispositions that make sense in a flow of reactions. The model of such a set of dispositions is a narrative character, a principal around which a story is told (Ricoeur 1992, 119–22). In this way, Ricoeur’s theory of personal identity rests on a conception of narrativity. But notice that narrativity here merely refers to the patterns of action and habit that make sense of a person’s character, where “sense” is given by common understandings operative in the person’s social setting. Again, narrativity becomes a placeholder for whatever organizing principle describes the pattern of experience and action produced by the person and interpreted by her with the use of socially produced and embedded rules of meaning. The concept of narrativity, then, can be jettisoned in favor of this more nuanced descriptor and, what is more important, in favor of a detailed analysis of what constitutes socially mediated self-reflection for a typical person. 22

Ricoeur 1992, 114n. Also, for Ricoeur no purely structural analysis of narrative will suffice, because it is essential to understand the narratives of persons (as well as fictional narratives) in terms of the “characters” or “roles” that are their center. On the other hand, theories of narrative that center wholly on character or role are themselves incomplete without reference to how a character profile becomes a story. In the end, accounts of narrativity must make essential reference to historically embedded traditions of storytelling. See Ricoeur 1992, 19–22; for discussion and adaptation of Ricoeur’s views see Atkins 2008.

The narrative self

83

So, what must be directly investigated by theorists of the self are the various dimensions of “making sense” of the particularities of lives. “Narrative” is merely the name of whatever results from this process, but one that must be shorn of the connotation that there is an independent condition of linear or semantically structured connectedness that experiences must conform to in order to constitute a self. While “tellability” indicates that the sense made of life-event sequences must be shaped by surrounding standards of meaningfulness, the process of interpreting itself will be the activity that persons – and only persons – can undertake. More must be said, then, about the capacity for self-understanding and interpretation that lies beneath the purported narrativity of selves’ experiences, leaving behind the view that the result of such reflection must have some particular form or other.23 If I am correct that “thematic unity” and “tellability” are the most plausible desiderata for narratives, the condition of narrativity will be met whenever a reflecting subject is able to interpret the events, memories, and impressions she experiences and make some sense of these according to socially mediated semantic rules. And so narrativity is defined in terms of the capacity for self-interpretation, where the latter is manifested whenever the conscious, psychologically connected personality can turn her attention to the experiences of which she is the subject and find them meaningful. Also, what is crucial is not that the sequence or collection of character traits is simply coherent but rather that it can be seen by the person as coherent as her own. This last attitude, seeing something “as one’s own” is, I think, primitive: it is a sense of a sequence of events, actions, and traits as ones I subjectively follow through, as making sense (in part at least) because I live through them, because I live through them.24 Hence the critical analysis of narrativity has motivated the view that meaningfulness within an established context is the mark of narrative unity, collapsing the condition of thematic unity into the “tellability” referred to by Leitch and, indirectly, by Ricoeur. What narrativity amounts to, then, is whatever results from the capacity for selfinterpretation mediated by socially embedded rules of meaning. Therefore, what unifies a self is the capacity for self-interpretation by way of socially mediated norms. 23 24

For a line of analysis that shares much with this view, see Kerby 1991. As Atkins has argued, seeing selves as the product of narrative helps show how identity is irreducibly first-personal: see Atkins 2004.

84

The Politics of Persons

“Social mediation” here operates at two levels. The first is the use of cognitive structures and semantic rules to formulate the terms in which selfreflection takes place. When an individual reflects on her personal characteristics, actions, memories, and the like, she must use concepts gleaned from the language(s) she speaks and the subtleties of meaning provided by the social world within which that language is developed. Second, the degree to which that self-interpretation is unified or meaningful turns on the degree of understanding achieved in one’s interaction with actual and potential “interpreters” of that account. As I have argued, stories are not set out apart from a community of interpreters; hence the internalized biographies that unify selves are similarly not insulated from such a community. This picks up on conclusions we reached in chapter 2 concerning the social constitution of the self. What is required for unified personhood is that the subject of a life is a reflecting subject whose self-interpretations make enough sense of those events that a consistent character can be seen at their center. But what are the constraining conditions of self-interpretive reflexivity necessary for a unified self or personality, beyond what has been said by narrative theorists themselves? First, the process of making sense of our experiences is more flexible than the narrative theorists suggest. Constraints on the “sensemaking” that is involved in the realization of people’s self-concept are obviously socially relative, but they are also individually relative: dream stories often make perfect narrative sense to their subjects, even if they are boringly disconnected to the rest of us. Also, reflection on the events of a life has a particular phenomenological character, and purported “experiences” that a person cannot make full sense of will be ones from which she is in a strong sense alienated. That is, reflection upon them will not issue in the usual affective and intentional responses appropriate to authentic self-reflection – a sense of internal memory with attendant proprioceptive sensations; feelings of regret, pride, embarrassment, and the like, will be lacking. The events and experiences will also lack certain motivational concomitants of authentic first-person memories, motivations connected with responsibility or regret, for example. They will feel external and foreign. And they will not make sense as part of a collection of other experiences with a common theme (cf. Strawson 1999; for commentary, see Sheets-Johnstone 1999). Finally, self-interpretation must serve the pragmatic purpose mentioned by Gergen and Gergen and others. While it is important to note the segmented nature of much of self-interpretation (few of us have woven an over-arching model of our entire lives), each component of this

The narrative self

85

interpretation must serve the purposes of interacting with, and explaining ourselves to, surrounding others. Providing reasons for actions, explaining judgments and decisions, will utilize self-interpretations in ways that are constrained by norms governing the adequacy of such reasons and explanations. Without a full-blown theory of action, one has to be extremely abstract here of course, but the important point is that selfreflective activity serves the additional purpose of explaining actions and decisions and hence is further constrained by norms operative in those settings. Of course, a full theory of self-interpretation is beyond the scope of this chapter; when it is developed it will have to make much use of psychological (and social-psychological) theories of the self. But unpacking the narrativity condition in the way I have done should refocus theoretical attention on the process of self-reflective meaning making, rather than on its structure or organization.25 The process of sense-making is, as the narrative theorists attest, an essential element of constructing the self-concept. But there is clearly a multi-dimensional interplay among a variety of dynamic elements that form parts of this process, and the question of whether such a process is one of retrieval, interpretation, or pure construction is left to be answered by analyses that take up where my criticisms leave off. Here, however, I hope at least to have moved that analysis in a more fruitful direction. Finally, the view of socially mediated reflection sketched here is spelled out in strikingly atemporal terms, as if a self is simply a being reflecting on her characteristics in a time-slice manner. But of course the processes of introspective sense-making I have been describing have as their object experiences, actions, feelings, and judgments that are grasped through memory, events that form a coherent collection in time. In the following chapter, then, I want to turn direct attention to these mechanics in order to show how the psychological structure of remembering helps fill out the picture of the socio-historical self. 25

For an example of work that attempts to flesh out the psychological underpinnings of self- interpretation, see Turner 1996. However, Turner makes much use of the concept of narrativity without unpacking the notion in the manner I urge here.

chapter 5

Memory, agency, and the self

We all walk backward into the future Maori proverb

Our discussions of various approaches to conceptions of the self have taken a few turns now. We noted how the standard model of the self passed down from the Enlightenment tradition of political and moral philosophy is notoriously individualistic. It was necessary, then, to consider ways to revise that model in light of the social constituents of selfhood that undeniably play a role not only in our self-concepts but in any useful conception of agency or autonomy. This mirrors much work in the philosophical literature to revise traditional models of agency in order to highlight the ongoing connections we all have with surrounding social dynamics and with particular significant others, connections that mold our identity and shape our values. Curiously, however, much less notice has been taken of the ways in which that standard account of autonomous agency has severed our connections to our past. On many standard models, rational agency involves the survey of current options, desires, and beliefs, and the capacity to choose the optimal path forward. But little is said directly about how those choices arise out of an ongoing historical narrative, access to which is gained through the powers of memory.1 Therefore, I want to consider how reflective self1

There is much discussion of memory in theories of personal identity of course (e.g., see Schectman 1996). In discussions of practical reason, memory is treated less often. An example that does consider memory in that literature is Velleman 1996; see also Margalit 2002. Theories of autonomy that include discussion of memory include Bratman (especially in Bratman 2007, ch. 2) and Cuypers 2001. Outside of analytic theories of agency and practical reason there is, of course, greater attention to memory; see, e.g., Ricouer 2004. I should make clear that in this chapter I am discussing psychological and philosophical conditions of being a continuous subject or self as that is generally experienced. This is not the same as giving metaphysical conditions for the identity, over time, of persons. Indeed, the traditional criticism of memory criteria of personal identity corresponds to the point I make in the text, namely that memory cannot establish personal identity since having veridical (autobiographical narrative) memories presupposes sameness of self.

86

Memory, agency, and the self

87

construction must take into account not only current elements of one’s person but aspects of our past, access to which is given to us by memory. As we will see, the particular way that memory works will complicate, though hopefully in fruitful ways, any views of the self that emerge as useful in social and political theory. Specifically, I want to suggest how viewing memory and remembering in the right light illuminates a model of agency that sees selves as temporally extended, diachronic projects rather than fixed repositories of interests or modes of identity. The procedure here will be first to look at the psychological structure of memory and to focus on the kind of remembering that is most relevant to our inquiry. Then, in order to investigate the value of memory in the construction of the self-concept, I want to investigate those cases where memory is lacking. This will enable us to get clearer on the precise relation between remembering and one’s sense of self. As we will see, memory is related to self-concept but not in the most straightforward way that might at first suggest itself. I will then close with a discussion of the implication these observations have for a workable conception of the self.

i. autobiographical memory There are many phenomena that count as memory, and while a detailed survey of the vast literature on the psychology of memory is not feasible here, we should get clear about the precise kind of recall that is of interest. (For an overview, see Brewer 1986; Conway 1990; and Tulving 2000.) Three large categories of memory are memories as skills, recognition memory, and experiential memory. Cutting across these groups, though, are kinds of memory that are either explicit or implicit. Skills are typically implicit, in that remembering how to do something often does not require conscious consideration of that skill (in fact explicitly thinking about how to perform the act will often disrupt it, as musicians and athletes can attest). Although these are called semantic memories, since they structure our abilities to carry out everyday actions and speech, they are not confined to linguistic skills. All manner of bodily capabilities are remembered in this way. As I said, we do not normally have to bring such memories into consciousness for them to operate. Other types of recall, however, involve bringing things explicitly to consciousness. While I will explain below that this “explicitness” is not a simple or continuous experience of selfsufficient scenes or narratives, the distinguishing feature of these memories is that they contain elements which are brought to consciousness. Without

88

The Politics of Persons

such conscious components, we would not say we remember the fact or event in question.2 This points to another relevant subdivision: recall of facts as such and recall of experiences of which we were the subject. There are things about my life that I remember that they happened – my birth for example – but not ones I remember as the subject (I can’t remember being born). Other facts of course don’t involve me at all, and I remember them as part of the broad knowledge base that makes up my epistemic field. I want to focus on those memories with an experiential content, though, and so those that are “personal” in this way, while keeping in mind of course that personal memories may or may not be explicit in the way I’m using that term. (Indeed there is much discussion and some controversy about how to classify memories as “personal” at all.3) Personal memories can be extended experiential sequences or mere “snapshots” of events or actions. The latter are single experiences whose meaning is fully contained in the snapshot or episode itself. Some of these may contain rich content, such as the recall of a significant event, or they can be quite simple, such as the memory of seeing what time it was the last time you looked. Finally, there are memories that are structured as extended narratives, sequences of experiences, actions, and events that occurred over time and which partly constitute the story of one’s life, as we discussed in the previous chapter. The segments of that story that we do or can explicitly call to mind are the narrative autobiographical memories which will be the object of our analysis here. It is not that other sorts of recall are not important for discussions of the self, but narrative personal memory is particularly crucial, as we will see. These are memories that can be brought to consciousness that are, in part at least, experiential and personal and extend over a span of time whose length and structure is essential for their meaning. My memory, for example, of getting my first academic job is composed of experiences and actions that extend over at least months if not years and whose beginning point will in some ways be arbitrary, depending on the defining conditions of that event. That memory cannot be understood as simply a series of disparate snapshots or facts about me. What makes this a unique personal memory is that it is composed of those complex elements which I bring to 2

3

A similar distinction in types of stored memory has been drawn between “declarative” and “procedural” information. For discussion see Klein et al. 2004, 462. See also what Wollheim describes as “event memory” (Wollheim 1984, ch. 4). For discussion of this issue, see, e.g., Barclay and Smith 1992.

Memory, agency, and the self

89

mind now and which I experienced first-hand earlier. This description of memory makes use of a distinction which will be important in what follows, namely the distinction between simply knowing I did or experienced something and remembering it. It is not possible here to adequately survey the vast array of psychological work on memory, although the systematic study of autobiographical narrative memory of the sort that will concern us here is of fairly recent vintage. Various aspects of remembering are studied in this literature, including questions of accuracy of memories and the possible phenomenon of suggested memories.4 However, of particular interest for us is what has come to be the accepted model of remembering in this literature, at least in its broad outlines. It is generally accepted now that narrative memory does not involve simply a passive experience of a structured phenomenological sequence that is somehow stored in our brains which we call up and, as it were, “watch.” Memory is not simply retrieval of encoded moments whose structure and meaning are set and which our current selves simply access in their original form.5 Just as the original encoding into memory is not simply a recording of events akin to filming the scene, the retrieval process is not a passive apprehension of structured material. In both cases the process is active – the brain arranges and encodes material in a complex and nonlinear manner – and reconstructive – the remembered narrative is assembled, as it were, at the moment of recall. This is not to say that a memory “trace” – a causal chain of some sort bearing isomorphically structured elements of the original experience – is not involved in veridical memories. But it is clear that recall is not simply passive storage and retrieval on the order of a replayed film. When we remember we construct a narrative based on stored data, as well as current cues, cognitive and affective frames, and external stimuli, all as part of a process to frame the features of the events being remembered into a coherent sequence (if, that is, we want to talk about the “facts” of the original events independent of the later representation of them6). As Daniel Schacter puts it: “We now know enough about how memories are stored and retrieved to demolish [the myth that] memories are passive or literal recordings of reality … it is now clear that we do not store judgment-free snapshots of our past experiences but rather hold on to the meaning, sense, and emotions these experiences provided us” (Schacter 1997, 5 cf. also Bartlett 1932; Brewer 1986). 4 6

For a recent overview, see Tulving 2000. 5 For a similar point, see Schechtman 1996, 124. See Hacking 1995, ch. 8 for discussion of this issue.

90

The Politics of Persons

These points can be seen readily by imagining the structure and content of simple narrative memories. Recall a recent sequence of events; in “replaying” the sequence, say of eating dinner last night, you will leave out myriad details from the original experience that clearly were recorded as part of the sensorial array – the smell of the food, the feel of the utensil in your hand – but are not currently part of the memory. This is because such details are not, we can assume, relevant to the aims of currently remembering the event. But it is clear you can recall some of these details if prodded – perhaps you just thought about the smell of your food last night when it was mentioned – but given that the social structure of the recall event did not call for a focus on that detail, it was not part of the memory. Similarly, we “fill in” details into a current memory that may or may not have been part of the original experience. We may not have recorded the color of the table upon which we ate, but when we remember the scene, such details are filled in to give the remembered sequence coherence and sensory density (after all, what would a memory of eating on a colorless table be like?). Elements of memories are added and subtracted as we reconstitute the experience which is being recalled.7 This shows two things, at least: first, memory is not, as I said, a passive replaying of structured data, and, second, memory structure is shaped by the pragmatic and social factors that spur the particular act of remembering. Why we remember something is fundamentally important for what we remember. Further evidence for the reconstructive nature of memory comes from realizing that we often shift perspective in remembering. It has been clinically shown that subjects often recall events which they experienced from a third-person point of view, showing that the memory is clearly a restructured version of the original scene (Brewer 1986: 42; see also Wollheim 1984, ch. 4). This reconstructive account of memory, sketched here in very broad terms, has become a commonplace element in psychological theories of remembering and recall. Psychologist Martin Conway provides what might be useful as a general model: autobiographical memory is a complex interaction among several cognitive processes and affective functions, including generative retrieval of encoded phenomenal records, hierarchically organized thematic knowledge structures that provide coherence to such phenomena, and a complex self-schema that is both imposed upon and 7

The novels and screenplays of Alain Robbe-Grillet offer an example of portraying events and images in isolated, snapshot fashion. The disorientation that results makes the point, I think. (See, for example, the novel La Jalousie or the film Last Year at Marienbad.)

Memory, agency, and the self

91

arises out of that retrieval process (Conway et al. 1992). Note how affect and emotion are also intrinsic to the process, as remembering both cues and depends upon feeling in relation to recalled narratives. Moreover, the entire process is embedded in socially constituted tasks, responding to requests or expectations, guided by coherence standards for event sequences, and so on. The process is active and constructive but constrained by encoded phenomenal records, standards of coherence and plausibility, and affective reactions. More importantly for us, and a point to which we will soon return, the process both presupposes and produces an ongoing sense of self, a selfschema, which, on the one hand makes the memory autobiographical (per se) and provides standards for memory veridicality (see, e.g., Bluck 2003; Klein et al. 2004). The individualist tone struck in the account just sketched should not obscure the important fact that remembering is often undertaken amidst interpersonal interactions and social expectations, factors which directly shape the structure and content of memories (see Fivush 1994; McLean 2005).8 These social factors shape the conditions of plausibility for memories, though of course veridicality cannot simply be reduced to criteria of social acceptability. Indeed, the interaction between social and interpersonal factors and faithfulness to actual past events is at the nexus of complex debates surrounding so-called recovered memory syndrome. (For discussion see Hacking 1995.) We know that memory is not fully voluntary or simply a matter of decision, as that would fail to distinguish it from acts of imagination.9 Yet, the specific components (images, sequences) that constitute a given memory to some extent depend on the interpersonal setting which sets the standard for whether reported memories make sense. The main point here is that memory is a socially structured, active process, an interpretive understanding of one’s experiences with attendant affective responses and judgments, organized by a working self-schema. In this discussion I am largely skirting issues of veridicality for memories. That question is quite complex, of course, as criteria for valid remembering involve issues of self-respect (considerations of respect may ground an obligation to take claims of remembering seriously) as well as standards of justified belief. I will only say here that valid memory narratives must at least

8

9

It is also cast in a largely mentalistic register, underplaying the way our bodies remember. This is a result of my focus on explicit narrative memories. For an alternative approach to the self that might approach memory from a different angle, see Meyers 2005. On memory and the imagination, see Velleman 1996. While memory is not voluntary in this sense, there may nevertheless be obligations to (try to) remember. For discussion, see Margalit 2002.

92

The Politics of Persons

conform to current facts which are the causal progeny of original remembered events. This is a very weak criterion, of course, as innumerable current stories are causally consistent with what we might know about the past. And that is the point: what creates memory is more than reporting some event or events; it is constructing an interpretation guided by a sense of oneself and the social expectations and meanings that shape the standards for such interpretations. Much more could be said about the psychology of memory, of course. Let us move on, however, to begin exploring further the relationship between remembering and the self-concept. One way to approach that question is to ask what it is that is valuable about remembering for a person. If we locate the particular and unique value of memory we might hone in on the way that memory and self are related. Of course, memory has instrumental value in providing information about current options, causal sequences, and so on. Indeed, in a way all learning is a function of memory and so it is valuable in the way knowledge is. But clearly there is something intrinsically important about remembering that is independent of these functions. Not merely the pleasures of reminiscing – though much remembering is painful and we would not say that means it should be erased for that reason – what is it that makes the capacity for and practice of remembering our pasts valuable for us in itself? How should we understand the non-instrumental value of personal memories and/or the constitutive role it plays in phenomena that are themselves intrinsically valued? By extension, what implications does this value have (whatever it turns out to be) for plausible conceptions of the self? One straightforward answer to this question is this: remembering is both necessary for and produces our self-concept. The value of remembering simply inheres in our valuation of ourselves, or to put it less tendentiously (and to make room for regret and shame), memory is necessary for having a self-concept, so insofar as having a self-concept is non-instrumentally important, memory inherits that value. However, as we will shortly see, things are not so simple. More must be said about the structure of both memory and the self-concept in order to grasp the meaning and import of this hypothesis. But to first get clearer on the value of memory and its relation to self, let us take a slightly different tack: let us continue this examination by looking at cases where memory is lacking. In this way the proposed relation between memory and self-concept will be explored and made more complex in what I hope are productive ways.

Memory, agency, and the self

93

ii. amnesia, memory, and selves Our attempt to gain insight into the structure and value of memory invites examination of cases where it has been lost, and hence of amnesia in its various forms. Loss of recall ability occurs in a variety of ways and under different conditions, including, perhaps most familiarly, memory losses due to aging and/or Alzheimer’s disease. Many sufferers of amnesia might also lack various other cognitive skills that render them less able to concentrate and reflect, not only due to memory loss strictly speaking but from inability to focus or from constant frustration or other conditions. However, in order to keep clear on the question of the value of memory itself, I want to focus on amnesia patients who seem to have maintained a range of cognitive abilities relevant to decision-making and action but lack memory capabilities per se. There have been important psychological studies of memory loss to which I will turn in a moment. First, though, let us examine a set of cases presented outside the clinical context. In a BBC documentary Living Without Memory a profile is presented of three individuals suffering from anterograde amnesia that resulted from disease or injury.10 All three have only limited short-term memory and must proceed via cues and reminders to get through some of the most basic daily tasks. In all three cases, the individuals are cared for by close relations and can generally but minimally function with the aid of such care. Debbie, for example, needs a list in order to bathe and dress herself each day. She could not recall a ride in a helicopter with her father 45 minutes after it occurred. Sally, who suffered from encephalitis at age 8 – she is about 25 in the film – has lost all ability to form and maintain new memories for longer than about 20 minutes. She keeps a diary and must write down everything that happens of significance in order to take note of it later. This, along with her parents’ care and the help of other friends, allows her to undertake various normal tasks, such as volunteering at a community center. She feels constant frustration in not remembering previous experiences and plans, but she also laughs at her own forgetting. She also is able to comment on others’ reactions to her and spontaneously report that her life is unusual because of her “memory problem.”11 10

11

“Retrograde amnesia” refers to the inability to recall events in one’s life prior to the onset of a particular condition (what might be thought of as loss of long-term memory). “Anterograde amnesia” refers to the inability to remember events subsequent to such an onset (loss of short-term memory). Much more should be said about these cases of course. For example, these individuals retain a set of skills, semantic memories, body-schemas and other elements of a sense of self. For discussion of such issues see Gallagher 2005.

94

The Politics of Persons

In watching these individuals it is clear that their lives are difficult and impaired. But one would never say that they lacked a sense of self. In fact it is clear that they all have a quite robust understanding of themselves as persons struggling with memory issues. As I said, in many cases of memory disability, patients exist in such a state of confusion and without normal abilities to concentrate that it would be hard to say that they maintained a coherent concept of self in any sense. But it is also clear that in other cases when pathologies are restricted largely to memory loss amnesiacs are at least able to maintain a conception of themselves in the minimal sense of knowing who they are, what they want, and often what is needed to get it, even if they often cannot satisfy these needs because of their condition. Laboratory experiments also suggest that often amnesia involves failure of recall abilities but not storage capacity, since patients can often recall events when sufficiently cued at success rates approaching normal levels, though temporal ordering of such events is systematically inaccurate. This suggests that narrative reconstruction of stored material is a separate cognitive function in the process of remembering.12 Other clinical research on amnesia suggests a similar conclusion. For example, Hirst studied brain-damaged patients whose cognitive deficits seemed confined to memory and claims that their understanding of themselves is remarkably developed (Hirst 1994, 255). Klein and his colleagues similarly argue that memory involves a variety of cognitive functions and these can fail separately as well as together. But he argues also that “[e]ven in severe cases of episodic memory loss covering a person’s entire life … individuals so afflicted appear capable of appreciating the present and anticipating the future” (Klein et al. 2004, 467). Hirst also argues that amnesiacs lack the ability to participate in the social dynamics that in part constitute remembering (collective recall). Memory contributes, he claims, to the social construction of the self. He stresses that social interaction is fundamental both to remembering (internalizing those externally generated cues from others and the environment that secure a memory, for example) and to development of the self. Self-representation as well as self-presentation is structured by social expectations and dynamics. Amnesiacs show the cleavage between these two phenomena when they represent themselves in ways that cannot be sustained by their current presentation (such as the retrograde amnesiac who wore three-piece suits

12

Hirst concludes that “[a]mnesia is not a complete breakdown in memory functioning. Certain specific and circumscribed aspects of memory are preserved” (Hirst 1994, 263).

Memory, agency, and the self

95

but kept describing himself as a working lumberjack, something he hadn’t done for several decades). What these cases show is that the initially plausible answer to our central question – that autobiographical memory is valuable because its function partially constitutes our self-concept – cannot be maintained in its most straightforward form. Insofar as the amnesiacs in question do have a concept of self, then autobiographical memory is not strictly speaking necessary for such a sense. The cases in question show more pointedly perhaps that it is the idea of a self-concept that is more complex than we first assumed. For while these amnesiacs maintain an idea of themselves in one sense, they clearly lack a conception of themselves as temporally extended beings whose life experiences are narratively structured. In Hirst’s discussion of amnesiacs, for example, he claims that “amnesics’ [sic] narratives of post-onset events resemble plain, unpatterned cloths. They do not have the texture that particular and concrete remembrances give to a narrative” (Hirst 1994, 271). But he is quick to stress that close study of such subjects indicates that we must go beyond any simple idea of self-concept to understand the relation between memory and self. For these patients not only show a working self-concept but are also, he argues, capable of change in that self-concept. One patient he describes went from expressing a confident determination that his memory condition will be treatable to (eventually) a resigned stoicism about living with his disease. Ulric Neisser suggests as many as five separate notions of “self” that should be distinguished in psychological studies of persons and their idea of themselves. These include such structures as the private self; the ecological (embodied) self; the interpersonal self; the conceptual self; and the temporally extended self (Neisser 1988). The amnesiacs just described had a concept of themselves in some of these senses but not others. They have a conceptual self, for example, in that they can form a working idea of their central characteristics. But what they specifically lack is what Neisser calls the “temporally extended self” – the understanding of themselves over time and as the subject of ongoing plans and remembered actions and states (Neisser 1988; see also Baumeister 1998).13 Therefore, memory is necessary for our self-concept in this diachronic, temporally extended sense.14

13 14

They may also lack or at least have only a constricted version of what he calls the interpersonal self, since their relations with others are highly restricted due to their inability to share remembered events. Goldie comes to a very similar conclusion in his analysis of narrative thinking, memory, and the construction of the self-concept (see Goldie 2003).

96

The Politics of Persons

This conclusion may strike many as banal, since it merely points to the relationship between memory and personal identity (see Parfit 1984). But I am not referring to the metaphysical question of what accounts for the sameness of a person over time, but rather to the relationship between autobiographical remembering and one’s sense of oneself as an agent.15 Moreover, theorists of personal identity (with some exceptions) have assumed a rather uni-dimensional model of memory. (Theorists who discuss the classic memory criterion of personal identity, for example, often see memory as merely a connector among psychological states, akin to anticipation: see, e.g., Parfit 1984.) In the view I want to develop here, the constructive nature of autobiographical recall, as well as its social and affective components, play a central role. What this shows, then, is that autobiographical memory provides one with the basis of one’s temporally extended self-concept and that the role it plays in this process is not simply passive recall of intact representations but active construction of self-narratives. What amnesiacs lack is the ability to formulate representations into conscious and organized mental constructions (even when they are affected by the affective or other motivational implications of those past experiences). The value of memory that arises from this understanding of its function is, as I mentioned, its role in the construction in this sense of the self, something important for its own sake.16 However, as we discussed in the last chapter, construction of the self as narrative is not simply the telling of one’s personal story which is already in the form of a tale or canonical narrative. It is an ongoing process of socially mediated self-interpretation. Similarly, remembering is not simply calling up a fixed, narratively structured, sequence of representations or episodes; it is an active construction of a valid sequence out of basal data formed in light of standards of coherence, which include normative standards. Therefore, identity, as this is produced and filled out by memory, is not simply a set sequence or array of characteristics, but involves an ongoing project of selfinterpretation, including interpretation and reinterpretation of our pasts. To have a narrative self-concept then, is to have a sense of one’s actions, 15 16

However, the discussion does overlap with some recent approaches to the issue of personal identity, for example in Schechtman 1996. Another way to put this is that normal autobiographical memory and reflective, temporally extended agency, are mutually constitutive (see Klein et al. 2004). That is, one cannot remember in the normal fashion without being able to self-reflect, think of oneself as a causally effective agent, and have a sense of temporally located subjectivity. Memory involves the assumption of agency as well as constitutes a sense of oneself as an agent over time. What is clear, however, is that remembering in this way actively constructs one’s sense of self as a diachronically structured agent.

Memory, agency, and the self

97

commitments, and traits as these have developed over time understood by way of the capacity to (re-)interpret these factors in light of new experiences, reflections, and interactions. That said, I would like to turn now to some (admittedly exploratory) reflections on the implications these conclusions have for the philosophical understanding of practical reason, and in further ways, the self. iii. memory, value, and practical reason A straightforward way in which memory is valuable non-instrumentally, then, is that it is constitutive of various processes of self-understanding and self-appraisal that are intrinsically important for us. Introspective reflection with the aim of appreciating how our lives are going, how were are fairing in regard to our experiences, accomplishments, relations with others, and countless other self-directed attitudes, involves autobiographical memory, since this is necessary to bring to mind the crucial elements to be evaluated. Self-reflection is modeled as introspective appraisal of the beliefs, desires, principles, and motives that currently constitute the person’s value profile.17 Taking an evaluative stance toward one’s life requires that one access memories of experiences and choices in the past as well as current conditions and prospects for the future. Indeed, our current condition and future prospects, in some respects, have no meaning apart from the historical trajectory of which they are a part.18 The question, though, is what particular relationship memory has to practical reason, in addition to being simply the source of information about our current options and desires. The way we should pose the question at issue here is this: what effect on the structure of reasons for action does remembering X have, where X is a temporally extended sequence of life events, that simply knowing that X occurred (without remembering it) does not? The beginning of an answer relates to the connection between remembering and (currently) deciding. Consider, for example, Michael Bratman’s account of intentional action which he calls the “planning model.” A rational intentional action, for Bratman, is one which is a part of a plan (or what he calls a “partial plan”) that is rational for the agent to adopt and 17

18

For a discussion of the role of memory in the construction of the self-concept with which my account has much overlap, see Wollheim 1984. For an interesting discussion of similar issues, see Mackenzie 2000. For an argument that such appraisals involve more than merely a summation of the happiness experienced in life, see Margalit 2002, 131–139.

98

The Politics of Persons

not irrational for her to fail to reconsider. In that way, first-order desires that are not reflectively considered at the time of action are nevertheless rational because they fit with a longer-term plan that it is rational for the agent to have adopted and to continue to follow (Bratman, 1987, 1999). Bratman adopts this approach to capture the obvious fact that ongoing or recurring reflection is not a plausible requirement for rational action. Most action involves habit and automatic response that not only fails to involve reflection, it sometimes precludes it. Such an account, on my view, plausibly underscores this unreflective element of everyday actions (even ones quite important to us).19 But note, it must at least be possible in principle to reconsider ongoing plans for actions based on them to be rational. One must be open, that is, to reconsideration when conditions arise that make such questioning called for. Therefore, it seems obvious that one must remember the initial motivation and justification for the plan itself. Without memory, one cannot justify actions as part of ongoing strategies, an aspect of virtually all action. In this way, memory is an operational component of intentional, rational action, and not merely an instrumental source of information with regards to it.20 Though perhaps we have drawn this conclusion too quickly. Is it necessary to actually remember making and confirming plans in order to currently know that such plans remain rational (to not revise)? Does this not show only an instrumental role for memory, in providing the information that supports a justification for my plan? And can I not rely on other sources of information about what is best for me to make this determination? The key to these questions will involve making use of our earlier conclusions about memory and narrative identity. Remembering is necessary for me to know that it is my plan, that it fits into a coherent or currently acceptable selfnarrative (my temporally extended self-concept). It seems that I must be able to remember not merely making the decision to adopt the plan (since not all rational plans were adopted by discrete acts of choice)21 but to at least remember the actions and experiences that make sense to me only on the assumption of the guiding influence of such a plan. That is to say we must either remember making a (rational) decision to adopt the plan or 19 20 21

For discussion of this point, see Christman 2001. This is a point developed by Bratman in his more recent work: see Bratman 2007, for example. This last claim marks a disagreement with Bratman, in that I do not think that authentic and rational plans need to have ever been the product of a conscious decision to adopt them (see Christman 2001; see also Berofsky 2003). This may also mark a slight disagreement with David Velleman, who sees memory as requiring merely perspectival commonality of the nominal subject of experiences and not identification in a stronger sense. See Velleman 1996.

Memory, agency, and the self

99

remember a sequence of intentional actions whose rationale necessarily involves the having of such a plan. If that is the case, we have to show that the role played by memory in constructing and maintaining our temporally extended self-understanding is important for current reasons for actions and our autonomy relative to those actions. What the amnesiacs we considered earlier lack is the ability to consider plans of action as their own in the way that those with normal memory capabilities can. They certainly can construct a set of plans and life trajectories for themselves, even a life narrative after a fashion (though they will have to be continually reminded of it). Their attitude toward such plans and projects, however, will be one of responding to what they will currently take to be good reasons for acting for a person like them or in their position. Such a process will be especially dependent on their caretakers and other external cues to remind them that this is a plan that they should follow. What they cannot construct for (and by) themselves is a sense of ongoing activity grounded in their sense of narrative identity that is not only recommended for a person like them but is theirs. As I argued earlier, memory is constitutive of a process of constructing, considering, and utilizing our narrative self-concepts. Without memory I cannot engage in the project of interpretive self-understanding, including the process of re-evaluating that self-understanding in light of new experiences, reflections, and interactions. Knowing who I am includes being able to remember experiences, actions, and decisions, that make sense in terms of the normative commitments and other characteristics I see as part of my diachronic self-concept. Without this capacity to remember in light of such commitments, I cannot rest assured that my current plans fit into a coherent ongoing narrative in which I find myself (from which I am not alienated). This connects to the conclusion of the last chapter that narrative models of the self require an ongoing sense of the constructed story of my life to be mine in a basic sense. To see this further, consider again the ways in which narrative self-understanding and memory construction interact. For example, I clearly remember engaging in various political activities over the past several years, most recently in the 2006 US elections. Clearly, an organizing idea at work in my entertainment of those memories is my understanding of myself as committed to certain political ideals and values. Imagine, though, that it was revealed to me, perhaps through psychotherapy, that my real motive in doing that work was trying to get back at my conservative father and that in effect my “commitment” to political ideas turns out to be an act of complex self-deception. (I would have easily worked for a right-wing candidate if my father had expressed

100

The Politics of Persons

leftist ideas, and so on.) If such a thing were true and I became convinced of it, I would have to understand the events of that period completely differently. I would certainly recall, as I do now, the behavior of going door to door, attending political meetings, talking strategy, and the like. But I wouldn’t understand that behavior in the same way (or in one sense I wouldn’t understand it at all). In short, my memories would change. That is because to remember that activity I must utilize an ongoing sense of my commitment to certain values as a hermeneutic lens through which to grasp the meaning of the mental images and sequences that comprise the memories. If that lens is discarded and replaced the memories in question would themselves radically change. A word of clarification should be added here, however. For in the model of remembering we have borrowed from the psychological literature it is clear that the process of recall of even long sequences of events and actions is not simply a replay of phenomenologically continuous sense data. The experiential elements of remembering may well be piecemeal, shifting, and sewn together in a cognitive map that I supply. To say that remembering differs from simple justified true belief (say) cannot imply that the former is phenomenologically continuous, as something we experience in our memory of it, and the latter is not. We must say, more precisely, that memory contains ineliminable experiential elements, the intensional (mental) properties of which are essential to their status as memories. This is consistent with understanding memory as a construction of disparate propositional, experiential, affective, and other elements of different psychological modalities, as long as some of those are experiential and this experiential quality is essential to their classification as memory. This understanding of the mechanics of memory helps us to understand how what we are calling identification (what I would prefer to understand as simple acceptance or non-alienation) is both an active and a passive process. Just as with memory, I am passively remembering events but actively understanding and interpreting them with a sense of an acceptance of them as mine; I identify myself with commitments that define me in that I accept the prospect of continuing to be guided by them. I thereby recognize them as part of my diachronic self-image. Self-interpretation, like interpretation generally, is neither value free nor completely spontaneous. I accept something as authentically me in interpreting it as something that defines me and which I embrace as such. This may help us understand how the practical identity that functions in this context is neither discovered by us upon reflection nor chosen

Memory, agency, and the self

101

ex nihilo.22 If upon reflection we can fit our actions and commitment into an autobiographical narrative without alienation (including resistance and repulsion) then such actions are part of what we can call our authentic selves (in a special sense of that term). If we wish to characterize our relation to this “self” as identification, then this involves the acceptance of this narrative as ours, just as memory is constructed by us in a way that must minimally cohere to make the memory ours in the right way.

v. implications of a socio-historical view of the self Before discussing the lessons we can draw from these claims, I want to discuss a set of arguments to the effect that the diachronic view of the self of the sort I have sketched here is wrongheaded, or at least a normative ideal masquerading as a neutral account of human psychology. As I mentioned in the last chapter, Galen Strawson has argued that not only is the narrative conception of the self mistaken, the diachronic view is equally misguided. As a descriptive account of the structure of human experience, the diachronic view of the self, he claims, is simply inaccurate, at least as a general account. As part of what he calls the “psychological narrativity thesis,” the diachronic account of self-experience (experience of ourselves as a subject and not merely a biological human being), claims that “one naturally figures oneself … as something that was there in the (further) past and will be there in the (further) future.” An “Episodic,” on the other hand, does not figure herself, as a self, as something that was there in the past and will be in the future (Strawson 2004, 430). Since being a Diachronic is necessary to seeing one’s life in narrative terms (his main target in this article), he argues against the general view that we all must experience ourselves diachronically. This, of course, is the view I have defended here. Therefore, I want to protect the psychological claim – that what he calls the Diachronic personality is not only true of some people some of the time but is deeply and pervasively true of virtually all people – against his criticisms. However, Strawson does not provide much in the way of argument for the existence (and psychological tenability) of an Episodic selfunderstanding as much as put forward what he takes to be examples (like himself). He quotes the Earl of Shaftesbury for example: 22

For a similar point see Noggle 2005.

102

The Politics of Persons

[But] what matter for memory? What have I to do with that part? If, whilst I am, I am as I should be, what do I care more? And thus let me lose self every hour, and be twenty successive selfs, or new self, ’tis all one to me … (Strawson 2004, 438)

If this claim is meant literally, that selves can have memories but not identify with the subject of those memories from one’s current standpoint, then I am at a loss to understand how autobiographical memory would be possible at all, on this view. According to the psychological literature I have sketched, personal memories are constructed precisely around a self-schema that organizes the disparate phenomenologically structured scenes given to us in recall as a memory of our actions. Without understanding a self as the subject of those scenes, indeed a self that is psychologically continuous (and in that sense identical) with my current self, I could never see my purported memories as memories at all, rather than say fantasies, dreams, or retrospective wishes. I was there is an essential element in the organizing process of remembering, as I understand it. Now Strawson is clearly aware of this. But he denies it as a necessary component of remembering. He rightly denies, on my view, that persons necessarily organize their lives as a narrative (in a strict sense). But he also denies (what I affirm) that persons are necessarily diachronic as well as being what he calls “form finders” – searchers for meaning and coherence (in a sense looser than strict narrativity). He argues that Episodics can experience the effects of the past “simply in so far as it has helped to shape the way one is in the present” (Strawson 2004, 432), but that does not amount to experiencing the past as my own. Using himself as an example, he writes, “when I am figuring myself as a self … I have no significant sense that I – the I now considering the question – was there in the further past” (433). Addressing memory directly, Strawson argues that he (as an Episodic) remembers things; his memories also have a “from the inside” character (what I have called the ineliminable phenomenological elements). What he denies is that I have to experience an identity between my current selfexperience (with its “from the inside” character) and those earlier experiences. What Strawson is claiming, in effect, is that one can have knowledge of one’s past, and one can know that one’s current states are causal consequences of that past, without having irreducibly phenomenal elements that amount to narrative memories of that past that are structured by the presupposition of sameness of self.23 23

Strawson concludes from this that all of the considerations in favor of a diachronic (and narrative) account of self are really normative, and he rejects those. I do not want to rely on normative

Memory, agency, and the self

103

As I have argued, however, personal plans and projects, which are diachronically structured in their own right, could not be seen as my own were it not for first-person memorial (retrospective) endorsement of them: one could only say that “yes, that project which my former subjectivity adopted, remains an optimal one for my present self, but not because of my subjective memorial embrace of previously living by it or adopting it in the first place.” This mode of expression, though, seems obviously to reject the use of plans and projects (of my own) completely. And I have insisted (parasitically on the work of others) that plans and projects are central to practical reason. I can agree with Sartre, for example, that the tendency to see lives as stories is problematic, often a sign of bad faith, and exhibits an obsessive attachment to an illusory sense of authenticity. But seeing one’s life as a narrative in the sense I am employing here is not to see one’s life as a story, per se, with a beginning, middle, and end and a thematic integrity which one is honor-bound to live up to. (That was a major theme of the last chapter.) I also reject that conception of self, at least as a necessary condition for the unity of a self or, more strongly, for a flourishing life. Narrativity in the sense I utilize here means nothing more than the project of making meaning of one’s ongoing experience and action. Such “meaning” need not be unified in any formalist literary sense; it need not be teleologically structured or unified by normative themes to which I, as its nominal “hero,” must keep faith. Now a claim weaker than the one I am defending in this section should also be mentioned, namely that even if there are “Episodics” in Strawson’s sense, implying that a univocal conception of selves as diachronic is wrongheaded in its most general form, it is still the case that many people are diachronic in their personality and self-understanding. Their interests and perspective will therefore be misunderstood if the model of personhood – in conceptions of justice for example that represent the interests and perspectives of citizens in the formulation of principles – is understood only in an episodic form. I have argued at least that conceptions of person in the literature on practical reason (a literature that has much influence on theorizing about justice) is, while not exactly episodic, decidedly futureoriented. Agents modeled in that literature have a present and a future (defined by their interests, plans, and beliefs) but they do not have a past. I at least want here to extend their identity conditions backward in significant considerations here, so I continue based on what I take to be relatively reliable psychological and intuitive accounts of experience and sense of self (as well as, later, on considerations having to do with the requirements of social and political interaction, which are normative in a different way).

104

The Politics of Persons

ways and thereby broaden the array of interests and the dimensions of perspective that arise out of models of the self.24 Nevertheless, I want to maintain that seeing selves as essentially selveswith-pasts (as remembered by them as their own) is important as a general view about people as subjects. I want to conclude by emphasizing the conception of self that has emerged and its relevance to social philosophy. I discussed earlier (in chapter 2) how the dynamics of social relations and connections can figure in the constitution of the self. Interpersonal connections, dominant public meanings, one’s social location, family history, and so on figure prominently in the full list of components that make up the self. Although I have not emphasized it in this chapter, I would additionally stress the ways that interpersonal and social elements operate in the construction of memories which, in turn, constitute the temporally extended self. Social dynamics operate in the current and diachronic self-conceptions that orient our thoughts and ground our judgments. What can be further added is that public activity and modes of public expression provide the terms in which self-definition and memory are articulated. Such public dynamics also express evaluative stances about the fundamentals of such articulations. If I define myself as Latino in a world where all associations with, say, the Latin American world are denigrated and marginalized, I have a fundamentally different experience of my self-conception than in places where such things are celebrated. Similarly, the events, experiences, and actions that make up my diachronic self (and my personal world over time) carry in their social forms interpretive and evaluative valences that interact with the public understanding of those elements. For example, an African-American adult’s memory of the events in the history of her childhood home in, say, Mississippi, may well conflict with the public memorialization of those events articulated by the (white) majority population. So memory has a social element which interacts with and depends upon various public interpretations of events. The extent to which that public interpretation conflicts with or affirms one’s own interpretive gloss on such events will determine crucially whether one is able to see oneself as an equal autonomous citizen in such settings. The object of equal respect in standard principles of justice is the “person,” but the fundamental interests of such 24

In some of the psychological literature on memory, the claim that having normal veridical memories presupposes continuity of a sense of self is explicit: as was mentioned earlier, autobiographical memory and reflective, temporally extended agency, are mutually constitutive according to some theorists (see Klein et al. 2004).

Memory, agency, and the self

105

persons shift when one conceives of them, as I have argued, as beings with a past whose elements are open to public interpretation. One need not assume that there are facts of the matter that should mold both those public interpretations and the personal self-conceptions that depend on them to argue that people’s ability to at least participate in the ongoing negotiation about those public meanings is a basic human interest grounded in a conception of self. Therefore, seeing persons diachronically in this way foregrounds the importance of participation in public negotiations and conversations about the past in ways that theorizing about justice in mainstream liberal theory (at least) has not acknowledged. We will reprise this theme in chapter 9. What I have hoped to have accomplished here is to have shown that where persons are conceptualized without sensitivity to the diachronic nature of their experience of themselves, their interests that are intrinsically tied to this connection with their pasts are systematically undervalued if not ignored. Social conflict that arises from negotiations with dominant meanings of the past will be, by that same token, misunderstood. Much is at stake in shifting perspectives in the way I have urged to this socio-historical conception of agency, selves, and the subjects of justice.

part ii

Autonomy

chapter 6

Political persons

Am I then really all that which other men tell of? Or am I only what I myself know of myself?

Dietrich Bonhoeffer (from “Who Am I?”)

We have traveled over complex terrain in part I in attempting to lay out a map of factors to which models of the self must be faithful. The current chapter will serve as a bridge to part II, where the concept of autonomy will be the focus. Autonomy, in this discussion, however, will be a characteristic of individual persons, rather than groups or relations among people, and it will serve as the fundamental model of the person in the context of democratic political theory, as I will suggest. So to make those two points at all plausible – that autonomy is a central political value and, as such, it attaches fundamentally to individuals – we must summarize the aspects of selves surveyed in part I which pointed toward a much more social conception of the self, indeed a “socio-historical” (and embodied) conception of self. What will be left is to argue that seeing autonomy as a characteristic principally of individuals does no violence to that complex model of selves. It will also be necessary to show that this serves acceptable theoretical aims in the construction of democratic political principles. I will first discuss what theoretical purpose models of the self serve, or at least serve in the particular context that interests us here. I will then survey the various conclusions we reached about the self in part I and attempt to organize these in a useful manner. In addition to factors discussed earlier, I will also focus on the ways in which aspects of the self and its condition are beyond the agent’s control, though not (necessarily) by that token alien to the self. This will also serve as the occasion to emphasize aspects of the self relating to its essential embodiment. I will next try to begin the argument that a model of the self is necessary in the construction and articulation of normative political principles, specifically the pluralist democratic theory of the sort we are concerned with here. 109

110

The Politics of Persons

This will involve attempting to show how individual autonomy can be seen as a fundamental (and fundamentally valued) characteristic of such selves without rendering irrelevant the social, temporally extended, and embodied elements of personhood outlined in part I. i. why (and what is) a political model of the self? As we saw repeatedly in the discussion up to now, there are any number of ways in which the self can be described. Indeed, talk of “the” self or even “the” theory of the self is almost certainly misguided, as the different ways of understanding what the self is are likely not reducible to one single conception. As I said, in different contexts the way in which features of ourselves are described as central vary, as do the levels of abstraction and modes of description used to capture such features. However, merely because we reject a single philosophical account of selves that can apply equally in all contexts does not mean we may not require a model of the self or selves that can function for theoretical and practical purposes. In particular, we are interested in what the person or self is that is represented in democratic political theory. More specifically, I will be discussing theories of autonomy, a term which, after all, means “self-government.”1 What is this “self ” then that is doing (and is subject to) this governing? What elements of people’s lives and bodies should be included or excluded in the conception at work in this setting? What is the theoretical connection between such a model of the self and normative political principles? A political model of the self serves the purpose of representing the perspective and interests of those to whom normative political principles apply. This is to utilize what Rawls has called a “model conception” in order to construct principles that map citizens’ perspectives in order to establish and maintain the legitimacy of institutions they must live under (Rawls 1999a, 307–308). This political conception of the person is meant to represent persons without describing them literally. What must be true is that actual citizens cannot be alienated from the picture of themselves exhibited in the basic constitutional principles of their society (and the more abstract principles of justice which structure that constitution). Even if such citizens would 1

Traditionally, the idea has also been rendered as self-determination, but as I discuss below, we do not determine, in the sense of create or choose ex nihilo, many central elements of our identity and condition, despite these elements being authentic aspects of ourselves. It is for this reason I would prefer the idea of “self-management” as the most accurate descriptor in these contexts.

Political persons

111

not describe themselves in such terms, or would use alternative modes of expression or levels of abstraction to express their own self-concept, such models can be useful in representing citizens’ basic interests. What must be the case, however, is that such models must be general and flexible enough to account for the variety of particular selfunderstandings found in complex, pluralistic societies. If an argument can be made that the model of self systematically privileges or excludes some types of self or identity, then such a model is defective. An example of an inadequate conception is the traditional liberal individualist idea that citizens are essentially separate from close ties to others or to traditions or social groups, that their identifying feature is (only) their cognitive capacities, or that reference to their pasts and memories are irrelevant to their core self-conceptions and hence their basic interests. But there is much room between inadequate conceptions of this sort and a fully “accurate” conception of the persons that populate society. Therefore, the conception of autonomy and the self proposed here must make room for the socio-historical and embodied elements of the person that we identified earlier as arguably central to the self-concept of many or (at times all) modern citizens. This is to reiterate the ecumenical approach to the person that we are developing here. I will return to this idea below when I discuss Rawls in more detail. First, we should take stock of the various social, historical, and bodily aspects of selves we discussed in part I that were thought to be relevant in such a modeling of the self for the purposes of political theory. In our rejection of the classic liberal individualist conception of the self, we said that the various ways in which identities can be specified with essential reference to social connections must be part of a replacement, non- or less- individualist conception of the person. This means that room must be made for the various (and varying) ways in which social relations constitute selves and their self-conceptions in the models utilized in political principles. What we found, though, is that such connections can be seen either at a very general and abstract level, such as the constitution of the self by way of the social practices of language, or at a specific one, such as the relation to particular significant others in one’s life. We found, or so I claimed, that the latter kinds of relations are not plausible elements for any general conception of the self, for such relations are subject to change in structure and significance. No matter how much one identifies oneself, say, as this person’s mother or that person’s spouse, the centrality of that relation to one’s sense of self may well shift as one’s life circumstances change, through death, estrangement, or renegotiation of the relationship.

112

The Politics of Persons

For example, the, by now well-discussed, communitarian critique of liberalism of the 1980s was interpreted as claiming that a communal conception of the person, where social ties of particular sorts are essential to identity, should replace the pared down individualist account of traditional theory. That claim, whether or not it was intended as such by so-called communitarians, was widely criticized as overly narrow and exclusionary itself, since it implied that only if we maintained ties to particular communities were we fully human, clearly a problematic implication, especially in the case of oppressive groups.2 Alternatively, the communitarian objection should be taken as the claim that conceptions of justice that rest on models of the person that ignore or exclude social ties which define the identities of some of us or all of us some of the time are problematic. Models of the self must be flexible to include all manner of self-conception and identity structure, at least among citizens who are subject to and who bestow legitimacy upon political institutions. The ecumenical approach to the self-concept I take here heeds that lesson. Secondly, conceptions of the self must be sensitive to the diachronic nature of identity. The model of the self represents the person over a life, not merely at a particular juncture. As we noted, selves can be seen as narrative constructions which include past patterns, actions, experiences, and traits as well as current commitments and capacities. This is why memory is crucial to agency, as we saw. Therefore, persons must be modeled as concerned with not only their current status and future prospects but also the way in which they have access to and can understand their memories, both public and private. This means that in order to grasp the meaning and importance of one’s most fundamental experiences – the constituent element of one’s autobiographical narrative – one must reflect on those events by way of concepts and symbols in the public discourse. In rejecting the idea of a private language,3 we assume that the construction of a self-narrative proceeds by way of a discourse that is publicly practiced and constructed. Interests connected to the construction, apprehension, and evaluation of that narrative will relate to those discursive practices. That is to say, how one’s own or one’s culture’s experiences are represented in current discussion, and in the institutional presentation of historical narratives, will affect how one is able to “remember” without deep disorientation or alienation. 2 3

See, e.g., Kymlicka 1989, ch. 5 and Buchanan 1989. For a more recent discussion, see Frohok 1997. This is, of course, a reference to Wittgenstein, but for a reference to this observation in the context of discussions of identity and normativity, see Korsgaard 1996, 137–139.

Political persons

113

We also noted in chapter 2 that selves are and should be seen as very much embodied creatures, as identified in reference to their physicality and (gendered) bodies. This means that physical abilities, and variations thereof, are relevant to the interests attaching to persons, so that one’s physical relationship to the environment in which one moves and acts is crucial for gauging those interests. Those whose physical abilities clash with this environment, whose ambulatory capacities are restricted because of lack of physical access to certain spaces (because of lack of wheelchair ramps, for example) will face decidedly different social opportunities from those who are not so restricted. Moreover, we described the ways in which bodies are socially “marked” in ways connected to identities. Certainly we (most of us) are gendered in clear ways, though it is equally clear that such marking is variable, in many (if not all) ways socially constructed, and subject to variable conceptualizations depending on social setting. A conception of the self, then, which somehow occludes or downplays the role of bodily identity in the shaping of one’s interests is by that token deficient. Though we will also note below the ways in which embodiment represents relatively unalterable elements of the self as well. We also surveyed various critiques of the “modernist” commitments of traditional conceptions of the self. One such line of critique concerned the assumption of a “true” self that could be referenced or consulted when reflecting on ourselves or discerning whether our desires are manipulated, distorted, or oppressively formed. It should be obvious, however, that in sculpting a model of the self in theoretical contexts of the sort sketched here no commitment to a true or authentic self is included. In constructing a conception of autonomy, it will be important to describe the aspects of the self from which the person is not alienated, but even there, there is no implication that what is left, the non-estranged elements of the personality, is somehow ontologically pure as an internal kernel of our real self. We can accept many of the observations of so-called post-modern thinkers that point out the limitations and “distortions” involved in introspection. Selftransparency need not be postulated in conceiving of the self as social and temporal. As we will shortly see, the extreme variability of self-conceptions which is acknowledged in this model of citizens emphasizes the rejection of any single metaphysical account of a person as a single and transparent entity to be discovered upon introspection. All this, however, was by way of reminder that thinking of selves in complex pluralistic societies requires that we be sensitive to the multiple dimensions that can make up particular people’s self-conceptions and identities. These elements need not be required, from a metaphysical

114

The Politics of Persons

point of view, as if persons as such were social or diachronic in particular ways, but such factors should also not be excluded or downplayed. But to see further what this flexibility in the construction of the model of the self implies for political principles, let us take a moment to discuss another famous use of a model conception of the self in political philosophy.

ii. rawls and the political conception of the person As I mentioned, the best known use of a model of the person along the lines being traced here is found in Rawls, especially in his political liberalism. That project makes use of a model conception that is not meant as a metaphysical account of agency as such, as the earlier Theory of Justice was taken to rest upon. Rather, persons are modeled as part of an interpretation of traditional concepts in the public political culture (the values of freedom and equality) and put forth as part of an overall view whose acceptance in an overlapping consensus of citizens provides its legitimacy. It should not matter, according to Rawls, if citizens’ own self-images are not mirrored in such a model conception, as long as they see their fundamental interests as represented in such a way that they can affirm the overall political view of which the model is a part. Now whether or not Rawls actually avoids any metaphysical commitments in the model of the person he proposes or not may be beside the point,4 since the project aims to construct conceptions which can garner approval by a highly pluralistic population. Rawls claims that the idea of the person comes from an interpretation of what it means to be free and equal, concepts dominant in the traditions of constitutional democracies of the modern age, or so he says. That model sees persons as having two moral powers that mimic such freedom and equality; these are the capacity to frame and revise a conception of the good and a sense of justice (Rawls 1993, 19). As such, they are free in seeing themselves as self-authenticating sources of valid claims for which they take responsibility. The moral powers correspond to highest order interests the person is assumed to have, interests in having the means to develop a sense of justice and a conception of the good. Moreover, the social primary goods, which Rawls provides as the index of social advantage needed in principles of distributive justice, model these interests. Primary goods are those 4

See Rawls 1999a, 388–414. For discussion (and the claim that he does not avoid metaphysical commitments), see Frohok 1997.

Political persons

115

all-purpose goods that would be needed by anyone seeing herself as free and equal in this way. These goods include basic rights and liberties, freedom of movement and association, the powers and prerogatives of office, income and wealth, and the social bases of self-respect (Rawls 1993, 75f.). This provides what Rawls calls the “institutional identity” of the person (1993, 30–31), meaning that the formal representation of the person and her interests is expressed by way of these two moral powers; but this is not intended to capture any person’s actual psychological, social, or cultural identity. In molding the interests that principles of distributive justice track, this identity provides political guidance about the social advantages citizens are assumed to pursue. This identity also models the perspective citizens are assumed to have in the construction of the conditions of legitimacy, according to which, for Rawls, an overlapping consensus can be established among such individuals who are guided by conflicting comprehensive moral, religious, and philosophical views. Further details of Rawls’s project need not trouble us here. But it is important to ask what alterations would need to be made in this profile of the citizen if we took more deeply into account the socio-historical and embodied aspects of the self underscored in previous chapters here. As touched on earlier, considering persons as (variably) structured by both social and temporally extended elements puts consideration of social connectedness into the forefront in specifying citizen interests. While income and wealth are often considered all-purpose means to securing desirable outcomes for people, they clearly do not represent the full range of interests we can conceive, and for many reasons.5 As a brief example, and to come back to an earlier point, public symbols and cultural forms are produced and endorsed as a collective product, often with state support, and cannot simply be purchased with private income. Not only memorials and churches (which themselves are often supported by state revenue) but social organization of cultural life (which holidays are recognized, whether businesses close on Sunday or Saturday, for example) directly affects the ways people can find elements of their identity supported or denigrated. One of the most under-theorized of Rawls’s primary goods is “the social bases of self-respect.” Rawls argues that the interest in such bases is satisfied by the public acknowledgment of the principles of justice themselves, along with the commitment to reciprocity and social cooperation that underlie them. Self-respect, for Rawls, is not an attitude but a social, institutionalized 5

Consider analyses by writers such as Simmel (1907/1990) and Titmuss (1970) for discussion of why outcomes secured via purchase are not equivalent to those gained by cooperation and gift.

116

The Politics of Persons

expression (Rawls 2001, 59–60). It is achieved for free and equal persons when the principles of justice, including the difference principle which determines the fair share of (other) social resources, are publicly acknowledged and accepted by co-citizens. Respect for ourselves as free and equal persons with the two moral powers is secured, on his account, when the principles that express the value of those aspects of our identity are recognized. In A Theory of Justice, however, Rawls had much more of interest to say about self-respect and its place in principles of justice. It is indeed instructive to see what Rawls said about this topic in the earlier work and why talk of this sort was abandoned, or at least pared down, in the later political version of liberalism. An element of self-esteem or sense of self-worth that underlies this primary good, for Rawls, meant having a rational plan of life that satisfies what he called the Aristotelian principle, which declares that, other things being equal, “human beings enjoy the exercise of their realized capacities … and this enjoyment increases as the more the capacity is realized, or the greater its complexity” (Rawls 1971, 374). It also involves “finding our person and deeds appreciated and confirmed by others who are likewise esteemed and their associations enjoyed” (386). Rawls emphasized the ways in which such esteem is established in virtue of one’s participation in social “associations,” “within which the activities that are rational for him are publicly affirmed by others” (387). But the support for such associations, and the self-respect that they engender, cannot proceed according to the “excellence” of their endeavors or the success of their pursuits. To do so would be to rely on a “perfectionist” principle of justice – the idea that social institutions should be supported in order to promote excellence in arts, the sciences, and culture – a principle rejected by liberal justice due to the latter’s pluralist commitments (285–292). Concern for social associations supportive of self-esteem cannot rest on an external evaluation of the accomplishments or value of those groups.6 Moreover, the enjoyment of self-respect (and the related feeling of self-esteem) is relative to the particular pursuits and social associations by which one defines success in life, Rawls notes, so social support for particular practices or institutions must be sensitive to the variations in membership, identification, and participation in such groups. Later Rawls dropped detailed discussion of self-respect along these lines, concluding that self-respect is not an “attitude” that can be promoted 6

Compare Charles Taylor’s discussion of the valuation of cultural forms required for a thoroughgoing multiculturalist politics (Taylor 1992, 65–68).

Political persons

117

publicly among citizens but a public condition, and that concern with achievement in line with the Aristotelian principle must be abandoned by political liberalism because such a principle rests on a contestable philosophical conception of the good life. A political conception of justice, in this sense, expresses commitments to success in social life only as a model for gaining consensus (and hence political legitimacy of institutions) but not as mirroring or attempting to promote actual citizens’ aspirations. The question here, however, is whether seeing persons as fundamentally (though variably) social in their identities, and as having self-conceptions that are diachronically structured, alters this approach to the social value of self-respect and the conception of the person in political liberalism. The standard to be used in evaluating possible alterations of this sort will be pluralism of the sort Rawls’s later views emphasized. Only if the conception of the political self maintains consistency with the wide variety of selfconceptions and conceptions of the good for citizens can it serve in principles of democratic justice in acceptable ways. This relates closely to what Kymlicka called the “social confirmation argument” (1989, 61–63; though see also Kymlicka 1995). We will discuss some of Kymlicka’s views in more detail in a moment, but here let us look at his comments about the need for social structures to reflect this interest in self-respect. Kymlicka considers the claim that social conceptions of citizens require that “considerable social confirmation” is needed to have confidence in their abilities as reflective agents. This entails that governments should “encourage certain communal values, and discourage non-conforming values, in order to ensure that … judgments are confirmed by society” (Kymlicka 1989, 61). Individualist liberalism, critics contend, not only protects choicemaking and the questioning of social ties, it supports social institutions that encourage such individualized questioning, thereby weakening the very social connections that give meaning to the lives of those citizens.7 Kymlicka’s reply is that this is really a sociological question and cannot be determined up front by political theory. Moreover, liberal theory implies that supporting people’s sense of their own effective sovereignty takes place through the empowerment of their abilities to rationally reflect upon and appraise their social connections, while alternative, communitybased methods of shoring up social connections bypass those reflective processes and work “behind the backs of the individuals involved – i.e., it generates confidence via a process which people can’t acknowledge as the grounds of their confidence” (Kymlicka 1989, 62). 7

I again refer back to the discussion in chapter 2 of these issues.

118

The Politics of Persons

While we will take up this issue and its kin more directly later, let me comment here that there is no reason that political institutions designed to promote the socially structured self-respect that citizens as social beings may require need to operate “behind the backs” of these citizens. Clearly, a conception of the person which includes basic interests that require social support, as when we postulate that citizens are socio-historically structured, can be a publicly acknowledged provision in the design and normative structure of political institutions. It can be part of the design of a democratic polity, for example, that social groups can receive special representation so as to support those social or cultural practices that give the needed context for the pursuit of basic interests guaranteed by state institutions. What I will suggest in chapter 9 is the construction of democratic practices that allow special weight to be given to people’s interest in (participating in the process of) defining and maintaining social practices that undergird their own identities. This is public and open and does not operate without the full acknowledgment of those participating. Moreover, the traditional conception of the person as sovereign emphasizes her abilities to “rationally” appraise connections with this or that social group, treating, as Rawls did, all social identities as species of “associations” – as groups one enters into voluntarily and from which one can depart at will. But some social groupings, such as those defined by gender or race, are not “associations” in this sense and are not appraised from a rational perspective like one appraises a neighborhood one might move to. Clearly the kind of critical reflection that I emphasize has much in common with this rational appraisal, but it is broader, including affective elements as well as cognitive calculations. The point is a subtle one, but in defining the basic interests of the person represented in political principles, if one stresses only the cognitive abilities to evaluate the pluses and minuses of an association one ignores the broader abilities to emotionally embrace, or to find a language to express deep alienation from, social connections that define one’s identity. The second set of capacities is broader than the first and more adequately reflects, I argue, the basic interests of those people whose self-conceptions include (emotionally colored) identifications with social groups or historical patterns. But with a socio-historical conception of the person, the second of Rawls’s two moral powers – the capacity to frame and revise a conception of the good – would have to be refashioned to take direct account of the social and temporal connections we grant in people’s self-conceptions (variable though those connections are in the population and across the life span). We need to revise this moral power, then, to read: the capacity

Political persons

119

to frame and revise a conception of the good in light of social dynamics that structure one’s identity. Similarly, to be self-authenticating, sources of valid claims must be understood as a capacity for authentication in concert with social dynamics and public understanding of one’s own and one’s society’s history. These qualifiers are, of course, variable phenomena, depending on the degree and dimensions of one’s involvement with these social and historical factors. But the interest must be specified in order to reflect sensitivity to these connections, where they exist. In later discussions we will see further implications of this alteration. After developing a concept of self-government – autonomy – to build upon this model of self, we will turn to conceptions and design of democratic institutions that must mirror the perspectives and interests of such autonomous agents. Acknowledging the presence (to varying degrees) of these non-individualist factors constituting the political self, but also stressing that variability, will mean that having one’s voice heard in the determination of the social forms that structure identity will be of fundamental interest for citizens, an interest not fully captured in traditional liberal conceptions of justice, or so I shall argue. iii. the liberal stance on the social self While our aim is not to specifically defend a liberal conception of justice per se, these discussions echo long-standing debates about the alleged hyper-individualism of liberal theory, particularly whether the conception of the person presupposed in liberal principles is incompatible with socially structured conceptions of the subject of the sort surveyed (approvingly) here. Again, many might understandably think at this point that this terrain has been adequately covered already, in particular regarding the liberal response to the communitarian “challenge” of the 1980s. As I mentioned, Kymlicka advanced detailed arguments similar to these some time ago, claiming that no matter how “social” citizens are or view themselves to be, individualist liberal values can be defended in ways consistent with this observation. Let me therefore comment on this briefly to show that the tack being taken here is importantly different from Kymlicka’s and, more broadly, subtly different from the standard liberal position on such matters. In reconstructing the communitarian critique of liberalism – specifically of liberal philosophy that sees the protection of individual autonomy as the fundamental good – Kymlicka distinguishes several variations of the general claim that selves are fundamentally social in their structure and

120

The Politics of Persons

self-perception and that (therefore) liberal ideals wrongly portray them as unencumbered by social ties and so falsely represent their (our) basic interests in pursuing the good. Kymlicka claims, contrary to critics such as Charles Taylor, that it is not true that liberalism places value on freedom (autonomy) without specifying the content of such freedom. Rather, freedom is valuable for people as part of their pursuit of valued ends, but ends that they can reflectively reassess and reject if no longer valuable: Rather, it is our projects and tasks that are the most important things in our lives, and it is because they are so important that we should be free to revise and reject them, should we come to believe that they are not fulfilling or worthwhile. (Kymlicka 1989, 48)

Kymlicka reviews Sandel’s claim that liberalism is committed to a view of the self as unencumbered, a view which clashes with our deepest self-conceptions. Sandel claims that liberalism runs afoul of these selfconceptions in requiring that (free) agents be able to abstract from all their values and commitments and review them dispassionately. Kymlicka replies, rightly I think, that liberal freedom merely requires that one be able to review any particular end in this way, not to review all of them in toto. Sandel would have to claim that there are particular commitments that are so self-defining that they cannot be reappraised and (in principle) rejected. As Kymlicka famously says, “[n]o matter how deeply implicated we find ourselves in a social practice or tradition, we feel capable of questioning whether the practice is a valuable one – a questioning which isn’t meaningful on Sandel’s account” (Kymlicka 1989, 54) I will return to this point and object to it (see also Christman 2001). But let us continue. Kymlicka also argues that despite his claims to the contrary, Sandel is himself committed to a view that says the “person” is able to re-examine her ends and commitments and hence, even on Sandel’s own view, the “person” must be definable independently of her ends after all. This is because Sandel admits that the subject can make choices about which of the “possible purposes and ends” it will pursue (Kymlicka 1989, 55, quoting Sandel). Now as I will explain below, it is one thing for a person to ask herself how and to what extent she might pursue a commitment or value, but it is another to ask, from a position outside of the orienting framework of that value, whether she should be committed to it at all. Further, Kymlicka considers Taylor’s claim that there is always an unquestioned “given” that orients our value judgments and provides

Political persons

121

horizons of meaning for us, a fact that liberalism must (implausibly) deny. Kymlicka’s reply is that since the given can shift over time for a person and varies among individuals, freedom to question and revise any particular commitment remains paramount (Kymlicka 1989, 51f.). But as Kymlicka considers later under a different heading, liberal institutions do more than simply protect the ability to change (that is, they do more than merely allow change). Rather, in setting up social institutions and encouraging those shared social values that make societies stable, liberal political institutions must encourage self-critical reflection, including questioning one’s most deeply held commitments and values. This encouragement shows itself in educational policy, promotion of social goods, as well as the constitutional structure that prioritizes some sorts of rights over others. This clearly will make it more likely that such questioning will occur, and thereby ties to specific communities, faiths, and relationships will be weaker. This clearly has its costs, as Kymlicka himself admits: “revising our ends is often a matter of a crisis in deeply felt beliefs and commitments” (Kymlicka 1989, 59). The claim ought to be that we should be able to effectively alter our commitments if, upon reflection, we are alienated from them. Social conditions ought not to obstruct a person’s ability to escape oppressive commitments and they should even encourage examination and reflection in those cases where alienation is likely. But this is not to say that state policy should encourage reflection and questioning when such ties are not problematic for the person. In these arguments, Kymlicka refers to “freedom of choice” – the capacity to reflect on one’s ends and select them from among alternatives – as the fundamental value that liberal institutions protect. To put it differently, freedom is the opportunity to choose among competing values. But there really are three components to freedom in this sense that should be separated. One is the requirement of open options, the social opportunity to change one’s life pursuits in favor of alternatives. If one is barred from practicing anything but a single religion, then one lacks this element of freedom. The second component is the meaningful personal ability to reject the values one holds, to effectively move from one set of commitments to another. This is not always available to people, as some commitments are so deeply embedded in their personality and view of the world that “rejecting” them is only a theoretical possibility for the person, not a psychologically viable option. Those that have this capacity, in effect, have the psychological ability (and the social support needed to sustain that ability) to step back from some commitment and seriously consider whether it is worthwhile.

122

The Politics of Persons

The third component of freedom operative in Kymlicka’s claims is the ability to reflectively appraise one’s values to any degree. Not everyone has this ability, namely to step away from a value or commitment long enough to ask whether it is really fulfilling or worthwhile, or at least what the social implications are for pursuing or continuing to pursue that lifestyle. The third is not equivalent to the second in that people may be able to appraise their commitments, even condemn them as debilitating (like an addiction) but lack the psychological capacity to change. I want to suggest that these three elements of freedom should be teased apart, and examined as separate elements of what I will call autonomy. The first – socially effective open options – is required for autonomy but only as a function of a person’s particular commitments, not as objectively given or philosophically specified for all persons as such.8 A person will require the opportunity to pursue a variety of life paths but only insofar as this is required for authentic pursuit of the one(s) she in fact reflectively accepts. Having options will be value-less for me if they contribute nothing to what I truly want to pursue and even distract me needlessly from that pursuit. The traditional liberal celebration of opportunity for its own sake can be easily questioned (see Dworkin 1982). Second, the psychological ability to revise one’s commitments is generally required for autonomy but only when and because we are sometimes deeply alienated from certain values or conditions. Only insofar as we are estranged in this way from a value orientation or life condition must we be psychologically able to change. But it is important to point out, that insofar as we are not so alienated, if our commitments to our values are unproblematic, deep, self-defining, and fulfilling, it is not the case that we must also have the ability to reject them. The ability to reject a commitment is not without its costs and will be of no use to those who happily see themselves as fully devoted to an uncomplicated pursuit or connection. To repeat a familiar example: imagine a parent of an older child to whom the parent is deeply devoted and with whom she has a wonderful relationship. To say that she is not autonomous unless she can psychologically and realistically spurn that relationship is, I contend, implausible. Finally, there is the capacity to reflect. Let us take this more slowly. On the one hand, I want to argue that critical reflection is indeed required for autonomy (as a political value), but it is reflection in the narrow sense, the ability to give reasons in support of the commitments one has, to 8

This is the position on this question that Raz takes, as I read him. See Raz 1986, 373–378. As with all of these points, this will be discussed further in later chapters (in this case, chapter 7).

Political persons

123

consider costs and difficulties regarding those commitments, especially for others, and to express those reasons publicly. The broader sense of critical reflection entails the ability to reject the commitment in question, which I just claimed is not required for autonomy. That is, the stronger requirement of reflection is to ask oneself whether one’s commitment to a value really has any basis at all, whether, from a point of view of a person without that commitment, one really should have it. But insofar as many people simply cannot imagine not having this or that commitment, since it defines them so completely, requiring reflection of this sort is an unrealistic demand for autonomy as a social value, or so I claim.

iv. variability, flexibility, and permanence in selves In order to make clearer this departure from Kymlicka’s analysis of the social self view, an additional set of observations will be necessary. We have discussed in detail the social and temporally extended aspects of selves. As a companion thought, we have repeated the liberal insistence that variability among citizens and flexibility in a single citizen must be acknowledged as well: while in general, social connectedness can be a constitutive element of the (political) self, any particular connection or anchor of identity is subject to change and variation, and as such cannot be cemented into the identities of citizens whose interests are represented in principles of justice. To assume such inflexibility would be to enshrine oppressive enforcement of group membership or faithfulness to being a person “of one’s own kind” in ways that deny fundamental freedoms of association and self-definition (see Appiah 2005; Barry 2001; Sen 2006).9 At the same time, however, it is a mistake to think that aspects of identities must be always subject to alteration and revision. This is equally oppressive, though in a different way. It not only problematically exaggerates the interests people are said to have in change and mobility, it also misrepresents the interests of some people to protect and maintain their social identities and the practices and institutions that support them. For these reasons it is important to note the various ways in which aspects of the

9

We can often make choices about whether to foster or retain a trait, allow it to express itself (and do so in this or that manner), or allow it to take on any number of real or symbolic manifestations (in response, perhaps, to social environments that invite or discourage such manifestations). Cf. also Audi 1991.

124

The Politics of Persons

self-concept that are relevant to social interests might not be realistically subject to self-alteration.10 The general contours of our bodies – our physical frame, abilities involving movement, perception, and personal interaction, as well as more general features of our physical existence – structure our experience of both the world and our self-conceptions.11 Such background factors concerning bodily make-up do not enter the field of consciousness and so are not subject to normal exercises of critical appraisal and self-alteration, yet they pervasively orient judgment and structure choice.12 The extent to which one has what are considered the normal range of physical abilities, for example, will pervasively structure the way one judges options and formulates values. Consider also biological sex. As we mentioned, social patterns and institutions can do much to ascribe meanings to the categories of “male” and “female,” and such patterns are, indirectly at least, under collective control. Moreover, individuals may well be able to voluntarily negotiate meanings and matrices of social significance that biological sex is afforded in the larger culture. (Indeed, these may well be completely socially constructed categories.) Nevertheless, the way in which most people construct their life plans, preferences, dispositions, habits, and values is simply set against the background of her or his sex; choices are made given that one is male or female, never in determination of it. Related to such factors is sexual orientation. While we can sidestep the question of the psychological or genetic basis for sexual orientation (or whether the typical categories of orientation are at all stable or well defined), we can observe that for many or most people, sexual orientation, in general, is not a psychological phenomenon over which they have much ongoing control. And while many do go through periods of searching and choice concerning the direction of their sexual lives, and for such people choice may indeed be involved in what sexual definition eventually emerges for them, many others describe their overall orientation as something discovered or assumed 10 11 12

For further discussion of this issue see Christman 2001 (from which some of what follows is drawn). This observation is part of the core of “ecological” psychological theories which build on the work of J. J. Gibson. For discussion, see the essays in Bermúdez et al. 1995. Psychologists refer to a “body schema” for example which, in a non-conscious manner, filters, structures, and orients ecological experience based on aspects of one’s embodiment, and this occurs in a way not subject to ongoing reflection and awareness (and hence choice): see, for example, Shaun Gallagher 1995, “Body Schema and Intentionality” in Bermudez et al. 1995, 225–244. Also, the importance of embodiment generally is often ignored in conceptions of the self that appear in analytic philosophical discussions of persons, the mind, practical reason, and autonomy. But as philosophers such as Merleau-Ponty, Michel Foucault, and recent feminists have stressed, aspects of our physicality and bodily natures cannot be separated from the structure of our experience, the nature of our judgments, and the content of our values.

Political persons

125

and, most importantly, psychologically unalterable. Reports of coming-out by gay men and lesbians often contain phenomenological accounts of realizing what was, in effect, already operative in the person’s life.13 Also included among non-transformable aspects of the self are longstanding emotional ties and deep affective connections. Such connections come in a variety of types and vary over time, and certainly aspects of those relationships are often subject to direct deliberative control by their participants. But many emotional attachments are such that they can best be described as a psychological “fact” of a person’s life, as a set structure within which a person makes choices and guides her life but which is not subject to ongoing reflective control.14 The love of a parent for a child is a particularly powerful example of such an unconditional (or at least unsheddable) commitment. This leads us to consider next the kinds of cultural, ethnic, and racial identifications that are often emphasized in the anti-liberal critiques we have been discussing (though of course, sex and sexual orientation play such a role as well). While some may be in a position to subject their cultural and racial identity to rational appraisal, especially the meaning it has for them, and perhaps reject these meanings in light of such an evaluation, many others describe their relation to such identity as an all-enveloping background fact about their lives, an organizing matrix within which they choose and deliberate, not an aspect of themselves open to modification. In a similar vein, we can refer back to our discussion of judgments concerning values in chapter 2, recalling how values often have the form of deep and visceral aspects of our psychology rather than that of deliberate choice. While unthinking devotion to a moral code can be obsessive, often attitudes toward certain values (or evils) are simply structural conditions that frame judgments rather than being up for consideration in light of reasons one has independently of those values. Harry Frankfurt, for example, discusses the category of “the unthinkable” which includes that realm of actions that simply could not be carried out by a person, no matter how overwhelming the reasons to do so (Frankfurt 1988b).15 Similar to this, for many, is religious faith. 13

14 15

Shane Phelan describes her experience this way: “What is it to become a lesbian? …. on a day in June, on a street in Los Angeles where I had always felt at home, I became a stranger. I looked around at others, and I felt them looking at me, and realized that I had crossed a line: I was a lesbian. I experienced that moment partially as discovery: so this was the difference I had always felt and never had a name for” (Phelan 1996, 52–53). For a similar discussion, see Mele 1995, 150–156. This indicates how this debate overlaps with a long-standing discussion in theories of practical reason and morality concerning the degree to which moral obligation is secured only if one can reflect rationally on one’s character and conform its elements to universal moral strictures: see, e.g., Williams 1983.

126

The Politics of Persons

Now clearly, merely being a stable aspect of body or personality, or merely being experienced as a background condition against which reflection and choice takes place, does not alone imply that the trait in question is unalterable. However, this inference does often hold – the fact that a characteristic is experienced as so central to one’s identity and value structure that changing it is virtually unthinkable amounts to the fact that for the purposes of practical reason and pursuit of a life plan, such characteristics (and hence the effect they have on shaping that life plan) are unalterable. These observations imply, then, that any conception of the self or its capacities operative in political principles that require complete revisability of all significant aspects of the self will be unrealistic if not implausibly restrictive. In light of these observations we should keep in mind the implications of a distinction discussed earlier between the “I-self” and the “Me-self,” between the self-as-knower and the self-as-known. In this context this translates to a distinction between what we could call “orientation” and “commitment.” The former refers to the perspective and moral orientation that a person exemplifies and manifests in her judgment and reflections. This may well involve social relations and connections, as would be the case, for example, if a person’s religious views or cultural identity color the very way that she thinks of things.16 In this way, reflection in general can be structured by socio-historical relations. “Commitments,” on the other hand, refer to those values and obligations that a person sees herself as having, and which can be the object of conscious reflection. The question can sensibly be raised, in such cases, whether “I” should continue to support this or that value. But in the case of moral (or social, etc.) orientation, the question of whether such a perspective should continue is more complicated. The issue must be raised indirectly, in that one must ask what it means to continue to view things through these lenses. While a person may be able to step back and consider this sort of broad orientation, she might not be able to do so in ways unaffected by it. In such contexts, dialogue and interpersonal interaction will help provide the alternative perspective that the first-person viewpoint cannot reach. In these ways the function of reflection cannot easily be separated from the process of social deliberation. Now, the degree to which such aspects of any given self will have to be subject to revision, and under what conditions, is a question that relates 16

There are affective as well as cognitive elements to such orientation. A member of a cultural group may always experience complex emotions such as pride, shame, or longing when considering questions that relate to that group and its history.

Political persons

127

specifically to the requirements of autonomy. Here I merely wanted to make explicit the ways in which self-alteration is, psychologically speaking, an implausible demand to make of many people, and a conception of the self that requires such manipulation and distancing is, by that token, deficient. In these ways, for Kymlicka to claim that liberalism demands that we are able to take every aspect of our self-schema and normative identity and subject it to the kind of critical reflection that presupposes possible rejection is similarly deficient. Now, as we will suggest, critical reflection of a particular sort is not too much to require of (autonomous) selves, and the capacity to reject or revise our self-schema (or the conditions under which we live and act) will be required under specific conditions, so the view developed here will be only subtly different from the standard liberal conception. But the subtleties are important, for they allow us to develop the kind of fully ecumenical model of the political self that is our aim.

v. self-reflection: why and what kind? In our various investigations into the components of selves, we repeatedly pointed out that while elements of the self-concept are often operative in ways that bypass our conscious control or awareness, there are many ways in which our own reflections about those elements (when cued) will be authoritative in fixing the meaning and public significance of such factors. What is necessary, though, is further discussion of both the pitfalls and the necessity of self-reflection in the political understanding of persons being developed here. For while our abilities to look into our own souls, as it were, are seriously exaggerated and often misunderstood, and indeed the philosophical tradition that places our “essence” in this ability is a serious sign of both that misunderstanding and that exaggeration, it will nevertheless be necessary for self-reflection to play a role in our public persona and hence the political model of the self operative in social principles. But to claim that, I need to make a few things clearer. First, the tradition of Enlightenment thinking about the self has placed the capacity not only to reflect generally, but to introspect, and understand ourselves in the pure light of reason, as the essence of (human) being. But as we saw in chapter 3, there are serious counter-currents to this trajectory, ones that want to deflate (to the vanishing point) the functioning of the reflective element of our being. Indeed, the capacity to “accurately” grasp

128

The Politics of Persons

our motives and values operative in our words and deeds is clearly quite dubious as a source of self-knowledge.17 There are many initially compelling reasons to worry about seeing reflection (and its outcomes) as central to the self. Two families of reasons can be given on this score: one is that reflection itself is often costly and carries with it effects on commitment and devotion that raise questions about its role in self-determination; a second is that the reflective voice in all of us often does not speak for our most settled and authentic personae in that such voices can cover over or misdiagnose the inner workings of our psyches. Let us look at these worries more closely. The first set of problems involves the way in which reflectively questioning our commitments and motivations can often disrupt and undercut those very commitments. This problem of first-order motivational distortion can best be brought out in a two-person case: consider long-time spouses or romantic partners; one day one of them enters the breakfast room to announce that she has lately been reflecting on the value of the relationship for her and on her commitment to it. Now even if the result of such rethinking is to redouble the strength of her commitment, the partner hearing this may well be disappointed and shocked, and the ties between the two deeply shaken. Now if we collapse this dynamic into a single mental life, we have cases where self-evaluation leads to self-doubt and diminished motivation.18 There are other ways in which introspection fails to reveal, and in fact works to obscure from view, the motivational and emotional drives that move us to action. Often, such reflections merely give voice to a rationalizing super-ego attempting to quash the more central elements of our motivational system, elements which, if allowed to move us, would issue in action that is more truly our own. We can easily imagine cases of selfdenial and internal subterfuge, where love, jealousy, envy, or revenge move a person through a complex set of actions but are emotions that she never admits to herself. People in the throes of such emotions often use superficial introspections to find cover stories that aid their selfdenial. In addition to revealing the important place that emotions have in the specification of the self,19 such cases indicate how the voice of reflection may distort rather than clarify our self-conceptions. 17 18 19

For a wide-ranging discussion of such themes, see Meyers 1994. Also, in what follows I draw heavily on Christman 2005. A similar argument is made by Bernard Williams concerning moral principles, see Williams 1985. For discussion, see Mackenzie 2002.

Political persons

129

Moreover, it is clear that only a marginal proportion of the self implicated in behavior and social interaction can ever be said to be available to conscious reflection both generally and at any particular time. As we have noted, factors connected with embodiment, demeanor, habit, and the emersion of the self in the ongoing flow of events, operate outside of the purview of reflection and often completely beyond its scope (see Sullivan 2006).20 What emerges from these several angles is a picture of systematic selfdelusion, or, at best, a fundamental disconnect between introspective understanding and actual structures of motivation, thought, and behavior. And as we discussed, this is a more empirically minded way of expressing what commentators writing in a post-modern mode have been saying about the liberal conception of the self for some time, namely that such a conception wrongly assumes a transparent, unified, fully rationalized selfconception of a sort no one realistically can realize. Nevertheless, there are reasons that support the view that reflective selfpresentation will be the key function in the political model of the self we adopt, namely that social and interpersonal communication requires that one’s motives, feelings, judgments, and values be expressed from the firstperson point of view in order to discuss and collectively deliberate matters of mutual concern with others. The mechanics of communication requires self-presentation in order to get on with the process of mutual understanding (whatever the limitations of that goal). We cannot read each other’s minds (or bodies). Secondly, it may be the case that in theoretical and fictional accounts of people’s opacity to themselves, or in the context of therapeutic analysis, a full description of a person’s inner states can be constructed; this happens only in the rarefied context of theoretical model-building, fictional omniscience, or specialized interpretation of others’ hidden lives in a therapist’s office. Philosophers clearly use their own experiences as well as observations of others’ behavior and self-reports to construct views about the limitations of reflective access to motivational states. And psychologists outside of the psychoanalytic tradition conduct controlled experiments in carefully manipulated and structured environments. 20

Psychoanalysis, of course, trades on the idea that normal introspection fails to uncover our true motives. For an overview, see Eagle 1991. For a parallel discussion, see Wollheim 1984. However, from other traditions in psychology there are numerous research programs that support the systematic phenomenon of the self’s misunderstanding of itself. For discussion of recent social psychological work on the self which reflects this tradition, see Baumeister 1998. For a discussion of the historical development of theories of the self see Gergen 1971 and Harter 1996.

130

The Politics of Persons

All this points to the contrast between the settings or methods from which we can conclude that persons misdescribe, repress, confuse, and wholly construct inner motives and feelings and those public encounters in which individuals as citizens must conduct political deliberation. So while we can admit that under some settings we can confidently conclude that people’s understanding of themselves contains all manner of distortion and subterfuge, in public encounters with fellow citizens (and their representatives) in which values, needs, convictions, and judgments are discussed, it is problematic at best to dismiss or ignore the person’s own reflective grasp of her views (feelings, etc.) and not see her as the privileged spokesperson, as it were, for their validity. This merely sketches some general reasons for including a requirement of self-reflection in conditions of autonomy in political settings. But this so far fails to face head on the challenge raised in several places here that critical reflection presupposes a voluntarist conception of our relation to core elements of our identity, a conception that we have found various reasons to reject. How can we make sense of critical selfreflection of the sort required for public deliberation – the giving of an account of oneself in interactions with others to settle problems of conflict, coordination, and communication? If such commitments are internal to the self in the way I have been describing, however, such encounters amount to nothing more than self-presentation: the “justification” one gives of one’s values, given their constitutive role in the structure of the self engaged in deliberation, will always be a variation on Luther’s dictum “here I stand, I can do no other.” In what follows I map out a view of reflection that may help answer this challenge. Giving reasons: two models of critical self-reflection The crucial move in attempting to reclaim a role for public selfjustification here is to consider the kind of critical self-reflection involved in accepting value commitments as genuinely one’s own. For in critically appraising such values, or our condition in light of such self-defining values, we may well arrive at reasons the articulation of which can serve as such a public justification. To accomplish this, it is important to distinguish two kinds of critical self-appraisal: the first is foundational, and asks whether, for any commitment or relation X, there are good and sufficient reasons to maintain one’s connection to X. Variations on this range from the universalist form, which asks whether any person (or any rational person) should hold X, to whether any person “like me” should

Political persons

131

hold X – where the implicit comparison class is somehow specified. These formulations all imply that “I” can judge separately from the value in question and assess its merits. The second mode of self-reflection asks a different set of questions (again considering an agent’s consideration of some value proposition X). These include: What is the importance of X? What are the implications for myself and others of pursuing, respecting, enjoying (etc.) X? How central is X to my self-concept? For example, let X be a set of religious convictions, such as being Catholic. Imagine also that the agent asking these questions of herself has been Catholic all her life and comes from a deeply religious family and region. She would ask herself things like “What does my Catholicism mean to me, how important is it in choosing paths in life? How crucial is it that others respect or allow me to practice this religion? How disorienting or unacceptable (if at all) would it be for me to consider moving away from this faith?” These latter sorts of questions can be called “embedded self-reflection” in that they all can be coherently asked by a person who is in part defined by the very value commitments under review. Nothing in the manner in which these reflections are stated need imply that the agent can stand apart from the commitment in question, with an intact and fully defined sense of herself, and consider reasons in its favor. If this distinction is plausible, then we can maintain that public deliberation of the sort required by confrontations in democratic deliberation requires only embedded self-reflection in the interpersonal presentation of reasons required in such settings. Also, such a model is consistent with critical self-reflection in the “embedded” sense, since this sort of reflection questions only the meaning, importance, implications, and centrality to one’s sense of self of the value at issue. Those critical questions can all be asked, I submit, even if one’s commitment to that value is partially definitive of who one is. In this way, we can accept much of the moral psychology of those who have insisted on a conception of the self that is constituted by unsheddable connections, identity categories, bodily and psychological orientations and other constituents of our “I-self.” Nevertheless, critical reflection on such elements does not become incoherent as long as this reflection is assumed to be the embedded sort just described. The role such reflection plays in both a plausible conception of autonomy and, in turn, a conception of justice which places autonomy at its center is still to be described. In the next three chapters, then, we will undertake that project, only to return to the broader issues of democratic justice in the closing chapter.

132

The Politics of Persons

So the selves being modeled here are variable and complex. They include social elements as well as extend diachronically through a life span, anticipating a future but trying to make sense of a past. They are able to reflectively appraise the constructed narrative that defines them as subjects, at least in the sense that they can determine the depth, meaning, and implications of their commitments. But they also realize there is much about their persons, bodies, and social surroundings that is beyond their control. What is left, then, is to investigate in detail the question of how selves of this sort can govern themselves, how it can be postulated that social, historically structured, and embodied subjects can be self-managing. In short we ask what it means for them (us) to be autonomous.

chapter 7

The historical conception of autonomy

So far in our examination of the self, we have taken on board various postulates concerning the self’s constitutive connections to social factors and relations as well as its own history and narrative trajectory. We have also considered the way that social selves might be modeled in political principles underwriting modern democracies whose populations are marked by multiform pluralism. In that discussion we also saw that the protection of individual capacities to participate in such democratic forms could be valued in ways consistent with the assumption of the socio-historical self. The idea of individual autonomy, it will be suggested, can play the role of expressing the locus of that protection. Approaches to the idea of autonomy, until recently at least, have for the most part failed to consider an agent’s past as crucial in thinking about selfgovernment. Connections to one’s autobiographical narrative have been occluded in laying out what it means to be an independent, competent, authentic agent. But if I am right that the self should be seen as having a temporally extended structure, as diachronic, and autonomy is in some way self-government, then we must somehow build in reference to personal history in our conception of that latter notion. Recently, however, philosophers and other theorists have been developing views of moral responsibility and personal autonomy that have turned increased attention to the historically and socially embedded nature of the person.1 I myself put forward a conception of autonomy that specifically 1

See, for example, Mele 1995. For an historical account of moral responsibility, see Fischer and Ravizza 1998. These discussions concern autonomy as a metaphysical concept, related in particular to questions of free will and moral responsibility. While those questions are not directly related to my concerns here (see below), I think that reflections on the concept of autonomy in these contexts help fix our intuitive ideas about what self-government can be understood to mean. Also, Nomy Arpaly has argued that “autonomy” has as many as eight different meanings, which dilutes its importance to issues such as moral responsibility (Arpaly 2003). I hope the manner in which I explicate the concept in the text shows it to be relevant to the particular role I see it playing in practical and theoretical contexts.

133

134

The Politics of Persons

referred to historical aspects of agency (Christman 1991a; see also Christman 2001). That view was met with various criticisms, to which I have attempted to reply (Christman 2007). The current chapter lays out the basic components of that revised view in a way that will hopefully provide a defensible picture of autonomy that will be useful in an autonomy-based conception of democratic justice but which also fruitfully builds on the conception of the socio-historical self described in part I. i. preliminaries In general, autonomy is meant to manifest self-government, the ability of the person to guide her life from her own perspective rather than be manipulated by others or be forced into a particular path by surreptitious or irresistible forces.2 There is much disagreement, however, about the particular components of autonomy, and models vary according to the theoretical and practical context in which the concept is understood to function. In the recent philosophical literature, autonomy has been conceived as potentially embodying a variety of conditions. Some of these requirements relate to the agent’s ability to form desires and make them effective. Such conditions relate to cognitive and normative competence – rationality, self-control, absence of psychosis and other pathologies, and so on. In addition, some have argued that autonomy means not only being able to act effectively on one’s desires but also that such desires, values, or other springs of action are truly the agent’s own. These requirements relate to the authenticity of the agent’s desires and values and often include the requirement of critical self-reflection on the factors relative to which the person is autonomous. Now the word “authenticity” carries particular philosophical baggage that should be bracketed here. For example, authenticity sometimes refers to the obligation to live up to one’s “true” self, in some way, as a form of integrity.3 It has also been theorized in other traditions as being a central element in human experience in the world as in the existentialist account of freedom as authenticity. Here it is a term referring to whatever counts as acting from motives and circumstances that are truly “one’s own” in the sense accounts of autonomy try to capture. The further question of whether 2

3

Writers who have analyzed the concept of autonomy in detail recently include Berofsky (1995); Dworkin (1989b); Friedman (2003); Mele (1995); and Meyers (1989). For surveys of such literature, see Buss 2002; Christman 2002a; Mackenzie and Stoljar 2000b; May 1994; and Oshana 2006. Indeed, some writers argue specifically that autonomy should not require having to be true to oneself in this sense. See, e.g., Velleman 2004, 97 and Arpaly 2003, 122. For discussion see below.

The historical conception of autonomy

135

authenticity in any such sense is necessary for autonomy, however, will depend on the particular gloss on it that is provided. As I have made clear in this book, the context in which I see autonomy functioning will be as it applies to citizens living in a just society. Autonomy is the chief characteristic of the model citizen whose perspective and interests locate the source of legitimacy for just political institutions and who frames the basic goals which provide content to the principles governing such institutions. It can also be seen as the characteristic we assume of people as participants in collective decisions or one which we think important to bring about, if absent, in order to protect the integrity of such decisions. Finally, it is the locus of interpersonal respect operative in personal relations which, in turn, places limits on paternalistic interferences or manipulative dealings with them.4 So the autonomy of participants in collective deliberation projects onto the outcomes of those deliberations a particular legitimacy, in the same way that the autonomy of a single person grants status to the choices such a person makes. This distinguishes such decisions, in at least a minimal way, from pathological, addictive, manipulated, or compulsive ones. We might properly respond to the latter in any number of ways – with compassion, tolerance, strategic maneuvering, or resistance. But autonomous acts have a minimally self-validating imprimatur; they command a degree of respect that non-autonomous choices do not. When Rawls claims, for example, that citizens are assumed to be “self-authenticating sources of valid claims” (Rawls 2001, 23–24), it is their autonomy, on my view, that authenticates the validity of these claims. I talk here of choices, which raises the question of whether we are referring simply to particular desires, acts, or other localized aspects of the person in defining autonomy or to the person as a whole.5 Certainly, it makes some sense to ask about the autonomy of particular aspects of the person – what is usually called “local” autonomy. Indeed, many ways of characterizing that trait make it readily applicable to specific factors. But often it will not be possible to determine whether conditions defining autonomy are met for particular choices, or it will not be important to 4

5

I do not mean here that paternalistic interferences with autonomous agents are never justified or that lack of autonomy permits paternalism in all cases. The suggestion is merely that insofar as limits on permissible paternalism are morally required, it is respect for the autonomy of the agent in question that is the chief source of such limits. For discussion, see Dworkin 1989, 13–17. Diana Meyers discusses a similar point under the guise of “episodic” versus “programmatic” autonomy. See Meyers 1989, 48–49. For a view similar to the position I take in the text, see Friedman 2003, ch. 1.

136

The Politics of Persons

measure whether specific ranges of acts are autonomous separately from the agent in her entirety, at least for the contexts we have in mind here. For if we view autonomy as the status marker for entrance into collective deliberations in general, and the interests of gaining, maintaining, and exercising autonomy as the interests represented in designing procedures for such choices, then it will be the autonomy of the person as such that will matter. On the other hand, an otherwise autonomous person may have particular addictions and pathologies in part of her life that can, in fact, be determined and designated as such. My suggestion is that what matters for the social role that autonomy plays in the contexts just outlined is that the person is autonomous relative to her basic, orienting values and motivations, or indeed any factor about the person that pervasively and fundamentally motivates and guides action. That is, the fundamental structure of normative commitments and pattern of judgment is what must be “one’s own” in order for the person to be autonomous in the sense that matters here. This refers to those commitments and frameworks of judgment that ground a broad range of decisions, tastes, and actions for the agent. So while autonomy in the sense being considered here is local, it relates to those basic elements of our personality that are the most pervasive in guiding our deliberation, choices, and actions over time. It will therefore not be untoward to call a person heteronomous when she lacks autonomy relative to her basic value orientation. This is a variation on the “local/global” distinction familiar in the literature on autonomy, but it is necessary for reasons I have stated: if autonomy is the locus of interpersonal respect that grounds deliberative legitimacy, then those central aspects of the person’s normative outlook matters most to that respect (and the attendant legitimacy).

ii. historicizing autonomy: basic motivations There is much literature on autonomy and its vicissitudes that we need not review here, but a predominant approach to the notion has involved a “hierarchical” conception according to which autonomy involves secondorder reflection on first-order desires, characteristics, motives, and other aspects of the self and agency.6 An ongoing controversy concerns how to specify the conditions of this higher-order reflection that does not involve a 6

This model was developed by Harry Frankfurt and Gerald Dworkin and has become the chief focus of discussions of the idea in the literature of the last 25 years. See Dworkin 1989b, chs. 1 and 2, and Frankfurt 1988a.

The historical conception of autonomy

137

regress of requirements or which fails to rule out ways that autonomy can be disrupted at higher levels of thought and judgment as well as lower-order motivations. Problems also have been raised with the requirement that such higher-order reflections must issue in identification with such factors for autonomy to obtain.7 All these issues are complex and interesting, and what I say below will attempt to take a position that obviates the most serious worries, but I will skip over these for now in order to identify what I take as the point of departure for the approach to autonomy defended here. Whatever the conditions of higher-order reflection that have been spelled out in this literature, it remains the case that most such views of autonomy fix that characteristic in a “time-slice” manner. No reference to the processes of socialization (or manipulation) that may have figured in the development of these desires is made. Yet we want a theory of autonomy to somehow be able to distinguish those processes of development that secure authentic sets of desires for us from those that do not. That is, on dominant models of autonomy, we would still be autonomous even if our full set of desires (including the “unsheddable” ones that ground our deepest cares) were implanted in us two days ago by a clever hypnotist.8 In general, the impetus behind looking at autonomy historically comes from the rough intuition that a person who thoroughly embraces her current values but does so because of overt past manipulation, oppressive upbringing or severely constricted reflection lacks something essential to autonomy and freedom. The fact that she can currently act as an agent out of those values, and the motivations that stem from them, obscures the way that her narratively structured identity, over time, is not her own in a crucial sense. If selves have temporally extended elements that should be accounted for in social principles, as I have argued, then such a person’s “self” (using that term in its vague and unspecified form) is not auto-nomos, a law imposed on itself.9 The point is that, since such a person would not exhibit autonomy, and this is precisely because of the tainted history of her value constitution, then only if historical conditions of some sort are added to models of autonomy will such models be intuitively adequate. Given that a person is embedded in, if not constituted by, the flow of events that form her life, autonomy must be seen in relation to that flow rather than as independent of it. 7 8 9

For an overview see “Introduction” to Christman 1989, 6–12 and Mackenzie and Stoljar 2000b. Cf. also Berofsky 1995, 99–102; Friedman 2003; Thalberg 1989; and Watson 1975. See, e.g., Mele 1991; cf. also Zimmerman 1999. For a succinct version of this argument, see Mele 1995, 145–146. For similar arguments, see Christman 1991a.

138

The Politics of Persons

Secondly, consider again the argument developed in chapter 5 regarding persons suffering from amnesia. The argument was made there that such individuals lack an element of agency that we want conceptions of autonomy to capture, namely the ability to reflectively accept their commitments and values as they have developed over time. If this judgment is correct, it shows that reference to the personal, historical development of desires, values, and commitments is required in models of autonomy. Further, we should again consider the claims made about the narrative structure of persons and their self-reflections considered earlier. Often the very values and commitments being considered in the kinds of selfreflections involved in autonomy are themselves temporally extended phenomena, not merely spontaneous choices and judgments. This last point is similar to Michael Bratman’s description of “selfgoverning policies,” which for him have agential authority because of the cross-temporal organizing role they play in guiding decisions and actions (Bratman 2007, ch. 10). Plans, then, structure practical reasoning in that they guide decisions even if, at the time, the agent acts unreflectively, as is often the case. As such they are part of our temporally extended agency (Bratman 1999, ch. 1). On my view, narrative coherence of the sort described in chapter 4 is required so that agents can reflectively utilize past and ongoing dispositions to guide current decisions. Memory operates both as a constructive activity in giving us a sense of ourselves (as discussed in chapter 5) but also in grounding current decisions as our own: such decisions cohere with the ongoing autobiographical narrative that grounds our agency (cf. also Cuypers 2001). I will return to this point below. Now if this line of analysis is successful, it shows at best that some conditions which refer to the processes of character formation must be included in conceptions of autonomy; but it does not show which ones or what kinds. There are at least two sorts of conditions that should be distinguished: subjectivist and non-subjectivist conditions.10 Subjectivist (or internalist) accounts take the agent’s particular attitudes toward her own development as central to the test for autonomy. Non-subjectivist (or externalist) accounts posit processes of personal development that secure autonomy independent of the agent’s own appraisal of such processes. In general, the distinction concerns whether the specification of processes of

10

This distinction parallels (but does not precisely mirror) what have been called “proceduralist” (or “internalist”) and “non-proceduralist” (externalist) accounts of autonomy. For discussion of this distinction, see, e.g., Buss 2002 and Mackenzie and Stoljar 2000b.

The historical conception of autonomy

139

personal development which are consistent with autonomy must be made with crucial reference to agents’ own attitudes toward such processes or not. More broadly, this issue embodies the contrast between seeing free agency as a set of conditions specified philosophically, as part of an account of “normal” human development of a sort that manifests self-government, and construing agency (in part at least) from the point of view of agents themselves. In the latter case, what counts as “self” government turns in part on perspectives and judgments that agents embody and exercise. No account of autonomous agency can do without philosophical articulations of rational thought and action – what would “agency” mean without such specifications? – but subjectivist accounts give broader range to individual judgments concerning what counts as “normal,” self-governing life processes and development. The subjectivist account is motivated by a robust eclecticism both about what kinds of processes of self-development different agents utilize to achieve valued ends as well as about the values and goals themselves that different people pursue. Although some conditions would have to be specified independently of the agent’s point of view, the subjectivist here tries to secure a space for variability concerning types of development and change. Agents have any number of reasons to manipulate themselves and their thought processes, or to accept the ways that they have been manipulated, that would be incompatible with autonomy if such agents had not approved of them. We also engage in self-constraint in various other ways: setting alarms that jar us fully awake, medication that alters our moods or capabilities, granting enforceable authority to others, and so on. The broader lesson here is that a remarkably wide variety of selfconstraining processes can be specified that could, in principle, be adopted by a person for some limited time in order to achieve an otherwise rational and independently formulated goal. We could suspend or interrupt our capacity to respond adequately to reasons in order to fulfill some end we have (because, for example, we want to forestall backsliding and weakness of will). Many externalist accounts, for example, require an ongoing cognitive monitoring of our changing character. But we often are not so self-aware, nor do we always proceed in rational, self-controlling ways. Moreover (and we have seen this as a recurring theme), most of our commitments, tendencies, emotional predilections, and forms of thought, have developed in us without any ongoing reflective consideration or judgment by our own reasons-responsive faculties. We merely find ourselves with such traits and come to endorse them as constitutive of who we in fact are. Many developmental stages of childhood and adolescence exclude

140

The Politics of Persons

opportunities for critical reflection at crucial junctures, but the agent comes to personify the results of that upbringing and does so in a way that is autonomous and responsible. This indicates that the strict non-subjectivist condition of development leaves too little room for the variety of ways that characteristics are developed and values adopted. Indeed, many thinkers have stressed the ways that unchosen and unreflective ways of (always already) being in the world, in Heidegger’s phrase, or ways of going on, in Wittgenstein’s, belie the notion that we are able to reflectively choose or even endorse our commitments and rule-bound activities.11 The language that we use, the ways we were raised, even key “actions” such as falling in love or having children, may well be things about which we either exercise no overt control at all or may fall into unreflectively or by the simple inertia of events. Additionally, many ways of life are formed by rules and disciplinary practices that participants in them would never think to question and over which they have no control. Some writers who have stressed these aspects of our lives have used such references to mount a case against autonomy as a basic value (or against autonomy-based political views).12 However, as I will explain below, I think that considerations such as these point simply to a need to refine our conception of autonomy to leave some room for the unreflective and the automatic in life, as long as such unreflective ways of living do not amount to repressive or manipulated – hence clearly heteronomous – states. To make these observations more concrete, let us examine a particular historical view of autonomy that contains externalist elements (in our sense). Alfred Mele has put forward a complex and powerful account of autonomy that combines externalist conditions with a voluntarist override, so to speak. His view, put all too briefly, is that processes of character development in autonomous persons are ones that do not exhibit any form of “compulsion” – where to be compelled is to have been conditioned to adopt the characteristic in question in ways that bypass the usual kinds of mental control normal agents enjoy. Such compulsion disturbs autonomy unless (and here is the subjectivist override) it was “arranged” by the agent.13 11 12 13

The considerations in this paragraph follow K. Anthony Appiah’s discussion of autonomy-based liberalism in Appiah 2005, ch. 2. I make a similar point in Christman 2001. See, for example, Gray 2000. Mele’s fully specification of the relevant part of his view is this: If an agent S comes to possess a pro-attitude P in a way that bypasses S’s (perhaps relatively modest) capacities for control over his mental life; and the bypassing issues in S’s being practically unable to shed P; and the bypassing was not itself arranged (or performed) by S; and S neither presently possesses nor earlier possessed pro-attitudes that would support his identifying with P, with the

The historical conception of autonomy

141

This avoids some of the problems just mentioned, in that it leaves room for the fact that many or most value commitments, cultural identifications, religious connections and such are often developed in ways that do not involve ongoing reflective endorsement by the agent. However, it is unclear what the force of “arranged” here is (the term Mele uses). For as I noted we often do not engage in conscious reflection prior to taking steps which fundamentally alter our character – they merely happen to us first. Many fundamental aspects of our personality, for example, were shaped during childhood or adolescence and determined by conditions we had no control over, though we may come to accept the results of those processes without regret after the fact. As long as that further acceptance is not itself simply the result of those processes themselves, especially if they are such that reflective, retrospective, evaluation is distorted by them or made impossible, later endorsement of constraining processes would suffice for autonomy. Moreover, requiring “non-compulsion-unless-arranged-bythe-agent” reduces to claiming that when processes have been approved by the agent, compulsive or not, autonomy obtains: subjective acceptance is sufficient (along with other competence criteria) for autonomy. Another attempt to give historical conditions that count as nonsubjectivist in my sense is included in an account of moral responsibility developed by Fischer and Ravizza that could be applied in this context. Theirs is an example of utilizing the condition of “responsiveness to reasons” as a desideratum for (for our purposes) autonomy.14 This states generally that agents are responsible if they are generally able to respond properly to good reasons to alter their behavior or character. To use the categories named earlier, this is to turn “authenticity” (whether one’s motivating states are truly one’s own) into competency (whether one is an adequate decision-maker). As an historical model of autonomy, the reasons-responsiveness approach requires that current motivational states of the agent had to have come about by processes that were reasons-responsive. In other words, a person is autonomous if she developed in a way that involved being able to respond exception of pro-attitudes that are themselves practically unsheddable products of unsolicited bypassing; then S is compelled* to possess P (Mele 1995, 172).

14

And a person is autonomous only if (among other conditions) she was not compelled* in this way to develop her pro-attitudes. (The label “compelled*” refers to being compelled in the way described by all the previous qualifying factors listed.) For discussion, see Fischer and Ravizza 1998, chs. 2 and 3 and Wolf 1990. I should note that the models I am discussing here were not originally meant to characterize the autonomy of the agent, but rather her freedom or moral responsibility. I am therefore adapting the language of these tests to the issue at hand. For further discussion of these issues, see Christman 2001.

142

The Politics of Persons

adequately to reasons. This capacity amounts to saying that in a “wide variety of possible circumstances” the person would exhibit an “understandable” pattern of reasons-recognition, and under some of those scenarios (though not all) she would exhibit the power to effectively act on those reasons (Fischer and Ravizza 1998, 75ff.). Now this may well suffice for an account of moral responsibility, but as the lynchpin in an account of autonomy, it invites the response given earlier, namely the fact that we can easily imagine those cases where people voluntarily suspended their abilities to respond (occurrently) to reasons so that they could achieve a desired end. Perhaps they underwent hypnosis to quit smoking, which involved their being relatively unable to respond to considerations in favor of having a cigarette (the hypnosis made such an idea viscerally repugnant say). Or imagine a person head over heels in love: any considerations that would count validly against continuing to love this other person would fall on stone-deaf ears. But insofar as there are no such reasons (the person is really a dreamboat), then we would be remiss, I think, to call such a person lacking in autonomy, as long as the way in which all this came about is not itself so repugnant to the person that she would then be alienated from this love. Being able to respond to reasons depends, I think, on one’s attitude toward the motivational states one has and their history. Without some specification, even a limited one, of the agent’s own attitude toward her life trajectory, we would always face these untoward examples. So the subjectivist line appears the most attractive here.

iii. a subjectivist historical account In earlier work I developed an historical account of autonomy that maintained a fully subjectivist core. A key condition in that view was that a person is autonomous relative to some characteristic if the influences and conditions that gave rise to the desire were factors that the agent approved of or did not resist, or would not have resisted had she attended to them (Christman 1991a, 11; for a slightly different formulation see Christman 1993). But this approach has been criticized by several writers who correctly note that the person’s attitude toward the processes that spawn her values is separate from, and less relevant to autonomy than, the person’s attitude toward the value itself. Someone can curse the restrictive religious upbringing they were forced to undergo but come to embrace, for independent reasons, the values of that religion nonetheless (see, for example Mele 1995, 138–139). So if the historical account is to be subjective, it cannot focus

The historical conception of autonomy

143

primarily on the agent’s evaluation of the processes themselves as a separate test for autonomy. I will return to this point shortly. In typical hierarchical accounts of autonomy mentioned earlier, it was required that the autonomous person identify with her first-order characteristics in order to preserve her autonomy. As has been widely noted, the notion of “identification” is problematically ambiguous between acknowledgment and endorsement. We identify with those ego-ideals to which we aspire but failure to achieve them should not disqualify us from being autonomous. Identification can also mean simply accepting a factor as true of us, but this is true of addictions and compulsions so clearly this kind of identity does not establish autonomy.15 More importantly, elements of the self include many factors whose appraisal leaves us ambivalent, so clearly not ideal from our point of view, but also which are not compulsions to which we are resistant. We acknowledge them and accept them in a minimal sense but do not value them wholeheartedly, especially given, perhaps, the realities of our life situation. One’s attitude toward, say, one’s children (and one’s feelings for them) may well be filled with ambiguities, complex conflicts, guilt, love, disappointment, and hope. Many people have children through processes that were not chosen or planned, for example, and the feelings they have about them (and the motivations which surround that relation) can be highly complex. But such a person, I am imagining, does not fully reject that part of her life nor does she actively resist the motivational and emotional implications of it. I would like to capture that reality, then, in constructing a conception of autonomy that would not render such a person heteronomous because of those ambivalences and inner tensions. Consequently, I suggest that the proper test for the acceptability of the characteristic in question is one where the person does not feel deeply alienated from it upon critical reflection. Alienation is not simply lack of identification, in that I can fail to identify with a trait but not be alienated from it. I can simply be indifferent to it or undecided about it.16 Alienation is a stronger reaction; it involves feeling constrained by the trait and wanting decidedly to repudiate it. Second, alienation is not simply a cognitive 15 16

For recent discussions of this issue, see Berofsky 1995, 99–102; Velleman 2002; Watson 2002. See Paul Benson for cases where (he argues) indifference about trivial aspects of myself cause problems for this sort of model (Benson 2005b). See also Michael Bratman on cases of undecided reflective agents (Bratman 1987, ch. 10). I should add that the position I take here reduces (though does not eliminate) the difference between my view and Frankfurt’s latest construal of “identification,” which he takes to be simple acceptance or satisfaction with one’s desires. See Frankfurt 1992. Though of course Frankfurt’s is not an historical account. For criticism of the role alienation plays in Frankfurt’s views, see Ekstrom 2005.

144

The Politics of Persons

judgment, it is a combination of judgment and affective reaction. To be alienated is to experience negative affect, to feel repudiation and resistance. It is, moreover, to feel a need to repudiate that desire or trait, to reject it and alter it as much as possible, and to resist its effects. If, while feeling alienation and self-repudiation of this sort, a person is unable to rid herself of the characteristic in question, she is heteronomous in relation to it. A similar notion has been utilized in psychology. Kenneth Gergen describes the phenomenon of self-alienation as [A] noxious feeling arising when overt actions are detached or inconsistent with underlying conceptions of self. That is, self-alienation can be viewed as estrangement of the concept world from the daily activities of the individual. The individual might feel, “What I’m doing doesn’t reveal the real me,” “My behavior is a sham,” or in its extreme form, “I hate what I do.” (Gergen 1971, 87)17

As this explication shows, alienation in this sense also brings in an affective component to the requirements of autonomous agency, but it weakens the condition of self-identification in order to allow more flexibility regarding the ways that people relate to their often conflicting and necessarily complex internal lives. Standard models of autonomy are often overly cognitive, for example, stressing the ability to make rational, detached, and calculative judgments about the acceptability of a trait. But as we discussed concerning social identities, the relation one has to various aspects of one’s self and life are often much more than simply calculation of the factor’s favorability for oneself. It involves deeply emotional and affective components of judgment. So to be alienated is to resist and reject values in light of one’s history and social situation. The key element of such alienation is this resistance, the anxious sense that the factor in question is constraining, that it undercuts one’s settled motivational frame and sense of the validity of that frame (given surrounding conditions). This resistance, again, has both affective and cognitive components, and both are required. Indeed, I’m not sure they are easily separated in these kinds of judgments, since one is considering whether the factor in question fits into a coherent and acceptable selfnarrative; such questions are neither simply questions of belief nor questions of desire or feeling, they are all of these. This last point raises an important issue, for upon self-reflection a person may judge a factor to be repellant or unacceptable on normative grounds, 17

The concept of self-alienation is also analyzed in different form in certain areas of psychoanalytic theory: see, for example, Horney 1945/1999.

The historical conception of autonomy

145

say, but neither desire to change it nor feel particularly bad about it. Examples commonly given involve sexuality, where because of a person’s upbringing and (still sincerely held) traditional values, she judges having certain sorts of sexual feelings as immoral, but the feelings not only remain in force, they are not unpleasant and not the cause of conflicting desire. She might for example be entirely in love with her friend but everything she has been taught counsels her against such a love. For this reason it will be necessary to see the reflection in question to remain invariant over a variety of conditions. This invariance is meant to establish that the self-acceptance in question is not idiosyncratic or fleeting. I will return to this point. Note also that the reflection that this conception of autonomy calls for is both hypothetical and piecemeal. It does not require that any actual reflection take place on the characteristic in question; nor does it demand that wholesale self-evaluation from a disembodied standpoint take place. On this view, a person is autonomous (relative to some characteristic) if, were piecemeal reflection in light of the history of the factor’s development to take place, she would not feel deeply alienated from the characteristic in question.18 Again, taking this position is quite intentional, and puts the view at odds with other theorists’ explicit claims. As I argued in chapter 6, many central aspects of our lives, bodies, and social circumstances were never chosen by us nor were they consciously accepted as desirable. Yet, I would submit, they form an uncontroversial and authentic structure that defines us and relative to which we can be autonomous if we are not alienated from them over time. That we are male or female, for example, can fundamentally define and structure our plans, but that element of our personality, for most people, was never chosen or reflectively appraised. At one time, Michael Bratman expressed criticism of Frankfurt for the latter’s claim that one can be autonomous if one identifies with a desire in the sense of being “satisfied” with it based on one’s most basic cares. Bratman insisted that satisfaction in Frankfurt’s sense could stem from depression or express mere indifference and so lack the agential authority autonomy demands. Consequently he claimed that such identification had 18

Some theorists have insisted that reflection or evaluation (on one’s first-order desires for example) is necessary for autonomy, but have added that such reflection need not be frequent or even conscious. See, e.g., Friedman 2003, 8 and Ekstrom 2005. This is close to my view, but I cannot quite fathom what it means to engage in “unconscious” reflection. I say this not simply out of skepticism about the operations of an unconscious mind per se but from wondering how one could ever tell that such reflective evaluation ever took place, other than other subsequent conscious reflections along the same lines as the hypothetical ones I am requiring.

146

The Politics of Persons

to involve a decision, specifically a decision to treat the desire as a reason (Bratman 1999, 185–206). As I’ve argued, however, many attitudes and motives that we are guided by and express our settled sense of ourselves were never actively chosen by us, though in being so guided we accept them without conflict (upon reflection). This marks a disagreement with Bratman’s earlier view that desires must emanate in some way (perhaps via plans and policies) from a decisive act by the agent.19 The object of the reflection here, it should be stressed, is not the process of character formation itself, but the trait such formation produces in light of that development. That is to say, autonomy requires that one would not feel alienated from the aspect of oneself in question upon reflection given the conditions under which that factor came about. This avoids the worries raised about my earlier view where a separate approval of one’s developmental history itself was required. It is important to note that the reflection involved here would induce the alienation in question, and that such alienation, or the lack of it, is not the result of happenstance or accidental correlation. I assume, therefore, that reflection would have effects, specifically an agent coming to embrace or resist the object of that reflection. Reflection here – including the affective elements of it – is a kind of judgment about the person’s values, condition, and desires. But it is not simply a judgment about the quality of the characteristic in question (we clearly can accept aspects of ourselves about which we are ashamed); it is a judgment about the characteristic as it functions in our narrative sense of ourselves.20 To repudiate the factor in question is to judge it unacceptable as an element in our autobiography. In addition to reflective non-alienation, autonomy demands that such reflection is not itself merely the result of manipulative forces – this is to avoid the regress problem noted earlier. What must be added then, are conditions on the type of reflection being imagined so that it is both minimally competent and it is itself authentic, though not in a way that requires a further level of reflection. Therefore, the hypothetical reflection we imagine here must be such that it is not the product of social and psychological conditions that prevent 19 20

In more recent work, Bratman has abandoned his insistence on decisiveness of this sort: see Bratman 2007, 6 n. 6. Bernard Berofsky worries that such phenomena (shame) imply that critical self-acceptance should not be a requirement of autonomy (Berofsky 2003). My intuition, though, is that the specification of alienation outlined in the text is strong enough so that simple shame or disappointment in ourselves does not defeat our autonomy. (I also think that this understanding of alienation avoids doubts about it as a requirement for autonomy raised by Ekstrom: see Ekstrom 2005.)

The historical conception of autonomy

147

adequate appraisal of oneself. This requires that the person have the general capacity to reflect adequately without constriction, pathology, or manipulation. This capacity includes the ability to assess the various aspects of one’s self and conditions, and the freedom from those factors and conditions that we independently know effectively prevent minimal self-understanding. A person who endorses his decisions while in an uncontrollable rage, or while on heavy doses of hallucinogenic drugs, or from having been denied minimal education and exposure to alternatives, does not adequately reflect in this way. Picking up on the conclusions drawn in chapter 4 concerning the narrative conception of the self, we saw that while components of one’s life trajectory need not take the form of a traditional story, such components must be amenable to coherent interpretation by one’s reflective eye. One must be able to find a “fit” between the events and elements of life – past, present, and future – with the other factors of one’s social existence. When such a fit is unavailable, and this unavailability is fixed and invariant over a variety of conditions, then one cannot form or maintain a settled motivational structure; one is fractured and divided in ways that undercut agency. Further, one cannot use the construction of such a narrative to give reasons to others for one’s point of view, one’s needs, and one’s value perspective. What is relevant is that the cares, commitments, and values that structure the reflective judgments constitutive of a person’s perspective – the self of self-government – are not simply a static set of propositions, but a temporally extended matrix of evaluation and obligation that is comprehensible (both to herself and to others) as operating (repeatedly) over time. It means also that the judgments that such structures ground are not one-off events, unconnected from similar, and similarly justified, reflections which make sense as a part of a sequence. If the person makes a radical shift in value commitment – through a conversion experience, say, or by repudiating a long-standing loved one – the psychic reverberations of such a shift can be explained by this diachronic approach.21 When one comes to the realization that one can no longer remain committed to a fundamental value framework, such as a religion, one does not simply make a choice (to change religions), for in rearranging one’s value perspective in this way, one must then reinterpret the value-coded elements of one’s world as well as the multitude of judgments and choices in one’s past that were shaped by the now defunct viewpoint. 21

Cf. David Velleman’s discussion of Christine Korsgaard (Velleman 2006, ch. 12), where he discusses how radical shifts in one’s practical identity involves a change in who one is.

148

The Politics of Persons

In this light, recall that we are assuming that the psychic and bodily components of the self can plausibly be said to have irreducibly social and/ or interpersonal components (as discussed in chapter 2). This means that the practical identity guiding reflection here – the person’s point of view – operates in relation to external factors, such as one’s relation to a tradition, a set of social practices, particular relationships and so on. This perspective can become fractured, then, when those relations are destroyed or disrupted. If one were to be excommunicated from a church, for instance, one may well feel a deep disorientation in one’s ability to deliberate and judge religious issues for oneself. In many ways one’s place in a social space structures the mode of reflection and judgment that guides action and hence forms our personal point of view. I will say more in a moment about this idea of the personal perspective. But these reflections indicate the ways in which alienation can sometimes be more like total disorientation. That is, we can contrast cases where a person reflects on a fundamental value, commitment, or motivational factor and feels repulsed by it, as with an addiction. This amounts to considering an aspect of the “Me-self” as we discussed it earlier. But in addition, one’s “Iself” – the diachronic practical identity that shapes one’s perspective, structures memory, and guides the (re-)construction of one’s autobiographical narrative – can itself be so fractured or full of conflict that one is no longer able to feel at home in the world. These are cases where external social or personal aspects of one’s life are so radically changed that the guiding values that shape one’s perspective and structure one’s sense of self no longer make sense. This is also a mark of a lack of autonomy, though the source of that lack may well be surrounding social relations rather than anything internal to the self. I point this out because room should be made for asking about our autonomy relative to the most basic value commitments that ground our identity. Both in the piecemeal manner we discussed above but also in this sense of disorientation I am here describing, we can account for ways in which an individual can lack autonomy relative to the most fundamental factors of her personality and identity while not assuming a completely separated “self” that stands behind all such commitments (as we discussed in chapters 2 and 6 for example). That said, for the most part we are considering a person with a settled and fully functioning value framework who is able to critically reflect on particular aspects of herself and ask whether they fit into this value framework over time without alienation. In addition, it is important that the reflective process in question, hypothetical though it is, can guarantee the “agential authority” that autonomy

The historical conception of autonomy

149

requires. As others have discussed, our reflective selves often act merely as a cover-up for our more authentic urges and needs.22 Moreover, reflection at a particular time may not reflect our settled and more developed character and personality – we may be thinking impulsively for example. Imagine a person who is rejected by a long-term romantic partner who, let us say, taught her to play the piano. In a rage, the person throws out all her piano music and reflectively feels alienated from all things associated with music, even though music had clearly become an important part of her life. Such a person may well not be acting autonomously even if at the time she gives reflective reasons for throwing out the music. What is needed, then, is one further condition to ensure this agential authority. This necessitates a condition that captures the requirement that our reflections echo our personalities, our characters, our embodied selves, and other settled (and central) modes of our identities. Authentic reflection Standard cognitive psychological theories describe how thinking occurs by way of scripts, schemas, heuristics, and frames. These filtering and orienting devices allow us to process information and impressions in clusters instead of having to deductively infer an endless array of propositions about the world we encounter in our perceptions.23 Recall also what we said in chapter 2 about the “self-schema” by which persons organize information about themselves.24 This is not the schema of ourselves that we can produce upon reflection – the content of our self-representations – but rather the frames and modes of thought that we utilize in making those very reflections. This is the “I-self” and not the “Me-self ”. This view refers to the mode of thinking, reflecting, and feeling that expresses a person’s unique way of being in the world. For several reasons, this way of picking out the agential authority of reflective judgments is superior to approaches that focus on the content of self-reflection in picking out the seat of agency for the person.25 22 23

24 25

See Friedman 1986 and Thalberg 1989. For a parallel line of criticism, see Benson 1990. For a general discussion of these concepts, see Tversky and Kahneman 1982. For an overview of psychological material on the self, see Baumeister 1998. For a survey of recent theories of self-concept, see Ross 1992. See, e.g., Markus and Kitayama 1991. Frankfurt, for example, refers to the “content” of our deepest cares and commitments as the seat of agency, as discussed above. Cf. also Christine Korsgaard (1996) who sees agency as grounded in a person’s “practical identity,” a point of view which orients reflection on values and manifests an aspect of our self-concept.

150

The Politics of Persons

The use of frames and scripts to process information about the world and ourselves is highly individualized according to the autobiographical narrative we all embody and experience. What we can simply label our “personality” is the characteristic manner in which we organize thought, orient judgment, and react to stimuli. Such a personal style of thinking can be marked in numerous ways (the degree to which one is impulsive, deliberate, studied, emotional, and so on). What is important here is that each of us has a certain pattern of thinking and reacting which, generally speaking, is ours alone; it marks our character and personality.26 This pattern develops over time and reflects the social components of our lives as well as the shifts and renewals that characterize our approach to decision-making, emotion, and judgment. It is a composite construction based on social and cultural inputs (the internalization of the “generalized other” as some have put it).27 Such cognitive structures, when operative in evaluative reflection about either ourselves or the world, serve to orient and order our judgments in ways similar to what Korsgaard has labeled our “practical identities.” For Korsgaard, our practical identity is comprised of those values and principles that constitute a viewpoint that, for the agent, guides a life that she regards as valuable. It is a normative conception of the self containing one’s guiding principles: it is “a description under which you value yourself” (Korsgaard, 1996, 101) Korsgaard suggests that such a normative framework is one that the agent thoroughly endorses as valuable. This amounts to seeing one’s basic commitments as part of the “I-self ” – the executive function by virtue of which reflective judgment is made, rather than the set of factors which one can consider introspectively and perhaps dispassionately. Now this is not to say that one’s basic commitments are immune from reflective evaluation, but that it is a mistake to think that in their most basic form, such commitments can be brought before our mind’s eye, like the fact that I am this height or have this eye color – and considered from a point of view apart from them. Except in special cases (and in a piecemeal fashion), such commitments form the point of view that engages in this reflective activity. Velleman gets close to this point: he writes that the “self” of selfgovernment is like the visual point of view from which all (normally sighted) viewing of the world extends. It is a point, like one’s center of gravity, which is only an abstraction, it is not a functioning organ or place,

26 27

Cf. Richard Double’s reference to a “management style” (Double 1992). See, e.g., Mead 1934. For discussion of this approach to the self concept, see, e.g., Burkitt 1991, 28–54.

The historical conception of autonomy

151

exactly. It is “where we are” when we look out onto the world. Similarly, he writes: If there is a part of your personality with which you necessarily think about things, then it will be your mental standpoint, always presenting a reflexive aspect of your thought. You will be able to think about this part of your personality as “it,” but only from a perspective on which it continues to function as the thinking “I” – just as you can find a reflection of your visual location “over there” only from a perspective in which it is also “back here.” (Velleman 2006, 358)

In addition, the practical identities that guide our reflective evaluations in this way do more than provide a propositional basis for arguments in support of our lower-order judgments, such as our decisions to treat particular desires as action-guiding. Practical identities often do perform this function, for example when they are defined by principles that can be used in practical syllogisms comprising our reasons for actions. But more than this, basic commitments of this sort perform what we can call an orienting function: they order the moral world in a way that sets the stage for our evaluations themselves. Moreover, since these operate over time and include memories as well as plans and temporally extended reflections, they should more accurately be referred to “diachronic practical identities.” However, it should again be stressed that embodiment and physicality often have as much to do with identity and uniqueness as do modes of thought. A person who claims to be unafraid but shivers and cowers at the same time is not expressing an authentic self-judgment. Consider also the way one’s gender, one’s overall physical abilities, one’s race, sexuality, and so on, may shape one’s perspective and hence one’s reflections. As we have noted, embodiment is a part of our self-understanding, self-presentation, and self-schema that cannot be eliminated and should not be minimized.28 However, it is a mistake, I think, to go as far as making bodily comportment, style, and physicality the seat of agency itself, at least insofar as this might conflict with the person’s overt expression and settled inner reflections. We certainly know of cases where people’s words and reflections belie their true motives, and that such motives are expressed by bodily movement and form. But taking a person’s announced (and reflective) presentation of her values and desires over whatever alternative message her body might be expressing is crucial for interpersonal respect. Merely consider the phrase “your words say no but your body tells me yes …” to see the danger in concluding otherwise. So while it must be acknowledged that bodily 28

See, for example, Baier 1985. See also Code 1991. For discussion, see Mackenzie and Stoljar 2000a, 6–7.

152

The Politics of Persons

comportment is a key expression of our personality and hence our agency, we should take the reflective understanding of ourselves, flawed though that is, as definitive of that agency in the end.29 This is to restate the postition I laid out in chapter 6 above, where I claimed that the capacity and the practice of self-reflection is crucial in specifying autonomy despite the ways in which models of the self operative in these contexts assume a thoroughly embodied self. Not only do we allow our physicality to remain central to what counts as the “self” in these arenas, we also allow room for the ways that social markers of identity function in the specification of that embodiment. But the implications of our intentions, commitments, values, and desires can conflict, so that if bodies convey (to ourselves and to observers) certain meanings but our reflective self-reports convey others, the approach to autonomy defended here takes the latter as (for these purposes) authoritative. In addition to the intuitive argument just made, we have to recall the functional role that attributions of autonomy play in political theory of the sort enacted here: self-reflection is crucial in the process of determining and communicating our reasons (for commitments, values, and so on), reasons that interact with others’ reasons in processes of social negotiations. The legitimacy of such negotiations, as part of the process of democratic deliberation crucial for justice, is dependent on the autonomy of participants and hence their capacity to reflect in producing and communicating those reasons. How, then, might we express the requirement that autonomy-preserving reflection arises from the agent’s characteristic mode of thinking and feeling? The test might be best expressed this way: autonomy concerns reflected judgment repeated over a variety of circumstances. Such repeated (or repeatable) reflections instantiate the characteristic mode of thinking and moving that the self-schema embodies. We must, therefore, require that the reflection definitive of autonomy be such that it would yield the same result if repeated over a variety of conditions. We could abbreviate this sort of sustained critical reflection as SCR. Critical self-reflection repeated in a variety of contexts with similar evaluative results counts as SCR. Agential authority is secured via SCR, I submit, because such reflection manifests personality and dispositions, 29

In other work I defend the view that reflective self-expression should be taken as the seat of agency even though we know that people are systematically self-deluded about their own desires. The core of the argument is that outward expressions of value commitments function in the give and take of reasons that constitute deliberation and social interaction, and it is in this process that the characteristic of autonomy is designed to function. See Christman 2005. This marks a specific contrast, for example, with Meyers’s view. See Meyers 2005.

The historical conception of autonomy

153

patterns of judgment, emotional tendencies, and bodily comportment (insofar as bodies are both the object of and partially structure such reflection). The idea is that when a person reflects on a trait over time and in a variety of settings and contexts, always yielding neither alienation nor rejection, such reflection indicates the kind of settled character that autonomous agency manifests. That is, these sustained reflections express the characteristic mode of thinking, embodiment, and feeling that the agent’s self-schema manifests. This view allows for the obvious fact that our self-schema is somatic as well as mental, qualified by the point just made that self-reflection trumps bodily expression when they conflict. So the conceptual test for autonomy in this regard is non-alienation upon sustained critical reflection. But this expresses, in the broad array of cases, our embodied identity and characteristic mode of thinking. For these reasons, self-reflection of the sort demanded of autonomy carries with it the agential authority demanded of such a view. Consider again, by comparison, Bratman’s planning view of agency. For Bratman, certain higher-order attitudes operate so as to determine which desired ends to treat as justifying considerations in motivationally effective deliberation (Bratman, 2007, 101). We mentioned earlier that at one point he claimed that this operation amounted to “deciding” to treat a lower-order desire as a reason (Bratman 1999, 185–206). But as we have argued (and Bratman now agrees) motivationally effective reflections that confer agential authority on lower-order motivations in action need not have ever involved a decision to adopt them. Indeed, reference to a decision always raises the specter of a regress, since the authority behind that decision must be modeled in a way that invites another order of reflection. But the regress is avoided here not only because choice is not necessarily involved but also because, for Bratman, these higher-order attitudes represent the temporally extended agency of the person.30 The way I put roughly the same point here is that reflections (hypothetical though they are) both rely upon and construct in an ongoing manner the autobiographical narrative that forms the self, our diachronic practical identity. Now some have argued that the historical conception of autonomy implausibly requires that we are transparent to ourselves, that we have an accurate picture of our true motivations and psychological histories, when 30

They do so because Bratman adopts a broadly Lockean view of personal identity such that attitudes that are connected by way of memory or anticipation ground the identity of the person over time (Bratman 2007).

154

The Politics of Persons

clearly we do not.31 But the view I am defending here need not require such self-transparency, at least not in any manner that other accounts requiring self-reflection would not also demand. The person’s conception of how she came to adopt a desire or value commitment need only be minimally adequate as an account of her self-development, in the sense that it be consistent with accepted evidence and known causal sequences. It need not be the final or most basic description of her psychology over time. So the historical component in its full form would read “a person is not alienated upon sustained reflection in light of the history of the development of the factor and this history is a minimally adequate account.” We have noted that critical reflection of the sort being demanded here is a kind of judgment that involves both cognitive and affective elements. It is a kind of broad consideration of aspects of oneself in light of one’s history. But let us press further: is this a judgment about what would be best, or a determination of what in fact we are? Insofar as the critical reflection I am describing here involves a grasp of one’s history, then the judgment in question is whether the factor being considered can be taken in as part of the person’s overall self-narrative; whether the person can accept this part of herself (without alienation) as part of an autobiographical narrative she embodies and enacts. One is not simply asking whether this factor is best for a person to have in general, nor is one asking merely whether one in fact embodies it. Rather one is asking whether I can take this as part of my ongoing autobiography, looking at my past as well as projecting into the future, and avoid the feelings of resistance and rejection characteristic of alienation. Therefore, “critical reflection” here refers to the consideration of the factor as an ongoing part of one’s self-narrative. Now before rounding out the components of the model, we must add a condition that captures the idea that autonomous agents are competent agents, in the way that several theorists of autonomy have stressed.32 There are a variety of skills and abilities that are fundamental to acting intentionally that must be included as part of the requirements for autonomy. The ability to effectively form intentions to act, then, along with the various skills that this requires, must be seen as necessary for autonomy. It must be noted further that the ability to act – successfully and as planned – cannot be what we mean here. I am often prevented from acting or completing my plans because of the happenstance of my surrounding circumstances. Such circumstances certainly make me less free (in 31 32

See, for example, Mackenzie and Stoljar 2000a, 16. See, e.g., Atkins 2004; Mele 1995; Meyers 1989; and Oshana 2006, 76–86.

The historical conception of autonomy

155

a certain sense of freedom), but they do not make me less autonomous, at least if this latter term is to retain any of its conceptual distinctiveness. So the competence conditions for autonomy merely refer to the effective ability to form intentions to act but not to complete such actions. My focus in this chapter is more on conditions of authenticity rather than competence, so the account of competence I give here is clearly incomplete.33 However, suffice it to say that a competent agent is minimally rational (where, for example, her desires and plans contain no manifest contradictions that could be easily brought to consciousness), she displays minimal self-control, and is generally able to form effective intentions that in the absence of external barriers lead to completed action. To summarize the conditions discussed here, autonomy can be specified as obtaining if the following conditions hold (as elaborated in the previous discussion): Relative to some characteristic C, where C refers to basic organizing values and commitments, autonomy obtains if: (Basic Requirements – Competence): 1. The person is competent to effectively form intentions to act on the basis of C. That is, she enjoys the array of competences that are required for her to negotiate socially, bodily, affectively, and cognitively in ways necessary to form effective intentions on the basis of C; 2. The person has the general capacity to critically reflect on C and other basic motivating elements of her psychic and bodily make-up; and (Hypothetical Reflection Condition – Authenticity): 3. Were the person to engage in sustained critical reflection on C over a variety of conditions in light of the historical processes (adequately described) that gave rise to C; and 4. She would not be alienated from C in the sense of feeling and judging that C cannot be sustained as part of an acceptable autobiographical narrative organized by her diachronic practical identity; and 5. The reflection being imagined is not constrained by reflection-distorting factors.

If one prefers prose (albeit rather inelegant prose), we can summarize the view this way: Autonomy involves competence and authenticity; authenticity involves non-alienation upon (historically sensitive, adequate) selfreflection, given one’s diachronic practical identity and one’s position in the world. To elaborate: “alienation” is active resistance to the factor in question; “sensitivity to history” involves the requirement that we accept the 33

For a general discussion of the relation between rationality and autonomy, see, for example, Berofsky, 1995, chs. 6 and 7. For a discussion of self-control and autonomy, see Mele 1995, part I. For a discussion of other autonomy-related competency skills, see Meyers 1989, part II, section 4 and Atkins 2008, ch. 7.

156

The Politics of Persons

factor in light of how it came to be part of our psychic economies; “upon … self reflection” refers to hypothetical reflection in the narrow counterfactual sense of reflection were we to turn our attention to the factor and all else stayed constant; “adequate” self-reflection involves competent critical appraisal that would remain constant over a variety of conditions; “diachronic practical identity” refers to one’s action-guiding attitudes, values, and principles that orient reflection and are central to our autobiographical narratives – the “I-self” – and are structured by (variable) social elements as well as one’s physical embodiment.34 These are meant generally as sufficient conditions for autonomy. They should be understood as elaborated in the longer discussion of each condition above. To further shore up support for this approach, however, let us consider some lines of criticisms that have been raised about similar views or earlier versions of my own view. This will allow us to discuss still further the meaning of these desiderata.

iv. further defense of the historical approach Two lines of argument for an historical dimension to autonomy have been broached here, one specific and one quite general. The general consideration is this: human agents are diachronic, developing beings whose lives are structured in a kind of narrative which unfolds over time; to capture the idea of self-government properly, attention must be paid to this diachronic nature of selves. The specific claim is that “time-slice” (non-historical) accounts of autonomy all face the difficulty that a person who meets all the current conditions of self-endorsement or competence they set out nevertheless could be placed into that condition by manipulative, external (hence heteronomous) factors; so historical considerations are crucial. Yet there have been several writers who have specifically targeted historical accounts of autonomy for criticism. Bernard Berofsky, for example, has argued that even if we demand rational self-reflection as a condition of autonomy, such reflection need make no reference to the history of the person whose autonomy is in question. If current choices obey a fully worked-out requirement of what I am calling here competence (Berofsky labels it “objectivity”), there is no need to refer further to the processes by 34

Putting this last element in this form makes the similarities with Bratman’s planning model of selfgovernance more apparent, in that diachronic practical identities here described are notably similar to what Bratman refers to as self-governing policies. See Bratman 2007.

The historical conception of autonomy

157

which the agent came to adopt the trait in question and, therefore, to her attitude toward those processes (Berofsky 1995, chs. 6 and 9).35 Part of the case he makes for this conclusion parallels points made above concerning the original formulation of the historical account: a person can despise the particular manner in which she came to develop a particular desire but nevertheless accept it as a valued part of her current personality. The revised version of the view stated here, however, avoids this difficulty, since the person must be alienated from the current desire in light of its tainted past, but her judgment of the past simpliciter is not dispositive. But Berofsky also raises a more general worry about historical conditions which can be brought out this way: we can imagine both that a person can lead an autonomous life that arises from heteronomous processes and that a person can lead a heteronomous life that results from autonomous processes. The first possibility is exemplified by any person who develops by way of oppressive, reflection-inhibiting processes – strict childrearing techniques for example – but who comes to exhibit fully independent and rational capacities for flexible choice as an adult. The course of a person’s development contains many twists and turns which may not embody open reflective self-endorsement.36 The case of autonomous agents arising from constricted and even manipulative processes has been touched on earlier. In fact I tried to emphasize how requirements of ongoing reflective awareness and control of our development ignore the innumerable ways that we grow, change, and engage in important and valued activities without enjoying such awareness and control. My insistence that critical reflection need only be hypothetical is meant specifically to accommodate this fact. There is, however, a variation on these kinds of cases that does give me pause and, as far as I know, has not been brought out in the literature in quite this way: We can imagine a person who finds out she has been severely abused in her childhood by someone who is responsible for several of the proclivities and skills she has developed. In coming to grips with the memories of the abuse, she has repudiated many of those proclivities. In addition, let us say, she doesn’t realize or remember that this person also taught her to play the piano, which she still loves to do and does well. If she were told that her piano playing was also rooted in her time with this abuser, she would feel alienated from that part of herself also; but in her ignorance 35

36

Others, such as Frankfurt and Velleman, have also specifically rejected the relevance of agents’ histories to their autonomy. See Frankfurt’s “Reply to John Martin Fischer” in Buss and Overton 2002, 27, and Velleman 2002. For a parallel line of criticism, see Meyers 1989, 28–30.

158

The Politics of Persons

she plays on contentedly. Does this person count as autonomous? On my view she does not since, were she to reflect on the trait in light of its origins she would be alienated from it. I worry about such a case, but my own intuitions about it are not strong enough to induce me to amend the view to account for it. If people are under illusions about their own character – illusions that would cause them severe internal distress leading to selfrepudiation if made clear – they are not really autonomous relative to them. So I do not, in the end, think such cases provide reason to alter the account. However, such a case does bring to light an element of this approach that might give people pause in its general form. That is, the view I take relies heavily on the subjective, affective, responses of the individual, something that may well be too variable, unpredictable, and unavailable to interpersonal appraisal to function well as a component of autonomy. Autonomy is the seat of interpersonal respect, as we have noted, but having that characteristic hang on the fluid and perhaps idiosyncratic feelings people might have about their inner states may well render the concept useless in public settings of the sort we imagine here, as in the design and regulation of political institutions. In response let me note that the judgments of (non-)alienation that are part of the test for autonomy range over time and are considered as invariant over a variety of circumstances, as noted earlier. This is what gives these judgments their agential authority: they reflect what I called the person’s diachronic practical identity. This means that fleeting and superficial reactions to one’s commitments and values won’t disturb autonomy. Also, these judgments concern our deepest normative commitments and related social and personal circumstances, so that self-reflection here does not refer to simple tastes and desires (which may well change with the wind) but to those pervasive normative structures that shape our self-narrative.37 The threat of making our commitments to those a matter of psychological happenstance is far less severe I think. Also, some have argued that authenticity in general is otiose with regard to autonomy. They argue that we need only imagine a person who thoroughly conforms to others’ expectations and/or social norms while repres-

37

Recall that in chapter 4 we defined “narrative” in terms of “socially mediated” norms. In this way, selfreflection of the sort that secures autonomy, on my view, involves social values that the person has internalized as part of the mental language of self-reflection. On this point my view has much in common with an aspect of Stefaan Cuypers’s view of autonomy in that he claims that autonomy must involve what he calls “social self evaluation” (Cuypers 2001, 146).

The historical conception of autonomy

159

sing all of his own principles and motives. We are asked to imagine, for instance, Sartre’s paradigmatic exhibitor of bad faith: the waiter in the café who strives with his every movement to act as a waiter is meant to act while never regarding such action as grounded in his values or ideals. Such a person lacks all integrity in relation to his own ideals. But, these critics claim, as long as he is self-controlling, otherwise rational, and a competent agent, it seems artificial to say he lacks autonomy (Arpaly 2003, 122; Velleman, 2002, 97). However, on the interpretation of authenticity spelled out here, lacking integrity or general faithfulness to one’s central values (as in this kind of case) does not spell a lack of autonomy. The person putting on this charade would need to be thoroughly alienated from the ideals she is mimicking; she would have to judge them as not only external to her own practical identity but also feel resistant to them, as intolerable as part of her temporally extended self. If despite all this, the person continues to put on the act of conforming to these external expectations, it is not clear she is even rational, as her deliberations appear to involve manifest inconsistency; so calling her heteronomous seems apt. On the other hand, if we imagine that she has overriding reasons for going through this charade – say she wants to impress some influential people who will be useful to her somehow – then it is not clear she is being inauthentic, since her artificial conformity fits into a plan which, we are imagining, is coherent and acceptable from her point of view. The other direction of criticism of the reformulated historical account I mentioned earlier is intended to show that processes which are autonomous (that is, acceptable to the person) may well lead to heteronomous states, exemplified by the willing slave or the voluntary addict. In such a case the person’s acceptance of the development of a constraining situation is irrelevant to the verdict that such a situation is, indeed, constraining to the point of heteronomy. We can imagine persons who meet the conditions I set out for autonomy but, for reasons peculiar to them and their situation, allow themselves to become severely constrained, compulsive, obedient, or the like. The criticism is that any view of autonomy that implies that such persons are autonomous while under such constraints is defective. This charge is often made of any subjectivist or “procedural” account of selfgovernment.38 Now we should clarify the cases we’re imagining by attending properly to issues of time indexing in the attribution of autonomy to the allegedly 38

Indeed, many use the possibility of an autonomous slave as simply a counter-example to any view of autonomy that has this implication. See, for example, Waller 1993.

160

The Politics of Persons

contented-but-constricted person: when in the person’s life are we being asked to evaluate her autonomy, and for what span of time? In most cases, one is judging a person’s autonomy at a particular time in light of decisions she may have made in the past that bring about or accept the present restrictions. We can imagine, for example, a psychologist who voluntarily addicts herself to heroin in order to experience its effects. It is clear, one might claim, that while the process leading up to the addiction is free from the taint of manipulation, the person is nonetheless addicted as a result and is hence heteronomous.39 But it is obscure what the locus of her autonomy is in this case: in a sense she has autonomously reached her goal of experiencing the pangs of addiction; but on the other hand, her current craving for the drug is not subject to ongoing deliberative control of the sort autonomy seems to demand. The question, simply, is what factor are we trying to characterize as autonomous? Should we not be able to say: a person at t1 autonomously chose to relinquish her autonomy at t2 (and hence is not autonomous at t2); or she autonomously accomplished a goal at t2 using techniques that involve her relinquishing, during the period from t1 to t2, reflective control of her environment and her desires, hence instantiating a longer period of unorthodox but nevertheless autonomous planning and achievement? Moreover, concerning her basic motivating and orienting values – her practical identity – this self-constricting plan of action is fully endorsed; she wants to be the kind of person who constrains herself during these periods. It is clear that a general ability to do otherwise, or even to desire otherwise, is not in itself a plausible requirement for autonomy. As I have stressed, we are often in states that are not revisable by us without tremendous pain, effort, or even outside assistance. We might be madly in love, for example, or unquestionably devoted to our children, and having such desires, values, and commitments could not be shed without great psychological cost, to an extent easily comparable to the effort it takes to resist the pangs of a heroin addiction. The requirement of an ongoing ability to choose or that our current states are the result of choice and decision is simply too stringent to account for the complex constraints we all live under in our normal lives. Describing such cases as “slave” cases often begs the question, for it masks the fact that by assumption these are people living out a reflectively endorsed autobiographical narrative that reflects embedded values operative over time and across conditions. The fact that the values 39

For a criticism along these lines, see Oshana 2006, ch. 3. Cf. Berofsky 1995, 127–129, 211–214.

The historical conception of autonomy

161

embodied in those reflective self-narratives are disturbing or self-abnegating does not gainsay the judgment that they are truly the person’s own. At points such as these, however, further meditation simply on the concepts involved ceases to yield philosophical results, as the core notions of “self-government,” “authenticity,” “independence,” and the like are not fine-grained enough for us to settle on their precise applications. What must be considered is what practical or theoretical work the attribution of autonomy will be doing in the context in which it operates, a context that should, ideally, be spelled out in fairly detailed terms. The context where autonomy plays a central part in the project described in this book concerns specifying the grounds of general moral respect whereby the scope of principles of justice is specified, in particular what kinds of agents are subject to such principles. In these contexts, sensitivity to pluralism and difference compels us toward a maximally flexible notion of autonomy, so that a broad variation of conceptions of the good, ideals of personhood, and approaches to personal and social life receive full respect.40 In this regard, we may well have to consider cases where persons have developed spiritual or traditional value conceptions of which revision or reconsideration is psychologically untenable for them. Analogous (though not identical) to the “willing” slave, such persons have internalized restrictive value systems that include, for example, codes of strict obedience that do not allow moments of detached reflection and possible rejection of those codes. Some have argued, in fact, that subservience to the dominant modes of value and hierarchy, which marks the experience of members of marginalized groups, are reflected in not only the first-order actions of the persons in question but their reflective “cover stories” as well. In this way, a view of autonomy that merely checks whether the person endorses (or is not alienated from) her values will fail to mark as heteronomous (and oppressive) those conditions that produce agents of this sort. The emphasis on the subjective, individualized reflection that views such as mine include continues the problematic tradition of individualized self-definition that has been the proper object of suspicion and critique in recent decades. However, there is a way in which this worry is simply a version of the “regress” problem discussed earlier, in that it asks how we can tell that patterns of reflection are themselves authentic and speak for the privileged seat of self-government that autonomy reflects. I have tried, however, to obviate these worries by adding conditions for adequate and authentic 40

For a similar point, see Dworkin 1989b, 21–23.

162

The Politics of Persons

reflection in the model of autonomy here developed. Such reflection must be sustained, hence repeatable over a variety of conditions. Also, only if the reflection we are imagining (since it may be hypothetical) is free of factors which we know independently distort introspection and inhibit reflection, can we say that the lack of alienation resulting from that reflection marks autonomy. In this way, a regress is avoided and we can forestall the possibility that heteronomous manipulation is as responsible for the reflective judgments of the person as they are for her first-order values. If we still want to insist that a person who can meet all these conditions – nonalienation, adequate and authentic reflection – is not autonomous because of the taint of restrictive conditions she has put herself in, then we will have to accept the way in which such a position disqualifies those who have accepted obedient, constrained, and, to us, oppressive social positions out of sincere and personal conviction from being counted as political agents worthy of respect. To function in the manner I indicated, autonomy in this sense must pick out all and only those agents whose capacities and point of view should matter as the sources of valid claims in collective decisions and toward whom paternalistic intervention would be disrespectful. Recall also that autonomy is understood as relative to the person’s basic organizing commitments and values. So when a person either is unable to reflect adequately at all (in the manner I’ve described) or, were she to reflect on her basic motivating values she would judge them alien to her diachronic sense of herself (and conditions that make this possible are inescapable), she is not autonomous. What this implies is that there will be two kinds of people being ruled out here as heteronomous: one is unable to reflect adequately at all and hence should not be counted in collective decisions, at least as a self-representing participant in the usual way; the other is a person who feels trapped or alienated from some central commitment or other but is generally able to reflect. The former might well be open to paternalistic care rather than egalitarian respect in our social dealings with her. The latter kind of person, though, should certainly be included (nonpaternalistically) in the membership of collective decision-making bodies, but her situation calls for reform of the conditions that result in her alienation and constraint. Many issues are left unresolved here, and clearly the account of autonomy I present needs much further fleshing out. My aim in this chapter, however, has been to develop and reformulate a conception of autonomy that has distinctively historical components. The purpose of such a reclamation project is to help reconnect conceptions of the person at work in

The historical conception of autonomy

163

mainstream philosophical views with the facts of cultural identity and social embeddedness which various critics of autonomy-based conceptions of justice have used to reject that approach wholesale. If successful, the analytic moves made here can be part of a larger project that enlarges the purview of this approach to justice and calls such critics back into its fold.

chapter 8

Relational autonomy

With a conception of autonomy outlined, it will be helpful to consider alternative approaches to the concept that also see it in its role as a fundamental political idea. In particular, it may be illuminating to see another way in which the traditional hyper-individualism of autonomy-based principles have been rejected and replaced. Critical examination of this alternative will help us, in addition, to consider the issue of recognition of persons’ social identities in the models of citizens’ perspectives and interests in autonomybased principles of democratic justice. In the end, we will see how the concept of autonomy developed in the previous chapter, along with the model of (socio-historical) self spelled out earlier, will allow us to accommodate the most pressing concerns in these areas but avoid some prickly implications as well. i. relational selves and relational autonomy Feminists have been especially vocal in the claim that the idea of autonomy central to liberal politics must be reconfigured or abandoned so as to be more sensitive to relations of care, interdependence, and mutual support that define our lives and which have traditionally marked the realm of the feminine.1 The resistance to conceptions of justice based on the independent individual with no ties to family, children, and significant others reflects this suspicion of autonomy as an ideal geared toward the life experience, for the most part, of privileged males. Emerging from this discussion is an alternative view of the autonomous person that is structured so as to fully embrace this social conception of the self. “Relational autonomy” is the label that has been given to the conception of what it means to be a free, self-governing agent who is also socially constituted and who possibly defines her basic value commitments in terms of interpersonal relations and mutual 1

See, for example, Jaggar 1988, 29; for an overview, see the essays in Meyers 1997a, part 4.

164

Relational autonomy

165

2

dependencies. These conceptions underscore the social components of our self-concepts as well as emphasize the role that background social dynamics and power structures play in the enjoyment and development of autonomy. However, when conceptions of relational autonomy are spelled out in detail, certain difficulties arise which should give us some pause in the utilization of such notions in the formulation of principles of justice, especially those motivated by feminist and other liberatory concerns. In this section, I want to take a closer look at the conception of relational autonomy as it has been developed in some recent work and to suggest some friendly amendments to those views, amendments which share the call for greater attention to the social nature of the self but which, in the end (and with much qualification), direct us back to a kind of individualism in the concept of the autonomous person, a move that is necessary if that idea is to do the theoretical and normative work that both liberal theorists and some of their feminist critics want it to perform. Taking off from the various criticisms of the hyper-individualism of traditional liberal theories, many writers from different quarters have insisted that a better understanding of the subject of principles of justice sees persons as relational and socially constituted in fundamental ways. Insofar, then, as the autonomous person functions as a central idea in moral and political reflections, such a “person” should be conceived as fundamentally and irreducibly relational. Now we have discussed at length the ways that any plausible philosophical or political theory must take into account the various ways in which humans are socially embedded, intimately related to other people, groups, institutions, and histories, that they experience themselves and their values as part of ongoing narratives, and that they are motivated by interests and reasons that can only be fully defined with reference to other people and things. But it is equally important to understand the logical gap between the rejection of a metaphysical individualism in this realm and the embrace of a metaphysically relational conception of the self.3 For it is one thing to deny that persons can always or should always be conceived without essential reference to social context, and it is quite another to claim that they should be conceived with a particular reference to some aspect of social context. The latter, relational, conception of self – at least when posed as a metaphysical claim – runs the risk of ignoring the very variability, 2 3

The most often cited source for the call for a new notion is Nedelsky 1989. See also Oshana 1998 and 2006. For an overview, see Mackenzie and Stoljar 2000b. Also, see Barclay 2000, Sher 1989, and Wong 1988. For an interesting discussion of whether the logic of identity always and everywhere implies an unacceptable repression of difference, see Weir 1996.

166

The Politics of Persons

contingency, and temporally fluid nature of human existence that motivates the rejection of old-style individualism. Nevertheless, as I have argued (and others have noted), there is nothing about a social conception of “self” that is incompatible with an individual conception of autonomy.4 Indeed, insofar as autonomy requires some measure of internal integration of the disparate elements of the self, and the self is constituted by social elements, then one cannot be autonomous relative to those social elements unless one exists in environments that allow their full manifestation. For one can claim that I am autonomous just in case I can turn a reflective eye to certain aspects of my character, even if those aspects can only be defined relative to external relations I have (or have had) with others. If political institutions and social patterns make it impossible to pursue projects defined by the connections by which I define myself, I will surely experience alienation (upon reflection) relative to such conditions; hence those institutions and patterns that induce this phenomenon are inimical to autonomy. So not only is autonomy not resistant to support for communal and social structures that shape and undergird human identities, it in fact demands them. Still, several writers have taken up the mantle of the social view of the self by promoting a relational approach to autonomy itself. “Relational autonomy” does not refer to a single account but is rather, as Catriona Mackenzie and Natalie Stoljar put it, an “umbrella term” which refers to all views of autonomy that share the assumption that “persons are socially embedded and that agents’ identities are formed within the context of social relationships and shaped by a complex of intersecting social determinants, such as race, class, gender, and ethnicity” (Mackenzie and Stoljar 2000a, 4). What also must be true to make a conception of autonomy uniquely “relational” or “social” is that among its defining conditions are requirements concerning the interpersonal or social environment of the agent.5 In particular, to mark out such accounts from others in the literature, social conditions of some sort must be named as conceptually necessary requirements of autonomy rather than, say, contributory factors.6 4 5

6

Mackenzie and Stoljar 2000a, 8. See also, Kymlicka 1989, ch. 5 and Crittenden 1992. The terms “relational” and “social” do not mean the same thing, and it would be instructive to examine their different connotations and implications, given the variety of motivations for such non-individualized accounts. For example, “relational” views seem to express more thoroughly the need to underscore interpersonal dynamics as components of autonomy, dynamics such as caring relations, interpersonal dependence, and intimacy. “Social” accounts imply, I think, a broader view, where various other kinds of social factors – institutional settings, cultural patterns, political factors – might all come into play. See Nedelsky 1989. For an argument that autonomy may well not be able to shed its atomistic baggage, see Code 2000.

Relational autonomy

167

Jennifer Nedelsky has argued for instance that the process of “finding one’s own law” specified in traditional accounts of autonomy – the processes of establishing the authenticity here referred to – can only occur in social conditions that foster certain types of human relationships. While traditional accounts of authenticity refer only to the isolated agent reflecting on his or her own desires, relational accounts “think of autonomy in terms of the forms of human interactions in which it will develop and flourish” (Nedelsky 1989, 12). Notice, however, that this specifies the conditions that allow autonomy to develop rather than the conceptual conditions that define it. Writers like Nedelsky are surely correct that discussions of autonomy have focused on the individual agent as if “he” were able to develop an authentic set of values and desires without the caring support of various others and the social structures that contribute to human self-development. And moreover, feminists have powerfully claimed that relationships that are necessary for the growth and development of healthy personalities are often ignored in accounts of autonomous personhood, motivated most likely by a devaluation of the traditional feminine roles of educator, mother, and caretaker. In addition, feminist concerns of this sort point to the ways that many of us find our authentic selves only in relation to various others, those we care for as well as cultural traditions, communities, and causes. However, if these lessons were all taken to heart, as they should be, we might still define autonomy as an individual undertaking, as a set of capacities which a person, apart from others, might exercise.7 For reasons that will emerge as we proceed, I want to focus, then, on relational views that see interpersonal and social factors as conceptually necessary for autonomy. One of the most developed and powerfully defended accounts of “social” autonomy has been put forth by Marina Oshana, who insists that autonomy should be seen as a “socio-relational” phenomenon (Oshana 1998, 2006). This is in contrast to the purely “internalist,” “psychological” accounts that pepper the literature (accounts that, for our purposes, can be seen as equivalent to the subjectivist historical view I defend here). Oshana faults internalist accounts of autonomy for running afoul of our intuitions in cases where agents seem to accept social conditions that deny their dignity, stature as independent agents, and essential self-determination. 7

For a similar distinction, see Friedman 1997, 57–58. Notice also that, despite the label, relational autonomy is still meant as a characteristic of individuals, not the groups, relationships, or social collectivities in relation to which autonomy is enjoyed. To pursue the latter route would mean that only “we” are autonomous, not “I” in relation to you and them.

168

The Politics of Persons

Such “essentially subjective” accounts are also unacceptably individualistic (Oshana 1998, 81). Oshana defends this view by describing a series of cases which illustrate the way that internalist conditions of autonomy come up short: they consistently ignore the importance of various social conditions that, while at some level are “acceptable” to the person, are fundamentally oppressive and restrictive. Examples include voluntary slavery, a subservient housewife, a religious devotee, and a female member of the Taliban. The structure of these cases is familiar, in that they follow in line with the “happy slave” examples that are often mentioned in this context. We are to imagine persons that meet all the conditions of competence and authenticity that internalist accounts of autonomy demand, but choose to enter or continue in conditions which deny them the basic opportunities for self-determination that seem to mark autonomy. Such cases, Oshana claims, systematically offend our intuitions that the “autonomous person is in control of her choices, her actions, and her will,” that she is able to meet her goals without depending upon the judgments of others as to their validity and importance … [While she] may require the assistance of others in meeting those goals, she decides which of them are most important” (Oshana 1998, 82). Of the voluntary “slave,” she writes, the “slave may never actually experience treatment of the sort that provides plain evidence of a failure of selfdetermination” (Oshana 2006, 54). So what’s the problem? It is that the person “could be punished or mistreated at the master’s whim … When a person’s ability and right to exercise control over his life are incumbent upon the independent, effective effort of another person (or persons) who possesses superior and ultimate authority over him, the person’s autonomy is lost” (Oshana 2006, 55). Oshana even considers a member of a restrictive religious order who actually has a yearly opportunity to revisit his commitment to the order and leave if he desires. (In the interim he is subjected to the unquestioned authority of religious elders and has no possessions or prerogatives outside of the rules of behavior of the order.) Of such a case she says that it is “farcical to attribute autonomy to a person who is not at liberty to leave a situation antithetical to self-government, even if the person assents to the situation on a continual basis” (Oshana 2006, 63). It is interesting, first, that the motives and reflections that led the people in question to adopt their restrictive conditions are never spelled out. Clearly, Oshana thinks this is irrelevant, since she admits that such people meet the conditions of procedural independence and reflective selfendorsement laid out by internalist models (such as the one defended here).

Relational autonomy

169

But laying out such reasoning may blur our intuitions a bit, for it may well paint a picture of a person whose deeply and sincerely held value commitments, which give meaning to her life and without which the world would make little sense, lead her to what seems to us to be such a constrained situation. Nevertheless, Oshana claims that examples such as these show that sociorelational conditions that go beyond simply the subjective self-endorsement (or non-alienation) that views such as my own require are needed. She insists that substantive independence is also required for autonomy, for example. Now we have said much in the previous chapter to defend the proceduralist view against arguments of this sort, ones describing thoroughly restricted people who voluntarily and autonomously chose their conditions. Additionally, consider the following kind of person whose life condition structurally resembles the kinds of “happy slave” cases that theorists such as Oshana base their claims upon: As part of her life condition, Betty must respond to all of Bob’s needs and desires. When Bob beckons, she must rush to him; when he expresses a desire, even the most whimsical one, she must either satisfy it immediately or somehow negotiate her way around it (distracting or tricking Bob until he stops asking). If she refuses to meet these obligations, she suffers almost intolerable consequences that she is loath to expose herself to. She also thoroughly embraces this role, sees herself as somehow noble in pursuing Bob’s every need and regards her life as valuable to the degree she makes Bob happy. He has absolutely no corresponding attitude toward her, and in fact is completely indifferent to her needs and desires. Betty has canceled or curtailed most of her own pursuits that conflict with this role of serving Bob’s needs. This arrangement will last from 5 to 10 years (or more) without possibility of exit, at least not without incurring tremendous psychological cost to Betty.

This sounds like another happy slave case, and if we build into the description such loaded terms as “is subservient to” or “is the slave of” it would settle the issue, intuitively speaking. Consideration of such a scenario matches the strategy that critics of procedural autonomy have used. But our intuitions may point in exactly the opposite direction when we realize that the case was sketched to describe the relation of mother and young child! I will return to this issue in a moment.8 Let us continue for now the examination of the relational view. As a replacement for the individualist, 8

As we will see below, Oshana’s view avoids the implication of calling someone like Betty heteronomous. For a discussion of cases of this sort, where the issue of whether the person being tended to is him- or herself an agent, see Taylor 2003.

170

The Politics of Persons

internalist accounts of this sort, Oshana puts forth a provocative alternative view of autonomy. On her account, the conditions of autonomy include various individualist requirements – epistemic competence, rationality, procedural independence – but also normatively substantive conditions such as self-respect. But the important additions in her view are social and relational conditions, such as enjoying an adequate range of relevant options. Here she follows Joseph Raz in arguing that a person is not autonomous if she lacks sufficient and morally acceptable options; for Oshana, this means options that allow one to develop one’s capacities (Raz 1986, 373–378; Oshana 2006, 84–86). It is important to distinguish two modes of justifying the kind and range of open options that autonomy might require. The first is that having this or that kind of option is needed for the agent to pursue actions or activities that she values. This might include having more than those options that she in fact pursues, in that a choice of roads not taken may well be necessary for informational purposes (one can’t be sure of the optimality of one’s top choice, so it helps to know one has fallbacks) or as a way of underscoring the value of the chosen option for the person.9 The other mode of identifying the adequacy of options, however, is that having certain opportunities is more objectively valuable, as determined philosophically by theorists, independent of what the agent wants (even wants with full information, after due reflection, and so on). Now most of what the objectivist specifies as the kind and range of options required for autonomy can be captured by the proceduralist view by simply pointing out that such a range is generally required for all life pursuits. Having choices about what life path to pursue or where to live and with whom will be required by most anyone since such choices figure centrally in virtually any valued project. But the reason it is required has to do with the agent’s perspective and judgment, not a theorist’s. The argument for this approach, however, is that what will count as a restraint or an option will be a function of the agent’s range of values, for without reference to such a range, it is entirely indeterminate whether any particular object or open space should be classified as an obstacle or option. Are my options increased when the desk across the room is moved to a different spot? Perhaps, if events conspire so that I must move in a path that the desk blocks. But absent even oblique reference to plans, action sequences, and

9

Thomas Hurka argues, for example, that in choosing a one becomes responsible for the choice of notb, not-c, etc. (Hurka 1993, 150).

Relational autonomy

171

(hence) values and preferences by actual agents, the counting up of options will be meaningless.10 The specifically socio-relational conditions Oshana lists, however, all come under the banner of “substantive independence.” They include: social and psychological security (persons cannot deprive the person of de facto or de jure power and authority “characteristic of global autonomy”) (2006, 86); the person can pursue goals different from those who have influence and authority over her; the person is not required to take responsibility for another’s needs unless reasonably expected in light of her particular function;11 she enjoys financial self-sufficiency adequate to maintain independence from others; and the person is not deceived. Much can be said about all these requirements.12 What is clear, however, is that they describe clearly an ideal of an independent life, an ideal which is, despite its label, a thoroughly individualist one. For in cases where a person fundamentally defines herself as part of a social network, a network perhaps where she authentically and reflectively accepts a condition of strict obedience, a proceduralist will count her as autonomous, as long as the (rather stringent) conditions of authentic acceptance are met. This mirrors the assumption that selves, at least in some ways and in some instances, should be seen as constituted by the social and interpersonal dynamics that surround them. But Oshana’s view insists that to be autonomous, she must, as an individual, maintain the ability to “pursue goals different from those who have influence and authority over her.” This view is in some tension with the idea that persons should be understood to be constituted by social relations (in some ways or in some instances), at least when those identityconstituting relations are overly authoritative. This observation points to the fact that views like Oshana’s combine substantive, perfectionist, conditions for autonomy – that autonomous agents must have certain value commitments and/or must be treated in certain normatively acceptable ways – with socio-relational conditions. 10 11

12

For further defense of this point, see Taylor 1985c and Christman 1991b. I should note that this condition would rule out the mother, Betty, I described earlier. But it is unclear why certain “roles” should be exempt from being classified as heteronomous, unless, of course, they are so classified because they are freely chosen (but that is not the explanation Oshana would pursue). For example, the condition that autonomy obtains only when the person is protected from physical threat seems too strong, as most people in most societies are in principle the possible victim of a crime (especially women), and while we might want to say that this threatens their autonomy in a sense, such a judgment would show that we are using the term in a fully idealized manner. Also, many people such as firefighters, police officers, and military personnel take on roles that subject them to extreme danger, yet we would not call them non-autonomous.

172

The Politics of Persons

Proceduralists defend their views in part in order to be able to utilize the concept of autonomy in as broad a value terrain as possible; indeed the attempt to maintain a value-neutral stance for our model of autonomy, in that we define autonomy without direct reference to the content of the value systems that define and motivate agents, reflects this. What views like Oshana’s rest upon is the claim that certain substantive value commitments – such as the view that I must obey my superiors unconditionally – are conceptually inconsistent with autonomy. But there is in fact a tension between the perfectionist aspect of the relational view and its anti-individualism. The latter aspect shows itself in the need to not only make room for but reify the social nature of the person, her values, and her psychology. The perfectionist strain, however, appears in the form of the denial of purely procedural, that is, content-neutral conceptions of autonomy. Relational theorists who decry procedural views on the grounds that they would allow voluntary slavery to masquerade as autonomy are in fact supporting a conception of autonomy which is an ideal of individualized self-government, an ideal that those who choose strict obedience or hierarchical power structures have decided to reject. Those whose value conceptions manifest relatively blurred lines between self and other, who downplay the value of individualized judgments and embrace devotion to an externally defined normative structure (which may include obedience to particular human authorities) stand in defiance of the normative ideals that such relational views of autonomy put forward. It is one thing to say that models of autonomy must acknowledge how we are all deeply related; it is another to say that we are autonomous only if related in certain idealized ways. I will return to this issue in a moment. Below it will be made clear that most of what theorists like Oshana are pressing for is acceptable in fact on the particular proceduralist view I develop here, namely that in most cases of oppressive social relations, autonomy is squelched by those relations. But the particular reason why autonomy is lost here is that the victims in such relations are unable to see themselves (without alienation) as even part author of the social narrative of which they are a part. It is not social connectedness as such, and not even hierarchical and highly unequal social relations, that disturb autonomy, but those social relations where participants cannot see themselves as “selfauthenticating sources of valid claims” to use Rawls’s phrase, or as we have it here, as part of an individual and social narrative that they can authenticate from inside of a diachronic practical identity of their own. The greater inclusiveness of the conception of autonomy developed here is motivated by a need to extend as far as possible the normative

Relational autonomy

173

banner under which political participation in democratic society extends, the boundaries of which are defined by autonomy. Indeed, theorists who decry the potentially exclusionary implications of traditional liberal conceptions of the person – so-called “difference feminists” for example – should be doubly concerned about conceptions of relational autonomy that connect that concept, and the status of responsible agency, equal standing, and so on that connects with it, to modes of social life that not all women and men embrace. Liberation from oppression must be undertaken within a normative framework that leaves the most room for disparate voices, even those who endorse traditional and authoritarian value systems, for it must be accepted, in principle at least, that many women and marginalized people will embrace traditional conceptions of social life and cultural roles that offend Western, liberal ideals of individual self-sufficiency. While some version of that ideal is certainly worth defending, it is dangerous to couch that defense in the definition of the autonomous person.13 We should also be clear about another point: what I mean by “perfectionism” here is the view that values and moral principles can be valid for a person independent of her judgment of those values and principles.14 Perfectionism in this sense implies that there are certain objectively determined intrinsic values that should guide individual and social action independent of the endorsement of those values by minimally rational, autonomous individuals. Moreover, state policy should be guided by those objective values. Oshana embraces a form of perfectionism in arguing that relational autonomy of the sort she develops is indeed a contested value but nevertheless one that “is too important a characteristic for persons to be without” (Oshana 2006, 117). In chapter 10 I will further develop reasons for rejecting perfectionism of this sort. Here I want only to make clear that an overly restrictive conception of autonomy will fail to recognize the moral status of reflective, self-defining persons whose participation in hierarchical or other kinds of social relations that appear to us (perhaps rightly) as oppressive should not deprive them of the social and 13

14

In making this point, I am sensitive to the claim that definitions of autonomy are not at fault for robbing oppressed individuals of their voices, rather the social conditions that put such people in a position of strict obedience accomplished this (see Oshana 2006, 100–101). But the question of what counts as oppressive social conditions, as distinguished from lives of devotion to sacred authority or unquestioned tradition, must be answered in light of all authentic voices; and the purpose of an account of autonomy is to determine such authentic agency in order to proceed to conduct that very inquiry. This account is, I think, largely co-extensive with the view of perfectionism put forth by Thomas Hurka, though it is not equivalent to it. See Hurka 1993, 3.

174

The Politics of Persons

political status that autonomy underwrites in democratic practices and institutions.15 This response to relational views can be further illuminated by consideration of another attempt to discredit individualist accounts. Natalie Stoljar has argued that such models fail to capture the particular injustice involving patriarchal cultures and the women who have internalized values dominant in such cultures (Stoljar 2000). Stoljar identifies what she calls the “feminist intuition,” namely that agents (in particular women in patriarchal societies) should be regarded as lacking autonomy “because they are overly influenced in their decisions … by stereotypical and incorrect norms of femininity and sexual agency” (ibid., 98). She considers women (studied in a famous work by Kristin Luker) who took the risk of becoming pregnant by engaging in sex without contraceptives. These are women who had access to such measures and seemed to have adequate information about the risks they were taking, but nevertheless engaged in sexual activity resulting in unwanted pregnancies. These decisions for many of these women were not irrational in a procedural or formal sense, in that they engaged in risk management that, in accordance with a subjective account of rational choice, need not have violated standard procedural models of rationality. However, the attitudes and judgments that moved these women to make such choices were substantively problematic, in that they embodied internalized oppressive norms concerning their legitimate control over their reproductive lives. Stoljar contrasts the risky activities of such women with other risky behaviors, such as the decision to smoke (before becoming addicted presumably). We could comfortably label the latter kind of decision as autonomous because, we imagine, it would likely not involve a value structure that involved, say, reduced self-esteem or internalized oppression. But the women that Stoljar describes took risks with their personal lives which exposed them to very costly and regrettable futures because of internalized norms reflective of an oppressively patriarchal culture. Moreover, such women cannot be described as under the influence of reflection-inhibiting factors such as drugs or threats or mental defects; their decisions arose from reflections that are adequate in the proceduralist sense. “It would be rash to conclude,” Stoljar writes, “that Luker’s subjects have severely hampered critical capacities” (Stoljar 2000, 107). 15

Oshana accepts a version of this point, in that she notes that people of the sort she describes who voluntarily enter and maintain oppressive social relations might be granted “political” autonomy but that they still lack “personal autonomy” (2006, 102). In this regard she discusses the challenge brought by views such as those of Uma Narayan (Narayan 2002); see also Mahmoud 2003.

Relational autonomy

175

Yet Stoljar insists, based in the “feminist intuition” that such women lack autonomy in an important sense, that procedural accounts are inadequate for failing to capture this fact. My response here is similar to the comments I’ve made about relational theories generally, namely that the judgment that people who have internalized oppressive norms into their practical identities (but are still able to adequately reflect on their values and social conditions in a procedural sense) are not autonomous rests on a perfectionist account of autonomy. Such an account may well be acceptable on other grounds and in other contexts – it may serve as a social ideal for example in feminists’ struggles to move men and women to reform their social attitudes – but it fails to capture the sense of autonomy at the foundation of egalitarian democratic politics. The reasons for this can be brought out by quoting Stoljar at length in the concluding section of her article: Women who accept the norm that pregnancy and motherhood increase their worthiness accept something false. And because of the internalization of the norm, they do not have the capacity to perceive it as false … The reason that Luker’s subjects are judged not to be autonomous is that the reasons weighed up in the bargaining process – the costs of active sexual agency, as well as the benefits of pregnancy – are often derived from false norms that have been internalized, such as that women should not actively desire sex or prepare for sex in advance, that pregnancy is an expression of “real” womanhood, or that pregnancy is likely to lead to a marriage commitment from one’s partner and that this is a good thing … To vindicate the feminist intuition that the subjects are not autonomous, therefore, feminists need to develop a strong substantive theory of autonomy. (Stolar 2000, 109, italics in original)

First, I agree completely with this conclusion, but only if we are using “autonomy” to represent a feminist ideal of independent and egalitarian social lives. However, as an indicator of equal status in the discussion of social norms generally, this will not do, for many of the claims described here as “false,” such as that women should not desire sex or that pregnancy is an expression of “real” womanhood are (unfortunately) positions people hold in public forums relevant to social policy (fundamentalist religious people for example). To say that such people lack autonomy, when they can critically reflect on the values they express as well as anyone else, they are not acting from diminished (procedural) rationality, they are not alienated from their values, and so on (if indeed this is all true of them), is to disqualify people from the discussions about values that democratic processes are meant to embody.16 16

Numerous qualifications to this last idea are needed, such as that certain values that would disrupt the maintenance of democratic processes themselves (including respect for pluralism and equality of

176

The Politics of Persons

A timely example of this controversy involves a North African woman who recently applied for citizenship in France.17 The woman practiced a particularly restrictive form of Islam (Salafism) in which she dresses in fullbody veil, obeys her husband and other male members of her family in all important matters, and rarely leaves the house except to bring her children to school. France’s highest administrative body, the Council of State, ruled that the woman, identified only as Faiza X, had “adopted a radical practice of her religion incompatible with the essential values of the French community, notably with the principle of equality of the sexes, and therefore she does not fulfill the conditions of assimilation” listed in the country’s Civil Code as a requirement for gaining French citizenship.18 We can consider the “principle of equality of the sexes” mentioned in the Council’s decision as representative of the standards of relational equality that Oshana, Stoljar, and other theorists would insist upon as required for autonomy. Now it is unclear whether this particular woman would meet the standards of procedural autonomy set out in this work, and there is some evidence that she would not.19 But if we imagine that she has accepted this subservient position out of sincere and reflective religious devotion, without defects in her competence as a reasoner, then my claim is that denying citizenship to such a person is tantamount to silencing a voice in the moral landscape that a fully inclusive democracy, one which rejects perfectionism in the way I have described, should include. Of course she is oppressed, and the particular religious views she espouses reflect, to my mind, an unacceptably restrictive and misogynist view of human relations. But these are claims that should be made within democratic deliberation, not as part of a conception of the autonomous person that defines its limits. A final point, however, about the grounds that are usually offered to reject overly individualist views of autonomy: procedural accounts are most often taken to task for their inability to countenance social identity in virtue of their emphasis on detached reflection and self-endorsement

17 18 19

standing) may well be justifiably restricted, but not based on the lack of autonomy on the part of their proponents. For a parallel discussion of Stoljar’s argument, see Benson 2005b who uses a critique of Stoljar to defend a “weakly substantive” account of autonomy. This description will likely depart from some of the precise facts of the case, which is also developing further as I write, but I use it here for illustrative purposes. See “France Denies Citizenship to Muslim Woman,” Associated Press, July 16, 2008. For example, according to news reports, the Council claimed that “she leads a life almost of a recluse, cut off from French society.” “She lives in total submission to the men in her family … and the idea of contesting this submission doesn’t even occur to her,” the government report said (ibid.). The fact that the idea of contesting submission “doesn’t even occur to her” indicates that she may not meet the requirement of adequate reflection needed for autonomy.

Relational autonomy

177

(or non-alienation) from aspects of the self.20 These critiques, then, focus on the authenticity conditions in models of autonomy. However, the real failing of those accounts lies in the competency conditions that are typically (and often incompletely) laid out as necessary for autonomy. As feminists and other relational theorists have been insisting, many life patterns (particularly for women and marginalized groups, as some claim) crucially involve intertwined personalities, close relations of care and dependence, embedded cultural identities and values, and the like. For autonomy to pick out the ability of persons to lead lives that they can fully embrace as their own, then surely people need to develop those abilities that are central to interpersonal relations of a variety of sorts. So while I cannot spell out a list of such competences here (others have done so), let me note that it is competency conditions in proceduralist views of autonomy that are problematic insofar as they do not include or make room for the wide variety of capacities for care, intimacy, social interaction, and the like that will be crucial for socially embedded persons to flourish.21 It should be emphasized that the deleterious effects of hierarchical social relations on agency and well being are central to the social and institutional structure of democracy of the sort envisioned here. To underscore that point, let us take a brief historical side trip to examine the views of one thinker who is not often referenced in such discussions but who offers a unique interpretation of the connection between social power structures and self-concept that will be great use to both defenders of relational autonomy as well as those, like myself, who resist that notion in its strict form. Aside: lessons from Rousseau In Rousseau’s speculative anthropology, worked out in his two discourses, Emile and the “Origin of Languages,”22 he develops a view of how not only the virtues and a sense of justice might develop in relation to social forces but also how the self-concept emerges. A key element in his speculations is the variability of this process, and especially its sensitivity to the surrounding dynamics of social life. In modern (especially, for him, Parisian) society, indeed everywhere where man is “in chains,” he argues that the self-concept emerges from a process of internalization of the expectations 20 21 22

See, for example, Mackenzie and Stoljar 2000a, 13–17. Meyers 1989. As Virginia Held has said in a different context, we should view “maturity … as competence in creating and sustaining relations of empathy and intersubjectivity” (Held 1993, 60). In his letter to Malesherbes, Rousseau claimed that the first three of these works form a unified whole. For discussion see Cranston 1991.

178

The Politics of Persons

of others and the desire to live up to the image of oneself in others’ minds, in particular the minds of one’s social betters. This, famously, produces amour propre or prideful self-love.23 Crucial to the Rousseauian picture of the social self is, of course, the dynamics of amour de soi and amour propre. The latter refers to the selfunderstanding according to which our self-esteem becomes thoroughly tied to our relations with others in society toward whom we occupy positions of subordination. We internalize this (mostly corrupted, for Rousseau) picture of ourselves that we find in the attitudes of others regarding us as we discern such attitudes in the superficial details of social posturing. The domination involved in this process is deep and crippling, in that our very self-image is structured out of our subordination to our supposed “betters” in an artificially inegalitarian social order.24 It is important here to note two things: this process of subject construction by way of interaction with others (and in reaction to others’ regard for ourselves) is not a process that can be escaped, at least not for any people living outside the rarefied (and fictionalized) arcadia of the state of nature (as outlined in Book I of the Second Discourse – see Rousseau 1987). For all socialized individuals, self-esteem is a function of social interaction and social dynamics, on Rousseau’s view. What must be understood is the ways in which that interaction is corrupted in nearly every society (save for the pristine Geneva of Roussseau’s patriotic fantasies) as a result of the inequality of power and status that pervades them. Now there is, for Rousseau, a “good” kind of amour propre (to abuse the term), in that there is a healthy manner in which we can construct our selfunderstandings outside of the “chains” of hierarchical social realities. Rousseau surely denigrates amour propre, where “to be” and “to appear” become indistinguishable in us, but he also calls it (in Emile for example) “what is best and what is worst among men” (Rousseau 1762/1993, 184). In the notes to the Second Discourse he says: “Amour propre is only a relative sentiment, factitious, and born in society, which inclines every individual to set greater store by himself than by anyone else, inspires me with all the evils they do one another, and is the genuine source of honor.” His denigration is grounded in the especially pernicious form of dependence that amour propre involves, in all but the most rarefied social conditions; but what he says here also makes 23

24

This ground is also covered by Charles Taylor, where he discusses the dynamics of social recognition that Rousseau described and evaluated in tracing the threads of the modern conception of self. Taylor 1992, 35–36, 44–51. See also Gauthier 2006. See Rousseau 1762//1993, Book IV, Social Contract, Book II (in Rousseau 1987, 153–172).

Relational autonomy

179

room for a kind of interpersonal, self-constituting dynamic which affirms rather than destroys healthy self-esteem. This can be seen more clearly in a remarkable letter that has only recently been translated and published in English, Rousseau’s letter to one M. Moltou (probably written in the spring of 1757). In answer to the latter’s apparent question “what is virtue?” Rousseau explains the thoroughly social nature of moral reflection and self-understanding (again, for all us postarcadian social beings). He explains that whatever is “moral” in me “seeks relations outside myself.” That although our dependence on others in society is a kind of weakness, it is nevertheless the unavoidable weakness of socialized beings, ones whose “natural goodness” has given way (in an anthropological rather than a historical sense of “given way”) to social virtue. That being the unalterable lot of all socialized selves, they must appropriate “goodness of another sort, suitable to this new existence.” Speaking as one such social being, he adds: [b]ecause my life, my security, my liberty, and my happiness today depend on the cooperation of others like myself, it is clear that I must look upon myself as no longer an isolated individual but as part of a larger whole, as a member of a larger body on whose preservation mine depends absolutely … Such are the indissoluble bonds that unite us all and make our existence, our survival, our reason [lumières] … dependent on our social relations.25

This shows how Rousseau pictured moral self-reflection as constituted by social connections and civic identifications. What is important to note here is the way Rousseau sees social power hierarchies as essential in producing, for him, a corrupted and pathological self-concept for modern citizens. This all may suggest, then, that insofar as self-government or autonomy is essential to human dignity, something Rousseau clearly thought, then egalitarian social relations are a necessary condition for autonomy. And this is precisely the sort of view I want to contest. In response, let me make two brief observations. The first is that Rousseau’s claim that modern citizens are “in chains” because of their internalized social subordination is part of a claim about not only freedom but the ideal character of the person guided by virtue (as “strength of soul”) and, by extension, by “nature.” The education of Emile “as a man,” which is needed before he can be educated “as a citizen,” is aimed at producing a person who is not only self-governing per se but who also exhibits the stoic virtues of the self-reliant individual who is in touch with the inner voice of morality and in tune with his natural 25

“A Letter from Jean-Jacques Rousseau,” New York Review of Books (May 15, 2003).

180

The Politics of Persons

passions. This is clearly an idealized picture, a perfectionist (and we should add quite sexist) view guided by what Rawls would call a comprehensive moral doctrine: a broad philosophical account of the virtuous life meant to shape our deepest values and commitments. Saying this reveals at the same time the contestability of such an ideal. Not that we need to reject it out of hand, but it clearly must be admitted that there are other reasonable competing ideals. (To see the contestability of Rousseau’s picture of the ideal life, one need only pay further attention to his views on the role of women in the life of the self-governing (male) citizen!) This means that while Rousseau’s diagnosis of the importance of hierarchical power in the construction of the modern citizen’s self-image is illuminating and persuasive, any conclusions linking autonomy to egalitarian social relations relies on more than merely the idea of self-government itself. It also relies on ideals of personal development and social relations (the latter embedded in a conception of justice) which other reasonable citizens with different philosophical outlooks may well contest. My point is to use Rousseau as an object lesson in attempting to tie a social conception of self to a social understanding of autonomy, and in doing so he exemplifies the use of substantive value conceptions in making such a link. As we saw earlier, that is precisely the step that attracts our suspicions when talking about the concept of autonomy in modern pluralistic democracies. That said, I want to suggest that there is a further lesson to be culled from Rousseau’s understanding of our social self-conceptions, one which will allow us to take on board many of the urgent claims made by relational autonomy theorists about oppressive social connections. In cases where the inegalitarian nature of the social relation (which in part constitutes my view of myself on this picture) suffocates my independent voice and suppresses my own self-government, when I am made to bow and scrape and thereby come to see myself as owing deference to my social betters, it is often the case that I am thereby unable to see myself as “author” of my own life. That is, in the social structures I find (and define) myself in I may feel that my own voice and judgment is so occluded in the social dynamic that moves me along in life (say in the commands of others with authority over me) that my life path is really not in any sense my own. I cannot see myself as a selfauthenticating source of the biographical narrative that defines me. In such cases, on my view, I have lost my autonomy. I cannot, without alienation “take in as my own” the narrative of my life, together with the diachronic practical identity that motivates decisions within that life. In this way I cannot see myself as “author” of my own life (and “author” is in quotes because of my repeated reminders that we are never self-creators

Relational autonomy

181

in any deep sense, we are only self-managers). I think this will help to classify many oppressive social conditions that relational autonomy theorists insist are heteronomous as such, though it is so classified on proceduralist grounds. It is not from an external perspective that any given social relation (of authority, care, dependence, or devotion) is deemed autonomyconferring; rather it is still from the point of view of the socially constituted agent within those relations. In cases where, as Rousseau outlines, our amour propre has turned us into self-defined lackeys and self-effacing second-class citizens, we lack true self-government because such conditions prevent us from seeing ourselves as the authenticating voice in the dynamics that move such conditions along. So much for Rousseau. To complete our treatment of relational conceptions of autonomy we must now consider an additional attempt to motivate that model which makes specific use of the conception of social recognition, a consideration that will occupy us thoroughly in the following chapter.

ii. social relations and social recognition Another relational approach to autonomy that comes from a different direction is one which combines considerations of equality of the sort that move Oshana with dynamic, interpersonal factors arising from a certain understanding of the development of the subject, specifically Hegelian ones. As we will discuss in more detail in a moment, Axel Honneth has done valuable work in underscoring the importance and complexity of recognition in critical political theory generally and in conceptions of the subject in particular. Drawing on this work, Honneth and Joel Anderson have sketched a conception of autonomy that includes relational, in particular recognitional, elements (Anderson and Honneth 2005) that are of relevance to our concerns. For Hegel, of course, mutual recognition is a developmental precondition for self-consciousness itself. And one might claim, as Honneth does, that recognition is central to conceptions of social justice that guide critical appraisals of current conditions.26 But the view I want to consider argues that personal autonomy involves recognition in ways that are stronger than merely a contingent connection. Honneth and Anderson argue that 26

Honneth 1996. On the concept of recognition, Honneth writes: “social recognition represents the necessary condition for subjects being able to identify with their valuable qualities and, accordingly, develop genuine autonomy” (Honneth 2001, 515).

182

The Politics of Persons

in at least three spheres of recognition relations – self-respect, self-trust, and self-esteem – the vulnerabilities to which individuals in the current age are subject (and against which autonomy is meant to protect them) require that particular relations of recognition are conceptually required for autonomy in a full sense. As they put it, proper relations of self-respect, trust, and esteem are needed so that “full autonomy – the real and effective capacity to develop and pursue one’s own conception of a worthwhile life – is facilitated by relations-to-self (self-respect, self-trust, and selfesteem) that are themselves bound up with webs of social recognition” (Anderson and Honneth 2005, 137). I will say more about this claim in a moment, but one wonders at this point why the apposite phrase in this last quote does not itself provide a definition for autonomy – the real and effective capacity to develop and pursue one’s own conception of a worthwhile life – while the other factors “facilitate” this condition. But Anderson and Honneth specifically argue that proceduralist accounts of autonomy (they cite Rawls as a source of such a view) fail to capture the way that human vulnerabilities require certain interpersonal and social conditions to enable individuals to enjoy autonomy in a stable and robust manner. This point relates to the claim made by others that autonomy requires a degree of self-trust. Indeed some have argued that a normative component of autonomy enters by way of a requirement that one have regard for oneself, as a chooser and deserving of respect from others on this score, to be truly self-governing. Paul Benson, for example, argues that part of what it means to “take ownership” of one’s intentions and actions (of a sort required for autonomy) is to maintain a degree of self-worth (Benson 1994, 2000; see also McLeod 2002). Benson argues that without a sense that she can use the power of her own judgments sanely, a person is not autonomous. My general reaction to these types of cases is to accept that a degree of self-confidence and self-trust is in fact required for autonomy, but this does not imply that autonomous persons must adopt a particular, substantive, value scheme, except in the sense that having this sort of confidence will often involve valuing oneself in a minimal sense. The cases Benson and others rely on to show that self-trust is necessary really show a failure of competence, not a failure to have a certain value commitment one ought to have. For example, I might have perfect trust in my physical trainer – I do all and only what she says because I have absolutely no confidence in my own ability to choose this exercise over that one. No one would say my autonomy is compromised under such conditions. This is because, in granting her this authority, I retain the competence necessary to carry out

Relational autonomy

183

my intentions to exercise effectively, since I transfer trust from myself to her concerning what is an effective exercise. In that case I have competence and hence autonomy (assuming other conditions of authenticity are met) but without self-trust. Admittedly this is a lack of localized self-trust – trust in a particular capacity – and not self-trust generally, for I did at least have to choose this particular trainer. But when the person who lacks even wholesale self-trust lacks autonomy it is not due to an absence of a particular value commitment per se, but because she lacks an ability, a kind of motivational competence to form reflective judgments and carry them out in action. Now I know that taking this position will open me up to so-called happy slave cases, for a person who lacks self-trust entirely, who gives herself completely over to the rule of another, will be autonomous on my view if she came to this position autonomously (and is not so distant from the original decision that the authenticity conferred earlier has lapsed). But this leads to the disagreement about such cases discussed in the previous chapter, a challenge I hope to have answered. To return to the issue of autonomy and recognition: as we will discuss in more detail in the next chapter, recognition is always enacted under terms of identity: that is, one expresses or experiences recognition as … this or that kind of person. Indeed, part of the reason that mis- or mal-recognition comprises injustice is that one’s identity and related commitments to a horizon of value are ignored, occluded, or marginalized in the broader dynamic of social power relations. Therefore, identity markers define the terms of the relations instantiated by recognition. However, the more general case against seeing recognition as a conceptual component of autonomy can be mounted in straightforward terms, and mirrors my earlier comments about relational autonomy. First, insofar as recognition operates under categories, the descriptors under which such recognition extends itself connect directly to the identity categories that underwrite the politics of identity as a specific mode of struggle and political action and which provides the nomenclature for the experience of injustice when such recognition is denied. Of course recognition need not express itself in any of the familiar identity categories, of race, ethnicity, gender, or sexuality. Recognition can be more general than this.27 Nevertheless, 27

Honneth refers to this variability when he writes that his claim about recognition remains “a claim within philosophical anthropology, even though I now emphasize much more than previously the historical alterability of forms of recognition; it is still a matter of the invariant dependence of humans on the experience of recognition, even though its forms and contours can become differentiated in the course of historical transformations” (Honneth 2001, 515).

184

The Politics of Persons

recognition of fundamental interests and capacities of this sort differs from the more generalized form of interpersonal respect precisely in the former’s dependence on categories of classification through which it is conveyed and the latter’s focus on only rational agency as such in a broad and amorphous sense. Second, such identity categories carry with them arrays of value connotations – behavioral expectations, psychological associations and the like. To be acknowledged as a male, or a homosexual, or a worker, or a mother, is to be both defined and evaluated according to myriad associated criteria of behavior and character. To be called a mother is to allow the question of whether one is a good mother. Identity categories are value categories. Indeed, the power of recognition resides precisely in the value of respectfulness for such categories, a recognition made necessary, according to its defenders, by the comparatively vacuous expression of respect for agency that it is meant to replace. Further, categories of classification are always contestable. Social typology is forever under revisionary scrutiny, and the particular public meanings associated with identity classifications are subject to ongoing renegotiation and sometimes resistance. This is clearly a complex point, and it brings to mind the myriad debates about essentialism, exclusionary tendencies of strictly defined social groups, feelings (and charges) of disloyalty and selfhatred, and the like. The point I put forward here, however, is rather weak and straightforward, namely that social categories by way of which recognition expresses itself are subject to question and ongoing evaluation. This meager claim, though, will play into the argument that processes of democratic social deliberation will be necessary to negotiate the dominant meanings of such social categories, and such deliberation proceeds with autonomous participants, ones whose autonomy must be defined without reference to the categories as such under dispute. I suggest that we distinguish closely related but importantly different aspects of relational views. It is one thing to claim that social conditions that enable us to develop and maintain the powers of authentic choice and which protect the ongoing interpersonal and social relationships that define ourselves are all part of the background requirements for the development of autonomy. This is a powerful contribution to the discussions of autonomy made by feminist (and other) defenders of relational concepts. (Indeed, I have tried to suggest that greater attention to these sorts of factors should be given in the “competence” conditions of procedural accounts of autonomy.) It is another thing, however – and a problematic and ultimately dangerous move, I have argued – to claim that being autonomous means

Relational autonomy

185

standing in proper social relations to surrounding others and within social practices and institutions. Taking this position, as we’ve seen, turns the concept of autonomy into an unacceptably perfectionist idea that carries with it the danger of exclusion and overarching paternalism that attention to autonomy should well protect against. The lesson of relational theories, then, is that socially constituted and interpersonally embedded selves, in all their varieties and complex perspectives on value and justice, are autonomous only when that variable and multiplex position in those social relations reflects the authentic and self-imposed standards of the free person. In this way, relational theories have taught us how far we should go in the direction of seeing justice as concerning social dynamics as well as individual choice, but they also point to the dangers of reifying any particular set of such dynamics as the only ideal to be taken into account in the public deliberations constitutive of justice itself. To adopt a thoroughly relational view of autonomy is to see it as a property, not merely of an individual and her capacities, but of the relations that comprise those conditions. To protect autonomy in this way is to protect those relations. What is powerful about these views is the emphasis they place on securing the social conditions that are required for the enjoyment of autonomy, conditions relating to education, social structures and opportunities, access to basic resources, housing, and so on.28 But relational views that see social conditions as not only supportive of autonomy but definitive of it carry with them a danger that autonomy-based principles of justice will exclude from participation those individuals who reject those types of social relations demanded by those views. Let me be clear, however, that various sorts of social relations, often involving power and authority, will bear directly on whether a person enjoys autonomy in an individualist sense. For as we saw in discussions throughout this book, insofar as certain structural relations are functional in the conceptualization and activity of any given self then autonomy as self-government will depend on those relations being realized and protected. More to the present point, specific conditions of autonomy may well require particular interpersonal and social relations to obtain for these conditions to obtain. For example, we discussed in chapter 2 the ways in which interpersonal relations can be seen as fundamentally operative in the way people think and feel. We talked about the disorienting effects of moving from one environment to another and how these 28

For a discussion of the way that some poverty-related welfare policies systematically undercut the provision of such conditions, see Christman 1998.

186

The Politics of Persons

disorientations can severely impair a person’s ability to think, reflect, and remember. Therefore, in any such cases, socio-relational conditions will have to obtain for any such person to maintain the ability to engage in the kind of reflection required of autonomy. I suspect, then, that the policy implications of relational accounts of autonomy will dovetail quite neatly with those of my own view. The difference, and it is an important one I think, is conceptual.

chapter 9

The dynamics of social identities

Discussions in political theory of the importance of social and cultural group membership have been plentiful and multi-faceted in recent decades. This has been spurred by several factors, including growing realization of the effects of globalization, confronting the increasingly multicultural nature of populations in modern democracies, and the continuing effects of identity-based struggles in the 1960s and 1970s in Europe and North America. These discussions gained new energy with attention given to multiculturalism, theories of social recognition, and the politics of difference in recent years. The central issue might be described as asking whether special weight must be afforded to “identity-based interests” on the part of members of particular social groups characterized by commonalities of culture, identity, history, or social position. If such weight must be specified in the design of political principles, how can reference be made to social groups in question while recognizing the multiple differences among members of any such group, without, that is, running afoul of problematic essentialism about social categories which is in tension with the variability of lived experience and the social constructedness of those categories themselves? This shows how in attempts to articulate this challenge to traditional liberalism, a subtle balancing act must be performed between finding a language which expresses the special weight that identity-based interests allegedly should receive in political principles but without relying on problematically static, reified, or essentialized categories in defining those claims. Theorists in various fields grappling with these issues have attempted to walk this tightrope. In this chapter we will look at some of those attempts (without attempting anything like a full survey). In the end I will suggest that there is a way to consider the special importance of identitybased claims, one which avoids the pitfalls of essentialism (and related problems), indeed one that connects these claims with the autonomy of the person. 187

188

The Politics of Persons

The literature on these issues is vast, and it has proliferated of late by both defenders of identity politics in various forms as well as by liberal voices raised in response. Some theorists have focused on the political importance of social-group membership as a way of rejecting the Enlightenment tradition of liberal political philosophy as a whole. These various thinkers have claimed that social-group membership adds a crucial element to considerations of justice, that the liberal tradition has not accounted for this importance (and, they argue, cannot); for that reason, they argue, we should depart from that tradition (see Young 1990a, for example). Others offer terms of reconciliation, either from within the liberal framework (Kymlicka 1995; Margalit and Raz 1990) or from an alternative perspective which looks for reconciliation with the liberal emphasis on the autonomy of individuals (e.g., as with Taylor 1992). Many in this last camp focus on what has come to be called the politics of recognition (and of course not all theorists following this line seek a robust reconciliation with liberalism). Finally, liberal philosophers have also addressed this issue but not in order to gather in the concerns of group identity, at least not on the terms of those raising the issue, but in order to dilute or diminish the threat that group-based concerns pose for traditional theories of justice (Appiah 2005; Barry 2001; Kernohan 1998; Sen 2006). In this latter group, as with some critical theorists (Benhabib 2002; Habermas 1999), emphasis has been placed on the deep and complex hybridity of culture, social groups, and hence identity. Writers such as Appiah, Sen, and Benhabib have recently analyzed the question of identity and group rights but have stressed the historical and current cultural complexity of various identity groups. This emphasis on complexity is meant to undercut claims on the part of multiculturalists that special claims (or rights) arise from attention to culture. I will have various things to say about some of these theorists and trends, both directly and in passing. In particular, I will suggest that some antiliberal theorists of identity conceptualize identity-based claims in ways that fail to support their wholesale rejection of the language of liberal justice. Also, I will point out that thinkers stressing the internal hybridity of cultures are, first, pressing a normative claim about how people should regard their connection to certain traditions and cultures, one which is disguised as a descriptive or analytical claim about history or ethnicity or culture. Second, it is unclear that such approaches can account for the special importance that identity-based claims have, when in fact they merit that importance.

The dynamics of social identities

189

However, I want to stress that, for the most part, these theorists have failed to advance a plausible account of the person’s relation to cultural groups – the very identity that operates at the center of their arguments. At least they have not put forward an account of social identity that (a) grounds the special weight that identity-based interests should garner in those cases where this is merited (and I will suggest there are such cases), and (b) avoids problematic essentialism about the nature of social groups. In short, a model of the political self is what is needed here. My more positive aim in this chapter, though, is to make out a case for why identity-based claims are related to autonomy in a manner that has not been adequately noted in the literature I will discuss. Insofar as autonomy involves non-alienation in the way I have described, and the “self” of self-government can be seen as constitutively, though variably, related to surrounding (and historical) social conditions, then avoiding alienation will require giving special weight to identity-based interests but in a manner which avoids permanently fixing the features of that identity once and for all. My plan is to examine first the overall structure of identity-based claims where I will bring out the important role that self-interpretation must play in the specification of social identities (in ways that theorists in this literature have not all seen clearly). Next, I will consider attempts by liberal theorists to take account of the special status of identity-based claims of this sort, and I will conclude that neither they nor their critics have adequately captured the varying role that choice plays (and does not play) in grounding such claims. I will also argue that such theorists have not given an adequate account of the “identity” in question here, a lacuna I try to respond to in section IV. I conclude by discussing how such claims can be grounded in individual autonomy when that notion is understood in the manner I have worked out here. i. overall structure of identity-based claims To view the landscape in its full light here, let us distinguish various positions that have been taken on these issues and the various controversies arising from them. We begin with the traditional liberal conception of the subject (citizen) of justice, the person represented in the abstract normative principles of democratic states. This model citizen is described in a way that makes no reference to particular characteristics, such as race, gender, sexuality, or the like, nor to membership in social groups or cultural traditions. Given that, it

190

The Politics of Persons

was thereby assumed that such principles based on that model simply applied universally.1 When an indigenous people, for example, makes a claim to traditionally sacred ground, the controversy plays itself out as something more than simply a conflict of property rights. When claims of this sort are debated, based as they are on litigants’ connections with earlier generations of culturally connected people, special importance is afforded to the depth of that connection and unique standing that (in this case) tribe members have in pressing the case, in comparison, say, to a housing developer’s desire to use the land for a new residential area. In a slightly different but related register, women, members of racial or ethnic minorities, and gay people, among others, often couch claims of past and ongoing injustice in terms of attacks on them as members of such groups, rather than as individuals. Hate crime legislation in the USA, for example, is supported in part by the understanding that crimes motivated by opprobrium toward people because of their membership in such groups are especially heinous. Anti-discrimination and equal-protection law also designates specially protected categories of individuals whose interests garner special attention by way of the increased scrutiny that is required of any legislation that adversely affects them. The particular weight given to these kinds of claims, when or if they are valid, rests on an assumption of the importance of group membership that the politics of identity (as I am using that term) attempts to capture.2 In the literature alluded to here, a variety of claims and counter-claims are at issue, some dealing with the status of social groups (and within that, 1

2

We all know, however, that the actual historical theorists purveying those views either explicitly or by implication did not intend the model citizen to be without race, gender, etc. Rather, the model citizen was a “white” male of a certain class position. Various studies have drawn out this exclusionary tendency and of course it is now taken as given that canonical liberal theorists assumed a narrow conception of the citizen in this way. Such studies include Lloyd 1984, Mills 1997, and Pateman 1988. I use the term “social group” as a generic phrase to refer to culturally organized groups, such as the Ojibwa of North America or the Maori in New Zealand, and other groups that share very little in common that can be called “cultural,” such as women, gay men and lesbians, the disabled, and racial groups. This is not to say, of course, that these latter groups do not tend to share certain commonalities relating to social practices, modes of behavior, associations, modes of treatment by others, and the like (at least in certain localities). But there seems an obvious surface difference between the factors that determine membership in those groups. For cultural groups, participation in social practices, or at least a tendency to acknowledge the personal importance of such practices, is part of the identity that grounds one’s membership. For ascriptive identities, such as sex, race, and physical ability, what determines membership are merely those facts about one (as defined by social classification schemes), independent of whether one’s mode of behavior, social practices, language, way of speaking, and so on are shared with other co-members. (See Barry 2001 for an argument that ascriptive identities share no common culture.)

The dynamics of social identities

191

both cultural groups and non-cultural identity groups), some concerning claims by individuals but with particular connection to group membership. At times, these claims are in the form of rights, as with Kymlicka’s discussion of polyethnic rights and the rights of self-determination by national minorities (Kymlicka 1995). However, the language of rights carries with it conceptual baggage whose controversial nature complicates the discussion in ways I’ll want to avoid here.3 I will talk here of special interests on the part of persons and groups, though even in this case, I don’t want to imply that justice considerations are all a matter of attending to (or simply balancing) citizen interests. As will emerge later, also at issue is the perspective by which a person makes judgments about her interests. The shorthand for all such consideration will be “identity-based interests.” The question will then be: what non-reducible consideration must be made to such interests in the structure and justification of principles of justice? Put differently, how should the model of the citizen as the subject of principles of justice in modern democracies include reference to membership in and identification with social groups (whether these be cultural groups or other ascriptive identities that arguably lack a common culture)? Another way to make this point is to simply postulate a difference between a preference and an identity-based claim. A preference may well be quite weighty, but an identity-based claim connects with various other value commitments, social practices and habits, and relations with others. In addition, connections to social groups of certain kinds have a historical dimension that is often overlooked. This is certainly true for cultural groups, whose claims for special protections refer specifically to past practices in reference to which current members define themselves. But also for ascriptive identities, in addition to ongoing discrimination and violence, overt injustices of the past, often involving legally sanctioned exclusion and discrimination, are relevant to the weight of current demands. Examples include the relative recency of the affordance of full voting rights to African Americans in the USA, as well as the countless legal exclusions of gays and lesbians, ethnic minorities, women, and other groups in the political and public sphere. This means that current claims of identity carry with them historical shadows that are very much constitutive of their power. 3

In general I have in mind here claims that rights assume overly individualist interests, overly oppositional social relations, and a generic conception of the rights-bearer that occludes reference to differences among members of social groups that, some claim, deserve theoretical attention. See, e.g., Brown 1995 and Williams 1991; for a defense of rights language, see Waldron 1987.

192

The Politics of Persons

But the issue remains whether membership in certain social groups affords special status to that aspect of one’s political personality, such that claims made in connection to such membership are given greater social weight than preferences and other needs. One influential manner in which claims of this sort have been advanced is couched in terms of the social recognition required for agency. As we saw in chapter 2, Charles Taylor has developed a complex conception of the self and its relation to culture that specifically relies on the dialogical nature of identity. In particular, the hypothesis that we develop a sense of self and identity by way of the language of a culture and in interaction with other speakers of that language, makes the ongoing practice of that language define those cultural forms crucial to our identity. “Recognition” is the term for the kind of external acknowledgment afforded to us by significant others that cements the sense we have of ourselves, which in turn underwrites its value. In addition, as we discussed in the last chapter, Axel Honneth has argued for the importance and complexity of recognition in critical political theory generally and in conceptions of the subject in particular. Honneth writes: “social recognition represents the necessary condition for subjects being able to identify with their valuable qualities and, accordingly, develop genuine autonomy.”4 In a disagreement with Nancy Fraser, who claims that the injustice involved in misrecognition cannot be removed from particular socio-historical struggles by identity groups, Honneth relies on a Hegelian conception of the establishment of our subjectivity by way of the acknowledgment by others of the terms of that subjectivity. Honneth has written a great deal about the dynamics of this development, both in its Hegelian and psychoanalytic registers (Fraser and Honneth 2003; Honneth 1996). However, as we discussed, the crucial question in these claims (and one that may well replay the tension between Fraser and Honneth) is whether the terms by which this recognition occurs will necessarily be specific languages of identity and value, operative in particular times and places, or general expressions of embodiment characterized abstractly. If the latter is what is demanded, that we all require acknowledgment of our subjectivity and agency by significant others and by social structures and institutions, then it is not clear how this could ground an argument for any particular identity claim (such as the call that a burial ground be protected as an expression of the importance of a tribal history and identity). 4

Honneth 2001, 515. See also Honneth (1996).

The dynamics of social identities

193

On the other hand, claiming that justice requires specific recognition along lines of a fixed identity implies that the value terms in which one’s identity is established and recognized are unproblematic and fixed. In other words, if the social recognition of my standing as a person requires that social institutions and practices express the value of my being this specific kind of person (a woman, a Maori, a Muslim), then what it means to be such a person would have to be uncontroversial. All expressions of recognition are made in particular value terms, languages of respect and description that capture the dignity and standing of a group or an identity. When we say that “we value the proud history of the Maori people” we are using the word “Maori” as shorthand for a range of valued attributes and descriptive terminology that both describes and values a group of people with a certain history and make-up. But if what it means to be that sort of person – what the history and attributes of that group are such that pride (or condemnation for that matter) of the group can be expressed – is a matter of dispute, then calls for specific recognition of this sort will inherit that controversy (cf. Benhabib 2002, 68–71). Taylor is surely right that there is a special sort of unjust damage that people suffer when others mirror back to them a “confining or demeaning or contemptible picture of themselves” (Taylor 1992, 25). The question arises, however, whether this special damage is multiplied because of a connection with identities per se, as opposed to simply being a show of contempt, full stop. If the latter is the case, an adequate defense against such social assaults would be “but I’m not like that.” The politics of recognition must imply that the damage is deeper and the response must be “I am that kind of person but such identities are not to be denigrated in that way.” Control over the language of public valuation must be reclaimed by the offended groups so that the terms of identity are not commandeered by the forces of hate. Notice, though, that what is lacking by the injured parties is access to the social resources necessary to affect the public discourse about what it means to be a certain type of person and by extension whatever other resources are needed to flourish equally as that type of person. The claim that we must be able to speak effectively in the broad discussion of the contours of our identity is prior to the claim for the particular resources needed to pursue valuable projects in light of our identity, for it presupposes a settled sense of that identity. Insisting that resources such as support for the use of a particular language, the protection of a burial ground, the recognition of a national holiday or celebratory public ritual – as a stand-alone demand based on the sacredness of identities – assumes a fixed determination of the

194

The Politics of Persons

nature of that identity, not as an outcome of an ongoing social discussion but as a social, biological, or metaphysical “fact.”5 (Cf. Benhabib 2002, 50.) Appiah argues along similar lines when he asks whether this recognition allegedly required by justice is thought to ground my identity or to constitute it. If recognition is necessary for identity-constitution, then not only are we all incomplete before being acknowledged by others in their own terms, but, paradoxically, members of marginalized groups are then meant to appeal to dominant groups to afford them their identity. However, if what is meant here is that recognition simply grounds identity, then it is still possible (and valuable) to individually reflect on what it means to be this or that kind of person as that is defined by the dynamic of recognition by others (Appiah 2005, 32–33). But I would add here, the claim for recognition should take the form: to be fully respected as a subject I need recognition from others of the kind of person I am, as interpreted by me. Only if my voice is prominent or adequately represented in the discussion of the value terms by which recognition is afforded can that recognition avoid being essentializing as well as patronizing. Now others have attempted to defend justice claims based on group membership while avoiding the specifically Hegelian concern with recognition per se. Iris Young has argued for example that the ideal of justice as impartiality which dominates the liberal tradition functions to occlude the perspectives and social interests of oppressed groups. The ideal of impartiality, and the conception of the autonomous agent which is its object and guiding frame, serves only ideological purposes by facilitating our lack of attention to entrenched hierarchies of power and informal, interpersonal, and non-codified patterns of domination that members of denigrated groups experience (Young 1990b; 2000). In her “structural” view of identity, which she sees as required by the brand of deliberative democracy she favors, identity is characterized by one’s place in a network of obstacles and opportunities that condition one’s life possibilities (Young 2000, 94). The 5

An example of this might be the language of “queering” in regards to gay and lesbian life. As is well known in academic settings, it is perfectly acceptable to refer to gay and lesbian experiences as “queer life” (after all, Queer Studies is the name of programs of study in many universities). But calling gay people or lesbians “queer” was for some time understood to be insulting and denigrating, and part of a pattern of marginalization that had profound effects on the ability of homosexuals to pursue valued lives in settings of respect and recognition. But the fact that use of such language now (often) lacks those connotations has to do, not with the reference or (literal) meaning of such terms, but the fact that the designation can now be seen as emanating from the community itself. To refer to oneself and one’s identity group with a particular terminology that is under one’s discursive control, as it were, is fundamentally different from cases where that same language is imposed by others. The dynamics of social self-determinations, specifically, determination relative to public discourse and the representation of identity, is what is at stake in such cases.

The dynamics of social identities

195

cultural structures and practices that publicly express social attitudes can either support or suppress the social elements of that identity, either by recognizing the unique interests tied to them or, more typically, marginalizing and making invisible the specific needs and values internal to them. Political principles that postulate an allegedly impartial view of the citizen “forces homogeneity upon the civic public,” she writes, “excluding from the public those individuals and groups that do not fit the model of the rational citizen capable of transcending body and sentiment” (Young 1990b, 109). For theorists such as Young, political recognition for identity-related interests is based on the broader concern for the achievement of just social arrangements for marginalized groups. This is because of past and ongoing failures of liberal justice to capture the unique victimization that such group members experience and which cannot be captured by distributive conceptions of social justice, at least not ones (as with liberal egalitarianism) that measure just distributive shares only in terms of rights, opportunities, and resources. It may appear initially that this line of attack fails to mark as radical a departure as is claimed from traditional conceptions of justice, in that models of social advantage and disadvantage could simply be expanded to include consideration of marginalization, oppression, and domination of the sort Young outlines.6 But Young herself argues that the structural account of identity requires more than a greater share of social resources, however those shares are defined. For she claims, on the one hand, that distributive conceptions of justice are static, while relief from oppression and domination require procedural overhaul, where modes of decisionmaking, political representation, and social formation of dominant attitudes come under scrutiny, not merely the overall picture of advantage and disadvantage at any given time (Young 1990b; cf. also Anderson 1999). Nevertheless, Young departs from what she calls the politics of cultural difference in laying out her view of positional or structural difference. The former view rests on claims about the special relation persons have to cultural patterns and languages, while her structural view focuses on the way that “the operation of diverse institutions and practices conspire to limit [groups’] opportunities to achieve well being” (Young 2007, 63). Young is insistent that such a focus does not reduce to traditional liberal egalitarianism for the reasons just mentioned as well as that approach’s blindness to injustices that occur outside of state institutions in civil society. 6

See what Young describes as the “five faces of oppression” in Young 1990b.

196

The Politics of Persons

However, Young has consistently defined the fundamental terms of (in)justice in ways that arguably refer to a kind of individual autonomy, at least a sort that is sensitive to the relational aspects of that characteristic. Domination, on her view, essentially involves the systematic exclusion of people from participating in determination of their own actions and conditions, while “oppression” is defined as preventing people from realizing satisfying skills in socially recognized settings (Young 1990b, 38). While it can be granted that political and social action should go beyond the kinds of reforms of state institutions that liberal egalitarianism typically calls for, it may also be the case that a suitably nuanced and rich conception of autonomy will capture the injustice of oppression and domination as she defines them. For autonomy will mean, generally, the ability to direct one’s actions and the conditions of one’s actions, implying that one must live in social settings that allow and facilitate this. Note also that Young’s view of justice and (what I am calling) autonomy makes no essential reference to group identity as such.7 At least when spelling out the core concepts and principles she defends, Young refers to identity as important because of its role in unjust limitation on opportunities and selfdevelopment; but those latter notions are valuable for all people as such, not in different ways for different members of groups. Before continuing with the discussion of how considerations of social identity ought to affect conceptions of autonomy and autonomy-based principles of justice, let us look briefly at some of the ways that liberal political philosophers have attended to these issues, either to support special rights and exemptions for cultural groups or to argue that such demands are overdrawn.

ii. liberal accounts of identity-based interests One of the best known attempts to accommodate claims of cultural identity and group rights is that developed by Kymlicka, whose argument makes use of a claim connecting a person’s identity with her culture when that culture takes on a particular form. In a view echoed by Margalit and Raz (1990), 7

This appears to mark a departure from her earlier view of the constitutive relation between social groups and their individual members, where she argued that “groups constitute individuals” (Young 1990b, 45). In the conception of autonomy and autonomy-based justice developed here, we accept the social conception of the person (albeit not with reference to specific social groups in a static manner) but insist on an individualized conception of autonomy. This, again, is arguably consistent with Young’s overall approach.

The dynamics of social identities

197

Kymlicka describes what he calls a “societal culture” that is meant to exemplify the connection between group practices and individuals’ selfconcept. A “societal culture,” says Kymlicka, is “synonymous with ‘a nation’ or ‘a people’ – that is, as an intergenerational community, more or less institutionally complete, occupying a given territory or homeland, sharing a distinct language and history” (Kymlicka 1995, 18). Such cultures, Kymlicka argues, give meaning to the choices and goods in pursuit of which autonomous agency (freedom) is worth protecting. On this view, freedom is valuable because it is a component of the person’s pursuit of the good. Insofar as one’s societal culture forms a community that defines and gives meaning to such goods, then protecting the social forms that comprise the value of those pursuits is therefore intertwined with valuing autonomy. Liberals can accept that the self is “social” in ways insisted upon by its critics, but the ability to reflectively evaluate those social connections will always be a capacity that is fundamentally protected and valued under liberalism. No assumption need be made that the person can undertake this evaluation apart from all ends and values at once, as long as the critical reflection central to autonomy is possible for each such connection in isolation. Now others have claimed that the strategy pursued by Kymlicka cannot support the devotion of resources to any particular culture, since autonomy merely requires I have a choice among goods. Insofar as freedom means being able to step back from any particular commitment and revise it in the face of social possibilities, and there are several cultural avenues and traditions I could choose in my society other than the one I grew up with (let us imagine), there is no autonomy-based argument for the survival of my culture in particular (see Appiah 2005, 123; Taylor 1991). The problem here is that Kymlicka fails to specify the relationship between the individual and cultural practices in a way that grounds the identity involved in such contexts. For first, claiming that societal cultures provide a “context of choice” that makes autonomous pursuit of the good meaningful ignores the ways in which such cultural forms form part of the way agents understand themselves, order their values, and interpret the world. The identity involved here is the link that ties group members to their own cultures, a point that, as the critics just noted pointed out, is not fully sustained in the argument Kymlicka develops.8 In addition, Taylor has argued that while Kymlicka can support particular rights to protect the current generation of cultural minorities, he cannot ensure what Taylor thinks is essential for the 8

For an attempt to spell out criteria of social identity based on self-esteem, see Copp 2002.

198

The Politics of Persons

full meaningfulness of my pursuit of culturally tempered goods, namely the indefinite survival of my culture into the indefinite future (Taylor 1992).9 Moreover, Kymlicka makes a fundamental distinction between those groups whose culture is under attack by way of processes imposed upon them (such as national minorities that became surrounded and dominated by colonial movements) and immigrant groups who enter a country “voluntarily.” Now to say that such exclusion of ascriptive social groups is problematic requires an argument, which is sure to be controversial, that membership in social groups of this kind carries with it perspectives, modes of social life, and relations to others that consistently ground special interests that justice should respond to. Some have claimed this of course (Alcoff 2006; Cochran 1999) and we will discuss an example below, but the problem with Kymlicka’s focus on culture can be stated more generally. If the value of protecting cultural forms is its relation to autonomy, the reason to distinguish between national minorities and ethnic groups dries up, since that difference is not based on the depth of the connection between autonomy and culture but on the process by which that connection has been fractured (Benhabib 2002, 59–67). As I noted, the problem with Kymlicka’s argument is that he lacks a full account of the identity that membership in a culture underwrites. Such a lacuna also explains why Kymlicka focuses on cultural identities to the detriment of membership of social groups that have also been the focus of identity politics in recent decades such as those characterized by race, ethnicity, and gender. Now I agree with Brian Barry that attributing common cultural characteristics to groups with ascribed identities such as race or gender trades on very flimsy sociology (Barry 2001, 95–96). It is highly doubtful that African Americans, for example, who differ in class, geography, religion, sexuality, and other dimensions share habits and practices of the sort needed to define a culture. Indeed, feminists and others have been claiming for years that attribution of essential characteristics to such a disparate social group as women is always problematic. But it is important to point out here that denying that, for example, women have common social practices or modes of living that shape their perspectives consistently does not entail 9

Interestingly, Appiah criticizes Kymlicka’s position for having just the implication that Taylor claims it lacks, namely the requirement that resources be devoted to support the survival of one’s particular culture in perpetuity (Appiah 2005, 125). Appiah claims that having a right to the temporally unlimited survival of one’s culture makes an untenable demand on not only the rest of the population but on one’s own cultural co-members who might be tempted to stray from the traditions that define one’s way of life. The degree to which such straying is in evidence marks the degree to which the culture is not as robust as it was, thereby implying that traditional members have even less of a case for the role that culture is meant to play in their collective identities.

The dynamics of social identities

199

denying any identity-based interests on the part of members of such (ascriptive) groups. Historical and ongoing patterns of treatment that have placed women and racial minorities into constrained and denigrated (and dangerous) social positions lend much weight to claims of special interests linked to group membership. This is to reiterate Young’s view that marginalization based on social positioning serves as grounds for shaping principles of justice that have special reference and sensitivity to differences in group membership. More must be said on that score, but it is also worth mentioning other responses to claims arising from considerations of identity on the part of various political philosophers, especially insofar as an emphasis on autonomy is in play in those exchanges. Indeed, the most powerful recent line of attack waged against defenders of identity-based claims is that for any particular cultural form, strong protections for the continued practice of specific traditions, languages, rituals and other marks of societal cultures belies the complex heterogeneity of cultures and other identity groups (Appiah 2005; Benhabib 2002; Gutmann 2003a; Sen 2006). This accords with a robust skepticism about the assumption of a stable self that is defined by a static set of commitments, lineage, cultural identity or connection with a particular social marker. We are all complex beings with connections to cultural histories that are themselves more like threaded rivers than straight lines. And particular associations with social groupings, such as seeing oneself as most saliently a man or a gay man or a woman etc. all vary in intensity depending on variable settings that may call them to mind (see, for example, Phinney 1996). Anthony Appiah and Amartya Sen have pressed this case most forcefully recently. Appiah explains how very transitory and manipulable some identities can be, citing the Oklahoma Robbers Cave study where children were given artificial labels and almost immediately began fiercely attributing positive and negative characteristics to members of the groups and forming strong bonds and oppositions in association with them.10 He quotes Ian Hacking’s claim that identities often come into being when labels are invented for them and are given public purchase. More generally, Appiah claims that collective identities tend to have the following diachronic structure: 1. Terms of public discourse come to ascribe membership to individuals and a social conception of the group emerges; 2. Such conceptions are internalized by group members; 3. Members are then also treated as members by others (Appiah 2005, 64–70). 10

There are several other examples of this, such as Jane Elliot’s “blue-eyed/brown-eyed” experiment (see: http://janeelliott.com/videos.htm).

200

The Politics of Persons

Part of Appiah’s point is to show the illusion of taking identities as transhistorical, permanent, or natural. Identities arise from contingencies of sociology, geography, and other factors which, were they to have gone differently, would have produced different self-conceptions. Such identity markers can be “taken up” or not, which means that individual judgment and choices remain as part of the grounding of a social identity. The importance of choice is similarly stressed by Sen, who lays out in illuminating detail how various global cultural identities are each plural rather than singular in their history and inner dynamics. A Muslim, for example, may identify prominently with her religion but can also look to her gender, nationality, profession, and so on for the locus of her conception of herself; and this locus may well shift in different situations and/or over time. He writes: “Given our inescapably plural identities, we have to decide on the relative importance of our different associations and affiliations in any particular context” (Sen 2006, xiii). This last quote encapsulates both the central insight as well as the fundamental non-sequitur of Sen’s argument. The “fact” that all individuals have personal connections with a variety of contemporary and historical groups cannot be overstated, though I put “fact” in quotes because it is unclear what the status of this observation is exactly. A person may identify herself and be identified as an African American, say, but this seems unrelated to the exact contours of the person’s ancestry, which may well include many people of European descent. This leads to the non-sequitur: even if we accept the multiple identities we are all composed of, it does not follow that it is a matter of choice which one is predominant in which situations. Sen uses as an illustration one Cornelia Sorabji, who came to Britain from India in the 1880s. She variously described herself over the years as an Indian, an English woman, a Christian, a lawyer, a supporter of the British Raj, and so on (Sen 2006, 159–160). Sen concludes that “Cornelia Sorabji’s choices must have been influenced by her social origin and background, but she made her own decisions and chose her own priorities” (ibid., 160). First, although the list of Ms. Sorabji’s identities is long, it is a specific list. Even if we can say she was able to choose which of the many “hats” she could wear to allow to predominate in various situations, the store of hats from which she could choose was specific and finite, and it was not up to her to pick one that was not on the list. More importantly, some strongminded and resourceful people like Ms. Sorabji may well be able to hold to one of her dimensions of identity and insist upon its dominance in some situations, but there will be other people (and I venture even she in some

The dynamics of social identities

201

situations) who will not have such a prerogative. If everyone around her treated her as an Indian woman, no amount of protestation from her that she was in fact (also) English would have mattered. Even if she can internally “choose” from among her various identities, many cannot: both the social situation and their own psychological structure direct them to identify as, say, a white man or Chinese American in particular contexts. Saying that there are other historical dimensions to their ancestry that they could (counterfactually) see as their own does not mean that that is a real choice for them. What this shows really is not that Sen or Appiah are wrong in their central inferences but rather that their projects are straightforwardly normative. They cannot be saying that their descriptions of the multiform and ephemeral nature of our identities are simply facts to be discovered like historical facts. Rather, they are using historical and psychological observations to mount a case that we all should think of ourselves and our identities differently. We should stop thinking that our race or ethnicity is either pure or homogeneous or that they demand from us a particular behavioral profile. We should also stop treating others along these lines as well. As a normative argument about self-concept and interpersonal relations, these arguments have much to recommend them. But if people do not in fact see their ethnicity or race as so porous or malleable, are we then to conclude that they are simply mistaken or that the interests that (they might claim) are tied to those identities are now groundless? If the special ground that such interests rest upon is their connections to their self-concepts, it is no argument against the weight of those interests to say that such people ought to alter their self-concepts. More must be said, at least, to connect the “facts” about a person’s contemporaneous and historical connections to a group to the interests that she bases on her identity with those groups to make such a claim. One interesting line of argument put forward by liberals about strong cultural identities is that the protection of a right to exit – the right to extract oneself from associations and commitments – is fundamental in the protection of individuals in just societies, and this right must take precedence over any call for the protection of cultural groups (see, for example, Margalit and Halbertal 2004). Although some pluralists insist that even protecting the right to exit is itself culturally parochial (Parekh 1994), this right is generally seen as a universal limitation on whatever claims might be made for the recognition or protection of identity groups in a just society. That is, group interests can be identified and given special weight, but only if membership in such groups is in some way voluntary, if

202

The Politics of Persons

the right to disavow such connections is protected by the group itself (for discussion see Barry 2001). However, while for some social identities the idea that one can (and should be given the right to) separate from them is crucial, for others it is not only unimportant but virtually incoherent: the idea that we should be able to “exit” a social grouping that defines us, such as being a minority group member or a woman (that is, regarding most ascriptive identities) will appear absurd. In addition, which factor among the multiple aspects of our identity becomes prominent may often be pressed upon us by external treatment or social location. Indeed, a person’s insistence on not only the centrality of her (say) ethnic identity to her self-concept but also its importance may be a direct response to a denigrating social environment vis-à-vis that identity. Moreover, political structures which encourage questioning all commitments and an openness to revising them (as well as the geographical and symbolic mobility that this involves) will skew the priorities of political institutions from the point of view of those who have no need to make such changes and moreover are harmed, in a unique way, when openness to change is a social priority. This gets to what really is the heart of the matter. For after the voluminous debates about the metaphysics of identity categories (the “essentialism” debates in feminism for example),11 a rough consensus arose around the idea that such categorizations are human constructions – there are no natural kinds where identities are concerned. As Linda Alcoff observed concerning these essentialism debates, “the anti-essentialists won” (Alcoff 2006, 152). Social identity categories are clearly human constructions (though they may be built upon observable facts about people, such as skin color, genetics, or morphology). The question that concerns normative social theory concerns the relative value of those categories and, if justice demands a critical stance toward their existence and others’ recognition of them, the comparative costs of altering them. Most commentators agree, for example, that the idea of “race” is a construction of language and social practice (and self-concept) that arose contingently in particular form in (in this context) Western history sometime after the Middle Ages (see, for example, Appiah and Gutmann 1996; Mills 1997, ch. 1). In the North American context, then, the races are categorized and self-identified in a way that trades on this history, rather than, say, biological facts alone. But that clearly doesn’t settle the issue about whether we (and our political institutions and social practices) should continue to recognize racial categories in 11

For discussion, see the essays in Meyers ed. 1997b and Fuss 1989.

The dynamics of social identities

203

ourselves and others, not to mention whether avoiding them is psychologically and socially possible without great cost. That latter question turns on any number of personal, social, and political factors surrounding the issues in question. My only point concerning the liberal views considered in this section is that reducing identity elements to a matter of choice oversimplifies those factors a bit too much and thereby undervalues the costs or benefits that might be in play when considering any alteration in the recognition or self-identification of people as members of social groups. One last observation about the focus on hybridity of identities should be added. Both Appiah and Sen emphasize the role of choice in the determination of predominant identities in our motivational systems (and hence our values and interests). Appiah, in fact, points out how two different kinds of choice are involved in this determination: the choice of which identity should predominate and how important such an identity should be for one’s pursuits, values, and social connections. I have already cast doubt on the ways that such choice can be assumed in the general case, but it might also be useful to keep this distinction in mind in thinking about the possibility of self-alteration by way of reflective decisions and choices. For Appiah advances a key insight here, namely that it is one thing to say we have a choice among the social identifiers that define ourselves (and this is the dimension of choice I have resisted as a safe assumption about all agents); it is another, however, to say we can choose what it means to be this or that sort of person, what the implications are and should be, in our judgment, for being a member of a social group. It may be that I cannot choose whether or not to be a male,12 but what being a male means to me, what I might ask my fellow citizens to do because of that identity, how important I would like those requests to be, and, in turn, what others might do in response to all those judgments on my part, may well be seen as up to me, as a choice. This was the point we discussed in chapter 6. What I have suggested in this section is that these liberal responses to the challenge of multiculturalism have not adequately dealt with the role that choice plays, and does not play, in claims for special recognition of identitybased needs. Moreover, it has also become clear that more must be said about what it is we mean by identity in these cases. What are social identities such that they sometimes ground distinct interests but are not fixed by nature or metaphysics (assuming they’re not)? Moreover, what is the 12

Though of course, there is a way in which this is a choice. As I will discuss later, even sex/gender need not be seen as outside the purview of reflective judgment in all cases, though it is understood that way for most people most of the time.

204

The Politics of Persons

relation of identity involved here, as that concept has so many different meanings? In the next two sections I will attend to these questions. In so doing I want to sketch an account of the relation between social identity and autonomy that succeeds in the balancing act I described at the outset, namely an account that avoids these unpalatable dangers: (a) the specification of identities that are overly fixed and exclusionary (the problem of essentialism); and (b) reducing identities simply to objects of choice, thereby failing to give a proper psychological picture of the way that many people relate to these social groups and the centrality of these relations to their self-concepts. iii. identity, identification, and autonomy As Appiah observed concerning Taylor’s account of recognition, there are two dimensions of identity that should be distinguished conceptually (although they are closely related psychologically): the collective identity of a social group – the characteristics that pick out particular categories of people such as race, ethnicity, physical ability, and so on; and the individual identities of people, whose self-concept may to some degree contain essential reference to social group membership (Appiah 1994).13 An external recognition of group cohesion, that certain people are or can be treated as similar for the purpose of social policy, is different from the claim that the members of those groups have internalized those categorizations in a way that defines their interests. Appiah goes on to comment that there can be tension between these external and internal perspectives, in that membership fixed by external designation or by tradition or by a census might put negative pressure on individual members when their own aspirations, lifestyles, and self-conceptions run counter to those designations (Appiah 1994, 155). In order to claim, then, that people’s interests should be defined in ways that pay special attention to identities, there must be an argument of sorts that their self-understanding is shaped by that kind of group membership. In addition, it must be claimed that their values and perspectives – the elements of identity that political principles presuppose and build upon – are so shaped by those memberships that normative views that ignore them in effect ignore (or fail to respect) those people (equally). That is the kind of claim that is on the table. 13

Others have made parallel distinctions. Cf., for example, Benhabib 2002, ch. 3.

The dynamics of social identities

205

To consider the search for such an account, it is important to reiterate two different angles from which to view (models of) the self, the “I-self ” and the “Me-self,” as discussed in chapter 2. Most commentators on the politics of the self have focused on the Me-self, the set of commitments and values that a person takes herself as having but which she cannot reject, which “define her” in an important way. But where questions of identity and identity-based interests are concerned, I think the correct focus should be the I-self, the mode of thinking, feeling, and acting that the person instantiates when she thinks, feels, and acts. Insofar as membership in social groups shapes that operation, and insofar as society’s treatment of that grouping (and its outward expressions) affects that operation, to that extent identities are deeply politically important. The question, though, is whether or how that happens. Linda Alcoff has developed a phenomenological account of identity that she intends to range over cultural connections as well as ascriptive categorizations such as gender, race, and ethnicity that may well fit this bill, if it is successful and otherwise acceptable. On her account, “identities operate as horizons from which certain aspects or layers of reality can be made visible” (Alcoff 2006, 43). Our social locations, she argues, constitute a horizon of perception and experience that is constitutive of the self. Patterns of social recognition interact with subjective perspectives to construct a framework of understanding that shapes our interpretive and epistemological capabilities and hence our experience of the world. While she denies that this shaping of horizons creates a static and essentialized self, she argues nevertheless that such identity markers as race, ethnicity, and gender constitutively structure the embodied self ’s place in and experience of the world. Conceptions of rationality, autonomy, and selfhood which ignore this phenomenon fail to capture the unique interests and perspectives of embodied subjects whose relation to social histories and ongoing patterns of treatment shape their self-concept and basic values and needs (Alcoff 2006, ch. 4). To emphasize this last point, Alcoff argues that seeing rational autonomy in a purely procedural sense makes that a controversial locus of respect, recognition, and attention to basic interests that normative political critique (and questions of justice I would say) must respond to. Following Charles Taylor, she argues that such an account of agency fails to acknowledge the substantive nature of the self and its commitments. Indeed, accounts of (fully functioning) rationality, she continues, always make implicit reference to substantive values, in that commitment to highly destructive or idiosyncratic aims is always taken as strong evidence

206

The Politics of Persons

of lack of rationality (such as the desire to murder one’s own children) (Alcoff 2006, 49–54).14 What is crucial about an approach such as this is the combination of a positional account of social difference, resembling in many ways Young’s view (whom Alcoff cites), and a phenomenological/hermeneutical view of human social experience. This allows Alcoff to talk about ascriptive identity groups defined by gender, race, and sexuality and cultural identities. Members of these groups, she argues, may not share common cultural habits and practices (though they might to some degree), but they will typically share with each other a social experience marked by their visible identities, the manner in which they are socially marked and in many cases marked by exclusion, denigration, and violence.15 Alcoff is very careful, however, to emphasize that on her view the characteristics that define social groups, and hence any claims about their common interests and values, rely on fallible and revisable models of those groups, whose existence and commonalities are contingent and not metaphysically fixed. This implies, though, that claims made on their behalf in the formulation of basic principles of justice or social critique (Alcoff would see herself engaged more in the latter) must inherit this revisability. Claims about special needs, status, social position, or interests more generally cannot be made in advance from a philosophical (or even socialpsychological) point of view. We cannot say, for example, that women have common fundamental interests unique to them grounded in their distinct embodied phenomenologically structured perspective. This is because, among other things, that perspective is heterogeneous, malleable, and subject to varied interpretations. Similarly, the embodied phenomenological structure of women’s experience, upon which those interests are grounded, is also highly heterogeneous.16 Alcoff would agree, I think. But this implies that in arguments about the fundamental normative principles shaping and legitimizing social and political power, we cannot specify special weight for the interests of such 14

15

16

I comment on the import of this argument below, but I should note here that it would be successful only if we ignored the difference between providing “evidence” for lack of rationality and defining it. Proceduralists can accept the former phenomenon. This very brief encapsulation is unfair to Alcoff ’s detailed and highly qualified view. For example, her understanding of identity includes elements of existentialism, psychoanalysis, feminist relational views of the self, and post-modernism, all qualified and commented upon. Since my aim here is not to mount a critique of Alcoff’s views, and in many ways my own view is in harmony with hers, I take some central ideas in order to facilitate my own discussion. Alcoff’s discussion (2006, chs. 2 (section II) and 3 (sections III and IV)) follows a path quite in the spirit of this analysis.

The dynamics of social identities

207

groups in advance and fix them abstractly as we do with a right to free speech or an interest in primary goods. Whatever special weight such interests have are represented in other ways in such principles. What is interesting also about this account of the social self is that, despite Alcoff ’s rejection of a purely procedural account of rational agency, a revised “modernist” view of autonomy of the sort defended here may well be quite compatible with most of these critiques and observations. If autonomy requires only self-reflection, without assuming that such reflection abstracts from all values, social positioning, aspects of embodiment and physicality (as also marked by social position), and other aspects of the social self, then these views can accept the possibility of autonomy as a political value. Of course a distinct argument would have to be waged for the positive value of autonomy, and this would depend highly on the normative theoretical context in which that value is designated. But seeing the self as shaped in these ways by embodiment, social position and experience, recognition and identification by others focusing on visual aspects of the self, are all compatible with viewing the self as also able to reflect on aspects of one’s person and social position for the purpose of political dialogue and expression of interests. What I hope to bring out further in this discussion is the political importance of that capacity (as central to autonomy), and hopefully this can be done in ways that resonate with views like Alcoff ’s that bring out the rich and fundamental importance of social identities in our experience of our self and world. So we remain in search of an account of identity that helps ground the kind of special consideration that interests based on social-group membership garners when in fact such special weight is merited, but not an account that affords contestably stable or “essential” properties to group members. The heterogeneity of group characteristics, the contingencies of group (social) construction and recognition, and the malleability of group-related values for individuals all speak loudly in favor of a more flexible model. Let us return, then, to a discussion of autonomy in order to help supply the account needed of how interests tied to social identities can be afforded the importance they deserve. iv. autonomy and the politics of identity: a reconciliation As I explained in previous chapters, the point of sculpting a conception of autonomy in this way is to consider carefully the way in which this characteristic marks the basic interests and perspective of citizens which just

208

The Politics of Persons

political institutions are designed to represent. As I’ve made clear, in my view the conception of autonomy at issue and the political principles it supports should be anti-perfectionist in that the justification of values and principles should not rest on controversial conceptions of moral value, philosophical or religious outlook, or ideals of the good life, a position I defend in the next chapter. For this reason, “autonomy” does not refer necessarily to an ideal of (say) an independent and self-generated life, though this idea may be defensible and indeed has been notably embraced under the banner of autonomy by many in the tradition. But for my purposes autonomy is simply the status marker for citizens participating in collective deliberations whose legitimacy constitutes just manifestations of political power in a modern pluralistic polity. As we have said, autonomy attaches to individuals (only) but such individuals are what they are because of past and present social relations of a complex sort. This is to leave room for the ways in which the components of our value systems, self-concepts, orienting commitments, and so on, have meaning because of their place in an ongoing social narrative. This narrative has, in turn, essentially historical as well as social elements. As Bikhu Parekh puts it concerning liberalism: “liberalism is in theory committed to equal respect for persons. Since human beings are culturally embedded, respect for them entails respect for their cultures and ways of life” (Parekh 1994, 13).17 However, Appiah and others have argued that this inference is too quick, if one construes “respect” for cultures to entail the sorts of measures that Kymlicka and Taylor claimed are supported, for example rights of sovereignty, special exemptions from principles of equal treatment, or rights to enforce language use (Appiah 2005, 73 n. 20). What must be added is a premise that says, in effect, failure to adopt this or that social policy (protecting or supporting a cultural practice for example) will mean that the person’s ability to lead an autonomous life is somehow compromised. This is the premise that a theory of identity should help adjudicate in the question of when and if social policy should respond in this way. 17

Habermas makes the same point this way: “from a normative point of view, the integrity of the individual legal person cannot be guaranteed without protecting the intersubjectively shared experiences and life contexts in which the person has been socialized and has formed his or her identity. The identity of the individual is interwoven with collective identities and can be stabilized only in a cultural network that cannot be appropriated as private property any more than the mother tongue itself can be” (Habermas 1999, 221). However, Habermas argues that this fails to support a claim of right for the preservation of cultures.

The dynamics of social identities

209

However, it is crucial to avoid over-extending this point. For to say that our self-concepts are socially constituted is not to say – and indeed we should not say – that these self-understandings are necessarily composed of particular and fixed relations with other people or social roles. For this would mean that people (all of us) who go through radical change in their lives change who they are, literally. We might say in a moment of exaggeration that we cannot imagine living without X (a loved one), or not being a mother or brother; but loved ones die and the elderly version of ourselves who has moved on and survived is still us. Relations to social patterns more broadly considered, such as value traditions, cultures, modes of socially acceptable ways of being (for women or men, say), social habits and expectations, and so on, all frame and define the basic terms we would use in describing ourselves. All these factors are meaningful only in and as a function of a social context. Such contexts have current as well as historical dimensions. Even at the general level, however, including social elements in the makeup of the self will have important implications for what we say about autonomy and social identities, for the question of whether a person is or can remain autonomous will depend crucially on the social setting he or she exists in. Alienation from some element of one’s character may well ensue if the social constituents of one’s self-concept are unsupported, made inoperative or unavailable, or publicly denigrated (as might happen regarding one’s ability to practice one’s religion with others in public). This is why we mentioned that social elements of our selves are often part of the “I-self” – the orienting and executive functions of our reflective selves. For in some social situations our alienation will ensue because those social conditions do not support or in fact conflict with, work to denigrate, or actively try to crush, the value foundations of our way of viewing the world. Insofar, then, as autonomy requires that one not be alienated from the constituent elements of one’s practical identity, including the historical and social compliments to those elements, one is not autonomous in those conditions. Political and social institutions, then, if they are committed to respect for autonomy, should be designed to prevent or eradicate this sort of alienation. This is at least the abstract conclusion I want to reach here, leaving aside for the moment the particular policies that might be supported by such an argument in any particular social location. These claims concern the form that any defense of any such policies must take. Indeed, I would argue that seeing autonomy as relative to one’s history makes the claim for recognizing identity-based claims even more powerful, when they are well grounded in sociological fact at least. This is

210

The Politics of Persons

because attention to the historical conditions under which one’s identity has been socially treated and defined, where that involves patterns of oppression and degradation, would cause alienation among current members of certain marginalized groups if recognition of special needs (associated with that history) were not acknowledged. However, I could make the claim for special recognition without reliance on a specifically historical conception of autonomy, as long as my points about hypothetical reflection are accepted. In addition, seeing autonomy in the way I describe deflates one particular line of critique by defenders of identity claims against autonomy-based conceptions of justice, namely that seeing autonomy as the central object of respect in principles of justice privileges choice, mobility, change, and separation over solidarity and social group cohesion. As I have argued, autonomy requires only hypothetical reflection in the weaker sense I sketched, so that if a person is not alienated from her connections with social groups (or traditions or cultural practices) that constitute her fundamental values, then valuing autonomy does not imply devoting social resources toward encouraging separation and disconnection from such groups.18 We noted along with Appiah, Sen, and others that cultures and social groups should not be considered as homogeneous or static, and therefore if any special status is given to identity-based claims, it cannot be based on a single and strictly ordered set of cultural values that are resistant to change, even from within the ranks of the members of those very cultures (Appiah 2005). Even if we accept claims made by identity groups as legitimate this does not imply that they must take the form of rights, especially basic constitutional rights for specific cultural protections that are immune to revision. Considerations of justice can enter the institutional design of the state either at the constitutional level, which is fixed permanently or at least subject to revision only with super-majoritarian procedures, or at the legislative level, as a matter of collective choice about the common good. At the former level, there is the provision of basic rights, such as the US Bill of Rights, as well as provisions for the design of democratic procedures themselves. The argument that special protections 18

I should note that this argumentative tack is different from that taken by Kymlicka (1995) and Margalit and Raz (1990) in two important ways. The first is that the connection with cultural practices and group-related interests here is tied directly to autonomy itself, not to the goods that autonomous selves pursue, as they claim. Second, as I will explain below, the claims that this connection to autonomy underscores are not best articulated in the language of (constitutional) rights – to the survival of one’s culture for example – but rather as constraints on deliberation in the determination of social policy.

The dynamics of social identities

211

for cultures should not be seen as a matter of basic right (for the protection of a static cultural form) does not imply that such considerations cannot be given expression in the design of democratic institutions or enter into legislative debate. Also, in order to draw the proper conclusions about the connection between autonomy and social identities, it is important to emphasize certain background conditions against which discussions of principles of justice should be pitched. In order to see the force of claims for recognition of identity, we must take into account that discussions of principles of justice of any sort relevant to current conditions must also acknowledge that these principles are meant to apply to societies with a notable and notorious record of past and ongoing violence and power differences among groups. In particular, one must acknowledge the way that certain groups have been marked as objects of degradation, violence, and oppression for generations, which continues in the current age. One need not utilize a contentious conception of what counts as unjust treatment to specify that a history of violence and differential power (domination) is in evidence in the societies to which principles of justice are meant to apply, in order to avoid distorting the claims made on behalf of the identity groups in question. This point will be reiterated and expanded upon in the next chapter. Another sociological point should be mentioned: it is a commonplace that legal protections of basic rights are less than fully effective in the actual world, as violence, abuse, and discrimination are ongoing facts of life for many. And it bears emphasis that members of certain social groups face historical and ongoing patterns of such violence, discrimination, marginalization, and degradation that make such ineffectiveness all the more pointed. Legal protections are impotent without reform of civil society that ensures general support for them, including not only fear of punishment but wholesale support for the values reflected in those right protections.19 Discrimination in informal settings, most notably employment situations where attitudes and behaviors have an intangible but quite real effect on job opportunities, exacts an untold cost on minorities and other members of traditionally marginalized groups in ways that legal restrictions only begin to diminish. One need only mention as an example, albeit a tragically salient one, that over 100,000 women a year are raped in the United States, according to a conservative estimate by the Department of 19

It is in this framework that anti-perfectionist approaches to justice can postulate conceptions of the good that ought to be pursued for a society, namely those collective goods that are assumed as valuable in the principles of the right definitive of justice. Cf. Rawls 1993 on this issue.

212

The Politics of Persons

Justice.20 More alarming and horrific facts can be reported in other parts of the world and for other groups. The fear of such violence marks the life choices and opportunities of women. The relation between such victimization and the particular needs of identity groups will be brought out presently. This connects directly to the way that claims for special recognition of the interests tied to identity-group membership are at issue here. The disrespect of non-recognition involves the failure to acknowledge and show respect for the way that such an identity label has marked the person (or persons like her) in a way that has involved injustice in the past, in particular that has taken away from her the power to help determine the public and social meanings of such membership. Discrimination, marginalization, oppression, and overt violence (motivated by such discrimination) are instances of social patterns of degradation of the type of person an individual is, which is to say, denigration of the meanings of that group membership. The claim cannot be, as we have seen, that such principles fail to recognize the particular and essential interests that all members of that group permanently share, since such interests vary along with the heterogeneity of the group’s history and character. However, erasing all reference to social identities in the public specification of principles and policies threatens to exacerbate the social oppression of marginalized groups. Now Appiah argues, rightly I think, that seeing culture as a resource, of the order of primary goods, is wrongheaded because such resources can be enjoyed in comparable degrees, traded one for another, and possessed in discrete amounts (Appiah 2005, 123–124). But this shows only that some other calculus of social advantage (besides tradable resources) must be developed that lacks this narrow implication. If primary goods are used as such an index, namely those elements of the quality of life that are not held in degrees, cannot be exchanged for (or like) money, and so on, then those whose identity and commitments require other sorts of goods will not have their relative well-being measured adequately by that index. Social disadvantages due to discrimination, cultural suppression, ignorance, and devaluation of traditions and ways of life will not be reducible in all cases to unequal shares of exchangeable resources.21 Moreover, even if 20

21

US Department of Justice website: http://ojp.usdoj.gov/bjs/glance/tables/viortrdtab.htm; also see Department of Justice document: “Criminal Victimization in the United States, 2004.” Rape is often used as a form of retribution and torture in civil war and other ethnic conflicts, such as Bosnia and Darfur. For discussion of the relevance of such ongoing violence to theories of justice, see MacKinnon 2006. See Kernohan 1998. Also, this is to agree with Nancy Fraser (though for different reasons in the end) that neither the politics of recognition nor the politics of redistribution are complete by themselves. See Fraser 1997, ch. 1 and Fraser and Honneth 2003.

The dynamics of social identities

213

such treatment were correlated with particular material disadvantages or overt violations of equal opportunity (to which liberal principles are positioned to respond), patterns of informal and cultural denigration from longstanding practices of sexism, racism, and discrimination will continue to exacerbate these disadvantages in ways that redistribution of resources alone will never succeed in alleviating. The fact that the inner city poor in the United States are disproportionately African American is closely connected with social attitudes toward blacks as a group that simple redistribution of income will never eliminate. It should be noted that I am self-consciously conflating two rather different lines of argument made on behalf of identity groups. The first are arguments for the maintenance or protection of cultural forms such as special rights and exceptions relating to traditional practices tied to culture. The second are special interests tied to group identification that rest on the way that group has been typically treated, for example interests regarding the fight against discrimination and violence against women, gay people, or minorities. I intentionally corral these arguments under the following general contention, that claims in the public arena that rest on a person’s identification with a group and its collective interests, whether those be interests in the protection of cultural practices or recognition of the particular history (and by extension current perspective) of other social groups, should get special weight in the articulation of principles of justice, in particular principles that see equal concern and respect for the autonomy of citizens as basic. The demand for recognition can be seen, then, as shorthand for the following, long-winded, claim: I deserve respect as a person with interests tied to a certain group membership, interests that have systematically been denied and publicly denigrated, and whose meanings have been discriminatorily defined; I therefore demand to be recognized as fully able to define those interests, and the broader social narrative of which they are a part, for myself (or ourselves). The ability to define and express interests is fundamental to autonomy, as a derivative of the capacity for self-endorsement enacted in the public sphere, without which deep alienation from the social components of one’s identity is inevitable. So recognition of identity involves respect for the capacity to express interests associated with that identity in the public deliberation of policies that affect one’s well being. This implies that in debates over social policy and the common good, special weight should be given to claims made on the basis of autonomy as manifested in citizens’ relations either to cultural practices or the treatment of particular (historically marginalized) groups with which they identify.

214

The Politics of Persons

As I mentioned, claiming special weight for group-related interests need not issue in calls for basic constitutional rights protecting those interests. What is being defended here is that group-related interests should get special weight in democratic decisions when the conditions of citizen autonomy are at stake. This implies that special provisions should be made at the legislative level to allow claims made on behalf of identitybased interests to be given procedural priority in a way that gives them special weight in deliberation about policy. This is because the claims being made, we are assuming, are ones that are linked to the possibility of enjoying autonomy as we have defined it, to avoid the alienation that results when social conditions render the person’s identity with certain social groups (or with the interests associated with being a member of such a group) of lower social value the object of disrespect.22 As we have noted (along with critics of identity politics), such interests and such groupings are not fixed and unproblematic. This is the main reason that these interests are expressed in the design of democratic procedures rather than constitutional rights. So what must be claimed is that representatives of the interests in question have a defeasible claim to speak for the identity groups in question but that those represented are always able to reject that interpretation of their interests. Procedurally, representative associations that speak for identity groups will be answerable to their members, and the legitimacy of these representatives will be a function of the general support they receive from those members.23 22

23

Now I did say earlier that one of the deficiencies of standard defenses and critiques of identity politics is the lack of a plausible conception of identity, one that grounds the claims for identity-based interests when they are valid but which does not rest upon an overly ossified understanding of the nature of social groups. And I have not provided an alternative conception of identity that does this either. Rather, what I have done is urge that we shift our focus, from what identity is to conditions in which it is importantly constricted; that is, from questions of identity to questions of alienation. The latter disrupts autonomy, I have suggested, and hence insofar as a person is alienated from the social conditions in which pursuit of her identity-based interests are meaningful, to that extent respect for her autonomy is lacking and injustice is in evidence. Alienation, however, can arise in ways that do not assume a fixed and historically homogeneous conception of groups or cultures. It is not the protection of positive identities that is necessarily required of justice (as the respect for autonomy) but the avoidance of alienation. For alienation and identification are not perfect complements. Between them lies the condition of ambivalence, acknowledgment of one’s group’s complex social history, and fluctuations in the narrative that provides the orienting framework of one’s basic values. Autonomy as non-alienation allows that for some, maintaining cultural forms or special recognition of shared group interests may hinge on the kind of self-defining relation to one’s social group that writers such as Taylor and Kymlicka (and in an earlier stage, Young) have claimed. But autonomy does not assume that such self-constituting relations always must hold between the person and the social group for justice to obtain, so that the liberal charge that identity claims rest on ossified conceptions of social groups is avoided as well. For discussion of the role of identity groups in democracies, see Gutmann 2003a.

The dynamics of social identities

215

This is akin to Habermas’s view that the establishment and valuation of both private and public autonomy must be co-original, in that individual self-government is incomplete without collective self-government and vice versa. The requirements of democracy (and more generally, for him, communication and discourse) both presuppose and support the enjoyment of individual autonomy. Only if I can claim participation or representation in effective public forums, where my perspective and voice functions on equal footing with the competing voices of my co-citizens, can sense be made of my own ability to govern myself in a social setting that partially constitutes my identity. Insofar as that identity has group-oriented dimensions such as ethnicity, race, gender, or sexuality, only when the interests associated with that group membership get particular play in those public discussions can I be said to be afforded respect as an autonomous individual (Habermas 1996a, ch. 8, 1999; cf. also Benhabib 2002). The argument I have constructed here is complex – surely overly so – but it has the following general form: autonomy involves non-alienation from factors that function in our basic value orientations; such factors have social constituents, such that their meaning and value are part of social practices and relations of a particular sort; when those practices and relations are denigrated or destroyed, alienation is likely to occur hence restricting citizens’ autonomy; so protecting autonomy will mean giving special weight to the interests connected to those social groups. Correspondingly, the form that public claims on policy based on identity interests take must follow the contours of the requirements of autonomy set out: one must claim that failure to alter this or that social policy would prevent the person from pursuing a social self-narrative shaped by her practical identity without abiding alienation. This is not to say all claims with this form are valid on their face, but only that democratic institutions should be designed so that claims of this sort, expressive of such identity-based interests, get special hearing, however the details of the procedural practices expressive of such values are worked out. One major conclusion that emerges here, however, is that due attention to social identities and the recognition of them does not entail the rejection of autonomy-based conceptions of justice, once the notion of autonomy has been formulated in acceptable ways. Attacks on traditional (liberal) conceptions of justice can be seen, then, as overdrawn if the claim behind them is that the culprit is valuing autonomy as fundamental for justice. It may be correct that many liberal philosophers continue to defend a hyperbolic emphasis on choice and self-revision in their conceptions of autonomy and the person. But autonomy can also be seen as

216

The Politics of Persons

hypothetical reflective self-acceptance in light of surrounding social conditions rather than the ability to remake oneself like a living statue sculpting itself into a new form. I think also, however, that such a conception of the autonomous person is strong enough to ground claims of injustice for those social group members who find themselves deeply alienated from the social conditions in which they are asked to pursue their values and find meaning in their identities.

part iii

Justice

chapter 10

Justice over time: history, public reason, and political legitimacy

Now that we have taken a stand on what autonomy should mean in the context of normative political principles, it is time to pick up on observations made earlier and give a bit more shape to these principles and the role a model of autonomy might play in them. We return, then, to political theory in order to sketch in broad terms the implications of the conclusions we have reached so far. What we mentioned in chapter 6, and can now expand upon, is that seeing selves as “socio-historical” alters the way that persons are modeled in principles of justice. This view will be expanded to include examining some of the implications of seeing individual autonomy as a basic political value in the context of principles of justice. In this discussion, I have avoided the label “liberal” for reasons alluded to earlier; specifically, that this term is often used to refer to political regimes that see individual rights and liberties of a particular sort – rights to free speech, association, religion, property, and so on – as justified prior to, and independently of, other formal and informal social forms, including democratic practices and institutions. Whether that is plausible in general is an issue I want to leave open for now, so the family of views I am concerned with will be called “liberal” by some but are ones I generally understand as simply autonomy-based principles of democratic justice. The overall aim of this chapter, then, is to provide a broad sketch of a conception of democratic legitimacy that places autonomy, as we have conceived it, at its center, hopefully making clear how autonomy as we have fashioned the notion would fit into, and perhaps help shore up, the case for such a conception. While I won’t be able to say much in the way of detailed argument for this view – conceptions of justice and legitimacy are big topics – my purpose is to illustrate the usefulness of the concept of the self and autonomy worked out here in the construction of normative political principles, and in so doing I hope to at least show how some trenchant criticisms of traditional forms of autonomy-based democratic justice might be fended off. 219

220

The Politics of Persons

Before turning directly to that topic, I want to comment on a question that is orthogonal to these, and one that our attention to historical trajectories and diachronic identities makes salient, namely the past and ongoing social relations found in societies to which democratic principles at issue are meant to apply.

i. non-ideal theory, historical conditions, and the circumstances of justice A troubling criticism of mainstream political theory in the liberal tradition has been that it either says nothing about, or is irrelevant to, the plight of under-privileged groups and other victims of injustice. The assumptions of standard liberal democratic theory abstract away from the particular patterns of violence, oppression, and inequality that characterize modern societies, currently and in the past. In this way, the experiences of such injustice, as well as practical accounts of how it might be resisted, are occluded in these standard theories. Charles Mills, for example, has argued that the use of “ideal models” as descriptive starting points upon which normative principles are built serves problematic ends, specifically what he calls “ideological” ones.1 The more general question is whether and why it is necessary to consider ongoing and historical inequalities in shaping principles of justice and corresponding conceptions of the person, whether, that is, we should practice “ideal theory” when we conceive of legitimate political orders. What Rawls means by “ideal theory” is the derivation of principles of justice using simplifications and abstractions about the social conditions to which the principles apply. We have discussed the model conceptions, specifically of the person, that function in this manner, but Rawls also sees ideal theory as assuming “strict compliance,” which refers to the general pattern of conformity in the population to the public principles of justice. This is contrasted with “partial compliance” theory which “studies the principles that govern how we are to deal with injustice” (Rawls 1971, 8). As Rawls claims, the “reason for beginning with ideal theory is that it provides … the only basis for the systematic grasp of ” the “pressing problems” of injustice that plague modern societies (ibid.) Mills argues that using models as descriptive ideals in broader normative theories, when such models deviate systematically from the undisputed 1

Mills 2005; see also Mills 1997 and Pateman 1988. For general discussion see the essays in a special issue of Social Theory and Practice (34(3), 2008).

Justice over time

221

facts of social life as well as the experiences of those bound by those normative theories, undercuts the practical aim of such principles. Insofar as the simplifications operative in these policies and principles fail to apply to those lives and social settings bound by them, they cannot effectively guide action. More pointedly, such principles of justice will be useless in struggles against long-standing, systematic injustice that, presumably, they are meant to direct us to work against.2 Along several dimensions, such as an idealized understanding of individuals (abstracted from class differences, patterns of domination and oppression, and so on), their capacities (including cognitive capacities for self-understanding), and the social institutions they live under, ideal models of persons and societies deviate from readily acknowledged aspects of political life in so glaring a way that one is tempted to conclude, he writes, that there is nothing but an ideological purpose behind such simplifications, namely to leave uncriticized the patterns of exclusion and domination that characterize modern life (Mills 2005, 69ff.). Now there is an easy response to this argument which was stated clearly in Rawls’s view quoted earlier: one cannot describe current circumstances with reference to systematic injustice (or any cognate ideas which rest on an understanding of injustice) without already assuming a fully worked out theory of justice. Rawls does not discuss (in detail) the patterns of victimization and discrimination that abound in our world because his purpose is to articulate why any such patterns are indeed wrong. However, I agree with Mills’s general point that normative principles that are meant to apply to a social situation that is, shall we say, inaccurately described, will not only lack practical motivational force but will also give a distorted normative appraisal of that situation. A set of rules for a foot race will be different if the runners assumed to be the competitors are all able to run upright in the average manner. But if it is known that half the competitors in the field are victims of conditions that include debilitating leg injuries, it would certainly seem puzzling to adopt the old rules without amendment. The point of the race is lost when one set of idealized conditions is assumed to generate rules that are then applied to very different and non-ideal circumstances. Another way to come to a similar conclusion is to reconsider the standard background assumptions made in discussions (and justifications) of principles 2

Mills is keen to point out that in most mainstream (ideal) theory, including Rawls, there is virtually no mention of the systematic victimization that has plagued the developed constitutional democracies that are intended as the locales to which those theories apply, such as the history of slavery, discrimination, and racism in the United States.

222

The Politics of Persons

of justice, specifically the “circumstances of justice” which are assumed to hold for people and environments over which normative principles are to range. The specification of such background conditions were famously described by Hobbes and Hume and refer to the aspects of personal and social life that make justice “both possible and necessary.”3 They are meant as minimal conditions, in that other factors that arguably hold in the real world could vary without making the basic rules of social cooperation irrelevant or unattainable. The classic summary of these circumstances is that individuals have limited benevolence, live in a world of moderate scarcity (so that desires conflict to some extent), and are of roughly equal power. Rawls divides these into objective conditions, which include the assumption that “individuals are roughly similar in physical and mental powers; or at any rate, their capacities are comparable in that no one among them can dominate the rest” and moderate scarcity, which means that “[n]atural and other resources are not so abundant that schemes of cooperation become superfluous, nor are conditions so harsh that fruitful ventures must inevitably break down.” The subjective circumstances concern the individuals’ needs and desires and include the stipulation that while such people have some needs in common they also have different ends and purposes and “make conflicting claims on the natural and social resources available” (Rawls 1971, 109–110).4 Now Rawls knows, of course, that actual conditions in society vary markedly from these assumptions, but his point is that for abstract principles to apply generally, across a variety of social conditions, it is required that the background circumstances assumed in the models be minimal, meaning that any, more substantive, specifications would render the principles derived from them irrelevant when applied to circumstances that deviate from those assumed to hold. But the critique waged by Mills concerning ideal theory can be applied here in different form, namely that when principles are derived that assume rough equality of social and other forms of power and are applied to conditions of systematic dissimilarity along those dimensions, they can have the distorting effect I alluded to earlier. That is, past and ongoing victimization and domination of some groups by others may well be exacerbated when the facts of that domination 3

4

The traditional versions of these statements can be found in Hobbes’s Leviathan, ch. 13 and Hume’s Treatise, section 3.1; see also Rawls 1971, 109 – 112 and Barry 1989, 148, 152–163. For comment, see Vanderschraaf 2006. Rawls adds what he calls “formal principles of right” that specify that principles of justice (for him, those chosen in the Original Position) must be general, apply universally (without reference to particular individuals), be public and final Rawls 1971, 112–118).

Justice over time

223

are ignored. Just as the runners who have injured legs will be disadvantaged in a race whose rules assume uninjured competitors, victims of historical injustice will be worse off when the fundamental principles of justice rest on assumptions that occlude that victimization. More specifically, the interests of persons modeled in these procedures (for deriving principles) will not be adequately specified when the circumstances of justice assume away their experience of past and ongoing oppression. The model of the person I have been assuming in this book, for example, has included references to memory and history, so that people’s attitudes toward the success of their life’s projects include sensitivity to how things have gone so far, indeed how such projects are understood as part of an already structured autobiographical narrative. People who see themselves as victims of past oppression will have interests connected with (at least) addressing the conditions that arise from that past and, more importantly, will be disadvantaged by principles that see everyone as starting out as equals. This is largely what Mills and others have claimed is the bias in so-called impersonal (ideal) principles in favoring status quo power relations.5 Those who are relatively free of past domination and who (among themselves) enjoy relatively equal power relations will have basic interests of one sort (perhaps matching the traditional liberal interests expressed in basic freedoms for example), while those who see themselves as victims of ongoing domination will add other interests to the mix, such as addressing and/or accommodating the effects of that domination. We need, however, to avoid value-laden terms in specifying these circumstances, ones specifically that rely on a conception of justice for their meaning. There are ways to do this. First, there are notions of social power that can be specified that do not rely on a fully worked out conception of just relations to understand. When persons or groups are systematically better able to pursue their favored projects, shape social institutions, generate social policy, and so on, they can be said to have greater social power, and saying this does not depend on our understanding what “justice” means.6 Relatedly, it will be acceptable to describe the background circumstances of justice to include the stipulation of ongoing domination of groups, which is a function of the differential power they enjoy with the added dimension of relations between groups. Social domination of one group by others involves differential power, exclusion from generally favored social positions, and intentional and/or passive denial of equal 5 6

See, for example, Young 1990b, ch. 4; cf. also Kernohan 1998, 48–70. For discussion of the concept, see the essays in Lukes 1986.

224

The Politics of Persons

social and political status to those groups. Finally, we can describe widespread violence as an aspect of the social landscape to which justice applies, specifically violence aimed at groups that also experience differential power and suffer from domination. All of these phenomena are uncontroversially true of racial and ethnic minorities, women, and indigenous peoples in virtually all of the societies to which democratic principles are meant to apply. Even if these factors cannot consistently be described in value-neutral terms, it can be assumed that any principle of justice generated by ideal models would describe the conditions as unjust. So for these reasons, I propose that we add to the circumstances of justice the presence of measurable inequality of social power, the systematic exposure to unique forms of violence, and a pattern of past domination affecting identifiable groups in the society, conditions that all would be labeled as unjust by any plausible principles.7 This also aligns with the socio-historical account of persons that I have sketched earlier. As I mentioned in chapters 6, 9 and elsewhere, citizens’ conception of the social history with which they identify and of which they see themselves as a product importantly shapes their interests and need for proper representation in democratic institutions. These and other aspects of the non-ideal circumstances of justice that I here insist upon will affect standards of justice and legitimacy by putting pressure on institutions to allow citizens to gain redress from their disadvantaged position in the democratic procedures that establish and maintain that legitimacy. ii. autonomy-based justice and political legitimacy The overall picture of justice I envision here, which I will use to illustrate the basic value of autonomy in the context of the justification of political principles, takes the form of the fundamental conception of a constitutional democracy for pluralistic populations in the modern age. The view is anti-perfectionist, in ways I have alluded to and will discuss further, and generally resembles the proceduralist and/or political approaches to the justification of principles exemplified by the work of Rawls (with variations I will add) and, to some degree, Habermas and theorists of 7

Notice that I am not claiming that current inequalities or patterns of victimization would be labeled as unjust on all accounts (though I am fairly confident they would); only that such uncontroversially unjust conditions prevailed until recent times. Such things as overt discrimination, racialized and sexualized violence, attempts at extermination directed toward indigenous populations, and related phenomena, would be included here.

Justice over time

225

deliberative democracy. What I see as essential, however, is that democratic practices are not only complementary to just political institutions but are constitutive of the establishment and maintenance of the legitimacy of those institutions. First, recall the socio-historical model of selves sketched in part I, specifically the view that political models of the self must leave room for social and historical factors that may shape identity along with an understanding that such factors vary along several dimensions. The acknowledgment of a history of domination and power inequalities just mentioned, along with the special interests in recognition of such inequalities (and related social factors) by some groups, discussed in chapter 9, will figure in the specification of autonomy-related interests in principles of justice. Also, the view of autonomy we examined had as one of its central requirements non-alienation, whether from elements of one’s self and/or one’s social condition. So given these claims, an autonomy-respecting political order will require some mode of collective self-government, some power over conditions to shape and symbolically code the social factors that constitute citizens’ interests and identities; if there is not power over such factors there must at least be avenues of objection so that alienation of the sort that undercuts autonomy is forestalled. The view of democratic justice I envision here, then, rests on an argumentative structure along the following lines: Since selves are, in variable ways, socially constituted and values are defined in terms of interaction with others, our abilities to pursue valued ends are both defined and constrained by the social dynamics in which we engage in those pursuits. Political structures and other institutions of power shape and codify those dynamics in broad and robust ways. That power is justified only if it can be seen as harmonizing with our own judgments, our perspectives about what is valuable to pursue given the fact that we live among people with contrasting values and who (like us) are products of the contingencies of history, both their own and society’s.8 The legitimacy of these social processes must rest, then, on the way they are controlled and produced, and only if citizens’ perspectives and interests are properly represented in those processes will that legitimacy be attained. If our interests could be represented as static and objective – if we all had the same basic interests ordered in the same way – then such processes could be designed along those lines, but that would rest on a kind of perfectionism 8

This claim can be understood to be supported by Rawls’s observations about what he called “the burdens of judgment,” which will be discussed further below. See Rawls 1993, 54–58.

226

The Politics of Persons

that we want to reject (see below). That is, no specific set of social connections or markers of social identity can be specified permanently in advance, for the particular mode of social connection and historical narrative that constructs our identity will vary for each of us during our lives and across persons in a population in various and unpredictable ways. Therefore, only if the principles that guide and shape (and justify) those power structures model us and our co-citizens as autonomous – able to competently, independently, and authentically judge values for ourselves – will they gain the legitimacy that just institutions require. What we will see is that attention to social connections and narratives that shape identity must take the form of allowing citizens to speak for themselves in specifying the nature and importance of those connections. For these reasons, people’s autonomy should be promoted and protected in the operations of legitimate democratic institutions. This kind of protection will require, then, a mode of collective deliberation that allows citizen input into the processes that yield social conditions that shape those citizens’ prospects. If not actual participation and deliberation, practices of these institutions must include elements that effectively represent those citizens’ interests and judgments and, as an extension of this, the power to object when policies conflict with their deepest selfunderstandings. Democratic deliberation, then, also requires participants’ abilities to reflectively endorse, indeed publicly defend, the points of view, values, interests, and opinions that are the inputs to such deliberative processes (the “outputs” of which are social principles and policies). This provides further reason for the presupposition that the autonomous person is able to reflectively grasp and present her values and perspective. This accords citizens this kind of representational authority over those points of view but also necessitates their capacity to reflect on their values as part of the dynamic of social interchange that produces collectively justified principles. So autonomy as competent, self-reflective endorsement (nonalienation) is central to this understanding of justice and politics.9 Therefore, to establish legitimacy for political institutions and principles, self-reflection is a crucial mark of the autonomous citizen whose status is respected and whose interests are protected in just political arrangements. Only if a person is put in a position to speak for herself, can the collectively generated principles of justice claim the legitimacy required by democratic justice. Advancing her interests in a way which thoroughly bypasses reflective endorsement of them threatens to violate the requirement that values 9

The ideas in this section are taken in part from Christman 2005.

Justice over time

227

promoted in a society obtain validity only by being subject to the citizens’ endorsement of them. So legitimacy presupposes a model of the (autonomous) person able to reflectively endorse her interests, respect for which is reflected in the structure of the principles themselves. Recall that we discussed at several points questions surrounding the level and accuracy of self-understanding that our model of autonomy requires (in particular given the fluid and ecumenical conception of the self utilized therein). In the current context, we must ask what the standards of selfunderstanding and cognitive competence should be that autonomy, used in democratic procedures, requires. To answer this question, we must say a bit more about the epistemic standards of public reason, within which autonomous self-expression plays such a crucial role. This is a complex subject so our discussion will need to remain quite general.10 First, in order for public justification to proceed in a way consistent with the endorsement constraint, we must assume at least a modest “internalism” as our epistemic standard of justification at the individual level.11 What this means is that no value claim can be said to be valid for a person (or no belief about such a claim or its components) unless there is an inferential relation between such a claim and other elements of that person’s belief/value corpus. Pure externalism would deny this and claim that some beliefs are justified for a person wholly independent (in principle) of that person’s belief set. But the endorsement constraint implies that, ideally at least, a person could come to embrace (or at least not be deeply alienated from) the value in question. This is not possible unless there is a hermeneutic or otherwise inferential relation between that value and values or beliefs the person already holds.12 Second, a person must have a level of understanding of her own psyche so that she is a relatively consistent representative of a viewpoint. If manifest inconsistencies arise from or are involved already in her corpus of desires 10

11 12

I rely greatly on the detailed analysis of “public justification” and its role in political legitimacy developed by Gerald Gaus (1996). In addition, however, a line of analysis similar to this is developed by Habermas (1996a). For example, he writes: “To the extent that action coordination, and with it the formation of networks of interaction, takes place through processes of reaching understanding, intersubjectively shared convictions form the medium of social integration. Actors are convinced of what they understand and consider valid. This is why beliefs that have become problematic can be supported or revised only through reasons. Reasons, however, are not adequately described as dispositions to have opinions; rather they are the currency used in a discursive exchange that redeems criticizable validity claims” (ibid., 35). For discussion of the endorsement constraint, see, for example, Dworkin 2000, 17–18, Gaus 2009, and Kymlicka 1989, 12–13. Assuming some qualified internalism for the purposes of political philosophy is not the same as claiming this as the best epistemic account, full stop. Also, what is meant by “hermeneutic” here is that a coherent interpretation could be applied to the belief (or value set) which includes the contested element. For a critical discussion of such an assumption of internalism, see Wall, 2009.

228

The Politics of Persons

and values, then the process of deliberation and negotiation cannot fruitfully proceed. So absence of manifest inconsistencies – where fully contradictory beliefs or values are held in ways that could bring them easily to mind – is a necessary part of autonomy competence. But this is compatible, it must be stressed, with sincere ambivalence and measured changes of mind. I meet this requirement even if I am torn in two directions on an issue or if I alter my view in light of new information and deliberation itself. But a person who is notably pulled by inconsistent desires in ways she does not admit – acting on or expressing one at one moment and doing the opposite the next – is not a competent deliberator and hence not autonomous in the requisite sense. The endorsement constraint continues to operate here. For it implies that subjective embracing of a value is a necessary component of its validity (for a person) in the context of political principles. So a person’s act of embracing a view, or embracing a view as part of a process of publicly expressing it in the dynamics of public deliberation, makes it her own in this crucial sense. Even if I am somewhat out of touch with my motives or systematically mistaken about the psychological sources of my opinions and values, I commit myself to them as I advance them to others in public discourse. I therefore “construct” myself (in part) by committing to this or that belief. At least I construct and commit myself provisionally in that I am open to reasons from others and, as a sincere and non-strategic communicator, I listen to others in ways that may lead me to reconsider my own views. But as a participant in this process, I commit to views I judge to be right by expressing them, not (or not always) by simply discovering them as a settled aspect of my nexus of other beliefs, desires, and values. Returning to the broad outline of autonomy-based conceptions of democratic justice, let us take a moment to make clear why the view being developed here is so deeply pluralistic. Specifically, we should be clear about the grounds we have for rejecting perfectionism in politics, a view that has powerful defenders. An aside on that issue is therefore necessary before we can proceed. The rejection of perfectionism Many view political norms as requiring foundations that refer to objectively valuable ends, values, and principles whose validity does not in principle depend on their being accepted by people to whom they apply. The view that justified political institutions are guided by such objective standards is called “perfectionism.” Stephen Wall defines perfectionism as the view that

Justice over time

229

“political authorities should take an active role in creating and maintaining social conditions that best enable their subjects to lead valuable and worthwhile lives” (Wall 1998, 22). This presupposes that there are objective conceptions of what it means to lead a worthwhile life.13 There are many approaches to the justification of perfectionism of this sort. Most often this question is raised concerning the status of liberalism, specifically whether one should think of liberal principles as resting on universal and objective foundations or as justified by reference to the general political acceptance of its principles. While I have not claimed to be working strictly within the liberal framework in this work, holding that terminological question at bay for the moment, let us consider the merits of perfectionism within liberal theory in order to draw conclusions that apply outside of it as well.14 Theorists in recent decades have become divided between “perfectionist” and “political” approaches to the justification of principles. The former view rests on the contention that the justification of political principles (of justice for example) rests on particular values, such as autonomy, tolerance, equality, liberty, and so on that are seen as objectively valid, independent of their being endorsed by those who live under them. Political approaches, on the other hand, reject the assumption that there are universal, objective values that apply to people independent of their own viewpoint and therefore establish the legitimacy of political institutions by virtue of procedural or political means. Rawls’s conception of political liberalism exemplifies this latter view, and rests on the acceptance of deep divisions among reasonable citizens concerning the validity of comprehensive moral perspectives (see also Gray 1993 and Larmore 1987). Political liberalism, then, rests fundamentally on a principle of legitimacy, which can be stated in this way: “in a closed society of free and equal citizens maintaining diverse moral views, political power is legitimate only when such citizens may reasonably be expected to endorse it” (Rawls 1993, 217). This is a procedural account of legitimacy, according to which political power is acceptable because of its grounding pedigree, not its content. On the other hand, perfectionist theorists insist that political power cannot be justified in such a merely proceduralist or political manner. Steven Wall, to repeat, argues from a perfectionist standpoint that political 13 14

See Galston 2002. For an overview of the issue, see Charles Larmore, “Public Reason,” in Freeman 2003, 368–393. Cf. also Hurka 1993. For more detailed discussions of the relation between autonomy and perfectionist liberalism see Christman 2005 from which the following section is drawn.

230

The Politics of Persons

authorities should take an active role in creating and maintaining social conditions that best enable their subjects to lead valuable and worthwhile lives. The criteria for “valuable” and “worthwhile” here, in addition, do not include a requirement of acceptance by those forced to live under them; rather they are specified by abstract philosophical argument establishing their validity. Political conceptions accept that some moral ideals may be true, and that some members of the population hold such ideals, yet such commitment may have to be bracketed in order to achieve a procedural or political compromise with co-citizens who reject such ideals. This, perfectionists argue, has the paradoxical implication that public political principles must be accepted by people in ways that conflict with the manner in which they embrace their own moral values. Even when citizens accept values central to this political “compromise” they will have to do so for reasons other than those they hold to support their own view (because these latter will be held to be universal and objective, but that kind of philosophical support is not allowed in the public reasoning establishing legitimacy on the political approach). One preliminary point: there is an important difference between the status of people’s moral commitments and that of their political views, at least in the respect that principles that apply to social institutions, interpersonal relations, and other aspects of public culture necessarily contain provisions that aim to constrain, coerce, modify behavior, redefine social roles, and force compliance. The reasons I might have for believing in (for example) a set of religious teachings take on a special character when those teachings include requirements for the forceful constraint of other co-citizens. Traditional liberalism, which the proceduralism I defend here resembles in this regard, includes a presumption that power is always suspect, and that only when it can be justified must it be tolerated. That is, no reason, however valid for an individual, can count as a reason for others to be constrained or guided by it unless it can be a reason for them. Intersubjective validity of the sort envisioned is necessary for this (cf. Gaus forthcoming). The motivation for this move is that without this condition of legitimacy, the application and enforcement of principles is a function merely of power – the force of those who happen to have superior ability to enforce their view of the good. Whether or not such a view of the good is valid as a moral claim, its enforcement can only be successfully carried out if there are sufficient numbers or sufficient might available to enable such enforcement. Therefore, social enforcement of principles is the result of pure power unless

Justice over time

231

those affected by the principle can, in some fashion, accept the reasons for it. This is a view of social life that is fundamentally modernist and emerges first (in this manner) in Rousseau and later in Kant in a more purified form.15 Consider the person who reflectively rejects the objective standard that lies behind (let us assume) a valid perfectionist principle. Political policies that can only be justified with reference to such a principle would therefore have to be imposed upon this person without such justification counting as a reason for her. That is, even if the value upon which this imposition is based is valid (externally), its enforcement would be enacted without recourse to a justification that the subject of that enforcement could access. (Recall that the moderate externalism to which perfectionist politics is committed implies that values can be valid even if the person to whom they apply could not in principle accept them.) Whether or not this person obeys the enforced imposition of this principle will depend on whether the power lying behind that enforcement is effective, if there is enough might to make it right. But relying on the superior force of one side in an argument is not sufficient for justifying the enactment of the implications of that argument. Moreover, perfectionism rests on the contention that ideals for individual and social lives can be specified in objective terms, independent of the particular social geography and history of the people to which they apply. The contention is not only that such ideals can be determined and validated but that they can provide sufficient reason for others to comply with forceful constraints (coercively enforced legal rules) despite not being convinced of that validity and not being convinced for understandable reasons. A powerful response to this view of value, though, is that what will count as a worthwhile pursuit or ideal for a person is highly “path-dependent” in that the particular individual and social histories from which people arise and develop will very much shape what is valuable for them. Moreover, as we emphasized in our discussion of memory and the narrative structure of the self, the person’s own practice of interpreting that path or history is central to constructing her current and ongoing identity. So while certain ideals might be put forward as valuable objectively and defended as such, whether a person or group could take on such ideals as their own would depend very much on the variable social geography from which they emerge and which they themselves interpret. But perhaps the most forceful set of reasons for rejecting such a position relates to what Rawls calls the burdens of judgment. As part of the 15

For an interesting discussion of Kant’s views relating to this point, see Waldron 1999, 36–62.

232

The Politics of Persons

requirement of toleration and reciprocity that he thinks members of modern constitutional democracies generally accept as part of their public political culture, the burdens of judgment accept that others’ moral outlooks can differ from ours but that, even if we view our values as sound or true, we can accept how reasonable people could come to reject them (see Rawls 1993, 54–58). Various factors lead us to conclude this, for example the recognition that evidence for various moral views is mixed and assessed according to different criteria of relevance and weight, our concepts are vague and open to conflicting interpretations, and that our course of life up to this point (as well as the history of the social settings we call our own) bears crucially on the values we come to see as valid and such things differ radically for different people. In Rawls’s view, these observations are part of the epistemological and sociological structure of our and others’ moral commitments. To claim, as perfectionists do, that objectively determined ideals can be validated and forcefully imposed flies in the face of these observations. Now defenders of perfectionism have critically discussed this position, specifically claiming for example that such a view leads to skepticism about values generally (see, for eample, Wall 1998, 91–100). In other words, it is inconsistent to claim that someone can justifiably think that a certain value claim is true (as Rawls does) but then argue that others should not be bound by it because they have an alternative reasonable view and don’t see the truth. Unless I am really a skeptic about the truth claim for my own view, I have no reason not to see it as valid for others (similarly situated relative to the value claim in question) (Wall 1998, 92). But those who reject political perfectionism along the lines I do here need not be skeptics about the truth claims she makes for the values she pursues; she need only be a certain kind of fallibilist. That is, in discussions of the good – life pursuits and ideals that guide a life – our perspectives will always be heavily shaped by our social location and history. That is not to say that we need reduce all judgments of the good to that location – this is not a defense of individual relativism – but rather the weaker claim that we will not be able to tell how much our judgment about the worth of a certain pursuit is shaded by that history, especially in the face of other reasonable people who disagree with us.16 Another powerful line of critique, however, is that the reasons one has to reject the possibility of consensus on objective values in a pluralistic society 16

This is admittedly an incomplete response and concerns highly complex issues. For further discussion, see Wall forthcoming.

Justice over time

233

will apply equally to the possibility of a consensus on principles of justice. If considerations such as the burdens of judgment or some other set of assumptions about the deep pluralism of moral and political values leads one to be skeptical that even objectively sound values could ever be seen as valid by radically different groups in a complex modern society, then what grounds are there to think the alternative criterion of political legitimacy – some set of acceptable procedures or an overlapping consensus – could also be achieved? My answer to this challenge, as much as I can offer one, will be borne out in the conception of procedural legitimacy I will lay out in the next section. My point is not to give a final defense of this approach but rather to sketch in as compelling a manner as I can a conception of democratic justice that illustrates the social value of autonomy as I conceive it. Let us, then, proceed to the question of legitimacy. iii. autonomy, legitimacy, and political liberalism So we reject perfectionism in politics, where that is taken to mean that normative political principles and values are justified independently of whether or not those to which those principles apply can or would accept such principles and values as valid. As we said, this last requirement is the endorsement constraint. The discussion of this constraint has taken place most trenchantly in debates over the foundations of liberalism and more specifically about whether the Rawlsian project of political liberalism can succeed. The latter is a version of anti-perfectionist theory to which the framework I am guided by here owes much. Hence, it will be helpful to look more closely at Rawls’s framework in order to draw out some lessons about the role of (the value of) autonomy in principles of democratic justice. The project of political liberalism is to secure a basis for principles of justice that avoids claims of philosophical or metaphysical truth so as to take sufficiently into account the deep and abiding disagreements over such “truths” in the late modern landscape. “Legitimacy” replaces truth or full philosophical justification as the basis for sanctioning the coercive power that political institutions wield and that their underlying principles countenance. To gain such legitimacy, however, citizens must affirm those principles as well as the procedures and reasoning with which they are derived. “Justice as Fairness” is, of course, the name for the most prominent set of such principles and procedures, referring to the use of the “model conceptions” of the Original Position, primary goods, and the conception of the

234

The Politics of Persons

person in order to support the derivation of two principles of justice (the Basic Liberties principle guaranteeing basic rights and liberties and the Equality principle securing fair equality of opportunity and limitations of inequalities of primary goods specified by the Difference Principle). All this, then, becomes the subject of deliberation by citizens who see themselves as free and equal (in ways left to be explained) and who engage in debate about the constitutional essentials for political institutions (of the basic structure) that will operate under the principles in question. These deliberations – public reason – are aimed at securing, if possible, an overlapping consensus about Justice as Fairness despite the deep and contrasting overall moral orientations guiding the reasoning of each participant. Such a consensus, Rawls envisions, secures the legitimacy necessary to underwrite political power (Rawls 1993, 2001). Public reason, then, is the process of discussion and deliberation (chiefly) about constitutional essentials of a state which conforms to these abstract ideas (freedom, equality, reciprocity, and so on). Such deliberation aims at an overlapping consensus concerning the core principles of justice and grows out of each citizen’s own reasonable comprehensive doctrine. “Public reason” refers specifically to the constraints placed on such deliberation in this process of establishing legitimacy (via this consensus). In particular, public reason demands that citizens do not base arguments in defense of political principles on their particular comprehensive moral, religious, or philosophical views, insofar as others in the pluralistic social world will not share such views and hence are not plausibly bound by arguments based upon them. Therefore, liberal legitimacy is established through a process of deliberation that demands “epistemic abstinence” in the sense that deliberators are to bracket their public justifications so as to exclude direct reference to their particular moral viewpoint. Others following Rawls have claimed that basic institutions of justice reflect the autonomy of citizens in significant ways. In particular, theorists of deliberative democracy have stressed that the processes of public deliberation by which basic institutions are legitimized collectively embody the self-government that individual autonomy manifests. Joshua Cohen, for example, claims that “by requiring justification on terms acceptable to others, deliberative democracy provides for a form of political autonomy: that all who are governed by collective decisions – who are expected to govern their own conduct by those decisions – must find the bases of those decisions acceptable” (Cohen 1996, 416). One of the fundamental difficulties that critics raise here is that such a picture itself conflicts with the deep pluralism of moral vision and value

Justice over time

235

commitment that ultimately motivates it and which Rawls insists is a permanent feature of modern life.17 Pluralism runs deeper than this framework allows, critics claim, and hence liberalism fails to be a political compromise available to settle disputes among devotees of sectarian viewpoints and rather becomes one more such sectarian view itself. Recall, though, that on both Rawls’s view and the path followed here, autonomy is attributed to citizens not as a psychological fact but as part of the conception of the person and her “institutional identity” functioning in the process of legitimation of principles of justice. Rawls’s view assumes, that is, that citizens have rational and full autonomy in the sense that they are willing to accept a public conception of justice and that they have the capacity to formulate and revise a conception of the good based on a comprehensive moral doctrine. These “powers” are reflected in higher-order interests that must be reflected in the principles of justice that govern the basic structure as well as the procedures for the derivation of those principles. Such autonomous citizens are (and understand themselves to be) free and equal in that they are (and understand themselves to be) self-authenticating sources of valid claims and represented in procedures that do not privilege any particular moral viewpoint (such as in the Original Position). Critiques of the constraints imposed on deliberation by public reason take aim at this conception of the autonomous person (though they may not use the terminology explicitly). For instance, Stephen Wall argues that the “bracketing” that public reason demands assumes that one will always place greater weight on the political principles that must be shared with dissenting others than on one’s own comprehensive view (Wall 1998, 67–69). The crux of the difficulty is that public reason bars direct reliance on elements of one’s comprehensive view to justify principles under which one must live, yet that very view is what provides the grounds for one’s commitment to public principles, as well as the motivation to participate in those public deliberations (or support elected representatives who do so). Autonomy fails to be respected when one’s own governing principles are not only not reflected in the public reasons supporting social institutions but are positively barred from playing a role in the development and support of those reasons. It is not merely that one departs from the public principles in question, but one does not identify with the publicly allowable reasons that support those principles. 17

See, for example, Friedman 2000c; Gray 1993, 283–328; Larmore 1996, ch. 7; and Young 2000. For further discussion, see also Gaus 1996, 130–136. See also Young 1990b, ch. 4, though her focus is wider than political liberalism or Rawls specifically.

236

The Politics of Persons

There is no argument, then, short of invoking the very comprehensive moral and political “truth” that political liberalism forbids, to justify the ordering of values demanded by public reason, specifically the greater weight placed on political cooperation over an individual’s reasonable, indeed perhaps “sound” moral commitments. This line of criticism is powerful and well rehearsed. As I’ve framed it, it is especially challenging since it is framed as an “internal” critique showing that the commitment to autonomy of citizens as one of the fundamental assumptions of political liberalism conflicts with the constraints put on public reason in the process of legitimizing political principles. Indeed, Rawls’s conception of rational autonomy as embodying a sense of justice and a conception of the good runs afoul of those citizens whose self-governing strategies put relatively different weights on the commitments from their comprehensive conceptions of the good than political liberalism demands, placing, for instance, more importance on the pursuit of such comprehensive values than on the constraints of the sense of justice articulated in political liberalism. Such citizens can no longer see themselves as autonomous if coerced by institutions structured by principles they do not impose upon themselves. Let us attempt a response. First, Rawls makes clear that Justice as Fairness – the particular array of liberal principles and the procedures to derive and justify them (the Original Position for example) – is just one of a “family of political conceptions of justice” and not the only framework which could, through the operation of public reason, establish legitimacy. This family must contain, however, generic provisions reflective of the commitment to basic reciprocity by citizens, provisions which include a listing of basic rights and opportunities, with a commitment to their priority, and a set of measures ensuring all citizens with the means to make basic freedoms “effective.”18 Second, Rawls introduces what he calls a “proviso” into the account of public reason. This proviso allows the introduction of one’s comprehensive view into the substance of public deliberation as long as in doing so one helps to affirm the ideal of public reason itself.19 In what Rawls called the “wide view of public political culture,” it is permissible to make reference to one’s particular comprehensive view in participating in public reason provided that “in due course proper political reasons – and not reasons given solely by comprehensive doctrines – are presented that are sufficient to support whatever the comprehensive doctrines introduced are said to support.”20 18

See Rawls 1999c, 140–141.

19

Rawls 1999c, section 4.

20

Rawls 1999c, 152.

Justice over time

237

This proviso can be interpreted in several ways. The gloss on this expansion of public reason that I suggest here is that strictures on public deliberation – the prohibition on grounding the justification of a principle in one’s own sectarian view – are relaxed as long as arguments making reference to such views are consistent with ongoing democratic discussions which include alternative viewpoints and aim toward a public justification that is acceptable to all. In the case, for example, where one’s comprehensive value orientation dominates in the current round of discussion of some provision, one must never act (or argue) in a way that restricts participation and equal voice for those alternative views of political principles or alternative modes of moral justification for any given political principle. Alternatively, if one’s moral vision is not in the ascendancy during a period of public justification, one must always see ongoing opportunities to participate in the reconsideration of any dominant principle in continuing democratic debate. In this way, consequently, democratic deliberation is constitutive of legitimacy as an ongoing process of reflection and reconsideration, underscoring the fallibilist approach to basic principles that any political conception must adopt.21 It should be emphasized here that this view puts ongoing democratic processes at the center of the establishment of justice, construed here as legitimacy of the institutions of governmental power. Rather than seeing democracy as a secondary mechanism to determine the public good after more basic principles and rights are established (constitutionally, say), justice demands, on this view, that constant operation of free and fair democratic processes partly determines the content of these principles. (Below we will see, though, how basic individual rights of the sort typically named in constitutional provisions will be justified here as necessary presuppositions of effective democratic processes themselves.) Secondly, the historical approach to the person and autonomy taken in the conceptions of those notions defended here is mirrored in this conception of democracy. Along with the assumption that historical factors involving differential power and domination must be included in the conditions of democratic justice, we are here envisioning justice as an ongoing process of equal consideration of citizens’ interests. These are interests that will also reflect concerns for historically related aspects of 21

This interpretation brings into closer alignment political liberalism, with its basis in public reason, and theories of democratic legitimacy, which see ongoing political debate as constitutive of political legitimacy rather than simply an amendment to constitutional principles needed to formulate social policy. See, for example, Cohen 2002 and Gutmann 2003b.

238

The Politics of Persons

those citizens’ experiences, for example, in the recognition and possible rectification of past injustices. This arises out of the socio-historical conception of the self and the historical view of autonomy defended earlier. Yet the complaint that political liberalism is exclusionary and problematically utopian remains powerful, for one can still claim that the inability to publicly justify principles, and to demand that co-citizens share the public justification of principles, in moral language which describes their basic moral commitments, prevents the affirmation of those principles demanded by legitimacy. Further, the internal conflict concerning autonomy reappears insofar as autonomous citizens cannot see themselves as collectively selfgoverned when public justification of principles (even given open-ended reconsideration and ongoing debate) fails to resonate with their own moral horizons. As we noted, the fundamental moral motivations that spur participation and underwrite a willingness to comply with coercive principles will be in conflict with the dominant principles in a political culture constrained by public reason, and hence autonomous citizens will be unable to affirm such principles and legitimacy will be ephemeral. This line of critique, however, rests on the presumption that autonomy of citizens requires that they be able to “affirm” or “endorse” public principles in ways that resonate with the public lines of justification given in political forums. The self-contradictions of autonomy-based legitimacy alleged earlier rest on a certain idea of autonomy which connects with the demand for affirmation and endorsement. What I want to do now is to plug in the conception of autonomy defended in part II above into this debate in order to round out the picture of autonomy-based legitimacy we are developing and defend it against this charge of self-contradiction. Autonomy and public reason As we noted, political liberalism demands that citizens be able to affirm or endorse the core principles of justice and the public justifications given of them from within their own comprehensive doctrines and moral frameworks. Now Rawls actually gives different and possibly inconsistent accounts of the relation that must hold between people’s commitment to their own comprehensive doctrines and the principles of justice that are the object of the overlapping consensus (and so that which public reason is used to legitimize). At times he claims that citizens must “affirm” justice as fairness (Rawls 1993, 218), at others they “endorse” principles of justice (Rawls 1999c, 140), at others it is said merely that they “can affirm” those principles. Rawls also approvingly

Justice over time

239

cites Thomas Scanlon’s view that a principle is reasonable if it is such that others could not reasonably reject the principle.22 Public reason fails as an acceptable process of political legitimation, critics claim, because it restricts both reference to and public reliance upon comprehensive moral doctrines by citizen-participants in public debate, hence conflicting with the independence and integrity that such citizens are owed as free and equal persons. That is, this conflicts with viewing them as autonomous agents. But insofar as autonomy merely demands that we fail to be deeply alienated from conditions of our person and environment that shape our life choices, then the inability to publicly affirm principles of justice for reasons allowable under public reason need not conflict with respect for the autonomy of those committed to comprehensive views upon which they place great (even predominant) weight in the order of values. That is, the strictures of public reason will not conflict with autonomy insofar as those strictures still allow participation in public deliberation in ways that do not cause alienation in our sense in participating citizens. A better test for legitimacy, then, is that (reasonable) citizens do not actively reject the basic principles of justice in the sense indicated here: they do not feel alienated from them by viewing them purely as restrictions, imposed externally, and out of keeping with the rest of their value and motivational set.23 Moreover, this view captures the central affective component of the acceptance or rejection of the guiding values of public life that many emphasize the importance of. If I am deeply religious, and the basic principles of my society contain elements which explicitly denigrate my religion, my inability to accept those principles will rest on a reaction that is as much visceral as it is cognitive or calculating. To see legitimacy as nonalienation we capture both a cognitive calculation as well as an affective reaction (see, for example, Young 1990b, ch. 4) Even in such a case, my rejection cannot merely be a subjective, psychological reaction that no one else could understand. For such a response to ground the rejection of (otherwise) shared principles, it must involve judgments that I can expect others to grasp at some level, considerations that can be made effectively public. For this reason, I propose that the test for legitimacy of political principles should be the following: principles are legitimate only if the (reasonable) citizens to whom they apply would not be understandably alienated from them. To be “understandably” alienated 22 23

See, for example, Rawls 1993, 49 n. 2 and 124. This is indeed similar to Thomas Scanlon’s view that something is morally wrong/unjust if it cannot be justified to others on grounds they cannot reasonably reject. See Scanlon 1998.

240

The Politics of Persons

is to definitively reject the principles but in ways one can share with others, reasonably expect them to understand (though not necessarily accept), and which can be generalized to other cases. To be “alienated” from such principles is (as in the case of autonomy) to vehemently reject them, resist their motivating force, and actively disavow them. This is a weaker condition than that one fails to “affirm” such justifications or principles. The implications of this weaker standard are notable. In at least two kinds of cases, constraints of public reason guided by this sense of autonomy would allow legitimacy where the stronger requirement (of endorsement or affirmation) would not. The first are cases where people are relatively indifferent between competing public principles (perhaps one of which conforms better with their comprehensive view). As Rawls points out, most citizens do not have fully worked out comprehensive moral doctrines, and hence they have not traced out the various implications of their views.24 Now such citizens may not be in a position to affirm public principles, since they are not given explanations of how those principles connect with the aspects of their own moral commitments that they have reflected upon. However, such citizens remain autonomous since they are not (we assume) deeply alienated from such principles. Secondly, and more in tune with the lines of criticism laid out earlier, those who could accept the public principles themselves but reject the reasoning given in public forums for them (since such reasoning fails to make reference to the fundamental values that guides their very participation), will often not feel deeply alienated from the principles themselves, in the sense defined. That is, the person described previously as having a commitment to toleration but for specifically religious grounds could not perhaps affirm the secular justification of toleration given in political liberalism. Yet, she may well not be alienated from such a principle as long as she can provide her own reasons for support of it. This picture of citizens accepting the content of political principles but not their full justification is touched on by Rawls himself but the full picture of legitimacy sketched in political liberalism demands that affirmation be made of specific principles as well as the mode of justification (the Original Position, and the like) used to derive them.25 Therefore, the range of people that could rightfully object 24 25

See, for example, Rawls 2001, 33. Rawls claims that “[i]t is left to citizens individually to decide for themselves in what way their shared political conception is related to their more comprehensive views” (Rawls 2001, 186). But he also says that a “liberal conception of political legitimacy aims for a public basis of justification and appeals to free public reason, and hence to citizens viewed as reasonable and rational” (ibid., 186), implying that the public support of political principles is fully non-sectarian.

Justice over time

241

to the constraints of public reason on the grounds that these conflict with the motivations by which they participate in public discourse and feel obligated by its results – and hence are in conflict with their autonomy as citizens – is severely reduced. Moreover, and importantly, we must recall the expanded conception of public reason mentioned earlier and in particular my specific gloss on it. To object to the constraints placed on public deliberation one must be able to claim that one’s particular moral viewpoint is not sufficiently operative in public discussions (though given the “wide” view of public reason, reference can be made to such viewpoints), and one must also claim that further, ongoing deliberation will not produce a result which pays sufficient homage to that viewpoint to the extent that one feels alienated from the dominant principle as well as the democratic processes that are open to one to demand reconsideration of it. One must not only reject the public principles dominant in one’s political culture (despite the fact that one can provide one’s own independent reasons for such principles), one must also reject and feel alienated from the process of ongoing deliberation used to reconsider and challenge those principles. Legitimacy fails, on this account, when populations reasonably lose faith both in the normative structure of political institutions as well as in the promise of democracy itself. This weakening of the conception of autonomy and, in turn, the requirements of legitimacy does much, I hope, to take more seriously the deep abiding dissensus that characterizes modern political life than did political liberalism in its earlier guises (see, for example, Dagger 2005 and Brink 2005). In keeping with the conception of autonomy and the self in which factors that conflict and escape the agent’s control are accepted as part of the complex constitution of the person, justice is envisioned here as an ongoing process of collective social management of ongoing conflicts, many of them at the deepest levels and with no foreseeable end. As long as the conditions of non-alienation of the sort described here obtain – frustration of attempts to reject the results and processes of democratic procedures – citizens’ equal status as autonomous beings is respected. I have used the defense and expansion of Rawlsian political liberalism as a means to fill out the picture of democratic justice into which our conception of autonomy and the citizen plays a crucial role. Also, that role illustrates the fundamental value of autonomy in that it is a central element in the operation and legitimacy of just political practices. The defense of Rawls was also meant to show the strong connection between justice and democracy in that the latter is necessary to secure the ongoing legitimacy of

242

The Politics of Persons

political institutions and such legitimacy is central to justice in the political sense envisioned here. Two issues, however, must also be addressed to further illustrate both the place of autonomy in such mechanisms of justice and the way such mechanisms presuppose and help advance the value of individual autonomy. iv. democracy, autonomy, and basic capabilities The value of autonomy resides in its place in a complex of other values all of which are organized in a broad framework of democracy.26 As such, I am not claiming that autonomy has objective, intrinsic value outside of this context, as a personal ideal, say, or a valuable way of life no matter where one lives or what one’s broader moral perspective is. Though I should add, I am also not denying this – it is simply a different issue. But given that approach, it might be thought that constitutional protections of basic rights and/or guarantees of basic resources such as housing, healthcare, education and nutrition, are not afforded independent justification (as they would be, for example, in certain versions of liberalism). So I should say a brief word about how and why I think such guarantees are indeed supported, as a claim of basic justice, in the democratic framework I am sketching here. First, as I argued in chapter 7, autonomy requires basic competence in one’s ability to reflect critically and to make one’s desires effective under favorable conditions. Insofar, then, as autonomy is required by a functioning democracy (and justice requires such a democracy), then justice requires that all citizens enjoy these basic competences. I have not gone into detail about what the complete list of basic competences are that autonomy includes, nor have I described the gamut of resources that a person needs in order to develop and maintain those resources. However, we can lay out some broad categories (ones that would need to be filled in in any particular social setting, with its own history, to which these principles apply). Certainly the ability to choose competently and reflect all require adequate nutrition, housing, education, health resources, a relatively safe home and neighborhood environment, and other basic necessities of 26

In this way, I am following the lead of various democratic theorists (see, for example, Cohen 1986, 1996 and Michelman 1988). This view also reflects Habermas’s approach, where he argues that “public autonomy” (meaning collective self-government or democracy) and “private autonomy” (meaning individual personal and political rights) are “co-original” – they are both justified together as a general mode of establishing just institutions (Habermas 1996a).

Justice over time

243

human development.27 The ability to authentically reflect on one’s personal life and the historical components of one’s condition would also require access to substantial educational and informational resources. One’s ability to adequately reflect on one’s values requires an ability to consider alternatives (even if one remains fully enmeshed in the traditional values, say, with which one was raised). Educational policies and curricula must certainly be guided by these considerations, supporting the need for universal basic education independent of income and social status. (For discussion see Brighouse 2000, ch. 1 and Levinson 2002, ch. 2.) The ability to autonomously participate in politics (or to guide one’s representatives) also demands basic resources and rights. In the latter category would surely be included free speech and press, association (which would include religious association), and access to political resources. Rights against violence and oppression and equality of opportunity (if not substantive equality of resources) will also be required. Basic principles of the sort defended by Rawls – basic rights and liberties, equality of opportunity, and the difference principle – would be an example of a set of requirements for the legitimacy of institutions established in the manner I lay out (though nothing I say here commits me to that particular set). The general point is that the fundamental principles of justice would be those that would be needed for a functioning democracy of the sort described to operate, a democracy where people’s interests are expressed and defined in their own voices (or voiced through effective representation) in a manner that protects and promotes their autonomy. We can also note further implications of the socio-historical conception of selves at work in the model of autonomy functioning here. As we discussed in detail above (and particularly in chapter 9), persons whose autonomy is being respected here are defined with reference to their connections with identity groups as well as particular histories associated with such groups. In addition, seeing the circumstances of justice as including acknowledgment of domination, power inequality, and violence, supports the idea that many individuals will have fundamental interests relating to those factors, for example in knowing about (and making public) their 27

One can use as a guide here Amartya Sen’s conception of basic capabilities needed to achieve valued functionings, which he takes to be a defensible measure of human social well being. On Sen’s view, our lives involve a combination of various “doings and beings” and our ability to achieve these valued states can be called our functioning. Combinations of various valued functionings comprise our “capabilities.” These range from the very basic, such as being adequately nourished and avoiding premature mortality and morbidity, to the more complex such as self-respect and social integration (Sen 1993, 31; see also Sen 1992). For a discussion of autonomy and poverty-related welfare goods, see Christman 1998.

244

The Politics of Persons

details, seeking recognition and perhaps redress for past victimization, and representing themselves as part of the social trajectory involving such phenomena. As we concluded in the last chapter, though, these interests will involve the ability to represent ourselves in the collective understanding of these interests, perhaps by way of special representation in deliberative bodies. In the context of the current discussion, this implies that social resources, for example educational practices as well as public discussion and deliberation, may need to be tailored to meet these needs.28 This is unfortunately all terribly brief. But the general contours of the policy implications of seeing autonomy as the central value of democracy should be clear. Along these lines I hope to have shown how the value of autonomy relies on the value of a procedural democracy that establishes and maintains legitimate political institutions for a pluralistic populace. Such a populace not only has deep and abiding conflicts about value and morality, they also understand themselves as products of a complex social history which includes persistent power differentials, victimization, and violence. 28

The point in the text is stated quite abstractly, so as to leave room for social variation and the particulars of concrete cases. What I have in mind here, though, is exemplified by practices of public memory (memorials, celebrations, public discussions) as well as, in some cases, efforts at dealing with past injustice that take the form of truth and reconciliation committees. For a discussion of public memory and its political role see Young 1993; for discussion of the latter see Rotberg and Thompson 1993.

Conclusion

This has been a wide-ranging study and many of the points examined were surely glossed over too quickly. The general thrust of the analysis has been to lay out and defend a model of the self that avoids the pitfalls of standard conceptions of the political person, specifically those that inherit the narrow liberal individualism that has dominated political theory since the Enlightenment. My broad goal was to take on as many of the critical observations about selves and social interaction pointed to by critics of that (liberal, Enlightenment) tradition, but to then reconstruct a model of the political self that can play a role in a conception of autonomy that itself is theoretically and politically viable. The rejection of autonomy-based conceptions of justice and democracy from many of these critics was based on the idea that such a view could only rest on a problematically narrow understanding of the person, often reflecting privileged social positions and identities. In many cases, the accusations were correct, in that many traditional models of the citizen and his interests reflected that privileged status. My goal here, however, was to explore as much as possible a conception of the autonomous person that absorbed as much of what was plausible in those critiques but keep alive the view of justice that rested on the ideals of individual and collective self-government. In many ways the model that results is highly deflationary, in that it admits that selves are often opaque, internally confused, embedded in unchangeable social situations, and living against the backdrop of injustice and oppression. Autonomy, in this context can, at best, mean a kind of selfmanagement, a minimal level of self- and social control that allows us to make sense of our existence, even if we cannot celebrate it fully. Sometimes we have full control over the details of our lives – we have full say over what happens and full command over our reasoning in deciding what to do. But at many other times and for many of us we are merely coping, we manage as best we can the conflicting and baffling aspects of our existence; we muddle along in a complex negotiation with the myriad forces that constitute our 245

246

The Politics of Persons

condition. In these latter kinds of case, as long as we can say competently and authentically that this muddling makes for a life narrative that we can live with, not necessarily contentedly because we may not currently be content, but in a way that is (alas) truly us, then we are not alienated, in my sense, from this difficult life. It is ours, for what it’s worth. One of the important implications of the model of autonomy I have defended here is that in such cases, we are still autonomous despite being pulled and pushed by the forces of such a life. But as a final answer to the question with which we began – the question of who you are – the conception of autonomy I defend here and the political order it supports implies that it is, indeed, you who should get the last word.

Bibliography

Alcoff, L. (2006) Visible Identities: Race, Gender, and the Self. New York: Oxford University Press. Allen, R. T. (1982) “Rational Autonomy: The Destruction of Freedom.” Journal of Philosophy of Education 16: 199–207. Allport, G. (1968a) “Is the Concept of Self Necessary?” in The Self in Social Interaction Vol. I: Classical and Contemporary Perspectives. C. Gordon and K. J. Gergen, eds. New York: John Wiley, Inc., 25–32. (1968b) The Person in Psychology. Boston: Beacon Press. Althusser, L. (1969/1996) For Marx. London: Verso. Altieri, C. (1994) Subjective Agency: A Theory of First-Person Expressivity and Its Social Implications. Oxford: Blackwell Publishers. Anderson, E. (1993) Values in Ethics and Economics. Cambridge, MA: Harvard University Press. (1999) “What Is the Point of Equality?” Ethics 109(2): 287–337. Anderson, J. (1996) “The Personal Lives of Strong Evaluators: Identity, Pluralism and Ontology in Charles Taylor’s Value Theory.” Constellations 3(1): 17–38. (2003) “Autonomy and the Authority of Personal Commitments: From Internal Coherence to Social Normativity.” Philosophical Explorations 6(2): 90–108. Anderson, J. and A. Honneth (2005) “Autonomy, Vulnerability, Recognition, and Justice,” in J. Christman and J. Anderson, eds., 127–149. Anthony, L. and C. Witt, eds. (1993) A Mind of One’s Own: Feminist Essays on Reason and Objectivity. Boulder, CO: Westview Press. Antony, L. (1995) “Is Psychological Individualism a Piece of Ideology?” Hypatia 10: 157–174. Anzaldúa, G. (1987) Borderlands/La Frontera: The New Mestiza. San Francisco, CA: Aunt Lute Books. Appiah, K. A. (1992) In My Father’s House: Africa in the Philosophy of Culture. New York: Praeger. (1994) “Identity, Authenticity, Survival: Multicultural Societies and Social Reproduction,” in Multiculturalism: Examining the Politics of Recognition. A. Gutman, ed. Princeton, NJ: Princeton University Press, 149–64. (2005) The Ethics of Identity. Princeton, NJ: Princeton University Press. Appiah, K. A. and A. Gutmann (1996) Color Conscious: The Political Morality of Race. Princeton, NJ: Princeton University Press. 247

248

Bibliography

Applebee, A. (1978) The Child’s Concept of a Story. Chicago, IL: University of Chicago Press. Arato, A. and E. Gebhardt, eds. (1982) The Essential Frankfurt School Reader. New York: Continuum. Arbib, M. A. (1986) In Search of the Person: Philosophical Explorations in Cognitive Science. Amherst, MA: University of Massachusetts Press. Aristotle (1958) The Politics. Oxford: Oxford University Press. Arneson, R. (1980) “Mill Versus Paternalism.” Ethics 90(4): 470–489. (1985) “Freedom and Desire.” Canadian Journal of Philosophy 15(3): 425–448. (1991) “Autonomy and Preference Formation,” in In Harm’s Way: Essays in Honor of Joel Feinberg. J. Coleman and A. Buchanan, eds. Cambridge: Cambridge University Press, 42–73. (2000) “Perfectionism and Politics.” Ethics 111(1): 37–63. Arpaly, N. (2003) Unprincipled Virtue: An Inquiry into Moral Agency. Oxford: Oxford University Press. Arpaly, N. and T. Schroeder (1999) “Alienation and Externality.” Canadian Journal of Philosophy 29: 371–388. Ashmore, R. and L. Jussim, eds. (1997) The Self and Identity: Fundamental Issues. New York: Oxford University Press. Atkins, K. (2004) “Narrative Identity, Practical Identity and Ethical Subjectivity.” Continental Philosophy Review 37: 341–366. (2008) Narrative Identity and Moral Identity: A Practical Perspective. New York: Routledge. Audi, R. (1991) “Responsible Action and Virtuous Character.” Ethics 101(2) (January): 304–321. Austin, J. L. (1976) How to Do Things with Words. Oxford: Oxford University Press. Avineri, S. and A. de-Shalit, eds. (1992) Communitarianism and Individualism. New York: Oxford University Press. Avnon, D. and A. de-Shalit, eds. (1999) Liberalism and Its Practice. New York: Routledge. Babad, E. Y., M. Birnbaum, and K. D. Benne (1983) The Social Self: Group Influences on Personal Identity. Beverly Hills, CA: Sage Publications. Babbitt, S. (1996) Impossible Dreams: Rationality, Integrity, and Moral Imagination. Boulder, CO: Westview Press. Badhwar, N. (1996) “Moral Agency, Commitment, and Impartiality,” in E. F. Paul, F. D. Miller, and J. Paul, eds. Cambridge: Cambridge University Press, 1–26. Baert, P. (1992) Time, Self, and Social Being: Temporality within a Sociological Context. Brookfield, VT: Avebury. Baier, A. (1985) “Cartesian Persons,” in Postures of the Mind: Essays on Mind and Morals. Minneapolis, MN: University of Minnesota Press. Bandura, A. (1982) “The Self and Mechanisms of Agency,” in Psychological Perspectives on the Self. J. Suls, ed. Hillsdale, NJ: Lawrence Erlbaum, 3–40. (1986) Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall.

Bibliography

249

Barber, B. (1995) Jihad vs. McWorld: How the Planet is Both Falling Apart and Coming Together and What This Means for Democracy. New York: Ballantine Books. Barclay, C. R. and T. S. Smith (1992) “Autobiographical Remembering: Creating Personal Culture,” in M. A. Conway, D. C. Rubin, H. Spinnler, and W. A. Wagenaar, eds., 75–97. Barclay, L. (2000) “Autonomy and the Social Self,” in C. Mackenzie and N. Stoljar, eds. 52–71. Barry, B. (1973) The Liberal Theory of Justice. Oxford: Oxford University Press. (1989) Theories of Justice. Berkeley, CA: University of California Press. (2001) Culture and Equality. Cambridge, MA: Harvard University Press. Barthes, R. (1982) “Introduction to the Structural Analysis of Narratives,” in A Barthes Reader. S. Sontag, ed. New York: Hill and Wang: 251–295. Bartky, S. (1990) Femininity and Domination: Studies in the Phenomenology of Oppression. New York: Routledge. (1997) “Foucault, Femininity, and the Modernization of Patriarchal Power,” in D. T. Meyers, ed. (1997b). London: Routledge, 92–111. Bartlett, F. (1932) Remembering: A Study in Experimental and Social Psychology. Cambridge: Cambridge University Press. Baumeister, R. F. (1986) Identity: Cultural Change and the Struggle for Self. New York: Oxford University Press. (1998) “The Self,” in Handbook of Social Psychology. Vol. I. D. T. Gilbert, S. T. Fiske, and G. Lindzey, eds. Boston, MA: McGraw-Hill, 680–740. Baxley, A. M. (2003) “Autocracy and Autonomy.” Kant Studien 94(1): 1–23. Beardsley, E. (1971) “Privacy: Autonomy and Selective Disclosure.” Nomos VIII: Privacy. R. Pennock and J. Chapman, eds. New York: Atherton Press, 56–70. Beiner, R. (1997) Philosophy in a Time of Lost Spirit: Essays on Contemporary Theory. Toronto: University of Toronto Press. Belenky, M. F., B. M. Clinchy, N. R. Goldberger, and J. M. Tarule (1986) Women’s Ways of Knowing: The Development of Self, Voice, and Mind. New York: Basic Books. Bell, D. (1993) Communitarianism and Its Critics. Oxford: Clarendon. Bellah, R. N., R. Madsen, W. M. Sullivan, A. Swidler, and S. M. Tipton (1985) Habits of the Heart: Individualism and Commitment in American Life. Berkeley, CA: University of California Press. Benhabib, S. (1987) “The Generalized and the Concrete Other: The Kohlberg–Giligan Controversy and Feminist Theory,” in Feminism as Critique. S. Benhabib and D. Cornell, eds. Minneapolis, MN: University of Minnesota Press, 77–95. (1992) Situating the Self: Gender, Community and Postmodernism in Contemporary Ethics. New York: Routledge. (ed.) (1996a) Democracy and Difference. Princeton, NJ: Princeton University Press. (1996b) “Toward a Deliberative Model of Democratic Legitimacy,” in Democracy and Difference. S. Benhabib, ed. Princeton, NJ: Princeton University Press, 67–94.

250

Bibliography

(2002) The Claims of Culture: Equality and Diversity in the Global Era. Princeton, NJ: Princeton University Press. Benjamin, J. (1988) The Bounds of Love. New York: Pantheon Books. Benn, S. I. (1967) “Freedom and Persuasion.” Australasian Journal of Philosophy 45: 259–275. (1976) “Freedom, Autonomy and the Concept of a Person.” Proceedings of the Aristotelian Society 66: 109–130. (1982) “Individuality, Community and Autonomy,” in Community. E. Kamenka, ed. London: Edward Arnold Publishers. (1988) A Theory of Freedom. Cambridge: Cambridge University Press. Benn, S. I. and W. L. Weinstein (1971) “Being Free to Act and Being a Free Man.” Mind 80: 194–211. Benson, J. (1983) “Who Is the Autonomous Man?” Philosophy 58: 5–17. Benson, P. (1987) “Freedom and Value.” Journal of Philosophy 84(9): 465–87. (1990) “Feminist Second Thoughts About Free Agency.” Hypatia 5(3): 47–64. (1994) “Autonomy and Self-Worth.” Journal of Philosophy 91(12): 650–668. (2000) “Feeling Crazy: Self-Worth and the Social Character of Responsibility,” in C. Mackenzie and N. Stoljar, eds., 72–93. (2005a) “Feminist Intuitions and the Normative Substance of Autonomy,” in J. S. Taylor, ed., 124–142. (2005b) “Taking Ownership: Authority and Voice in Autonomous Agency,” in J. Christman and J. Anderson, eds., 101–126. Berlin, I. (1969) “Two Concepts of Liberty,” in Four Essays on Liberty. Oxford: Oxford University Press, 118–172. Bermúdez, J. (1997) “Reduction and the Self.” Journal of Consciousness Studies 4(5/6): 458–466. (1998) The Paradox of Self-Consciousness (Representation and Mind). Cambridge, MA: MIT Press. Bermúdez, J., A. Marcel, and N. Eilan, eds. (1995) The Body and The Self. Cambridge, MA: MIT Press. Bernstein, M. (1983) “Socialization and Autonomy.” Mind 92: 120–123. Berofsky, B. (1995) Liberation from Self. Cambridge: Cambridge University Press. (2003) “Identification, the Self, and Autonomy.” Social Philosophy and Policy 20(2): 199–220. Bittner, R. (2002) “Autonomy, and Then.” Philosophical Explorations 5(3): 217–228. Bluck, S. (2003) “Autobiographical Memory: Exploring its Functions in Everyday Life.” Memory 11(2): 113–123. Bogue, R. and M. I. Spariosu (1994) The Play of the Self. Albany, NY: SUNY Press. Bordo, S. (1993) Unbearable Weight: Feminism, Western Culture, and the Body. Berkeley, CA: University of California Press. Bracken, B. A. ed. (1996) Handbook of Self-Concept: Developmental, Social, and Clinical Considerations. New York: John Wiley. Bransen, J. (1998) “True to Ourselves.” International Journal of Philosophical Studies 6(1): 67–85.

Bibliography

251

Bratman, M. E. (1987) Intention, Plans and Practical Reason. Cambridge, MA: Harvard University Press. (1999) Faces of Intention: Selected Essays on Intention and Agency. Cambridge: Cambridge University Press. (2003) “Autonomy and Hierarchy.” Social Philosophy and Policy 20(2): 156–176. (2007) Structures of Agency: Essays. New York: Oxford University Press. Brewer, W. (1986) “What is Autobiographical Memory?” in D. Rubin, ed., 25–49. Brighouse, H. (2000) School Choice and Social Justice. Oxford: Oxford University Press. Brink, B. van den (2005) “Liberalism without Agreement: Political Autonomy and Agonistic Citizenship,” in J. Christman and J. Anderson, eds., 245–271. Brison, S. (1997) “Outliving Oneself: Trauma, Memory and Personal Identity,” in D. T. Meyers, ed. (1997a), 12–39. Brown, L. M. and C. Gilligan (1992) Meeting at the Crossroads: Women’s Psychology and Girls’ Development. Cambridge, MA: Harvard University Press. Brown, W. (1995) States of Injury: Power and Freedom in Late Modernity. Princeton, NJ: Princeton University Press. Bruner, J. (1983a) Child’s Talk: Learning to Use Language. New York: Norton. (1983b) In Search of Mind: Essays in Autobiography. New York: Harper & Row. (1986) Actual Minds, Possible Words. Cambridge, MA: Harvard University Press. (1990) Acts of Meaning. Cambridge, MA: Harvard University Press. Buchanan, A. (1989) “Assessing the Communitarian Critique of Liberalism.” Ethics 99, 852–882. Burge, T. (1988) “Individualism and Self-Knowledge.” Journal of Philosophy 85: 649–663. Burkitt, D. I. (1991) Social Selves: Theories of the Social Formation of Personality. Beverly Hills, CA: Sage Publications. Bushnell, D., ed. (1995) Nagging Questions. Savage, MD: Rowman & Littlefield. Buss, S. (2002) “Personal Autonomy.” Stanford Encyclopedia of Philosophy (http:// plato.stanford.edu/). Buss, S. and L. Overton, eds. (2002) Contours of Agency: Essays in Honor of Harry Frankfurt. Cambridge, MA: MIT Press. Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge. (1993) Bodies That Matter. New York: Routledge. (1995) “Contingent Foundations,” in L. Nicholson, ed., 35–58. (1997) The Psychic Life of Power. Palo Alto, CA: Stanford University Press. (2005) Giving an Account of Oneself. New York: Fordham University Press. Campbell, J. (1994) Past, Space, and Self. Cambridge, MA: MIT Press. Carruthers, P. (1986) Introducing Persons. London: Routledge Books. Cassam, Q. (1997) Self and World. Oxford: Clarendon Press. Chodorow, N. (1978) The Reproduction of Mothering: Psychoanalysis and the Sociology of Gender. Berkeley, CA: University of California Press. Christiano, T. (1996) The Rule of the Many: Fundamental Issues in Democratic Theory. Boulder, CO: Westview Press.

252

Bibliography

Christiano, T. and J. Christman, eds. (2009) Debates in Political Philosophy. Oxford: Blackwell. Christman, J. (1987) “Autonomy: A Defense of the Split-Level Self.” Southern Journal of Philosophy 25(3): 281–293. (1991a) “Autonomy and Personal History.” Canadian Journal of Philosophy 21(1): 1–24. (1991b) “Liberalism and Individual Positive Freedom.” Ethics 101(2): 343–359. (1993) “Defending Historical Autonomy: A Reply to Professor Mele.” Canadian Journal of Philosophy 23: 281–290. (1994) The Myth of Property: Towards an Egalitarian Theory of Ownership. New York: Oxford University Press. (1995) “Feminism and Autonomy,” in D. Bushnell, ed., 17–39. (1998) “Autonomy, Independence, and Poverty-Related Welfare Policies.” Public Affairs Quarterly 12(4) (October): 383–406. (2001) “Liberalism, Autonomy, and Self-Transformation.” Social Theory and Practice 27(2): 185–206. (2002a) “Autonomy in Moral and Political Philosophy.” The Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/) (2002b) Social and Political Philosophy: A Contemporary Introduction. London: Routledge. (2004) “Narrative Unity as a Condition of Personhood.” Metaphilosophy 35(5) (October): 695–713. (2005) “Autonomy, Self-Knowledge, and Liberal Legitimacy,” in J. Christman and J. Anderson, eds. 330–357. (2007) “Autonomy, History, and the Subject of Justice.” Social Theory and Practice 33(1): 1–26. Christman, J., ed. (1989) The Inner Citadel: Essays on Individual Autonomy. New York: Oxford University Press. Christman, J. and J. Anderson, eds. (2005) Autonomy and the Challenges to Liberalism: New Essays. Cambridge: Cambridge University Press. Cochran, D. C. (1999) The Color of Freedom: Race and Contemporary American Liberalism. Albany, NY: SUNY Press. Code, L. (1991) “Second Persons,” in What Can She Know? Feminist Theory and the Construction of Knowledge. Lanham, MD: Rowman & Littlefield. (2000) “The Perversion of Autonomy and the Subjection of Women: Discourses of Social Advocacy at Century’s End,” in C. Mackenzie and N. Stoljar, eds., 181–209. Cohen, G. A. (1989) “On the Currency of Egalitarian Justice.” Ethics 99: 96–104. (1995) Self-Ownership, Freedom and Equality. Cambridge: Cambridge University Press. Cohen, J. (1986) “Autonomy and Democracy: Reflections on Rousseau.” Philosophy and Public Affairs 15: 275–297. (1996) “Procedure and Substance in Deliberative Democracy,” in S. Benhabib, ed., 95–119.

Bibliography

253

(2002) “Deliberation and Democratic Legitimacy,” in Deliberative Democracy. James Bohman and William Rehg, eds. Cambridge, MA: MIT Press, 67–92. Cohler, B. J. (1982) “Personal Narrative and Life Course.” Life Span Development and Behavior 4: 205–241. Colapietro, V. M. (1989) Peirce’s Approach to the Self: A Semiotic Perspective on Human Subjectivity. Albany, NY: SUNY Press. Collins, P. H. (1998) “It’s All in the Family: Intersections of Gender, Race, and Nation.” Hypatia 31(3): 62–82. Connolly, W. (1974) The Terms of Political Discourse. Lexington, MA: D. C. Heath & Co. (1991) Identity/Difference: Democratic Negotiations of Political Paradox. Ithaca, NY: Cornell University Press. Conway, M. A. (1990) Autobiographical Memory: An Introduction. Buckingham, UK: Open Court Press. Conway, M. A., D. C. Rubin, H. Spinnler, and W. A. Wagenaar, eds. (1992) Theoretical Perspectives on Autobiographical Memory. Dordrecht: Kluwer. Conway, M. A., J. Singer, and A. Tagini (2004) “The Self and Autobiographical Memory: Correspondence and Coherence.” Social Cognition 22(5): 491–529. Cooke, M. (1999) “Questioning Autonomy: The Feminist Challenge and the Challenge for Feminism,” in Questioning Ethics: Contemporary Debates in Philosophy. R. Kearney and M. Dooley, eds. New York: Routledge, 258–282. Copp, D. (2002) “Social Unity and the Identity of Persons.” The Journal of Political Philosophy 10(4): 365–391. Cranston, M. (1991) The Noble Savage: Jean-Jacques Rousseau, 1754–1762. Chicago, IL: University of Chicago Press. Crittenden, J. (1992) Beyond Individualism: Reconstituting the Liberal Self. New York: Oxford University Press. Crocker, L. (1980) Positive Liberty. The Hague: Martinus Nijhoff. Cuypers, S. (2001) Self-Identity and Personal Autonomy: An Analytical Anthropology. Burlington, VT: Ashgate Press. Dagger, R. (2005) “Autonomy, Domination, and the Republican Challenge to Liberalism,” in J. Christman and J. Anderson, eds., 177–203. Daniels, N., ed. (1975) Reading Rawls: Critical Studies on Rawls’s A Theory of Justice. New York: Basic Books. Davion, V. and C. Wolf, eds. (2000) The Idea of Political Liberalism: Essays on Rawls. Lanham, MD: Rowman & Littlefield. Dearden, R. F. (1972) “Autonomy and Education,” in Education and the Development of Reason. R. F. Dearden, R. S. Peters and P. H. Hirst, eds. London: Routledge & Kegan Paul, 451–452. Delaney, C. F., ed. (1994) The Liberalism-Communitarianism Debate. Lanham, MD: Rowman and Littlefield. Delgado, R. and J. Stefancic (2001) Critical Race Theory: An Introduction. New York: New York University Press. Den Uyl, D. (2003) “Autonomous Autonomy: Spinoza on Autonomy, Perfectionism, and Politics.” Social Philosophy and Policy 20(2): 30–69.

254

Bibliography

Dennett, D. (1988) “Why Everyone is a Novelist.” Times Literary Supplement (Sept. 16–22): 1016–1022. (1989) “The Origins of Selves.” Cogito 3: 163–173. (1990) “The Reality of Selves,” in Consciousness Explained. Boston: Little, Brown & Company, 412–430. (1992) “The Self as the Center of Narrative Gravity,” in Self and Consciousness: Multiple Perspectives. P. M. C. Frank, S. Kessel, and D. L. Johnson, eds. Hillsdale, NJ: Lawrence Erlbaum, 103–115. Derrida, J. (1982) Margins of Philosophy. Chicago, IL: University of Chicago Press. (1997) Of Grammatology, Baltimore, MD: Johns Hopkins University Press. Double, R. (1992) “Two Types of Autonomy Accounts.” Canadian Journal of Philosophy 22(1): 65–80. Downie, R. S. and E. Telfer (1971) “Autonomy.” Philosophy 46: 296–301. Dreyfuss, H. (1991) Being in the World: A Commentary on Heidegger’s Being and Time Division I. Cambridge, MA: MIT Press. Dryzek, J. (2000) Deliberative Democracy and Beyond: Liberals, Critics, Contestations. New York: Oxford University Press. Dworkin, G. (1970) “Acting Freely.” Nous 4(4): 367–383. (1982) “Is More Choice Better Than Less?” Midwest Studies in Philosophy VIII. Minneapolis, MN: University of Minnesota Press, 47–62. (1989a) “The Concept of Autonomy,” in J. Christman, ed., 54–62. (1989b) The Theory and Practice of Autonomy. Cambridge: Cambridge University Press. Dworkin, R. (1977) Taking Rights Seriously. London: Duckworth. (1989) “Liberal Community.” California Law Review 77(3): 479–504. (2000) Sovereign Virtue: The Theory and Practice of Equality. Cambridge, MA: Harvard University Press. Eagle, M. (1991) “Psychoanalytic Conceptions of the Self,” in The Self: Interdisciplinary Approaches. J. Strauss and G. Goethals, eds. New York: Springer-Verlag, 49–65. Eagleton, T. (1983) Literary Theory: An Introduction. Minneapolis, MN: University of Minnesota Press. Eccles, J. C. (1994) How the Self Controls Its Brain. Berlin and New York: SpringerVerlag. Edwards, R. (1981) “Mental Health as Rational Autonomy.” Journal of Medicine and Philosophy 6: 309–321. Ekstrom, L. (1993) “A Coherence Theory of Autonomy.” Philosophy and Phenomenological Research 53: 599–616. (2005) “Alienation, Autonomy, and the Self.” Midwest Studies in Philosophy XXIX. Minneapolis, MN: University of Minnesota Press, 45–67. Elshtain, J. B. (1981) Public Man, Private Women: Women in Social and Political Thought. Princeton, NJ: Princeton University Press. Elster, J. (1979) Ulysses and the Sirens. Cambridge: Cambridge University Press. (1983) Sour Grapes. Cambridge: Cambridge University Press. Elster, J. and A. Hylland, eds. (1986) Foundations of Social Choice Theory. Cambridge: Cambridge University Press.

Bibliography

255

Engstrom, S. “Conditioned Autonomy.” Philosophy and Phenomenological Research 48(3): 435–453. Eposito, J. L. (1987) The Obsolete Self. Berkeley and London: University of California Press. Epstein, S. (1973) “The Self-Concept Revisited: Or a Theory of a Theory,” in Personality: Critical Concepts in Psychology. C. L. Cooper and L. Pervin, eds. New York: Routledge, 453–473. Erez, M. (1993) Culture, Self-Identity, and Work. New York: Oxford University Press. Evans, G. (1982) The Varieties of Reference. Oxford: Oxford University Press. Fausto-Sterling, A. (2000) Sexing the Body: Gender Politics and the Construction of Sexuality. New York: Basic Books. Feinberg, J. (1986) Harm to Self. New York: Oxford University Press. (1989) “Autonomy,” in J. Christman, ed., 27–53. Felson, R. B. (1993) “The (Somewhat) Social Self: How Others Affect SelfAppraisals,” in Psychological Perspectives on the Self. J. Suls, ed. Hillsdale, NJ: Lawrence Erlbaum, 1–26. Festinger, L. (1957) A Theory of Cognitive Dissonance. Stanford: Stanford University Press. Fischer, J. M. and M. Ravizza (1998) Responsibility and Control: A Theory of Moral Responsibility. New York: Cambridge University Press. Fisher, R. (1995) Becoming Persons. London: Oxford University Press. Fivush, R. (1994) “Constructing Narrative, Emotion, and Self in Parent–Child Conversations about the Past,” in U. Neisser and R. Fivush, eds., 136–157. Flanagan, O. (1991) Varieties of Moral Personality. Cambridge, MA: Harvard University Press. (1996) Self Expressions. New York: Oxford University Press. Flathman, R. (1989) Toward a Liberalism. Ithaca, NY: Cornell University Press. Flax, J. (1993) Disputed Subjects: Essays on Psychoanalysis, Politics, and Philosophy. New York: Routledge. Fogel, A. (1993) Developing Through Relationships: Origins of Communication, Self, and Culture. Hemel Hempstead, UK: Harvester Press. Foucault, M. (1965) Madness and Civilization: A History of Insanity in the Age of Reason. New York: Vintage Books. (1984) The Foucault Reader. New York: Pantheon Books. (1988) The Final Foucault. Cambridge, MA: MIT Press. (1990a) The History of Sexuality: An Introduction. New York: Vintage. (1990b) The Use of Pleasure. New York: Vintage. (1995) Discipline and Punish: The Birth of the Prison. New York: Vintage Books. Frankfurt, H. (1973) “Coercion and Moral Responsibility,” in Essays on Freedom of Action. T. Honderich, ed. London: Routledge & Kegan Paul, 65–86. (1987) “Identification and Wholeheartedness,” in Responsibility, Character and the Emotions. F. Schoeman, ed. Cambridge: Cambridge University Press, 27–45.

256

Bibliography

(1988a) “Freedom of the Will and the Concept of a Person,” in The Importance of What We Care About. Cambridge: Cambridge University Press. (1988b) “Rationality and the Unthinkable,” in The Importance of What We Care About. Cambridge: Cambridge University Press, 177–190. (1992) “The Faintest Passion.” Proceedings and Addresses of the Aristotelian Society 49: 113–145. Fraser, N. (1997) Justice Interruptus. New York: Routledge. Fraser, N. and S. Bartky, eds. (1989) Revaluing French Feminism: Critical Essays on Difference, Agency, and Culture. Bloomington, IN: Indiana University Press. Fraser, N. and L. Gordon (1994) “A Genealogy of Dependency: Tracing a Keyword of the US Welfare State.” Signs 19(2): 309–336. Fraser, N. and A. Honneth (2003) Redistribution or Recognition? A PoliticalPhilosophical Exchange. London: Verso. Freeman, M. (1993) Rewriting the Self. New York: Routledge. Freeman, S. (2009) “Facts and Moral Justification,” in T. Christiano and J. Christman, eds. Freeman, S., ed. (2003) The Cambridge Companion to Rawls. Cambridge: Cambridge University Press. Freeman, W. and J. W. Watts (1941) “The Frontal Lobes and Consciousness of Self.” Psychosomatic Medicine 3: 111–119. Friedman, M. (1986) “Autonomy and the Split-Level Self.” Southern Journal of Philosophy 24(1): 19–35. (1997) “Autonomy and Social Relationships: Rethinking the Feminist Critique,” in D. T. Meyers, ed. (1997a). Boulder, CO: Westview Press, 40–61. (1998) “Feminism, Autonomy, and Emotion,” in Norms and Values: Essays on the Work of Virginia Held. J. G. Haber, ed. Lanham, MD: Rowman & Littlefield, 37–48. (2000a) “Autonomy, Social Disruption, and Women,” in C. Mackenzie and N. Stoljar, eds., 35–51. (2000b) “Feminism in Ethics: Conceptions of Autonomy,” in The Cambridge Companion to Feminism in Philosophy. M. Fricker and J. Hornsby, eds. Cambridge: Cambridge University Press, 205–224. (2000c) “John Rawls and the Political Conception of Unreasonable People,” in V. Davion and C. Wolf, eds., 16–33. (2003) Autonomy, Gender, Politics. Oxford: Oxford University Press. Frohok, F. (1997) “Conceptions of Persons.” Social Theory and Practice 23(1) (Spring): 129–158. Frye, N. (1957) The Anatomy of Criticism. Princeton, NJ: Princeton University Press. Fuss, D. (1989) “‘Essentially Speaking’: Luce Irigaray’s Language of Essence,” in N. Fraser and S. Bartky, eds., 94–112. Gallagher, S. (1986) “Body Image and Body Schema: A Conceptual Clarification.” Journal of Mind and Behavior 7: 541–554. (1992) “The Theater of Personal Identity: From Hume to Derrida.” The Personalist Forum 8(1992): 21–30.

Bibliography

257

(1995) “Body Schema and Intentionality,” in J. Bermúdez, A. Marcel, and N. Eilan, eds., 225–244. (2005) How The Body Shapes The Mind. Oxford: Oxford University Press. Gallagher, S. and A. Meltzoff (1996) “The Earliest Sense of Self and Others: Merleau-Ponty and Recent Developmental Studies.” Philosophical Psychology 9(2): 213–236. Gallagher, S. and J. Shear (1997) “Models of the Self.” Special issue of Journal of Consciousness Studies 4(5/6). Galston, W. (1991) Liberal Purposes: Goods, Virtues, and Diversity in the Liberal State. Cambridge: Cambridge University Press. (2002) Liberal Pluralism: The Implications of Value Pluralism for Political Theory and Practice. New York: Cambridge University Press. Gaus, G. F. (1996) Justificatory Liberalism. New York: Oxford University Press. (forthcoming) “The Moral Foundations of Liberal Neutrality,” in T. Christiano and J. Christman, eds. Gauthier, D. (1986) Morals by Agreement. Oxford: Oxford University Press. (2006) Rousseau: The Sentiment of Existence. Cambridge: Cambridge University Press. Geertz, C. (1973) The Interpretation of Cultures: Selected Essays: New York: Basic Books. Gergen, K. J. (1971) The Concept of Self. New York: Holt, Rinehart and Winston. (1991) The Saturated Self: Dilemmas of Identity in Contemporary Life. New York: Basic Books. (1994) “Mind, Text, and Society: Self-Memory in Social Context,” in U. Neisser and R. Fivush, eds., 78–104. Gergen, K. J. and M. M. Gergen (1997) “Narratives of the Self,” in L. P. Hinchman and S. K. Hinchman, eds., 161–84. Giddens, A. (1991) Modernity and Self-Identity: Self and Society in the Late Modern Age. Palo Alto, CA: Stanford University Press. Gilligan, C. (1982) In a Different Voice: Psychological Theory and Women’s Development. Cambridge, MA: Harvard University Press. Gilligan, C., J. V. Ward, and J. M. Taylor, eds. (1988) Mapping the Moral Domain: A Contribution of Women’s Thinking to Psychological Theory and Education. Cambridge, MA: Harvard University Press. Goldie, P. (2003) “One’s Remembered Past: Narrative Thinking, Emotion, and the External Perspective.” Philosophical Papers 32(3) (November): 301–319. Gould, C. (1988) Rethinking Democracy. Cambridge: Cambridge University Press. Gray, J. (1989) Liberalisms: Essays in Political Philosophy. New York: Routledge. (1993) Post-Liberalism: Studies in Political Thought. London: Routledge. (2000) Two Faces of Liberalism. New York: New Press. Greenawalt, K. (1995) Private Consciences and Public Reason. New York: Oxford University Press. Greimas, A.-J. (1983) Structural Semantics: An Attempt at a Method. Lincoln, NE: University of Nebraska Press.

258

Bibliography

(1987) “Elements of a Narrative Grammar,” in Twentieth Century Literary Theory. V. Lambropoulos and D. N. Miller, eds. Albany, NY: SUNY Press, 304–328. Griffin, J. (1986) Well-Being: Its Meaning, Measurement, and Moral Importance. Oxford: Oxford University Press. Grimshaw, J. (1986) Philosophy and Feminist Thinking. Minneapolis, MN: University of Minnesota Press. Gross, H. (1971) “Privacy and Autonomy.” Nomos XIII: Privacy. J. W. Chapman and J. R. Pennock, eds. New York: Atherton, 169–1. Grosz, E. (1994) Volatile Bodies: Toward a Corporeal Feminism. Bloomington, IN: University of Indiana Press. Grovier, T. (1993) “Self-Trust, Autonomy and Self-Esteem.” Hypatia 8(1): 99–119. Gutmann, A. (1985) “Communitarian Critics of Liberalism.” Philosophy and Public Affairs 14(3): 308–322. (1993) “Democracy,” in A Companion to Contemporary Political Philosophy. R. Goodin and P. Pettit, eds. Oxford: Blackwell Publishers, 411–421. (2003a) Identity in Democracy. Princeton, NJ: Princeton University Press. (2003b) “Rawls on the Relationship between Liberalism and Democracy,” in S. Freeman, ed., 168–199 Habermas, J. (1990) Moral Consciousness and Communicative Action. Cambridge, MA: MIT Press. (1995) The Philosophical Discourse of Modernity. Cambridge, MA: MIT Press. (1996a) Between Facts and Norms. Cambridge, MA: MIT Press. (1996b) “Three Normative Models of Democracy,” in S. Benhabib, ed., 21–30. (1998) The Inclusion of the Other: Studies in Political Theory. Cambridge, MA: MIT Press. (1999) “Struggles for Recognition in the Democratic Constitutional State,” trans. S. W. Nicholsen, in Multiculturalism. A. Gutmann, ed. Princeton, NJ: Princeton University Press: 107–148. Hacking, I. (1995) Rewriting the Soul: Multiple Personality and the Structure of Memory. Princeton, NJ: Princeton University Press. Hamaguchi, E. (1985) “Relational Selves in Asians.” Journal of Japanese Studies 11(2): 289–321. Hampshire, S. (1975) Freedom and the Individual. Princeton, NJ: Princeton University Press. Haraway, D. (1997) “A Manifesto for Cyborgs: Science, Technology, and Socialist Feminism in the 1980s,” in D. T. Meyers, ed. (1997b), 501–531. Hardin, R. (1988) Morality within the Limits of Reason. Chicago, IL: University of Chicago Press. Harter, S. (1996) “Historical Roots of Contemporary Issues Involving SelfConcept,” in B. A. Bracken, ed., 1–38. Harvey, D. (1989) The Condition of Postmodernity. Cambridge, MA: Basil Blackwell. Haugeland, J. (1982) “Heidegger on Being a Person.” Nous 16: 15–26. Haworth, L. (1986) Autonomy: An Essay in Philosophical Psychology and Ethics. New Haven, CT: Yale University Press.

Bibliography

259

Hekman, S. J. (1995) Moral Voices, Moral Selves: Carol Gilligan and Feminist Moral Theory. University Park, PA: Pennsylvania State University Press. Held, D. (1987) Models of Democracy. Stanford, CA: Stanford University Press. Held, V. (1993) Feminist Morality: Transforming Culture, Society and Politics. Chicago, IL: University of Chicago Press Hill, S. (1975) “Self Determination and Autonomy,” in Today’s Moral Problems. R. Wasserstrom, ed. New York: Macmillan, 118–132. Hill, T. (1987) “The Importance of Autonomy,” in E. F. Kittay and D. T. Meyers, eds., 129–138. (1989) “The Kantian Conception of Autonomy,” in J. Christman, ed., 91–105. (1991) Autonomy and Self Respect. New York: Cambridge University Press. Hinchman, L. P. and S. K. Hinchman, eds. (1997) Memory, Identity, Community. Albany, NY: SUNY Press. Hirst, W. (1994) “The Remembered Self in Amnesics,” in U. Neisser and R. Fivush, eds., 252–277. Honneth, A. (1996) The Struggle for Recognition: The Moral Grammar of Social Conflicts. Cambridge, MA: MIT Press. (2001) “Grounding Recognition: A Rejoinder to Critical Questions.” Inquiry (February): 499–519. Horkheimer, M. and T. Adorno (1944/1998) Dialectic of Enlightenment. New York: Continuum Publishing. Horney, K. (1945/1999) Our Inner Conflicts. London: Routledge. Hurka, T. (1987) “Why Value Autonomy?” Social Theory and Practice 13(3): 361–382. (1993) Perfectionism. New York: Oxford University Press. Husak, D. (1980) “Paternalism and Autonomy.” Philosophy and Public Affairs 10: 27–46. Jacobs, J. (2003) “Some Tensions between Autonomy and Self-Governance.” Social Philosophy and Policy 20(2): 221–244. Jaggar, A. (1988) Feminist Politics and Human Nature. Totowa, NJ: Rowman and Allanheld. James, W. (1890) The Principles of Psychology: New York: Henry Holt. Jameson, F. (1972) The Prison House of Language. Princeton, NJ: Princeton University Press. Kagan, J. (1989) Unstable Ideas: Temperament, Cognition, and Self. Cambridge, MA: Harvard University Press. Kaplan, H. B. (1986) Social Psychology and Self Referent Behavior. New York: Plenum Press. Kateb, G. (1992) The Inner Ocean: Individualism and Democratic Culture. Ithaca, NY: Cornell University Press. Keller, J. (1997) “Autonomy, Relationality, and Feminist Ethics.” Hypatia 12(2): 152–164. Kelly, M. (1990) “MacIntyre, Habermas, and Philosophical Ethics in Hermeneutics in Ethics and Social Theory.” Philosophical Forum 21(1–2): 70–93.

260

Bibliography

Kerby, A. P. (1991) Narrativity and the Self. Bloomington, IN: Indiana University Press. Kernohan, A. (1998) Liberalism, Equality, and Cultural Oppression. New York: Cambridge University Press. Kittay, E. F. (1999) Love’s Labor: Essays on Women, Equality, and Dependency. London: Routledge. Kittay, E. F. and D. T. Meyers, eds. (1987) Women and Moral Theory. Savage, MD: Rowman & Littlefield. Klein, S., et al. (2004) “A Theory of Autobiographical Memory: Necessary Components and Disorders Resulting from their Loss.” Social Cognition 22(5): 460–490. Kleinig, J. (1983) Paternalism. Totowa, NJ: Rowman and Allenheld. Kneller, J. and S. Axinn, eds. (1998) Autonomy and Community: Readings in Contemporary Kantian Social Philosophy. Albany, NY: SUNY Press. Korsgaard, C. M. (1996) The Sources of Normativity. New York: Cambridge University Press. Kourany, J. A., J. P. Sterba, and R. Tong, eds. (1992) Feminist Philosophies. Englewood Cliffs, NJ: Prentice Hall. Kristeva, J. (1980) Desire in Language. New York: Columbia University Press. (1986) The Kristeva Reader. New York: Columbia University Press. Kuflick, A. (1984) “The Inalienability of Autonomy.” Philosophy and Public Affairs 13: 271–298. Kukathas, C. (1997) “Multiculturalism as Fairness: Will Kymlicka’s Multicultural Citizenship.” Journal of Political Philosophy 5: 406–427. Kupfer, J. H. (1990) Autonomy and Social Interaction. Albany, NY: SUNY Press. Kymlicka, W. (1989) Liberalism, Community and Culture. Oxford: Clarendon. (1990) Contemporary Political Philosophy: An Introduction. Oxford: Clarendon. (1995) Multicultural Citizenship: A Liberal Theory of Minority Rights. Oxford: Clarendon. Lacan, J. (1977) Ecrits: A Selection. New York: Norton. Laclau, E. and C. Mouffe (1985) Hegemony and Socialist Strategy: Towards a Radical Democratic Politics. New York: Verso. Ladenson, R. F. (1975) “A Theory of Personal Autonomy.” Ethics 86: 30–48. Lakoff, G. and M. Johnson (1980) Metaphors We Live By: Chicago, IL: University of Chicago Press. (1999) Philosophy in the Flesh. New York: Basic Books. Larmore, C. (1987) Patterns of Moral Complexity. Cambridge: Cambridge University Press. (1996) The Morals of Modernity. New York: Cambridge University Press. Lau, J. 2002 “Externalism About Mental Content.” Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/entries/content-externalism/#Bib). Law, I. (2003) “Autonomy, Sanity and Moral Theory.” Res Publica 9(1): 39–56. Lehrer, K. (2003) “Reason and Autonomy.” Social Philosophy and Policy 20(2): 177–198.

Bibliography

261

Leitch, T. (1986) What Stories Are: Narrative Theory and Interpretation. University Park, PA: Pennsylvania State University Press. Leland, D. (1989) “Lacanian Psychoanalysis and French Feminism: Toward an Adequate Political Psychology,” in N. Fraser and S. Bartky, eds., 113–135. Levey, G. B. (1997) “Equality, Autonomy, and Cultural Rights.” Political Theory 25(2): 215–248. Levine, A. (1976) The End of Autonomy: A Kantian Reading of Rousseau’s Social Contract. Amherst, MA: University of Massachusetts Press. Levinson, M. (1999) The Demands of Liberal Education. New York: Oxford University Press. (2002) The Demands of Liberal Education. Oxford: Oxford University Press. Lindley, R. (1986) Autonomy. Atlantic Highlands, NJ: Humanities Press International. Lloyd, G. (1984) The Man of Reason: “Male” and “Female” in Western Philosophy. London: Methuen. Loftus, E. (1997) “Creating False Memories.” Scientific American 277 (September): 70–75. Lukes, S., ed. (1986) Power. New York: New York University Press. Lyotard, J.-F. (1997) The Postmodern Condition: A Report on Knowledge. Minneapolis, MN: University of Minnesota Press. MacCallum, G. (1967) “Negative and Positive Freedom.” Philosophical Review 76: 312–332. Macedo, S. (1990) Liberal Virtues: Citizenship, Virtue, and Community in Liberal Constitutionalism. Oxford: Clarendon. MacIntyre, A. (1981) After Virtue. Notre Dame, IN: University of Notre Dame Press. (1988) Whose Justice? Which Rationality? Notre Dame, IN: University of Notre Dame Press. (1997) “The Virtues, the Unity of Human Life, and the Concept of a Tradition,” in Memory, Identity, Community. S. K. H. Lewis and P. Hinchman, eds. Albany, NY: SUNY Press, 241–263. Mackenzie, C. (2000) “Imagining Oneself Otherwise,” in. C. Mackenzie and N. Stoljar, eds., 124–150. (2002) “Critical Reflection, Self-Knowledge, and the Emotions.” Philosophical Explorations, 5(3) (October): 186–206. Mackenzie, C. and N. Stoljar (2000a) “Introduction: Autonomy Refigured,” in C. Mackenzie and N. Stoljar, eds., 3–31. Mackenzie, C. and N. Stoljar, eds. (2000b) Relational Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self. New York: Oxford University Press. MacKinnon, C. (1989) Toward a Feminist Theory of the State. Cambridge, MA: Harvard University Press. (2006) Are Women Human? Cambridge, MA: Harvard University Press. MacLean, D. and C. Mills, eds. (1983) Liberalism Reconsidered. Totowa, NJ: Rowman and Allenheld.

262

Bibliography

Mahmoud, S. (2003) “Ethical Formation and Politics of Individual Autonomy in Contemporary Egypt.” Social Research 70(3): 837–866. Margalit, A. (2002) The Ethics of Memory. Cambridge, MA: Harvard University Press. Margalit, A. and M. Halbertal (2004) “Liberalism and the Right to Culture.” Social Research 71(3): 529–548. Margalit, A. and J. Raz (1990) “National Self-Determination.” Journal of Philosophy. 87, 439–461. Markus, H. and S. Kitayama (1991) “Culture and Self: Implications for Cognition, Emotion, and Motivation.” Psychological Review 98: 224–253. Markus, H. and K. Sentis (1982) “The Self in Social Information Processing,” in Psychological Perspectives on the Self Vol. 1. J. Suls, ed. Hillsdale, NJ: Lawrence Erlbaum, 41–70. Markus, H. and D. M. Wegner, eds. (1980) The Self in Social Psychology. New York: Oxford University Press. May, T. (1994) “The Concept of Autonomy.” American Philosophical Quarterly 31(2): 133–144. McCallum, G. (1967) “Negative and Positive Freedom.” Philosophical Review 76: 3112–3134. McLean, K. (2005) “Late Adolescent Identity Development: Narrative Meaning Making and Memory Telling.” Developmental Psychology 41(4): 683–691. McLeod, C. (2002) Self-Trust and Reproductive Autonomy. Cambridge, MA: MIT Press. Mead, G. H. (1934) Mind, Self and Society. Chicago, IL: University of Chicago Press. Mele, A. R. (1991) “History and Personal Autonomy.” Canadian Journal of Philosophy 23: 271–280. (1995) Autonomous Agents: From Self-Control to Autonomy. New York: Oxford University Press. Merleau-Ponty, M. (2002) Phenomenology of Perception: An Introduction. New York: Routledge. Meyer, M. J. (1987) “Stoics, Rights and Autonomy.” American Philosophical Quarterly 24(3): 267–271. Meyers, D. T. (1987a) “Personal Autonomy and the Paradox of Feminine Socialization.” Journal of Philosophy 84(11): 619–628. (1987b) “The Socialized Individual and Individual Autonomy,” in E. F. Kittay and D. T. Meyers, eds., 239–254. (1989) Self, Society and Personal Choice. New York: Columbia University Press. (1994) Subjection and Subjectivity: Psychoanalytic Feminism and Moral Philosophy. New York: Routledge. (2000a) “Feminism and Women’s Autonomy: The Challenge of Female Genital Cutting.” Metaphilosophy 31(5): 469–491. (2000b) “Intersectional Identity and the Authentic Self,” in C. Mackenzie and N. Stoljar, eds., 151–180.

Bibliography

263

(2005) “Decentralizing Autonomy: Five Faces of Selfhood,” in J. Christman and J. Anderson, eds., 27–55. Meyers, D. T., ed. (1997a) Feminists Rethink the Self. Boulder, CO: Westview Press. (1997b) Feminist Social Thought: A Reader. London: Routledge. Michelman, F. (1988) “Law’s Republic.” Yale Law Journal 97: 1493–1537. Miller, B. (1981) “Autonomy and the Refusal of Life-Saving Treatment.” Hastings Center Report: 22–28. Milligan, D. and W. W. Miller, eds. (1992) Liberalism, Citizenship and Autonomy. Brookfield, VT: Avebury. Mills, C. (1997) The Racial Contract. Ithaca, NY: Cornell University Press. (2005) “‘Ideal Theory’ as Ideology.” Hypatia 20(3) (Summer): 165–184. Mohanty, S. (2004) “The Epistemic Status of Cultural Identity,” in Identities. L. Alcoff and E. Mendieta, eds. Oxford: Blackwell, 392–405. Montifiore, A., ed. (1975) Neutrality and Impartiality. Cambridge: Cambridge University Press. Moon, J. D. (1993) Constructing Community: Moral Pluralism and Tragic Conflicts. Princeton, NJ: Princeton University Press. (2003) “Liberalism, Autonomy, and Moral Pluralism.” Political Theory 31(1): 125–135. Mouffe, C. (2000) The Democratic Paradox. New York: Verso. Mouffe, C., ed. (1992) Dimensions of Radical Democracy: Pluralism, Citizenship, Community. London: Routledge. Mulhall, S. and A. Swift (1992) Liberals and Communitarians. Oxford: Basil Blackwell. Nagel, T. (1991) Equality and Partiality. New York: Oxford University Press. Narayan, U. (2002) “Minds of Their Own: Choices, Autonomy, Cultural Practices, and Other Women,” in A Mind of One’s Own: Feminist Essays on Reason and Objectivity. L. N. Antony and C. E. Witt, eds. Boulder, CO: Westview Press, 418–432. Nauta, L. (1985) “Historical Roots of the Concept of Autonomy in Western Philosophy.” Praxis 4(4): 363–377. Nedelsky, J. (1989) “Reconceiving Autonomy: Sources, Thoughts, and Possibilities.” Yale Journal of Law and Feminism 1(1): 7–36. Neisser, U. (1988) “Five Kinds of Self-Knowledge.” Philosophical Psychology 1: 35–59. Neisser, U. and R. Fivush, eds. (1994) The Remembering Self: Construction and Accuracy in the Self-Narrative. New York: Cambridge University Press. Nicholson, L., ed. (1990) Feminism/Postmodernism. New York: Routledge. Nietszche, F. (1887/2003) The Genealogy of Morals. Trans. Horace B. Samuel. New York: Dover Publications. Noddings, N. (1984) Caring: A Feminist Approach to Ethics and Moral Education. Berkeley, CA: University of California Press. Noggle, R. (2005) “Autonomy and the Paradox of Self-Creation: Infinite Regresses, Finite Selves, and the Limits of Authenticity,” in J. S. Taylor, ed., 87–108.

264

Bibliography

Nussbaum, M. C. (2003) Upheavals of Thought: The Intelligence of Emotions. Cambridge: Cambridge University Press. O’Neill, O. (2003) “Autonomy: The Emperor’s New Clothes.” Aristotelian Society Supp. Vol. 77: 1–21. Okin, S. M. (1989) Justice, Gender, and the Family. New York: Basic Books. Oliver, K., ed. (2000) The French Feminism Reader. Lanham, MD: Rowman & Littlefield. Oshana, M. (1998) “Personal Autonomy and Society.” Journal of Social Philosophy 29(1): 81–102. (2003) “How Much Should We Value Autonomy?” Social Philosophy and Policy 20(2): 99–126. (2006) Personal Autonomy in Society. Hampshire, UK: Ashgate Publishing, Ltd. Parekh, B. (1994) “Superior People: The Narrowness of Liberalism from Mill to Rawls.” Times Literary Supplement (Feb. 25): 11–13. Parfit, D. (1984) Reasons and Persons. Oxford: Oxford University Press. Pateman, C. (1988) The Sexual Contract. Palo Alto, CA: Stanford University Press. Paul, E. F., F. Miller, and J. Paul, eds. (1986) The Communitarian Challenge to Liberalism. Cambridge: Cambridge University Press. Pearsall, M., ed. (1986) Women and Values: Readings in Recent Feminist Philosophy. Belmont, CA: Wadsworth Publishing. Pettit, P. (1997) Republicanism: A Theory of Freedom and Government. New York: Oxford University Press. (2001) A Theory of Freedom: From the Psychology to the Politics of Agency. New York: Oxford University Press. Phelan, S. (1996) Getting Specific: Postmodern Lesbian Politics. Minneapolis, MN: University of Minnesota Press. Phinney, J. S. (1996) “When We Talk About American Ethnic Groups, What Do We Mean?” American Psychologist 51(9) (September): 918–927. Prince, G. (1995) “Narratology,” in The Cambridge History of Literary Criticism, Vol. VIII. R. Selden, ed. Cambridge: Cambridge University Press. Putnam, H. (1975) “The Meaning of ‘Meaning.’” Philosophical Papers, Vol. II: Mind, Language, and Reality. Cambridge: Cambridge University Press. Putnam, R. (2000) Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Rachlin, H. (2003) “Autonomy from the Viewpoint of Teleological Behaviorism.” Social Philosophy and Policy 20(2): 245–264. Radden, J. (1996) Divided Minds and Successive Selves: Ethical Issues in Disorders of Identity and Personality. Cambridge, MA: MIT Press. Rawls, J. (1971) A Theory of Justice. Cambridge, MA: Harvard University Press. (1993) Political Liberalism. New York: Columbia University Press. (1999a) Collected Papers. Cambridge, MA: Harvard University Press. (1999b) The Law of Peoples. Cambridge, MA: Harvard University Press. (1999c) “The Idea of Public Reason Revisited,” in J. Rawls (1999b), 129–180. (2001) Justice as Fairness: A Restatement. Cambridge, MA: Harvard University Press.

Bibliography

265

Raz, J. (1986) The Morality of Freedom. Oxford: Clarendon Press. Reath, A., B. Herman, and C. M. Korsgaard, eds. (1997) Reclaiming the History of Ethics: Essays for John Rawls. Cambridge: Cambridge University Press. Ricoeur, P. (1986) Time and Narrative. 3 vols. Chicago, IL: University of Chicago Press. (1992) Oneself as Another. Chicago, IL: University of Chicago Press. (2004) Memory, History, Forgetting. Chicago, IL: University Of Chicago Press. Ripstein, A. (1999) Equality, Responsibility, and the Law. Cambridge: Cambridge University Press. Rorty, A., ed. (1976) The Identities of Persons. Berkeley, CA: University of California Press. Rorty, R. (1989) Contingency, Irony and Solidarity. Cambridge: Cambridge University Press. Rosenblum, N. L., ed. (1989) Liberalism and the Moral Life. Cambridge, MA: Harvard University Press. Ross, A. O. (1992) The Sense of Self: Research and Theory. New York: Springer. Ross, M. (1991) Remembering the Personal Past. New York: Oxford University Press. Rotberg, R. and D. Thompson, eds. (1993) Truth v. Justice: The Morality of Truth Commissions. Princeton, NJ: Princeton University Press. Rousseau, J.-J. (1762/1993) Emile. Trans. Barbara Foxley. London: Everyman. (1987) Basic Political Writings. D. Cress, ed. Indianapolis, IN: Hackett. Rubin, D. (1998) “Beginnings of a Theory of Autobiographical Remembering,” in C. P. Thompson, J. J. Skowronski, S. F. Larsen, and A. Betz, eds., 47–68. Rubin, D., ed. (1986) Autobiographical Memory. Cambridge: Cambridge University Press. Sandel, M. J. (1982) Liberalism and the Limits of Justice. Cambridge: Cambridge University Press. (1996) Democracy’s Discontent: America in Search of a Public Philosophy. Cambridge, MA: Belknap Press. Sanders, S. (1982) “Autonomy, Authority and Moral Education.” Journal of Social Philosophy 13: 18–24. Sartorius, R. (1983) Paternalism. Minneapolis, MN: University of Minnesota Press. Scanlon, T. (1998) What We Owe to Each Other. Cambridge, MA: Harvard University Press. Schacter, D. (1997) Searching for Memory: The Brain, the Mind, and the Past. New York: Basic Books. Schafer, R. (1981) “Narration in the Psychoanalytic Dialogue,” in On Narrative. W. J. T. Mitchell, ed. Chicago: University of Chicago Press, 25–116. Schechtman, M. (1996) The Constitution of Selves. Ithaca, NY: Cornell University Press. Scheman, N. (1983) “Individualism and the Objects of Psychology,” in Discovering Reality. S. Harding, and M. Hintikka, eds. Boston, MA: D. Reidel, 225–244. (1993) Engenderings: Constructions of Knowledge, Authority, and Privilege. New York: Routledge. (1996) “Reply to Louise Antony.” Hypatia 11(3): 150–153.

266

Bibliography

Schneewind, J. B. (1998) The Invention of Autonomy. Cambridge: Cambridge University Press. Schrader, G. (1969) “Autonomy, Heteronomy, and Moral Imperatives,” in Foundations of the Metaphysics of Morals. R. P. Wolff, ed. New York: BobbsMerrill, 117–133. Seidman, J. S. (2003) “Rationality and Reflection.” Ethical Theory and Moral Practice 6: 201–214. Sen, A. (1992) Inequality Reexamined. Cambridge: Cambridge University Press. (1993) “Capability and Well-Being,” in The Quality of Life. A. Sen and M. Nussbaum, eds. Oxford: Clarendon Press, 30–53. (2006) Identity and Violence. New York: Norton. Shank, R. (1990) Tell Me a Story: A New Look at Real and Artificial Memory. New York: Scribners. Sheets-Johnstone, M. (1999) “Phenomenology and Agency: Methodological and Theoretical Issues in Strawson’s ‘The Self,’” in Models of the Self. J. S. Shaun Gallagher, ed. Thorverton, UK: Imprint Academic, 231–252. Sher, G. (1987) “Other Voices Other Rooms: Women’s Psychology and Moral Theory,” in E. F. Kittay and D. T. Meyers, eds., 178–189. (1989) “Three Grades of Social Involvement.” Philosophy and Public Affairs 18(2): 133–157. (1997) Beyond Neutrality: Perfectionism and Politics. Cambridge: Cambridge University Press. Siegler, M. (1977) “Critical Illness: The Limits of Autonomy.” Hastings Center Report 8: 12–15. Simmel, G. (1907/1990) The Philosophy of Money. Trans. T. Bottomore. New York: Routledge. Smith, G. W. (1987) “J. S. Mill on Freedom,” in Conceptions of Liberty in Political Philosophy. Z. Pelcynski, and J. Gray, eds. New York: St. Martins Press, 182–216. Spence, D. (1984) Narrative Truth and Historical Truth: Meaning and Interpretation in Psychoanalysis. New York: Norton. Stoljar, N. (2000) “Autonomy and the Feminist Intuition,” in C. Mackenzie and N. Stoljar, eds., 94–111. Strawson, G. (1999) “The Self,” in Models of the Self. J. S. Shaun Gallagher, ed. Thorverton, UK: Imprint Academic, 1–19. (2004) “Against Narrativity.” Ratio 17(4): 428–452. Sullivan, S. (2006) Revealing Whiteness: The Unconscious Habits of Racial Privilege. Bloomington, IN: Indiana University Press. Sumner, L. W. (1996) Welfare, Happiness, and Ethics. Oxford: Clarendon Press. Sutton, J. (2004) “Memory.”The Stanford Encyclopedia of Philosophy (http://plato. stanford.edu/). Taylor, C. (1976) “Responsibility for Self,” in A. Rorty, ed., 281–299. (1979) Hegel and Modern Society. Cambridge: Cambridge University Press. (1985a) Philosophy and the Human Sciences: Philosophical Papers, II. Cambridge: Cambridge University Press.

Bibliography

267

(1985b) “What Is Human Agency?” in Human Agency and Language: Philosophical Papers I. Cambridge: Cambridge University Press, 15–44. (1985c) “What is Wrong With Negative Liberty?” in C. Taylor (1985a), 211–229. (1989a) “Cross-Purposes: The Liberal-Communitarian Debate,” in N. L. Rosenblum, ed., 159–182. (1989b) Sources of the Self: The Making of Modern Identity. Cambridge, MA: Harvard University Press. (1991) The Ethics of Authenticity. Cambridge, MA: Harvard University Press. (1992) Multiculturalism and the “Politics of Recognition.” Princeton, NJ: Princeton University Press. Taylor, J., executive producer (2004) Living Without Memory. BBC documentary. Taylor, J. S. (2003) “Autonomy, Duress, and Coercion.” Social Philosophy and Policy 20(2): 127–155. Taylor, J. S., ed. (2005) Personal Autonomy: New Essays on Personal Autonomy and Its Role in Contemporary Moral Philosophy. Cambridge: Cambridge University Press. Thalberg, I. (1989) “Hierarchical Analyses of Unfree Action,” in J. Christman, ed., 123–136. Thompson, C. P., J. J. Skowronski, S. F. Larsen, and A Betz, eds. (1998) Autobiographical Memory: Remembering What and Remembering When. Mahwah, NJ: Lawrence Erlbaum. Titmuss, R. M. (1970) The Gift Relationship. London: Allen & Unwin. Todorov, T. (1977) The Poetics of Prose. Oxford: Basil Blackwell. Tulving, E., ed. (2000) The Oxford Handbook of Memory. Oxford: Oxford University Press. Turner, M. (1996) The Literary Mind. New York: Oxford University Press. Tversky, A. and Kahneman, D. (1982) Judgment under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press. Van de Veer, D. (1979) “Autonomy Respecting Paternalism.” Social Theory and Practice 6: 187–207. Vanderschraaf, P. (2006) “The Circumstances of Justice.” Politics, Philosophy and Economics 5(3): 321–351. Velleman, D. (1992) “What Happens When Somebody Acts?” Mind 101: 461–481. (1996) “Self to Self.” Philosophical Review 105(1): 39–76. (2000) The Possibility of Practical Reason. Oxford: Clarendon Press. (2002) “Identification and Identity,” in S. Buss and L. Overton, eds., 91–123. (2003) “Narrative Explanations.” The Philosophical Review, 12(1) (January): 1–25. (2004) “Identity and Identification,” reprinted in Self to Self. Cambridge: Cambridge University Press, 330–359. (2005) “The Self as Narrator,” in J. Christman and J. Anderson, eds., 56–76. (2006) Self to Self. Cambridge: Cambridge University Press. Vygotsky, L. S. (1930/1978) Mind in Society. Cambridge, MA: Harvard University Press. Wagenaar, W. (1994) “Is Memory Self-Serving?” in U. Neisser and R. Fivush, eds., 191–204.

268

Bibliography

Waldron, J. (1987) Nonsense Upon Stilts: Bentham, Burke, and Marx on the Rights of Man. New York: Routledge. (1992) “Minority Cultures and the Cosmopolitan Alternative.” University of Michigan Journal of Law and Reform 25(3): 751–793. (1993) Liberal Rights: Collected Papers 1981–1991. New York: Cambridge University Press. (1999) The Dignity of Legislation. New York: Cambridge University Press. Wall, S. (1998) Liberalism, Perfectionism and Restraint. New York: Cambridge University Press. (forthcoming) “Perfectionist Politics: A Defense,” in T. Christiano and J. Christman, eds. Waller, B. (1993) “Natural Autonomy and Alternative Possibilities.” American Philosophical Quarterly 30(1): 73–81. Walzer, M. (1983) Spheres of Justice. Oxford: Blackwell. Watson, G. (1975) “Free Agency.” Journal of Philosophy 72: 205–220. (1987) “Free Action and Free Will.” Mind 96(384): 145–172. (2002) “Volitional Necessities,” in S. Buss and L. Overton, eds., 129–159. Watson, G., ed. (1982) Free Will. Oxford: Oxford University Press. Wegner, D. M. and R. R. Vallacher, eds. (1980) The Self in Social Psychology. New York: Oxford University Press. Weir, A. (1996) Sacrificial Logics: Feminist Theory and the Critique of Identity. New York: Routledge. White, S. (1988) The Recent Work of Jürgen Habermas. New York: Cambridge University Press. (1991) Political Theory and Post Modernism. Cambridge: Cambridge University Press. Williams, B. (1983) “Persons, Character, and Morality,” in Moral Luck. Cambridge: Cambridge University Press, 1–19. (1985) “Styles of Ethical Theory,” in Ethics and the Limits of Philosophy. Cambridge: Cambridge University Press, 71–92 Williams, P. J. (1991) The Alchemy of Race and Rights. Cambridge, MA: Harvard University Press. Wittgenstein, L., (1958) Philosophical Investigations. Trans. G. E. M. Anscombe. Oxford: Blackwell. Wolf, S. (1990) Freedom and Reason. New York: Oxford University Press. Wollheim, R. (1984) The Thread of Life. Cambridge: Cambridge University Press. Wong, D. (1988) “On Flourishing and Finding One’s Identity in Community.” Midwest Studies in Philosophy XIII. P. French, ed. South Bend, IN: University of Notre Dame Press, 324–341. Young, I. M. (1990a) “The Ideal of Community and the Politics of Difference,” in L. Nicholson, ed., 300–323. (1990b) Justice and the Politics of Difference. Princeton, NJ: Princeton University Press. (1990c) Throwing Like a Girl and Other Essays in Feminist Philosophy and Social Theory. Bloomington, IN: Indiana University Press.

Bibliography

269

(1997) “Gender as Seriality: Thinking About Women as a Social Collective,” in Intersecting Voices: Dilemmas of Gender, Political Philosophy and Policy. Princeton, NJ: Princeton University Press, 12–37. (2000) Inclusion and Democracy. Oxford: Oxford University Press. (2007) “Structural Injustice and the Politics of Difference,” in Multiculturalism and Political Theory. A. Laden, ed. Cambridge: Cambridge University Press, ch. 2. Young, J. (1993) Textures of Memory. New Haven, CT: Yale University Press. Young, R. (1980) “Autonomy and Socialization.” Mind 89: 565–576. (1986) Personal Autonomy: Beyond Negative and Positive Liberty. New York: St. Martin’s Press. Zimmerman, D. (1981) “Hierarchical Motivation and Freedom of the Will.” Pacific Philosophical Quarterly 62: 354–368. (1999) “Born Yesterday: Personal Autonomy for Agents without a Past.” Midwest Studies in Philosophy XXIII, 236–266.

Index

Adorno, Theodor W., 58 affect, 12–13, 33–36, 89–92, 96, 100, 118, 125, 143–144, 146, 158 agency, 9–10, 12–14, 30, 33, 57–61, 63–64, 80–81, 86–87, 112, 114, 136–139, 144, 147, 149–153, 173–175, 177, 184, 192, 197, 205–207 planning model of, 97 agential authority, 138, 145, 148–149, 153, 158 Alcoff, Linda, 198, 202, 205–207 alienation, 13, 100–101, 112, 118, 121, 143–148, 153, 154, 155, 158, 162, 166, 169, 172, 177, 180, 189, 209–210, 213–215, 225, 226, 239 Allport, Gordon, 32 Althusser, Louis, 56 Alzheimer’s disease, 93 amnesia, 93–97, 138 anterograde amnesia, 93 retrograde amnesia, 94 amour propre, 178–179 Anderson, Elizabeth, 195 Anderson, Joel, 181–182 anti-individualism, 33, 172 anti-perfectionism, 6, 16, 208, 224, 233 anti-perfectionist, 211 Appiah, Kwame Anthony, 123, 188, 194, 197–204, 210–212 Applebee, Arthur, 78 Arpaly, Nomy, 159 Ashmore, Richard D., 30 Atkins, Kim, 68 Audi, Robert, 123 authenticity, 14, 15, 25, 44–46, 62–63, 84, 101, 103, 113, 122, 128, 133, 134–135, 137, 141, 145, 146, 149, 151, 155–156, 158–159, 161–162, 167, 168, 171, 177, 183–185 autonomy and collective deliberation, 135–136, 208, 226 and democratic citizenship, 15–18, 135–136, 179–181, 208, 213–215, 240, 244

and value commitments, 13, 22–24, 39–46, 131, 141, 147, 148, 154, 164, 169, 171, 172, 182, 183, 191, 234 as a metaphysical concept, 9, 114 competence conditions for, 15, 133, 134, 141, 146, 154, 155, 156, 159, 168, 170, 176, 182, 183, 184, 226, 227, 228, 242 episodic and programmatic autonomy, 135 externalist accounts of, 34, 36, 138, 139, 140, 227, 231 global autonomy, 135–136, 171 hierarchical accounts of, 14, 136, 143, 172, 173, 177, 178, 180 internalist accounts of, 34, 138, 167, 168, 169, 227 local autonomy, 25–28, 135–136 proceduralist accounts of, 16, 17, 43, 169, 170, 171, 172, 174, 177, 181, 182, 224, 229, 230 protection and promotion of, 29, 38–41, 117–122, 181–185, 197, 215, 225–227 Badhwar, Neera, 32 Baier, Annette, 151 Bandura, Albert, 33 Barber, Benjamin, 39 Barclay, Craig R., 88 Barclay, Linda, 165 Barry, Brian, 123, 188, 198, 202 Barthes, Roland, 72 Bartlett, Frederic, 89 basic capabilities, 242–244, see also autonomy, competence conditions for Beiner, Ronald, 39 Bell, Daniel, 26, 44 Bellah, Robert N., 45 Benhabib, Seyla, 188, 193, 194, 198, 199, 215 Benson, Paul, 182 Berofsky, Bernard, 156–157 Bluck, Susan, 91 Bourdieu, Pierre, 24, 27 Bracken, Bruce A., 31

270

Index

271

Bratman, Michael E., 97–98, 138, 145, 146, 153 Brewer, William F., 87, 89, 90 Brighouse, Harry, 243 Brown, Wendy, 191 Bruner, Jerome, 70 Buchanan, Allen, 112 burdens of judgment, 231–233 Burkitt, Ian, 150 Buss, Sarah, 134 Butler, Judith, 52, 54, 56–61, 63, 64

embodiment, 3–5, 10, 36–39, 109–110, 113, 151–153, 205–207 emotion, 9, 12, 22, 24, 33–36, 71, 77, 81, 89–91, 125, 128, 139, 143–144, 153 endorsement constraint, 227, 228, 233 Evans, Gareth, 34 expression, 53–54, 61–65 non-discursive, 64 externalism, see autonomy, externalist conceptions of

circumstances of justice, 5, 18, 221–224, 243 Cochran, David Carroll, 198 Code, Lorraine, 176 cognition and social self, 33, 84, 90, 111, 118, 144, 150 Cohen, Joshua, 234 Cohler, Bertram J., 68 Colapietro, Vincent, 28 communities of memory, 45 conceptions of the good, 4, 114, 117, 118, 161, 235, 236, see also moral powers Conway, Martin A., 87, 90–91 Cooley, Charles Horton, 31 Copp, David, 197 Cranston, Maurice, 177 critical reflection, 26, 47, 118, 121, 122–123, 126–131, 140, 143, 157, 197 Crittenden, Jack, 166 culture, 38–39, 70–71, 115, 125–126, 166–167, 173–177, 187–214, 230–232, 236–238 Cuypers, Stefaan E., 27, 138

Faulkner, William, 73 Fischer, John Martin, 141–142 Fivush, Robyn, 91 Flax, Jane, 54 Foucault, Michel, 52, 54, 58 Fraser, Nancy, 192, 212 freedom, 114–116, 119–123, 196–197, 234 Freeman, Mark, 68 Freeman, Samuel, 229 Frohok, Northrop, 112 Frye, Northrop, 71 Fuss, Diana, 54

Dagger, Richard, 241 deliberation, 49, 58, 61, 64, 126, 130, 131, 135–136, 152–153, 159, 176, 184–185, 213–214, 226, 227–228, 234–237, 239, 241, 244 democracy, 3–6, 15–18, 47, 109–111, 172–180, 191, 215, 224, 237–238, 242–244 deliberative, 194, 225, 234 democratic theory, see democracy Dennett, Daniel, 78, 80 difference, 14, 65, 161, 173, 187, 195, 206 discrimination, 18, 190, 211–213, 221 dissociative identity disorder, see multiplepersonality disorder domination, 5, 18, 178, 194–196, 211, 221–225, 237, 243 Double, Richard, 150 Dreyfuss, Hubert, 62 Eagle, Morris, 129 Eagleton, Terry, 50 Elliot, Jane, 199

Gadamer, Hans Georg, 27 Gaus, Gerald F., 230 Geertz, Clifford, 32 Gergen, Kenneth J., 71, 84, 144 Gergen, Mary M., 71, 84 Giddens, Anthony, 10 Goldie, Peter, 68 Gordon, Linda, 32 Greimas, Algirdas Julien, 75 Grimshaw, Jean, 35 Grosz, Elizabeth, 50 Gutmann, Amy, 199, 202 Habermas, Jűrgen, 62, 64, 188, 215, 224, 227, 242 habits, 23, 26, 37, 82, 98, 124, 129, 191, 198, 206, 209 habitus, 24 Hacking, Ian, 68, 91, 199 Harter, Susan, 129 Heidegger, Martin, 27, 140 Hekman, Susan J., 27 Hirst, William, 94–95 Honneth, Axel, 192 Horney, Karen, 144 ideal theory, 220 identification, 41–43, 101, 143–146, 191–193, 200–201 identity as sameness, 82

272

Index

identity (cont.) hybridity of social identity, 188, 199, 203, 206, 207, 212 institutional identity, 115, 235 phenomenological account of, see Alcoff, Linda politics of, see identity-based interests identity-based interests, 11–12, 187–216 individualism, 21, 25–26, 164, 165 internalism, see autonomy, internalist conceptions of Jaggar, Alison, 164 Johnson, Mark, 62 Joyce, James, 75 Jussim, Lee, 30 justice, 114–119 autonomy-based conceptions of, 2–7, 46–47, 219–244 critiques of traditional liberal conceptions of, 11–18, 46–47, 164–165, 187–196, 210–216 Kagan, Jerome, 68 Kahneman, Daniel, 149 Kelly, M., 45 Kerby, Anthony Paul, 83 Kernohan, Andrew, 188 Kieslowski, Krzysztof, 76 Kitayama, Shinobu, 149 Klein, Stanley B., 91, 94 knowledge/power, 51 Korsgaard, Christine M., 150 Kristeva, Julia, 50, 52–54, 61, 63 Kymlicka, Will, 23, 43, 117, 119–123, 127, 188, 190–191, 196–198, 208 Lacan, Jacques, 51–53 Laclau, Ernesto, 16 Lakoff, George, 62 language, 31–34, 193–194 and the self, 9, 24, 48–55 indeterminacy of, 61–65 private language, 112 Larmore, Charles E., 16, 229 Lau, Joe, 33 legitimacy, 4–5, 17, 41–43, 110–117, 135–136, 224–244 Leitch, Thomas, 73, 75, 81, 83 Leland, Dorothy, 52 Levinson, Meira, 243 liberalism communitarian critique of, 21, 119–123 communitarian critiques of, 112 in the context of this work, 4–6, 16 individualistic nature of, 2–4, 39–41, 110–114, 119–120, 165–166, see also individualism

liberalism, critiques of, see liberalism, individualistic nature of and justice, critiques of traditional liberal conception of Lloyd, Genevieve, 190 Lukes, Steven, 223 Lyotard, Jean-François, 49, 52 MacIntyre, Alasdair, 75 Mackenzie, Catriona, 166 MacKinnon, Catharine, 18 Mahmood, Saba, 174 Margalit, Avishai, 188, 196, 201 marginalization, 12, 29, 104, 161, 177, 183, 194–195, 209–213 Markus, Hazel, 149 Marquez, Gabriel Garcia, 75 master-slave dialectic, 57 McLean, Kate C., 91 McLeod, Carolyn, 182 Mead, George Herbert, 27, 31 Mele, Alfred, 140–141, 142 memory autobiographical narrative memory, 10, 89–92 experiential memory, 87–89, 100 recognition memory, 87 veridicality of, 89–92 Merleau-Ponty, Maurice, 50 metaphor, 17, 24, 46, 50, 61–62, 65 Meyers, Diana Tietjens, 7 Michelman, Frank, 242 Mills, Charles, 202, 220–223 model conception, 7, 110, 114, 220, 233 Mohanty, Satya, 35 moral powers, 114–115, 116, 118 moral responsibility, 69, 133, 141–142 motivation, 22, 41–43, 49, 98, 128–129, 136–137, 153–154 Mouffe, Chantal, 16 multiple-personality disorder, 67, 68 Narayan, Uma, 174 narrativity causal narrative connections, 73–74 Diachronics, 78–79 directionality, 71 Episodics, 78–79, 101–104 functional (teleological) narrative connections, 74–76 tellability, 81–83 thematic unity, 76–83 Nedelsky, Jennifer, 167 Neisser, Ulric, 95 Nietszche, Friedrich, 49 Noggle, Robert, 101 non-alienation, see alienation

Index non-ideal theory, 221, 224 Nussbaum, Martha C., 35 oppression, 13, 173, 174, 195–196, 209–212, 220–223, 243 Oshana, Marina, 167, 168, 169, 170, 171, 172, 173, 176, 181 Overton, Lee, 157 ownership, 42, 182 Parekh, Bikhu, 201, 208 Parfit, Derek, 96 patriarchy, 18, 39, 52, 174 perfectionism, 14–16, 116, 171–174, 228–233 performativity, 54 personal identity, 67–70, 82, 96 personhood metaphysical accounts of, 4–5, 113 metaphysical conceptions of, 22–29 Phelan, Shane, 125 Phinney, Jean S., 199 pluralism, 9, 14, 116, 161, 228–233, 234 politics of difference, see difference post-modern thought and language, 48–55 power, 5–6, 49–61, 177–181, 183–186, 206–211, 223–226, 229–234 practical identity, 100, 148–156 primary goods, 114, 115, 207, 212, 233 Proust, Marcel, 24 psychoanalysis, 49, 51–54, 56, 129, 192 heretical ethics, 53 public reason, 227–228, 233–241 Putnam, Hilary, 33 Putnam, Robert D., 39 racism, 213 Radden, Jennifer, 68 rationality, 7, 12–13, 25–26, 44–46, 48–55, 62–63, 97–99, 139, 174–175, 195, 205–207 Ravizza, Mark, 141–142 Rawls, John, 5, 16–17, 41, 110–118, 135, 172, 180, 182, 220–225, 229, 231–243 Raz, Joseph, 16, 43, 170, 188, 196 recognition, see social recognition reflection, 33–37, 44–46, 78, 237 authentic self-reflection, see authenticity and self-reflection regress problem, 136, 146, 153, 161, 162 reidentification question, 69 relational conceptions of the self, 31, 32, 165, 184 responsiveness to reasons, 141 Ricoeur, Paul, 81–83 right to exit, 201–202

273

Robbe-Grillet, Alain, 75 Rorty, Richard, 27, 62, 63 Ross, Alan O., 68 Rotberg, Robert, 244 Rousseau, Jean-Jacques, 177–181, 231 Sandel, Michael J., 2, 120 Scanlon, Thomas, 239 Schacter, Daniel, 89 Schafer, Roy, 68 Schechtman, Marya, 69–70, 76 Scheman, Naomi, 35 Schneewind, Jerome B., 50 self, 118–119 as synchronic, 22 diachronic conception of, 8–10, 22, 99–105, see also narrativity, Diachronics I-self and Me-self, 30–33, 126, 131, 148–150, 156, 205, 209 non-propositional aspects of, 23–25, see also metaphor social constitution of, see relational conceptions of the self socio-historical conception of, 3–4, 6–12, 101–105, 115, 118, 219–225 self concept, 2–4, 30–32, 36–39, 67–72, 92–100, 111–112, 177–178, 200–205 self-constraint, 139 self-esteem, see self-respect self-governing policies, 138 self-interpretation, 45–46, 83–85, 96, 189 self-knowledge, 30–32, 53–54, 60–61, 97–100, 221, 227, see also self-concept inaccuracies of, see amnesia and post-modern thought self-reflection, 8–13, 22–33, 49–51, 56–61, 82–85, 126–131, 149, 156, 158 authentic self-reflection, 144–156 self-respect, 91, 115–118, 170, 182 self-schema, 31–32, 90–91, 102, 127, 149–152 self-trust, 181–183 self-understanding, see self-knowledge semiotic realm, see Kristeva, Julia Sen, Amartya, 123, 188, 199–203, 210 sense of justice, 114, 177, 236 Sheets-Johnstone, Maxine, 84 Sher, George, 165 Smith, Thomas S., 88 social embeddedness, 17, 26, 44, 163 social identity, see identity Social Learning Theory, 33 social mediation, 84 social recognition, 181 Spence, Donald P., 68 Stoljar, Natalie, 166, 174–177

274

Index

Strawson, Galen, 78–79, 84, 101–102 substantive independence, 169, 171 sujetissement, 55 Sullivan, Shannon, 129 sustained critical reflection (SCR), 152–156 symbolic interactionism, 31 symbolic order, see Lacan, Jacques symbolic realm, see Kristeva, Julia Taylor, Charles, 25, 27, 120, 188, 192, 193, 197–198, 204, 205, 208 Thalberg, Irving, 137 Thompson, Dennis, 244 Todorov, Tsevetan, 72 tradition, 45, 120, 197, 209–210 trauma, 51, 56 truth claims, 232 truth-functional analysis, 24, 62, 64–65 Tulving, Endel, 87 Turner, Mark, 76 Tversky, Amos, 149 van den Brink, Bert, 241 Vanderschraaf, Peter, 222

Velleman, David, 76–77, 80, 81, 150, 151, 159 violence, 5–6, 18, 211–213, 224, 243–244 Vygotsky, L. S., 27 Waldron, Jeremy, 191 Wall, Steven, 16, 43, 228, 229, 230, 232, 235 Waller, Bruce, 159 Watson, Gary, 137 Wegner, Dan, 31 Weir, Allison, 165 White, Stephen K., 49 Williams, Bernard, 125 Williams, Patricia J., 191 Wittgenstein, Ludwig, 27, 140 Wolf, Susan, 141 Wollheim, Richard, 90 Wong, David, 165 Young, Iris Marion, 188, 194, 196, 199, 206, 239 Zimmerman, David, 33–34