> C O N C E P T U A L I Z I N G A U T O N O M Y F O R E D U C A T I O N by D O N A L D K E R R B.Arts Sc., McMaster University, 1991 B .Ed . , The University of British Columbia, 1993 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department of Educational Studies; Philosophy of Education We accept this thesis as conforming to the required standard: T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A December 2002 ©Donald Kerr, 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of VuX\{ t Q A A ( ^1*1^4 The University of British Columbia Vancouver, Canada Date DE-6 (2/88) A b s t r a c t In this thesis I develop a conceptualization of autonomy that, I argue, is not only defensible within liberal theory, but does not.succumb to prominent criticisms raised of that theory. To do this I examine the requirements for a conceptualization of autonomy provided by our understanding of the tenets of liberalism, and also by the legitimate criticisms of liberalism raised by feminist and communitarian theorists. In particular I consider concerns about relational as opposed to atomistic conceptions of the self. I outline the criteria that a conceptualization of autonomy must meet if it is to be both useful and defensible, and I examine several prominent versions of autonomy against these criteria and show how I believe they fail to fulfill them adequately. I argue that conceptualizations of autonomy as sets of cognitive skills and abilities do not capture the criterion that a useful conceptualization reflect the liberal requirements of demonstrating an ability to act congruently with the demands of justice and equality. And conceptualizations of autonomy that discuss it in terms of being able and willing to make good decisions, or the kinds of decisions necessary of citizens in liberal democracies, are unclear on whether an action or decision must itself be seen as good in order to be considered autonomous. I suggest that autonomy is best understood as a descriptor of decisions and actions that meet specific criteria, and I show that such a conceptualization is useful in our discussions about education. ii C o n t e n t s Abstract ii Contents i i i Preface v Acknowledgements vi Introduction 1 Chapter 1: Arguments for Educating for Autonomy 6 Dearden and Reference to One's own Activity of Mind 7 Callan and Regulation of the Will and Liberal Virtue 13 Strike and Freedom to Choose 24 Conclusion 28 Chapter 2: The Liberal Context for Autonomy Liberal Theory Justice Criticisms of Liberal Theory Communitarian Criticisms Feminist Criticisms 29 30 30 34 44 53 56 57 68 Equality Liberty The Moral Status of the Individual Chapter 3: Criteria for a Conceptualization of Autonomy 82 Criteria From Liberal Theory 86 Criteria From Communitarian Theory 89 Criteria From Feminist Theory ' 93 Criteria From the Aims of Education 94 Summary of the Criteria for a Conceptualization of Autonomy 100 iii Contents iv Chapter 4: An Analysis of Current Conceptualizations of Autonomy 103 Dearden Revisited 103 Callan Revisited, and Brighouse 107 Strike and Callan/Rawls Revisited, and Gutmann 111 Chapter 5: Re-Conceptualizing Autonomy 117 The Liberal Conception of the Person 117 A Conceptualization of Autonomy 125 Discussion 129 Chapter 6: Autonomy and Education 141 Teaching for Autonomy 141 Application to Controversial Issues in Education 150 Autonomy's Relationship to Other Accepted Aims of Education 163 Bibliography 167 P r e f a c e Parts of the introduction and chapters 1 and 4 of this thesis have been published as: Donald Kerr, "Devoid of Community: Examining Conceptions of Autonomy in Education," Educational Theory 52 (Winter 2002): 13-25. A c k n o w l e d g e m e n t s This thesis could not have been written without the help and support of many people, nor would the writing of it have been as pleasurable. Jerrold Coombs has guided and challenged my thinking and my research for years, and to the extent that this thesis is organized clearly, is carefully thought-out and well defended, it owes much to his careful reading and constant feedback, and his support. LeRoi Daniels offered me much encouragement, help, and warm friendship throughout my graduate career. Murray Elliott offered me much kind support and guidance; he generously stepped in to co-supervise my project when I needed help. I wish he could have seen my work completed. I miss him. Pamela Courtenay-Hall also generously stepped in when I needed help, and has given much time and energy to my work, and important criticism. Linda Farr-Darling has always been very supportive and encouraging, and made me think deeply about the practical implications of my work. I have been blessed by many friends and a loving family. I cannot list all the friends who's friendship, company and good humour have made the last few years very pleasurable, but I wi l l name one or two who have especially given me much encouragement, support and love. Several years ago Sadeesh helped talk me into going back to school. I thank E d and L i z , Linda, Carol, Fiona and Lindsay, and many others. I am especially indebted to my own private benefactors, Mike and Jen. M y parents have always been my most patient teachers, and this thesis is dedicated to them. A n d I want to thank Melissa for giving the last few years much meaning, joy and love. vi For my parents v i i I n t r o d u c t i o n Autonomy is central to the liberal theory that underpins the provision of public education in democracies: a regard for autonomy is intrinsic to such important liberal values as freedom, rights, democracy, legitimacy, justice, and some versions of equality. And a belief in the right of all to participate in democratic life, and to be able to choose for oneself—at least in some sense—how to live one's own life, that is, what constitutes a good life and to pursue it, is a prime justification of education. It is therefore important that we have an understanding of just what we mean by the concept of autonomy, how it impinges upon our other liberal values, and what demands it places on education and educators in liberal democracies. Of course, this is not new to either philosophers of education or liberals; nonetheless the relationship described by autonomy, between the individual and the community, has come under strong criticism from communitarians and feminists. Communitarians claim that the liberal view of human nature is wrong: they say liberals deny that our knowledge of what counts as good, of what is valuable to us and our criteria for making value judgments comes from our history with others—our upbringing, our existence in a society and a culture, our social and cultural traditions. If liberals do indeed argue for this then they are arguing for a world in which there is little of value. As Charles Taylor says, "The self which has arrived at freedom by setting aside all external obstacles and impingements is 1 Introduction 2 characterless".1 But he is not alone in saying this; at least some liberals have made similar points. R. F. Dearden, for one, has argued that "The notion of 'choice' would seem to require as a condition of its intelligible application that there should be criteria by reference to which the choice is made, and which are not, for the moment at least, brought into question." 2 The important task is to know how individuals can be understood both as free and as existing in a context that provides us with our values, interests and morals. We need to know what would count as free and in what ways individuals can be helped to act freely without creating some kind of imaginary independent self. Alasdair Maclntyre has characterized the liberal position as claiming that "I am what I myself choose to be. I can always, i f I wish to, put in question what are taken to be the merely contingent social features of my existence." In contrast, Maclntyre suggests, "the key question for men is not about their own authorship"; he has famously claimed that the individual cannot answer the question 'What am I to do?' until they have first answered the question 'Of what story or stories do I find myself a part?' 4 But it is important to challenge this claim; Maclntyre is right that liberalism is empty if the liberal view of the person deciding what to do denies the importance of one's personal history. But we should also worry about just how prescriptive one's narrative and embeddedness is understood to be. We must find a way to understand the person as embedded in 'Charles Taylor, Hegel and Modern Society (Cambridge, England: Cambridge University Press, 1979), 157. Quoted in Wil l Kymlicka, Liberalism, Community, and Culture (Oxford: Clarendon Press, 1989), 47. 2 R. F. Dearden, "Autonomy and Education", in Education and the Development of Reason, eds. R. F. Dearden, P. H. Hirst and R. S. Peters (London: Routledge and Kegan Paul, 1975, 448-65), 457. 3Alasdair Maclntyre, After Virtue: A Study in Moral Theory, Second edition (Indiana: University of Notre Dame Press, 1984), 220. 4Ibid.,216. Introduction 3 a context and at the same time able to reflect on and make responsible decisions about this context and his or her own course of action. Among the many important criticisms feminists have raised of liberal theory are the criticisms of conceptualizations of autonomy as too abstract, and as premissed on a conception of the person devoid of any important, caring relations with others. Some of these criticisms take the mother-child relationship, and the close physical and emotional bonds felt between the mother and her child, as well as the child's biological dependency on the mother for nurturing and survival, as the protypical relationship. The argument is that any conception of the person that does not take into account the person's dependence on others for emotional and physical health, and ultimately survival, is abstracted to the point of leaving behind important features of what it means to be human, and therefore of not much use. Marilyn Friedman has summarized much feminist work as requiring that ethical theory in general, and conceptualizations of autonomy in particular, must "recognize the moral importance of emotions, close personal relationships, social relationships generally, and the non-impartial nature of any actual ethical standpoint."5 This is an important point; any notion of autonomy that is to be useful in liberal and educational theory must obviously envision persons as, for the most part, living together, in living and breathing human communities in which we find much of our sustenance and joy. This thesis is an attempt to present a conceptualization of autonomy that is clearly based in liberal theory, does not succumb to the legitimate criticisms of that theory, and is clear enough to be useful to people engaged in discussions about education. To do this I first show how we can understand individuals as deriving their identities from their communities and families, and the traditions 5 M a r i l y n Friedman, "Feminism in Ethics: Conceptions of Autonomy", in The Cambridge Companion to Feminism in Philosophy, eds. Miranda Fricker and Jennifer Hornsby (Cambridge: Cambridge University Press, 2000, 205-24), 212. Introduction 4 these are a part of, in a way that recognizes the importance of relationships to the individual, but that still allows us to understand the individual as capable of autonomous actions. In the first chapter I survey the arguments that have been advanced for a concern with autonomy in educational discourse; this should make clear the need for a clear understanding of the term. But considering the arguments that have already been put forward for autonomy leads us to a consideration of the ways the term has already been conceptualized. Examining these, makes it clear that difficulties remain in current discussions of autonomy. In the second chapter I discuss the liberal theory that makes sense of the idea of autonomy. I also consider the major criticisms raised of this body of theory, and I explain where I believe these criticisms fail, and where they raise serious concerns that liberal theory must contend with. Both the liberal theory and the relevant concerns with it leave us with criteria that any defensible conceptualization of autonomy must meet. I discuss and summarize these criteria in chapter 3. In chapter 4 I use these criteria to examine the versions of autonomy that I introduced in chapter 1. Each of these conceptualizations fails, when examined against these criteria, in some way. But, at the same time, examining these conceptualizations in this manner helps to clarify our thinking with regard to autonomy. In chapter 5 I provide a conceptualization of autonomy that is intended to meet the criteria I laid out in chapter 3. The thrust of my argument is that autonomy is best thought of as describing choices and actions as meeting certain standards. These standards are derived from liberal theory. In the final chapter I show that the conceptualization of autonomy that I have drawn is useful to discussions of education. I do this by showing that my description of autonomy helps to clarify and buttress some of our current educational practices and aims, helps to clarify our thinking about some Introduction 5 prominent controversies in education, and makes clear what teachers have to do in order to foster autonomy in their students. C h a p t e r 1: A r g u m e n t s f o r E d u c a t i n g f o r A u t o n o m y In this chapter I will survey the arguments advanced by several well-known contemporary philosophers of education for educating for autonomy. I have chosen these particular defences for, and descriptions of, autonomy because the authors clearly situate themselves within liberal theory, my own starting place for this project, and because each of these authors either gives important reasons for concerning ourselves with autonomy, or an interesting conceptualization of it, or both. As well, these authors have each significantly influenced the field of liberal educational theory and therefore their descriptions represent much of the consideration given this topic. Each argument for the need to concern ourselves with autonomy entails its own conceptualization of autonomy, and I will outline the conceptualization each author puts forward. This survey is intended to demonstrate several important features of the arguments put forward for autonomy. First, taken together, these authors suggest that there is much considered opinion supporting the idea that the concept of autonomy has an important role in educational discourse (and in political philosophy more generally), and each author provides some of the reasons for this. However, and second, these descriptions of autonomy also show that problems remain in conceptualizing it and in understanding its role in liberal and educational theory; indeed, the fact that these descriptions do not always agree with one another is one important indicator of these problems. 6 A rguments for Educating for A utonomy 1 I do not intend to suggest that this survey exhausts the current state of educational discourse on autonomy, or in any way reflects everything important that has been said about it. But these examples do offer a fair representation of the arguments extended for a concern with autonomy, and of current concept-ualizations of it. Examining them makes clear that, although good reasons have been given for concerning ourselves with autonomy, and there already exist carefully thought-out conceptualizations of autonomy with important implications for education, there remains work to be done, both in articulating autonomy's role within a liberal theory of education and in providing a clear and educationally useful conceptualization of it. DEARDEN AND REFERENCE TO ONE'S OWN ACTIVITY OF M I N D R. F . Dearden argued in 1972 that a host of changes that were then taking place within both educational discourse and practice reflected autonomy as a new aim in education.1 He cites examples such as the use of terms like 'self-direction' and 'independence,' and the increasing emphasis on such ideas as individualized learning and learning how to learn for oneself. Children were increasingly expected to be independent and self-directed in their schooling, and he suggests that this probably has its roots in a world in which adults are increasingly encouraged to make choices for themselves.2 Underlying these changes, he points out, is an assumption that every child wi l l want, or should want, more autonomy. He says that we do not put up with individual differences or predispositions for autonomy, and this is comparable to our expectations around such values as honesty or fairness. "If children are dishonest, we do not simply adjust to that as a 'R. F. Dearden, "Autonomy and Education," in Education and the Development of Reason, eds. R. F. Dearden, P.H. Hirst and R. S. Peters (London: Routledge and Kegan Paul, 1972), 448-65. 2Ibid., 448-9. A rguments for Educating for A utonomy 8 brute fact of individual difference, but try to explain and alter it. A n d so it is with individual differences in degree of autonomy." Is autonomy to be valued primarily for its own sake, or for its utility—for the other things it makes possible? Dearden says autonomy does have a utilitarian value, as "a quality which must be possessed in relevant ways in all roles which afford much independence of thought and action to their occupants." But, he says, to think of it this way "is to take an employer's view of it. To the individual agent it is the job which is likely to have utility, and precisely in affording opportunities for the exercise of autonomy."4 There is also an odd sort of utilitarian value that arises when there is no consensus as to what children should be learning; in this situation it may be felt that children "may as well choose for themselves." But this is to allow autonomy to arise in education by default, rather than as an aim. Autonomy's primary value, Dearden argues, is intrinsic. We notice early in children's lives how much they enjoy doing things for themselves, and we know how much satisfaction and pride we find in our own work. "Other things being equal, the exercise of such an autonomy wi l l be a source of considerable satisfaction. The accomplishment of what we want or intend, under the description embodied in the intention, is necessarily a satisfaction, and our satisfaction is the greater the more there is of what we intend in what we accomplish." 5 Dearden also claims that if we consider the examples of those around us, we readily accept that such autonomy wi l l necessarily be an important part of a person's self-concept. "As such, it wi l l be an important part of his dignity, or sense of personal worth, and its exercise wil l be claimed as a right to be respected by others."6 3Ibid., 449. 4Ibid., 459-60. 5Ibid. 6Ibid., 460-61. Arguments for Educating for Autonomy 9 So what does it mean to be autonomous in terms of making choices for oneself, or independent-mindedness? Dearden explicates autonomy as existing in the individual to the degree that what he thinks and does in important areas of his life cannot be explained without reference to his own activity of mind. That is to say, the explanation of why he thinks and acts as he does in these areas must include a reference to his own choices, deliberations, decisions, reflections, judgments, plannings or reasonings. This would not necessarily apply in every instance, since what a man thinks now, and how he acts now, may reflect an activity of mind engaged in hours, days or years previously, but without losing its force as the explanation. 7, He also adds that autonomy exists when what requires reference to one's own activity of mind are thoughts and decisions about relatively important areas of one's life: we tend not to think of as autonomous the person who only decides for her or himself what to wear each day. Intuitively, decisions about our "work or domestic life are normally very important," to which we might also add decisions about our values and the moral choices we are faced with. What is involved, then, in autonomy for Dearden is the ability to reason about one's own choices and the fact of doing so or having done so in the past. Indeed, his definition of a person's autonomy as explanations about his or her thoughts and actions requiring reference to his or her own activity of mind seems to be more a description of how to recognize someone as autonomous, where autonomy itself is the ability to reason about such things and the fact of someone's having exercised that ability in making important life decisions in the past. Implied here is having direction and control over one's own life in important ways. Certainly one's own rational activity does not exhaust the concept for Dearden, because he then goes on to examine whether reason can be a threat to 7Ibid., 453-54. 8Ibid., 453. Arguments for Educating for Autonomy 10 autonomy, and this only makes sense as a question i f autonomy is not identical with reason.9 For Dearden autonomy is a matter of degree. It also reflects a particular sort of evaluation of one's mental activity: we would not say, for example, that someone who worries incessantly over what her father would think about her choices, and therefore chooses accordingly, or who tries desperately to remember what his favourite comic book hero said in just such a situation, is acting autonomously. To judge someone as autonomous is to make the normative judgment that his or her reasons for choosing as he or she did were, in some way, both good and important ones. Dearden points out some other characteristics of autonomy. First, personal autonomy is not an intellectual or academic ideal manifested only in a few who put a lot of intellectual effort into considering their own values or opinions. Rather it is or can be displayed by almost everyone in all sorts of daily activities and decisions. Dearden admits that there is some difficulty here, as he has already said that many instances of personal choosing—the trivial ones—are not, for him, good examples of the exercise of autonomy. But many of our day-to-day activities do allow for the exercise of autonomy: "buying things, in choice of job or the way that a job is interpreted, in arriving at a particular sort of domestic arrangement, in the uses that are found for leisure and so on ." 1 0 The important point is that autonomy need not be seen as a rare ideal but rather something almost all can exhibit in our day-to-day lives. Nor, he adds, must our thoughts and actions be original, or even original to us. They "may have been learned from others, but [they] must have been made our own and not just acquiesced in or been merely 9Ibid., 456-59. 10Ibid., 454. Arguments for Educating for Autonomy 11 lodged in our minds by drilling or repetition." 1 1 We make such things our own by reflecting on them, and by consciously incorporating them into our own understanding or behaviour. Dearden claims that none of this implies that the autonomous person does not agree to live by laws, keep promises, observe conventions or accept any authority. How, then, does the autonomous actor internalize rules autonomously? Dearden does not explain this, nor does he suggest the extent to which one could be.expected to provide an autonomous defence of a statement such as 'I chose to follow the rules.' Gerald Dworkin has characterized some views on autonomy as requiring that persons "not defer independent judgment." 1 2 Proponents of this view sometimes insist that this means that agreeing to live by laws or keep 13 promises, for example, is actually to forfeit autonomy. ' Dearden obviously does not want to go this far, but he must clarify just what the mechanism is by which some people autonomously internalize rules while others do so heteronomously. 1 4 Second, autonomy cannot be absolute in the sense of the description of the activity in a person's mind being entirely describable in terms that solely refer to that individual. A l l descriptions wi l l , in turn, eventually be describable in terms of things a person learned. (To which Dearden adds the possibility that the description wi l l include a person's genetic makeup.) We might add to this that even in the case where an individual does arrive at an entirely unique thought or decision, the conditions that make this possible can be described in terms of a social history—the knowledge upon which this is built, the language learned that "Ib id . l 2 Gera ld Dwork in , The Theory and Practice of Autonomy (New York : Cambridge University Press, 1988), 22. l 3 Robert Paul W o l f f s is the most prominent of these. See his In Defense of Anarchism (New York : Harper & Row, 1970), 15. l 4 I f autonomy is taken to mean some version of 'self-rule,' heteronomy is generally understood to be its opposite, that is, some version of 'ruled by others.' Arguments for Educating for Autonomy 12 facilitates her thinking, and so on. So it is important to make the distinction between, on the one hand, an autonomous answer and its sequence of reasons leading back to external conditions and facts and, on the other, a non-autonomous answer and its reasons that also rely on facts or conditions that do not originate with the individual. Third, Dearden claims that a person's reasoning is autonomous even when she reasons with wrong information, or wrong criteria. He gives the examples of a person who has historical beliefs based on reading Genesis, or a person planning a crime without any thought of moral principles. So, for Dearden, truth and morality are not required for autonomy. Interestingly, though, Dearden adds to this another point that seems to constitute an exception to this last claim. Dearden claims that an increase in self-knowledge is necessary for autonomy and in this case, he argues, the beliefs must be true beliefs. This is because "the better we know our motives, wishes, purposes, typical reactions to others and so on then the greater is our possibility of bringing our thought and action under conscious control." 1 5 But in order to consciously take control of beliefs and impulses that would otherwise be unconscious we must have true beliefs about them—for if our beliefs are wrong, then we remain at the whim of what is actually happening in our unconscious. If this is an exception to the rule, it is worth using it to raise questions about the rule which gives rise to it. For instance, why is the person who makes a judgment based on historical beliefs influenced by Genesis to be seen as being able to give adequate reasons in support of her or his decision, instead of also remaining at the whim of other forces, in this case poor beliefs? This question is brought into relief when we contrast this person's ability to give reasons with the l5Dearden, 456. Arguments for Educating for Autonomy 13 ability of the historian who has given consideration to a wide range of sources, and can articulate defensible criteria for choosing one over another. C A L L A N A N D R E G U L A T I O N O F T H E W I L L A N D L I B E R A L V I R T U E Dearden's conceptualization is not the only way autonomy has been described based on arguments for its intrinsic worth. In his book Autonomy and Schooling Eamonn Callan also describes autonomy as an ideal with intrinsic value. 1 6 He fleshes out a conceptualization of it that is intuitively appealing, and uses it to explain why freedom should be considered a constituent of autonomy. This leads him to a slightly different description of autonomy. The point of Callan's argument is to defend the need to promote freedom in education, and to show that this is best achieved through schooling that reflects the child-centred tradition. But in arguing for the importance of freedom he provides two important arguments for a concern with autonomy in education and in society in general. The first is that it is intrinsically valuable: autonomy already has such a central value for us as a concept that we intuitively know its importance both to a liberal society and to education. Callan describes what he believes the important aspects of autonomy are for us and why it should play such a central role. The second argument for the importance of autonomy to education is that it is intimately connected with, and helps to provide a useful framework for evaluating, other important values we hold, such as self-respect, moral virtue and, most importantly for Callan's purposes, freedom. In other words, were autonomy not considered important, it would not be clear why such things as freedom should be important; nor would it be clear what kinds of freedoms should be taken most seriously, how they are to be evaluated, and so forth. A l 6Eamonn Callan, Autonomy and Schooling (Montreal: McGill-Queen's University Press, 1988), chapter 2. Arguments for Educating for Autonomy 14 conceptualization of autonomy thus has explanatory importance in our moral framework. Callan recounts Robert Nozick 's well-known argument for the intrinsic value of autonomy. Nozick suggests the mental experiment of imagining a machine we could plug into that could create any experience in our brain that we desired. "Superduper neuropsychologists could stimulate your brain so that you would think and feel you were writing a great novel, or making a friend, or reading an interesting book. A l l the time you would be floating in a tank, with electrodes attached to your brain." 1 7 Nozick assumes that many of us would find the idea of such a machine, or at least the idea of permanently hooking up to such a machine, repulsive. The point of the experiment is that there are many things we value for their own sake that are not simply experiences. "If the joys of winning a game, making love, or making scientific discoveries could be mechanically produced, one could still regard these experiences as a poor substitute for the reality of victory, love-making, or scientific discovery, even i f the joys were the same in either case." 1 8 Now we can expand on the capabilities of this machine, or add new machines, i f we felt that there are other things that would be lacking in a life hooked up to one. Suppose we would not want to be hooked up to such a machine because we felt there are things we would like to achieve, or become. Nozick writes: "There is no answer to the question of what a person is like who has long been in the tank. Is he courageous, kind, intelligent, witty, loving? It's not merely that it's difficult to tell; there's no way he is. Plugging into the machine is a kind of suicide." 1 9 Nozick goes on to suggest that perhaps our displeasure could be 17Robert Nozick, Anarchy, State, and Utopia (Oxford: Basil Blackwell, 1974), 42. 1 8Callan, 43. l 9Nozick, 43. Arguments for Educating for Autonomy 15 relieved i f we were also hooked up to a transformation machine that could make us the person we would like to be. But there still seems to be something lacking in such a life, and we assume it is something like our l iving our own lives for ourselves, however imperfect or imperfectly satisfying that may be: "Perhaps what we desire is to live (an active verb) ourselves, in contact with reality. (And this, machines cannot do for us.)" 2 0 That is, we consider autonomy (and not merely the experience of exercising autonomy or the achievement of becoming an autonomous person) to be intrinsically valuable, and considerably so. Autonomy also has considerable constitutive value for us. Callan argues that autonomy is essential to the exercise of moral virtue on the grounds that the morally virtuous person must act in ways that demonstrate realism, or "that the wil l is regulated in a realistic fashion." 2 1 This is because the virtuous person has to choose between virtuous actions and "propensities which are apt to conflict with that aspiration" to be virtuous. Without a sense of realism a person may conceive her choices in such a way as to obscure the true moral qualities of those choices, or she may justify past unjust actions on the grounds of their benefits to others. And, because our desire for the positive regard of others wi l l not always compel us to act in the most moral manner, a high degree of independence of mind is also required to be virtuous. Callan gives the example of a parent who must choose between doing what is morally best for his child and his desire to be loved and admired by his child at all costs. Public choices may similarly require that one choose in a way that does not meet with the approval of those whom one wants to please. For Callan, since both realism and independence of mind are components of autonomy, autonomy is necessary for the exercise of moral virtue. 2 0Ibid., 45. A l l italics appearing in quotations are from the original. 2 lCallan, 46. Arguments for Educating for Autonomy 16 To show that autonomy is necessary for self-respect, Callan differentiates between self-esteem and self-respect. " M y self-esteem is determined by the value I attach to what I am and have done, my self-respect is determined by the extent to which I can take credit for the value of what I am and have done." 2 2 So although a malicious criminal could appropriately value her own personhood, and therefore have self-esteem, she would not have grounds for a sense of self-respect. Given this distinction, "We can take credit for what is valuable in our lives, and thereby enhance the degree to which we have justifiable self-respect, only to the extent that we create this value through the exercise of personal autonomy." ~ Consider, Callan suggests, a person who had lost the ability to walk but then recovered it by means of his own heroic efforts. This person has reason for much self-respect in this case because her success depends greatly upon elements of her self-rule, or autonomy. For most of us, however, our ability to walk is not a source of self-respect because this has little to do with our self-rule. Autonomy is therefore partly constitutive of at least two other things we value for their own sake, the exercise of moral virtue and the achievement of justifiable self-respect; and because these things are very plausibly regarded as weighty intrinsic values, autonomy can be seen, with equal plaus-ibility, as a very weighty constitutive good. . . . The conceptual connections which make autonomy a constitutive good yield an especially strong justification because it is not the case that only minimal autonomy is necessary to evince moral excellence or achieve a robust and well-grounded sense of self-respect; it is rather the case that the degree to which these intrinsic values are realized necessarily depends on the degree autonomy is realized. 2 4 Callan says that this reinforces our feelings toward Nozick 's thought experiment—we could not fail to exercise our autonomy and still have everything z Ibid. , 47. 3 Ib id . 4 Ib id . , 45 . A rguments for Educating for A utonomy 17 else we desire because autonomy is partly constitutive of at least some of the things we value deeply. What, therefore, does autonomy look like for Callan? For him autonomy, or self-rule, "essentially pertains to the regulation of the w i l l . " 2 5 This regulation of the wil l occurs through the control one exercises over what Callan calls our 'motivational structure'—"that set of propensities which gives general shape to the individual self and the course of her l i fe . " 2 6 Our motivational structure largely has to do with our interests. Regulation of the wil l is also evident in how our propensities are brought to bear in choosing between options. And , in a related manner, it occurs in the control we exercise in forming our beliefs. We all have a variety of goals, interests and desires. Many of these wil l conflict—at least in their demands on our time. We have to choose how we wi l l divide our time and our efforts between them, as well as how best to pursue them. This choosing wil l itself reinforce some propensities and help to diminish others: "the propensities we cultivate . . . incline us toward specific options [and, in turn,] the disciplined choosing of certain things rather than others . . . strengthen[s] 27 some propensities and weaken[s] or preclude[s] the development of others." Realism and independence of mind, evinced in Callan's argument for the importance of autonomy to the realization of both self-respect and virtue, are important components of autonomy. But note that this implies a certain normative judgment about the control one is exercising over one's w i l l . I may spend a great deal of time daydreaming about a particular implausible event, but during that time I may come to convince myself that what I dreamed about was not only possible but likely. This in turn would have an effect on my interests, plans, 25 Ibid., 26. "Ibid. 27 Ibid. Arguments for Educating for Autonomy 18 decisions about dividing my time and so on that Callan ascribes to the autonomous individual—except that, in this case, I would not be displaying much realism and would likely only create frustration for myself. Callan's description of autonomy does not make explicit that the regulation of the wil l must be toward certain realistic and worthwhile ends, or done according to certain normative criteria, but it implies this. As it is largely in the regulation of our interests that, for Callan, we display self-rule, it is worth describing what he means by interests. Interests in his sense are not judgments of what wi l l benefit us—as something that is in my interest. Rather it is an expression of a feeling toward something, of a desire to give something one's attention and effort in specific ways—as something that I am interested in. Callan says that the particular desire that interests carry have a necessity that "distinguishes feelings of interest from mere l ikings ." 2 8 A n interest is an expression of our state of mind or a feeling, but it can also refer to tendencies to engage in certain behaviour, devote time and effort, and so on. "This is a sense of 'being interested' which requires a certain coherent pattern of mental activity, composed of intermittent feelings of interest and inclinations to act in ways which sustain these feelings." 2 9 Notice that to profess an interest is to make a claim about our own feelings, our own state of mind. This is something we (usually) are not mistaken about—the way we are sometimes mistaken about judgments of what is in our interest. Notice also that it is to make a value judgment: "One does not really have an interest in [something] unless one ascribes intrinsic value to experiences and achievements internal to [that thing]." 3 0 'Ibid., 27. 'ibid., 28. 'ibid., 29. Arguments for Educating for Autonomy 19 Callan interprets John Dewey's claim that we identify with the things we are interested in to mean that our sense of self is in large measure derived from our interests. A person who lost all her interests would be a very different person, and one for whom it is hard to imagine that life held much meaning. Indeed, it is hard to imagine a person without any interests at all. If it is true that our own identity and sense of meaning comes in large part from our interests, and we can choose, evaluate and pursue or act to diminish our interests, then, Callan says, we can rationally have an impact on our sense of self. "The shaping of a human being's interests and the way this determines decisions, and hence conduct, in interaction with other propensities are mental processes subject to reason."31 Thus the pursuit of our interests is not just one important area where we exercise our autonomy, it is the means by which we exercise our autonomy in determining who we are, or our own sense of ourself. Callan maintains, reasonably, that we can evaluate and shape our interests, and therefore shape important elements of who we are, and our sense of self. Interests presuppose desires, but they also involve certain beliefs about those desires. Callan uses Charles Taylor's term of 'strong evaluation' to describe the kind of beliefs we have—based on the judgments we make—about our different desires. "The strong evaluator assumes that 'some desires or desired consummations can be judged as bad, base, ignoble, trivial, superficial, unworthy and so on,' and the judgments she makes according to these criteria will profoundly affect whichever desires eventually prevail in her conduct."32 A person therefore brings strong judgments to bear on the interests that she will pursue and those she will curtail. 3 1Ibid., 30. 3 2Ibid., 29. Arguments for Educating for Autonomy 20 More recently Callan has provided a different argument for concerning ourselves with autonomy. In Creating Citizens: Political Education and Liberal Democracy he argues that certain virtues are necessary for the maintenance of liberal democracies, and that liberal democracies provide for many of us a way of -in life that is inherently worthwhile and good. These virtues are premissed on a conception of autonomy. Callan limits the scope of this argument by using Rawls ' notion of political liberalism to focus on the public, political virtues that demand autonomy. Granting the success of this argument, he says, gives us an important argument in favour of autonomy and leaves only to be shown that autonomy does not have negative consequences. This is a much more manageable task than attempting to show that autonomy "makes our lives better or worse all things —not just the political things—considered." 3 4 The argument underlying both Rawls' conception of political liberalism and Callan's use of it to support an education in liberal virtues is that the maintenance of justice in society requires certain virtues on behalf of its citizenry. In particular, the political morality is constructed from the political conception of the person. What does this involve? According to Rawls, in order for persons to be considered full participants in a fair system of social cooperation, we ascribe to them the two moral powers connected with the elements in the idea of social cooperation . . . : namely, a capacity for a sense of justice and a capacity for a conception of the good. A sense of justice is the capacity to understand, to apply, and to act from the public conception of justice which characterizes the fair terms of social cooperation. . . . The capacity for a conception of the good is the capacity to form, to revise, and rationally to pursue a conception of one's rational advantage or good. 3 3 Eamonn Callan, Creating Citizens: Political Education and Liberal Democracy (Oxford: Clarendon Press, 1997). See, in particular, the first two chapters, although the entire book is based on the connection Callan describes between a conception of autonomy and a necessary set of liberal virtues. 3 4 Ib id . , 42. Arguments for Educating for Autonomy 21 In addition to having these two moral powers, persons also have at any given time a determinate conception of the good that they try to achieve. Such a conception must not be understood narrowly but rather as including a conception of what is valuable in human life. Thus, a conception of the good normally consists of a more or less determinate scheme of final ends, that is, ends we want to realize for their own sake, as well as attachments to other 35 persons and loyalties to various groups and associations. Rawls claims this allows us to think of persons as free and equal: The basic idea is that in virtue of their two moral powers . . . and the powers of reason . . . persons are free. Their having these powers to the requisite minimum degree to be fully cooperating members of society makes persons equal.j6 These powers, therefore, are required to participate in creating and maintaining justice in the public sphere. But by themselves they will not necessarily ensure that citizens will be able or willing to co-operate; the range of values and beliefs embodied in individuals' comprehensive doctrines is often too much for society to bear. The existence of divisive pluralism within our democracies along with the criticism that his earlier work did not accommodate a very broad range of comprehensive doctrines led Rawls to reformulate his ideas in Political Liberalism. According to Rawls the existence of divisive pluralism demands a sense of reasonableness. This is the particular form of moral sensibility that underlies the desire to engage in fair cooperation as such, and to do so on terms that others as equals might reasonably be expected to endorse. I do not assume the reasonable is the whole of moral sensibility; but it includes the part that connects with the idea of fair social cooperation.'7 What does this entail? 'John Rawls, Political Liberalism (New York: Columbia University Press, 1993), 19. 'Ibid. 'ibid., 51. Arguments for Educating for Autonomy 22 The first basic aspect of the reasonable. . . is the willingness to propose fair terms of cooperation and to abide by them provided others do. The second basic aspect . . . is the willingness to recognize the burdens of judgment and to accept their consequences for the use of public reason in directing the TO legitimate exercise of political power in a constitutional regime. The burdens of judgment are the sources of disagreement between reasonable people. In at least their public lives, citizens in democracies must exhibit a willingness and ability to participate in discussion about how they are to get along, and they must be will ing to abide by agreed-upon rules in the knowledge that others wi l l too. They wil l also understand that they wi l l not always come to agreement on fundamental values but they wil l require that the terms of co-oper-ation accommodate this disagreement. Rawls says that in order for people to overcome these sources of disagreement—that is, to find ways to work around and accommodate the varying comprehensive doctrines—the "scope of what reasonable persons think can be justified to others" must be limited. Those comprehensive doctrines that accommodate his conception of public virtue are considered 'reasonable comprehensive doctrines' and it is upon these that Rawls ' Political Liberalism is based. Rawls intends that most of the comprehensive doctrines found within our plural societies wi l l be regarded as reasonable and therefore his conception wi l l provide a theoretical framework for understanding the common ground upon which the members of a plural society come together.4 0 J 8 Ib id . , 54. 3 9 I b i d „ 59. 4 0 Cal l an ' s disagreement with Rawls is that the range of comprehensive doctrines that wi l l actually accommodate Rawls ' framework for a just society is actually much narrower than Rawls intends. This is because Callan believes that the standard of reasonableness required of citizens by Rawls ' conception is in actual fact very high; he says that the "ideal embodies an onerous standard of public virtue rather than some negligible threshold of reason that virtually everyone exceeds. The ideal entails that any doctrine held in a way that repudiates the burdens of judgement, for example, pushes the believer beyond the borders of reasonable pluralism." Creating Citizens, 30. Arguments for Educating for Autonomy 23 According to Callan, the capacities and powers required for freedom and equality in the state, and the abilities and dispositions entailed in the notion of reasonableness required for public co-operation, presuppose a notion of autonomy. Callan does not explicate this autonomy other than as 'reasoned self-rule,' but he says that its exercise is not only necessary for the development of civic responsibility but that it wil l also at times work against i t . 4 1 This is because individual reflection wil l sometimes lead people away from (or, we can add, against) the terms of political agreement. So long as a liberal state (and its education) allows for the development and pursuit of individual conceptions of the good, it must also allow that some people wil l pursue conceptions that are, at least in appearances, inimical to it. Although Callan does not explicitly describe this form of autonomy, a list of the abilities and dispositions required by the liberal virtues described by Rawls and used by Callan wil l give some description of it: Autonomous citizens able to participate in a liberal society understand, and are able to reflect on and act within the norms prescribed by, a notion of public justice. They have the "capacity to form, to revise, and rationally to pursue a conception o f their own good and they hold such a conception. They are will ing to engage in public discourse, are able and wil l ing to propose and discuss fair terms of co-operation with others, and they are wil l ing to abide by the terms of co-operation so long as these terms are fair and others are also wil l ing to abide by them. Disagreement wi l l arise in society over worthwhile conceptions of the good, the ends that are to be pursued, and the support that various potential goods are to receive from the state. Autonomous citizens are wil l ing to accept this disagreement and its sources, as well as the fact that the terms of co-operation must accommodate the various beliefs as 4 l Ib id . , 11. Elsewhere in Creating Citizens Callan does provide a definition of autonomy (148). But it is a rewording of the conception presented in Autonomy and Schooling and described above, and not explicitly tied to the political virtues discussed here. Nor are additional reasons given in support of this particular description. Arguments for Educating for Autonomy 24 comprehensively as possible. And they have the general powers of deliberation, reason and judgment required to conduct one's own affairs and to participate in the state 4 2 The argument that autonomy is necessary because liberal democracies depend for their survival on it, and liberal democracies provide for many of us a way of life that is inherently worthwhile, is an unusual one for a liberal to take. It is a form of argument (if not in specifics) more common among such opponents of liberalism as communitarians. It is not, by itself, decisive in refuting the challenge posed by those who do not hold liberal democracy as the preferred form of political association or who value ways of life other than those acceptable in a democracy. Callan adds to this by arguing that one cannot have a sense of justice without some form of personal autonomy. He then argues that autonomy does not interfere with the pursuit of other worthwhile goods (such as personal attachments, relationships and commitments), thus attempting to counter criticisms of autonomy raised by others, such as communitarians. 4 3 I w i l l survey the arguments raised by communitarians in the next chapter; their criticisms of autonomy and its role in libera] theory require consideration before expanding on a conceptualization of autonomy. S T R I K E A N D F R E E D O M T O C H O O S E Callan's latter argument in favour of autonomy is, very generally, that it has good consequences. Kenneth Strike has also argued for autonomy based upon its 4 2 I should emphasize that I have not by any means described in entirety the treatment Cal lan gives autonomy in this book. I have only described one key argument he presents that is of particular use to my project, but his whole book provides a careful consideration of the implications of autonomy on education in a liberal society. These present us with further arguments tor autonomy, generally of the form: educating for autonomy has these ends, and these are desirable ends, therefore autonomy is desirable. In doing this he considers a broad range of educational issues such as patriotism and the conflict between private values and public schooling. 4 3 Ib id . , § § 1 6 - 2 1 . A rguments for Educating for A utonomy 25 desirable consequences; his argument is that having a notion of autonomy is the only way we can give meaning to the idea of moral responsibility. In his book Liberty and Learning Strike is concerned with identifying an appropriate and defensible notion of the authority of ideas and associated practices such as academic freedom and student rights. He argues for a particular understanding of the student-teacher relationship and the demand on teachers to give reasons to students for what they are being asked to believe. His arguments on behalf of both of these are premissed on both epistemological reasons and a belief in the autonomy of students.4 4 Strike claims that students, like everyone else, have a right to autonomy because they are responsible for their actions and decisions: Human beings are ends in themselves and are moral agents who are responsible to choose wisely on their own behalf and act justly with respect to others. They are morally responsible for what they choose and what they do. A moral agent who is responsible for his choices must demand both the opportunity and the resources to choose wisely 4 5 The fact that people are responsible, and can be held responsible, for their actions means that they must be given the right to make decisions for themselves. As he has written elsewhere, "Liberals assume that there are limits on social authority and that there is a private sphere of beliefs and conduct over which the individual should exercise autonomy." 4 6 This opportunity to make decisions for oneself, then, or self-government, is what constitutes autonomy for Strike. Strike claims it has at least three components. The first is psychological freedom: this is the capacity for independent choice, and it requires the capacity 4 4 Kenneth Strike, Liberty and Learning (Oxford: Robertson, 1982). 4 5 I b i d . , 4 3 . 4 6 Kenneth Strike, Educational Policy and the Just Society (Urbana, Illinois: University of Illinois Press, 1982), 3. Arguments for Educating for Autonomy 26 for rational judgement and for self-control. The second component is the right of self-determination in those areas of life that are properly left to the individual 's discretion: individuals should have the right to choose their own beliefs and their own lifestyle, and they have a number of other rights that limit a government's or a society's authority over them. Finally, individuals have the right to participate in collective choices. 4 7 Another way to describe these components is to say that in order for individuals to have the opportunity for self-government they must have both the necessary political conditions—the freedom to choose in the appropriate areas of one's life, and to participate in democratic governance—and the ability to make good choices—including such things as the ability to assess reasons and the knowledge relevant to choices. "The central point about the ability to choose responsibly is that having information that provides a satisfactory base for a decision is not the same thing as being able to interpret or judge that information in a reasonable 48 way." Thus, individuals need to have both the necessary information and the ability to understand and use that information. Although he does not describe them any further, Strike acknowledges that this account of autonomy entails "such virtues as a regard for and commitment to truth, honesty and fairness. [These] are intellectual virtues in the sense that they are presupposed by the commitment to have one's choices and actions warranted by available evidence." 4 9 Since they have such important implications for education—they require fostering and modelling the giving and assessing of reasons—an explanation of the role these virtues play in autonomy, and some description of them, are required. 47Strike, Liberty and Learning, 43. 4 8Ibid., 44. 4 9Ibid. Arguments for Educating for Autonomy 27 Strike's account stresses that knowledge and its assessment are important ingredients within the ability to make responsible decisions. His book develops the implications of this for teaching. But his conceptualization leaves us with the task of determining just what "areas of life are properly left to the individual's discretion" something we need to know if we are to determine what kind of education is to be considered adequate. As well, it wi l l be worthwhile for our purposes to explain and justify his claim that an individual's responsibility for her own choices implies that "individuals [necessarily] have the right to participate in collective choices." Like Dearden's, Strike's conceptualization of autonomy makes autonomy a matter of degree. But he does not explain exactly what the implications of this fact are on the extent to which people are to be held responsible for their actions or on the degree to which freedoms are to be extended to people. Obviously the connection between one's ability to responsibly exercise autonomy and the freedoms extended to allow one to take responsibility for one's actions must be a reciprocal relationship. But he does suggest some criteria forjudging individuals responsible when he describes the maturity necessary for making educational decisions for oneself: a lack of maturity can override the right to voluntariness. General maturity must be distinguished from intellectual competence. The novice in physics lacks competence in the standards of physical argument. Such a person may, however, be mature. Maturity is the general capacity to discover or choose a stable and rational set of goals, needs and interests and make choices that further them. A student who does not know physics may, nevertheless, know himself well enough and know enough about what physics is like to make a competent (if tentative) commitment to study it. A student who lacks maturity cannot do even this. 5 0 These criteria for maturity—knowing oneself well enough, a "general capacity to "Ibid., 51. Arguments for Educating for Autonomy 28 discover or choose a stable and rational set of goals, needs and interests and make choices that further them"—are very close to the descriptions others have given of autonomy, and may suggest usable criteria for autonomy as a goal of education. C O N C L U S I O N Several arguments have been given for a concern with autonomy among educators. Primary among these is an intuitive recognition of the considerable intrinsic value autonomy has for us. Also, it is an important constitutive value of other things that also have a high value to us including self-respect, dignity, our moral virtue and our freedom. It is a necessary condition for responsible action and it is necessary for those virtues that are in turn necessary for the maintenance of the liberal state. This is a weighty list that gives some suggestion of the educator's responsibility and the necessity of a clear articulation of the concept. At the same time, we are left with a few conceptual difficulties. There is an important normative aspect to the concept, both in judging the quality of the evidence that a person bases her reasoning on and in judging what counts as an autonomous action. There is some uncertainty over the role such virtues as truth, reason, honesty and fairness play in autonomy. A n d while we have reason to believe that it plays an important role in a liberal state, there is an important diversity of opinion about just what that role is, or indeed how we are to imagine the liberal state. If education is to adequately meet the demands of liberal democracies we need a firm understanding of how we are to understand autonomy, and what role it plays in our understanding of liberalism. The remainder of this thesis is an attempt to conceptualize autonomy in a way that wi l l meet the demands placed on it by our understanding of liberalism, and be clearly relevant to the making of educational policy. C h a p t e r 2 : T h e L i b e r a l C o n t e x t f o r A u t o n o m y In trying to conceptualize autonomy I am interested in a specific use of the term. As a concept 'autonomy' is used to refer to a variety of things; it is used to describe other concepts and objects that have some independent characteristic, groups use it to both describe their own members' actions taken outside the group and their relationship as a group to other groups or to society as a whole, it was originally applied to city-states and is still used to describe the status held by nations and national groups, and professional organizations use it to describe the freedom they believe they .deserve with regard to their work, thus professional autonomy. M y project in this thesis is to describe what autonomy means when it is used to denote a desirable attribute of citizens in a liberal state, and to put this description in terms useful to educators taking the development of such autonomy as a goal or motivating principle for their work. If a conceptualization of autonomy is to be defensible as a legitimate goal of education in a liberal democratic society such as ours, then it must be consistent with the aims and principles of liberal political philosophy, the philosophy from which autonomy derives its importance, and which gives the best moral justification for mandated schooling. In particular it must be consistent with the conception of persons that liberal theory is based on. The usefulness of a conceptualization wil l be increased to the extent that it is not just consistent with these but also makes it clear how we can understand the person as both 2 9 Liberalism: Theory and Criticisms 30 autonomous and existing within and deriving her identity from shared traditions, close relationships, and so forth. A n d a usable conceptualization of autonomy should take into account the legitimate criticisms raised of liberal theory. In this chapter I give a description of liberalism in order to better determine the role autonomy plays within it. Liberalism is often criticized for conceiving persons as too individualistic or atomistic, and for neglecting the role of society and culture, and our relationships within these, in forming our beliefs about the world and our set of values within it. I do not believe such criticisms are accurate, but they do make an important point that liberals may not have paid sufficient attention to—they remind us of the importance of considering persons in context, and of keeping in mind those things that are fundamentally important to persons. I consider such criticisms in the latter half of this chapter, both to help describe how we might conceive of people relating to their society and to show how liberal theory can accommodate such an understanding. L I B E R A L THEORY As with almost any social or political theory, it is hard to p in down a precise definition of liberalism. The best way to understand it may be as a set of principles that are taken to underpin the institutions of the liberal state. Any list of such principles would have to include, at a minimum, justice, equality and liberty. In order to describe the underlying theory, then, I give a brief description of what some of the major theorists have said about each of these principles, and of the fundamental moral status of the individual in liberalism. JUSTICE Justice is the fair treatment of individuals by the various institutions of society; this is generally taken to mean the same treatment of persons in the same circumstances, and the appropriately different treatment of persons when good Liberalism: Theory and Criticisms 31 reasons for different treatment exist. Thus a conception of justice has to set out the ways in which the state must consider people equal, in what ways people are to be treated the same in order to respect this equality; the kinds of reasons that require a person to be treated differently and the kinds of treatment that such reasons call for. John Rawls has been credited with reviving political philosophy in the twentieth century; he has done this by attempting to provide a thorough conceptualization of justice. 1 Rawls argues that what is fair are those institutions and principles chosen in a situation of fairness. "A practice wil l strike the parties as fair i f none feels that, by participating in it, they or any of the others are taken advantage of, or forced to give in to claims which they do not regard as legitimate. This implies that each has a conception of legitimate claims which he thinks it reasonable for others as well as himself to acknowledge." 2 Rawls describes the original position, in which we are to imagine a group of representative people sitting down to negotiate and agree to principles of justice which wi l l guide their society and by which they wil l have to l ive. 3 A 'veil of ignorance' covers these representative actors—that is, they do not know anything about the position they wi l l occupy in society that might influence their choice of the principles of justice, such as what jobs they wi l l hold or how wealthy they wil l be. They do not even know such basic facts as what gender or race they wil l be, or specific values that they wil l hold, such as religious ones. In this way Rawls is asking us to imagine a situation in which we are not influenced by special considerations resulting from our own situation in society, or what would be to our own benefit, but ratherdemands that we take into account the situation of all in society-—because as people debating in 'See, e.g., Will Kymlicka, "Community," in A Companion to Contemporary Political Philosophy, eds. RobertE. Goodin and Philip Pettit (Oxford: Blackwell Publishers, 1995), 366. 2John Rawls, "Justice as Fairness," Philosophical Review 67 (1958): 178. 3John Rawls, A Theory of Justice (Cambridge, Massachusetts: Belknap Press, 1971), particularly chapter 3. Liberalism: Theory and Criticisms 32 the original position, we are in the position of possibly occupying any role in society, and therefore must consider the impact of any principles or institutions we debate on all persons, and take the welfare of all persons equally seriously. The original position therefore models a fair starting point; it does not imply that any of the conditions by which we live, or that we propose, were actually considered under such a condition, but rather gives us a way of considering how we might give the claims of all persons in society equal consideration. From such a fair starting situation we would have a better claim that the principles we agreed to live by are fair; the extent to which the laws that do govern our society or that we propose mirror principles that would be chosen in the original position,.or an equivalent situation of fairness, is an argument for their fairness. Rawls believes that the persons in the original position would be motivated to choose the following two principles of justice: First: each person is to have an equal right to the most extensive basic liberty compatible with a similar liberty for others. Second: social and economic inequalities are to be arranged so that they are both . . . (a) to the greatest benefit of the least advantaged and (b) attached to offices and positions open to all under conditions of fair equality of opportunity.4 This tells us in what respects Rawls believes people should be treated equally: The basic liberties of citizens are, roughly speaking, political liberty (the right to vote and to be eligible for public office) together with freedom of speech and assembly; liberty of conscience and freedom of thought; freedom of the person along with the right to hold (personal) property; and freedom from arbitrary arrest and seizure as defined by the concept of the rule of law. 5 The principle of equality also underlies the distribution of the primary social goods, which is taken to include "rights and liberties, powers and opportunities, 4Ibid., 60, 83. 5Ibid., 61. Liberalism: Theory and Criticisms 3 3 income and wealth," and self-respect.6 Rawls stresses that the ordering of the principles is important; we must act to satisfy the first before considering the second, and inequalities allowed by the second principle must not create inequalities with regard to the liberties protected by the first. The primary mode of distribution of all social values is thus equality; it is only when an unequal distribution can be shown to be to the benefit of the least well-off is an unequal distribution permitted. This is known as a minimax distribution of goods, where an inequality is only permitted when it is to the benefit of the worst off, or maximizes the minimum, and it is restricted by the fact that we can never permit an unequal distribution of the basic liberties. The autonomous persons in Rawls' just, well-ordered society are abstractly represented by the actors, operating under the veil of ignorance, in the original position. These actors are taken to be rational, in that they have a coherent plan of life, and can plan and act in order to achieve their ends.7 Under the veil of ignorance the actors do not know what their preferences are, they do not know their plan of life, their ends or values. But they are aware that, whoever they are in society, they generally have a preference for more primary social goods rather than less. "Thus even though the parties are deprived of information about their particular ends, they have enough knowledge to rank the alternatives. They know that in general they must try to protect their liberties, widen their opportunities, and enlarge their means for promoting their aims whatever these are."8 Through the construction of the original position Rawls ' theory embodies a Kantian version of autonomy. The original position models the premissing of 6Ibid., 62. 7Ibid., 142-43. 8Ibid., 142. Liberalism: Theory and Criticisms 3 4 principles and institutions of justice on the best reasoning of free and equal rational beings, who are motivated not to choose according to what wi l l be only in their own special interests, but according to what wi l l be a fair choice for all. This is an abstract version of the idea of conducting ourselves according to our best reasoned judgments, that we would choose according to principles that would reasonably be acceptable to all, and that take into account the points of view of all involved. This requires that people can reason abstractly, can see situations and possible outcomes from several points of view, can consider moral principles and how they apply to particular cases that arise, can judge what the appropriate criteria for deciding just actions are, and understand that they are then committed to acting according to their mutually agreed-upon decisions and principles. Note that the ways that the representative actors in the original position are taken to be equal; they are not equal with regard to such things as their wealth and status in society. Rather they are each seen as rational and capable of proposing principles and judging their merit, of engaging in discussion about the principles of justice, of having a sense of the good and corresponding life plan in their actual lives in society, and, perhaps most importantly, having equal status with regard to endorsing the principles of justice. That is, the principles of justice that hold over society are valid because they have met with the agreement of the actors in the original position; implicitly, our own laws and institutions are legitimate only i f they would be endorsed by each of us acting on our own best judgments, and leaving aside such special interests as those entailed by our gender, wealth, status, religious beliefs, and so on. EQUALITY Equality is the accepted principle in liberal democracies that all persons are to be treated as having equal value. But such a demand for some form of univers-alizability is not premissed on any claim that all people have the same abilities, Liberalism: Theory and Criticisms 35 talents, or power—indeed, it could not be, for in no way are people the same with regard to any of these things. For this principle to count, then, we must carefully demark the ways persons are to be valued equally. In turn, such valuing wil l lead to equal treatment; but, since people are not the same and are never served by being treated identically in every way, equal treatment means that in some way people are to count the same, but it does not mean that people are always to be treated the same. It may result in people being treated the same, but then only in limited spheres. Perhaps the best-known theorist of equality, Ronald Dworkin argues that "Equal concern is the sovereign virtue of political community," and is best realized through an equality of resources.9 His philosophical work attempts to characterize an equality of resources, and to derive other liberal principles, such as liberty and political rights, from this. Dworkin admits that equality of welfare—an active concern by the state for the equality of people's general well-being—has much immediate appeal as a conception of equality. After all, welfare aims to "describe what is fundamental in life rather than what is merely instrumental." 1 0 But equality of welfare falls to two important difficulties. One is the difficulty in deciding just what constitutes welfare, and in whose eyes. Compounding this is the problem in valuing welfare in such a way that we can make comparisons between one person's overall welfare, or their perception of this, and another's. A second and related difficulty is the expensive tastes objection: this is the difficulty that arises when one person's welfare can only be attained at an inordinately high cost for others. Suppose, for instance, that I have a taste for very expensive wine, and find that I 9Ronald Dworkin, Sovereign Virtue: The Theory and Practice of Equality (Cambridge, Massachusetts: Harvard University Press, 2000), 1. 1 0Ibid., 14. Liberalism: Theory and Criticisms 36 am only happy when I have a regular supply of this. Would the state therefore be compelled to compensate me in some way for my more expensive taste? The intuitive answer to this objection is that no, the person with expensive tastes is not entitled to some form of compensation, or a larger share of resources than others. Dworkin explores this, but his objection to this remains that it is not intuitively satisfying that one person should get more than "an equal share of social resources [at his or her] disposal." 1 1 But note this presupposes some notion of an equal share. Dworkin believes that this notion of equality is equality of resources, which he does not give an explicit defence of per se, but rather intends that his conceptualization of it wi l l be convincing. Equality of resources obtains when resources are divided among the members of a community such that the cost to an individual of having a specific resource reflects the cost of not having that resource to others. Dworkin models this situation through an auction. He asks us to imagine that the survivors of a shipwreck have washed up on a deserted island, along with an assortment of goods from the ship. Unable to foresee a quick rescue, they decide to divide up the resources equally, since none can lay a prior claim to them. Dworkin says that the fairest method of dividing up the goods would be to give each person an equal quantity of something that is otherwise useless, like clam shells, and then to auction the goods off. In this way the cost of each good would reflect the relative value of what people had to give up in order to win it. In an ideal situation—one in which time was not a factor—once the auction is completed, all the participants would be asked whether they were happy with the results. If not, the auction would be repeated, and repeated again, until everyone was satisfied with his or her share. "Ibid., 58. Liberalism: Theory and Criticisms 37 In the auction, one person could bid on, and win, more of a particular resource that he or she desired, but only at the cost of his or her ability to bid on other resources. If many people desired a particular resource, say a large block of cheese, then the value in clam shells of that resource would increase to reflect the demand on it and the value placed on it relative to those things that the unlucky individuals, who desired the cheese but lost the bidding for it, must compensate themselves with. These features are best reflected in some form of free-market system, and thus Dworkin concludes that equality "is a matter of equality in whatever resources are owned privately by individuals" and that the liberal state must therefore encompass either a free-market in social goods, or a system that in some manner reflects the outcomes a free market would create. 1 2 Note that according to Dworkin equality is achieved when people are satisfied with their 'take' from the auctions compared to others. Dworkin calls this the envy test. Equality of resources as exemplified by the auction under certain assumptions uniquely satisfies the envy test in a way that reflects peoples' preferences; there are other ways of satisfying the envy test which do not necessarily do this. For instance, a strictly equal distribution of goods would be fair in an obvious sense, and would leave people little reason for envy, but would have little chance of satisfying the varying preferences among individuals to the same extent. But Dworkin is assuming here that envy is motivated by a particular conception of equality and an awareness of when this is being contravened. However some people may be envious because of mistaken beliefs or, leaving out mistakes in beliefs and perceptions, because of other conceptions of equality. For 1 2Ibid., 65. For example, Dworkin wrote in favour of B i l l Clinton's failed health care plan, which encompassed not only basic health care for all, but also the freedom to purchase from among different supplemental health insurance schemes. Ronald Dworkin, "Wi l l Clinton's Plan Be Fair?," New York Review of Books, 13 January 1994, 20-25. Liberalism: Theory and Criticisms 38 instance, although Dworkin uses the envy test to motivate a conception of equality as equality of resources, it may be that some people feel envy when they perceive an inequality of-welfare. So if you and I had the same talents and access to the same resources, but you were far more industrious than I and achieved much greater welfare with your share of resources, I might feel envy at your situation even though my own lesser welfare was only a result of my comparably slothful nature. Thus the envy test only works as an indicator of equality when the individual whose envy or lack thereof is to count is motivated by the conception of equality that we are testing. But this circularity makes this test unsatisfactory as a criterion for choosing an appropriate conception of equality. This is not to completely devalue the envy test as a test for equality—an examination of the envy that we imagine in certain situations may sometimes point us toward clarifying our notion of equality. Elizabeth Anderson has suggested a more practical, and perhaps more useful, way of deriving the important ways people are to be treated equally. She suggests that we start by considering the kinds of inequality that liberals and egalitarians have traditionally been concerned with eradicating. The relevant inequalities are those that base moral worth "on birth or social identity—on family membership, inherited social status, race, ethnicity, gender, or genes. There are no natural slaves; plebeians, or aristocrats."1 3 These inequalities result in oppression, in some treating others with violence, or contempt, or simply excluding them from social and political life, and severely curtailing their opportunities for personal and public advance. Such inequality is often manifested in ideologies such as sexism, racism orclassism. But how can liberals oppose sexism, say, without therefore requiring that everyone believe in the equality of all, regardless of gender? That is, how can they 1 3 Elizabeth Anderson, "What Is the Point of Equali ty?," Ethics 109 (January 1999): 312. Liberalism: Theory and Criticisms 39 avoid infringing on some of the most important freedoms we each hold, in this case in the area of beliefs? Liberals are concerned with the beliefs people hold, and with the reasons people have for holding them—but they do not believe that we can insist on the beliefs that others hold. However, they do believe that we can require people to behave in certain ways—specifically, in ways that do not infringe the freedom of others, in this case the freedom to participate in political life, or to share in the range of important freedoms available to everyone. Indeed, freedom requires just such limitations on each other's behaviour. Equality, then, primarily asserts that we each have a right to the opportunity to consider and pursue our own conception of the good, as well as the right to be treated with respect by each other and the state. Importantly, this means the widest range of freedoms consistent with the like freedom for all. This is consistent with the important ways we have already seen that people are to be treated equally: people have an equal right to freely decide the course of their own lives, and the facts of each person's interests count equally in importance to others. This wi l l also entail a list 'of specific freedoms that allow people to survive, support themselves, co-operate with others and grow. Anderson suggests one possible list of specific freedoms this might entail: To be capable of functioning as a human being requires effective access to the means of sustaining one's biological existence—food, shelter, clothing, medical care—and access to the basic conditions of human agency—knowledge of one's circumstances and options, the ability to deliberate about means and ends, the psychological conditions of autonomy, including the self-confidence to think and judge for oneself, freedom of thought and movement. To be capable of functioning as an equal participant in a system of cooperative production requires effective access to the means of production, access to the education needed to develop one's talents, freedom of occupational choice, the right to make contracts and enter into cooperative agreements with others, the right to receive fair value for one's labor, and recognition by others of one's productive contributions. To be capable of functioning as a citizen requires rights to political participation, such as freedom of Liberalism: Theory and Criticisms 40 speech and the franchise, and also effective access to the goods and relationships of civil society. This entails freedom of association, access to public spaces such as roads, parks, and public accommodations including public transportation, the postal service, and telecommunications. This also entails the social conditions of being accepted by others without shame, and not being ascribed outcast status. The freedom to form relationships in civi l society also requires effective access to private spaces, since many relationships can only function when protected from the scrutiny and intrusions of others. 1 4 This is a comprehensive list of the kinds of freedoms and requisite resources liberals are concerned with in the real world. Anderson adds three points that help to make clear some of the implications of this conceptualization of equality. 1 5 First, with regard to some rights, equality may mean that people are ensured effective access to the conditions necessary to realize those rights, not that the actual conditions necessary to fulfill those rights are provided for them. A right to housing might be one such right: if we agreed that the right to shelter and privacy meant that everyone had an equal right to housing, we might provide for this right by ensuring the existence of affordable housing for all. We might not actually choose to house everyone—indeed, most people in our world would probably not want to have their housing provided for them. A n d some may even choose to forego housing altogether at present for a variety of personal reasons. "Effective access to a level of functioning means that people can achieve that functioning by deploying means already at their disposal, not that the functioning is unconditionally guaranteed without any effort on their own part." So access to housing might require that those who can work to pay for their shelter have to do so. Educational attainment provides us with another example of this. While a certain level of education is necessary for citizenship in any number of ways, the 1 4Ibid., 317-18. Liberalism: Theory and Criticisms 41 state cannot guarantee attainment of this level—this depends, in part, on individuals valuing this, and choosing to work to achieve it. The state can, and must, ensure that everyone has effective access to education, but it cannot guarantee that everyone wi l l actually attain the requisite educational achievements. Second, equality does not mean that everyone has a right to the same level of achievement, or functioning, but rather to a level sufficient for functioning as an equal in society. Anderson gives the example of literacy: everyone requires the ability to read and write with understanding and the ability to consider what is being discussed, in the language of her or his social and political organization, to a reasonably high degree. So as a society we are required to provide an education that provides this. But to participate as an equal in our society does not require that one can speak Mandarin, or that one has the ability with the English language of a Ph. D . graduate in literary theory. Nor is everyone wil l ing to put forth the effort required to reach these levels of ability. Equality cannot therefore be dependent upon everyone having access to these higher levels of functioning. But everyone must have access to the level of functioning necessary for equal participation as a citizen. In some spheres this will require equal levels of functioning—the right to participate in democratic governance means that everyone gets one vote, for example. Third, equality means that everyone holds their rights throughout their lives. They may choose not to exercise them—an individual may choose not to vote, or speak, or live in a house—but they cannot trade off their rights to others, or sell them. For an individual to sell her vote or her right to access to health care would be for her to ask the rest of us to treat her in ways incongruent with the principles of universality and respect. If one person sold another his right to 15Ibid., 318-19. Liberalism: Theory and Criticisms 42 participate in our common democracy, for another example, would be for the one to ask the other to act in ways that denied his rights as a citizen, and these two together to ask the rest of us to treat them in ways that valued one of them as less than the rest of us in this regard and one as more. Each of these two would therefore have the right or expectation to treat the rest of us as different from him or herself—whether more or less. Let me add two more points to these. First, I have repeatedly used the term 'rights' to describe the various ways people are to be treated equally. The word is often used to mean things that are necessary for the protection of freedom, but if we understand equality as an equality of freedoms, then rights also denote the way in which we can demand to be treated equally. This is only true in a society where we can assume that speakers are not making claims dependent upon a hierarchy, of course. So when one person says "I have a right to X " ("I have a right to an education," or "I have a right to affordable housing," for example) she is implying that she has a right to X because we all do—that is, because we all stand in a relation of equality to one another, and the list of rights-claims that we each can make is precisely what describes this relation of equality. In a hierarchical society, the existence of rights is often premissed on inequality; in such a society an individual may have a right due her or him exactly because of her or his status in the hierarchy. In this situation to say "I have a right to Y " often implies "I have this right because I am in such-and-such a position," or "because I belong to this class." What would happen i f we decided that in order to be truly free a person required a certain minimum level of nutrition and, in order to meet the demands of equality, we were going to ensure that everyone had reasonable access to three full meals a day? Soon people who found it difficult to access three meals a day would claim that a right they held was being denied them—and they would be correct; i f we agreed that this is what equality requires, then we would have a collective obligation to provide this. So in a liberal Liberalism: Theory and Criticisms 43 democracy we demarcate and protect those ways in which people are equal—the conditions necessary for autonomy, perhaps—by enshrining rights. This goes against those writers who talk as if rights conflict with equality, and is a different explanation for the source of our rights than the one given by writers who claim that individuals (or, traditionally, men) have some kind of inherent rights, usually assumed to predate any sort of political union which recognizes this, and sometimes stemming from religious beliefs. 1 6 This also conflicts with those writers who describe equality and freedom as two competing claims. 1 7 Second, this points out the fallacy in the popular way of talking about rights versus responsibilities. 1 8 In a classless society rights exist on the premiss that they exist for all. But a right is a requirement that certain demands are met. 1 9 So the existence of rights places demands—responsibilities—on all of us. That is, 'rights' is another term for 'shared responsibilities to all individuals.' We cannot expect to have rights—to have our claims respected—if we ourselves do not understand these as existing in such a situation of equality, and therefore are not l 6 Richard Norman has surveyed writers who discuss rights as conflicting with equality, ranging from Hume to F . A . Hayek, in Free and Equal: A Philosophical Examination of Political Values (Oxford: Oxford University Press, 1987). Probably the most well-known example of assumed inherent rights, the second sentence of the Declaration of Independence of the United States states: " W e hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain unalienable Rights, that among these are L i fe , Liberty and the pursuit of Happiness." The first article of France's Declaration of the Rights of Man similarly claims: " M e n are born free and remain free and equal in their rights. Social distinctions can only be founded on public utility." l 7 See, e.g., Ronald Dworkin , "Liberal ism," in Public and Private Morality, ed. Stuart Hampshire (Cambridge: Cambridge University Press, 1978), 113—43. 1 8 For instance, in moral psychology, Carol Gi l l igan describes Lawrence Kohlberg 's system as "Regarding all constructions of responsibility as evidence of a conventional moral understanding, [whereas] the highest [post-conventional] stages of moral development [derive] from a reflective understanding o f human rights." Carol Gi l l igan , In a Different Voice: Psychological Theory and Women's Development (Cambridge, Mass.: Harvard University Press, 1982), 19. l 9 In Joel Feinberg's slightly different phrasing, rights are claims we each can make on others. Joel Feinberg, Rights, Justice, and the Bounds of Liberty: Essays in Social Philosophy (Princeton, New Jersey: Princeton University Press, 1980), 149-55. Liberalism: Theory and Criticisms 44 will ing to respect the equivalent claims by others. So the existence of rights demands responsibilities from all of us; we cannot demand the recognition of rights without simultaneously recognizing our collective obligation to one another. So far I have been talking about rights from the point of view of individuals who are in a position to extend such rights to each other—you and I agree that these are reasonable demands to place on one another if we are to exist side-by-side, and we are each will ing to respect each other's rights. It does not take much imagination to extend these rights (with the caveat: to the extent that we can) to all members of a society, including those who do not have the ability or equal power to openly and consciously enter into such a contract with us, including the young, the impaired elderly, the mentally i l l or disabled, and those who have failed to meet the demands of our collective rights in the past, to equally participate in making and respecting such a metaphorical contract. We do this for a number of reasons. We do it because we know that i f we were in their shoes we would still desire to be protected from harm or abuse, we would still want our wishes and desires to be respected in certain ways (and, again, can imagine that these desires are reasonable and universalizable), and we would want the best opportunity to develop into, or back into, persons who could consciously participate in society to the maximum extent possible, including entering into such an equal relationship with others. A n d protecting the rights of others in these circumstances is an important part of ensuring these for others and so also for ourselves in their situation. L I B E R T Y A third cornerstone of liberalism is a belief in liberty—the right of each individual to make a broad range of choices for him or herself, and so to conduct his or her life as he or she chooses. As we have seen, freedom requires for its continued Liberalism: Theory and Criticisms 4 5 existence that each person cannot curtail the freedom of others, and thus it is generally accepted that each should have as much freedom as is consistent with the like freedom for all. But it is not clear what rules we should agree to live by in order to ensure the greatest possible individual freedom. One way to approach this is to make clear why we value freedom, and therefore which freedoms we place more value in. This is the approach taken by John Stuart M i l l in his seminal essay On Liberty.20 M i l l offers several arguments for freedom. Although the categories are not entirely distinct, these arguments can be roughly divided into epistemological, pragmatic, empirical and moral reasons for valuing and defending individual liberty. Epistemologically, M i l l argued that "We can never be sure that the opinion we are endeavouring to stifle is a false opinion; and i f we were sure, stifling it would be an evil s t i l l . " 2 1 This is because, as Isaiah Berlin paraphrased, in a climate that stifles false opinion, "the truth wil l not, for lack of a free market in ideas, come to light," and even when the truth is known it needs to be 22 continuously challenged i f it is to be kept 'fresh and strong.' Pragmatically, M i l l believed that " A s it is useful that while mankind are imperfect there should be different opinions, so it is that there should be different experiments of living; that free scope should be given to varieties of character, short of injury to others; and that the worth of different modes of life should be 23 proved practically, when anyone thinks fit to try them." " Further, M i l l believed that without liberty civilization cannot advance, for 'the despotism of custom' always impedes human progress, since it curtails the disposition to improve one's 2 0John Stuart M i l l , On Liberty, ed. Gertrude Himmelfarb (London: Penguin Books, 1974). 2 1Ibid., 76-77. 22Isaiah Berlin, Four Essays on Liberty (Oxford: Oxford University Press, 1969), 127. 2 3 M i l l , 120. Liberalism: Theory and Criticisms 46 lot beyond what one is used to, and that the only "unfailing and permanent source of improvement is liberty," since this allows a multitude of ways to improve oneself. 2 4 Empirically, M i l l claimed that without liberty there would be no scope for spontaneity and originality, and "The same things which are helps to one person towards the cultivation of his higher nature are hindrances to another."" Morally, M i l l stated that "The only part of the conduct of anyone for which he is amenable to society is that which concerns others. In the part which merely concerns himself, his independence is, of right, absolute. Over himself, over his own body and mind, the individual is sovereign." 2 6 Thus no one can tell another how to live his or her life, for the individual is always the one with the strongest and most important interest in his or her own life, "the interest which society has in him individually (except as to his conduct to others) is fractional and altogether indirect," and so far as his own desires, feelings and interest go, the individual has far more insight and knowledge of these than anyone else can possibly have. 2 7 M i l l carefully outlined what he considered to be the appropriate domain of personal liberty: It comprises, first, the inward domain of consciousness, demanding liberty of conscience in the most comprehensive sense, liberty of thought and feeling, absolute freedom of opinion and sentiment on all subjects, practical or speculative, scientific, moral, or theological. The liberty of expressing and publishing opinions may seem to fall under a different principle, since it belongs to that part of the conduct of an individual which concerns other people, but, being almost of as much importance as the liberty of thought itself and resting in great part on the same reasons, is practically inseparable from it. Secondly, the principle requires liberty of Ibid., 136. Ibid., 133. 'Ibid., 68-69. 'ibid., 142-43. Liberalism: Theory and Criticisms 47 tastes and pursuits, of framing the plan of our life to suit our own character, of doing as we like, subject to such consequences as may follow, without impediment from our fellow creatures, so long as what we do does not harm them, even though they should think our conduct foolish, perverse, or wrong. Thirdly, from this liberty of each individual follows the liberty, within the same limits, of combination among individuals; freedom to unite for any purpose not involving harm to others: the persons combining supposed to be of full age and not forced or deceived. 2 8 But M i l l makes it clear that he believes liberty is based, at least in part, on an individual's ability; he stresses repeatedly that he is talking about people "in the maturity of their faculties." 2 9 Nor, M i l l says, should we live as if there is nothing to be learned from others, or as if there are not better and worse ways to do things. He says that [I]t would be absurd to pretend that people ought to live as i f nothing whatever had been known in the world before they came into it; as i f experience had as yet done nothing towards showing that one mode of existence, or of conduct, is preferable to another. Nobody denies that people should be so taught and trained in youth as to know and benefit by the ascertained results of human experience. 3 0 How we attempt to influence another is therefore important. There "are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him or visiting him with any evil in case he do otherwise." Berlin suggests that, of the many uses of the term liberty, there are two central ones. The first has to do with the sense in which we are free when we are not interfered with; he calls this negative liberty. In this sense, " Y o u lack political 2 l b i d . See, e.g., 69, 122. 3 0Ibid., 122. 3 1Ibid., 68. Liberalism: Theory and Criticisms 48 liberty or freedom only i f you are prevented from attaining a goal by human beings." j 2 In order to ensure this liberty for citizens we must define an area of negative liberty. This area cannot be unlimited, or we would all end up interfering with one another. So "the area of man's free action must be limited by law" but, at the same time, "there ought to exist a certain minimum area of personal freedom which must on no account be violated." J J Berlin places a very high value on personal liberty, but he believes it is not the only goal or value; we may, at times, give it up for other gains, like bread for others or greater freedom down the road. 3 4 The second sense of freedom that Berlin identifies he calls positive freedom. This has to do with the desire to be our own master, to be driven by reason and our own conscious purposes. Berlin says this sense of freedom "is involved in the answer to the question 'What, or who, is the source of control or interference that can determine someone to do, or be, this rather than that?'" 0 5 It is freedom understood as freedom from slavery to forces we cannot control. B y this distinction between negative and positive liberty, M i l l is arguing for negative liberty. Berlin points out that the connection between non-interference and the development of the kind of character that M i l l desired, one that was critical and imaginative, not hindered by 'the despotism of custom'—a connection that provided for M i l l an important argument for liberty—is at best empirical, and not necessarily historically accurate.3 6 Berlin also points out that negative liberty is not incompatible with at least the absence of self-government. A regime under a despot who allowed his or her subjects a good deal of personal liberty would qualify as free in the way M i l l desired. "Self-government may, on " B e r l i n , 121-22. 3 3 Ib id . , 124. 3 4 Ib id . , 125, 170. 3 5 Ib id . , 122. 3 G Ibid. , 128. Liberalism: Theory and Criticisms 49 the whole, provide a better guarantee of the preservation of civi l liberties than other regimes . . . But there is no necessary connection between individual liberty and democratic rule." 3 7 Berlin claims it is this lack of connection between a lack of interference and being ruled that drives the distinction between negative and positive liberty, and leads to the concern with positive liberty. Although Berlin has these concerns with M i l l ' s version of liberty, he is even more wary of the dangers associated with positive liberty. The desire to be one's own master, to be driven by one's own conscious purposes, can too easily, according to Berlin, slide into saying for others that 'they would choose this,' and then into 'this is their freedom, which they just don't know yet,' or even 'their real self knows and desires this, although their empirical self is still ruled by 38 1 irrational desires.'" This allows one to oppress others in the name of freedom, and even to claim that their real selves have willed their own oppression. Berlin 's conclusion is that we cannot have an all-embracing liberty, nor some final formula balancing the claims of liberty and such other values as justice and security. "The extent of a man's, or a people's, liberty to choose to live as they desire must be weighed against the claims of many other values, of which equality, or justice, or happiness, or security, or public order are perhaps the most obvious examples." But he does believe in a form of negative liberty that wi l l result in a diversity of lives, and that wi l l require us to continually work out how it is to be applied and how we are to get along under it. "Pluralism, with the measure of 'negative' liberty that it entails, seems to me a truer and more humane ideal than the goals of those who seek in the great, disciplined, authoritarian Ibid., 130. 'ibid., 132-33. 'ibid., 170. Liberalism: Theory and Criticisms 50 structures the ideal of 'positive' self-mastery by classes, or peoples, or the whole of mankind." 4 0 Perhaps a liberty that allows for a diversity of good lives to be lived alongside one another, and the resulting plurality we might expect this would lead to, as well as the existence of continued debate as to how the details are to be worked out, are necessary criteria for a liberal society. But accepting these may not necessarily entail rejecting positive liberty outright, as Berlin would seem to suggest. Without evidence that a concern for positive liberty must lead to un-liberal practices, there may still be reason for a concern with some form of positive liberty. Charles Taylor argues that we cannot understand the importance of liberty until we can see it in the context of what is significant to us, what motivates us, the goals that are truly important to us and that we take as central to what we identify as our true, authentic desire or purpose. 4 1 That is, "some discrimination among motivations seems essential to our concept of freedom." 4 2 He provides what has become a well-known example 4 3 Suppose, he suggests, there are far fewer traffic lights per person in Albania than in England. Then consider a defence of Albania as a freer country than England because, even though such things as religious freedoms do not exist there, people are far freer to drive unhindered by traffic lights. Not everyone in England engages in religious practices and most of those that do, do so but once a week. B y sheer quantity of restricted acts, therefore, people in Albania are less interfered with in their day-to-day lives and so enjoy greater liberty. Taylor's point is, of course, that the people 4 U lbid., 171. 4 1Charles Taylor, "What's Wrong with Negative Liberty?" in The Idea of Freedom, ed. Alan Ryan (Oxford: Oxford University Press, 1979), 175-93. 4 2Ibid., 183. Liberalism: Theory and Criticisms 51 in Albania are not more free but far, far less free, because it is not the number of acts interfered with that matter but the value we place in the activities those acts are a part of. Having to stop frequently for traffic lights may be a nuisance to London drivers but nobody sees this as a serious loss of freedom, in part because it allows the accomplishment of other goals—in this case driving securely—but also in part because no one places great value on the ability to drive at wi l l through intersections. On the other hand, the ability to choose and practise one's own religion is valued so highly that even those that do not exercise it defend it as a fundamental liberty. The point of this is that we have to talk about freedom in the context of those aims, purposes, goals and activities that "we value 4 4 There is a second point to be made here. While restricting ourselves to defending a negative conception of liberty might make it easier to avoid oppression in the guise of liberty, the aim is to increase liberty while avoiding oppression, not merely to avoid oppression. Perhaps, as Berlin would have it, given the horrendous abuses that have taken place in the guise of positive liberty, we should err on the side of safeguarding ourselves against oppressive uses of the concept of liberty. But this does not mean that we have to restrict ourselves to talking about freedom as a lack of interference by others with our activities. There are other important barriers to our freedom to do things than is captured by negative freedom. Berlin himself supplies the analogy of a mathematician or musician who becomes able to do mathematics or play an instrument only through first learning, and then internalizing, necessary mathematical proofs and the rules 4 4 M y concern with Taylor 's argument is not with what he is arguing for but the ends to which he puts it. Taylor believes that a core value for contemporary liberals is self-realization, and that we therefore would want to adopt a positive freedom that would facilitate this. But without a more concrete conceptualization of what Taylor means by this term I believe that we cannot know that it is not, as he says, "metaphysical hog-wash" rather than a clearly definable purpose, and its ambiguity invites speculation and similarly vague claims as to how it is best pursued.(178) I am wary of the sort of abuse this could lead to; I do not see how the promotion of such a positive freedom would clearly meet the criteria I discuss below. Liberalism: Theory and Criticisms 52 for constructing these proofs, or musical scores and the rules and techniques for playing an instrument. 4 5 Neither the mathematician nor the musician would say she was trapped by the rules for her discipline, rather she is made free—in the sense that she is enabled to do what she wants to do—through learning and internalizing the rules and norms of her discipline. Without these rules and norms she is not free to engage in her chosen pursuits. Similarly, if I want to read but cannot, and no one is wil l ing to teach me, it would seem that in a very real sense I am not free to read, although no one is actually stopping me from the act of reading. In a similar way many people are not free to send their children to private schools because they cannot afford to, even though there are no actual prohibitions on their doing so. Such barriers are just as real as people interfering with one's activities, or laws made to stop one from pursuing his or her goals. It seems that i f we agree that there are certain things that are in general of value to the choosing and pursuit of good lives, then we can increase our liberty (to pursue good lives) by ensuring that these other kinds of barriers—in the form of lack of ability or lack of resources—are not present. To do this requires working out which freedoms are the most important to us. At the same time we must be careful to heed Berlin's warning. Ensuring that people have the necessary ability to make choices among religious beliefs can very easily lead to making sure they make specific choices. It is hard to ensure that people have the ability to make good choices without insisting that they exercise that ability in particular ways, but this is what the liberal society that is going to increase people's freedom beyond simply minimizing state intrusion or laws must struggle to accomplish. The just society would desire that its members made.good choices, but would likely be marked by a diversity of expressions and even some conflict over the value of various choices. 'Berlin, 141. Liberalism: Theory and Criticisms 53 One way to ensure this would be to take the reasons offered by M i l l in defence of liberty as criteria for any 'freedom' we chose to promote. So, for instance, we might choose to ensure, so far as possible, that everyone can read, because we believe that an ability to read would enable people to make more and better choices for themselves. And we might expect that an increased rate of literacy, per se, should increase the variety of opinion available and debate over the truth and not stifle it, should not hinder such things as spontaneity and originality nor variety in lives, and should increase the ability of individuals to exercise their right to independent control over their lives without prescribing in any way how those lives are to be led. What sorts of things do we value in this way? It strikes me that the list of ways in which we want people to be equal suggested by Elizabeth Anderson, that I cited above, would likely attain wide agreement as a summary of the important freedoms people should have. M O R A L STATUS OF THE INDIVIDUAL There are many other principles that fall within a common understanding of liberalism, such as democracy, tolerance, respect, and neutrality with regard to choice among good lives. I wi l l not discuss them all here, because there is considerable overlap between the reasons in support of the principles I have discussed and these others, and because they are unnecessary as distinct steps in conceptualizing and defending a version of autonomy within liberal theory. But there is one other central element of liberal theory that bears mention. The individual has a moral status in liberalism that differs from most other competing, theories. This status is not so much a principle of liberalism like the others I have discussed as it is a central element that is reflected in all of them, but it is worth discussing briefly because many of the criticisms of liberal theory either want to deny this as necessary to good lives, or else reflect an impoverished understanding of it. Liberalism: Theory and Criticisms 54 Liberalism assumes that the sole object of moral concern is the individual. Communities, states, families, groups and traditions are important, but they do not elicit the same moral concern that people do—they do not have rights that compete with individuals' rights, for example. And of course, as I discussed under justice and equality, individuals share these rights equally. This is not to say that communities and traditions have no moral importance to the liberal; they are extremely important. But they are important because of—and their moral worth is measured in terms of—the value they have to individuals. Such things as communities and traditions have no moral importance before or beyond the importance they have to individuals. The moral status of individuals, of course, is in no way dependent upon, nor limited by, the value they have to communities or traditions. Therefore communities and traditions have no rights that come before, or can interfere with, the rights of individuals. One reason for this is that individuals suffer in a way that communities and traditions do not. People suffer pain and indignity, they can be shown disrespect and they can be oppressed. Communities do not suffer any of these things in the same way. Of course communities can be harmed in that their growth or future can be limited, or they can be broken apart. But it does not make sense to say that the community per se suffers in the same way that individuals do— although, as in the case of a community coming to an end, the individual members may suffer and lose a great deal. This, of course, may be morally bad suffering, but inasmuch as it is experienced by and affects individuals, not as it affects the community qua community. A second reason is that communities and traditions cannot enter into agreements and make commitments in the same way that an individual can, and therefore cannot be held responsible for their actions and the consequences thereof. No one or group can ever speak for every member of a community in a way that expresses all of their interest, all the time. Individuals can speak on behalf of groups, but only with the consent of the individuals Liberalism: Theory and Criticisms 55 involved. A community's desire cannot trump the consent of the individual members of that community. Harry Brighouse calls the moral status of the individual in liberal theory 'ethical individualism,' and points out that it is important to distinguish this kind of individualism from at least three other forms of individualism, none of which have a necessary connection to l ibera l i sm 4 6 There is methodological individualism, the belief that all social phenomena are explained by accounts of the behaviour of individuals. This is an empirical claim, not a moral one. There is what Brighouse terms social-policy individualism; this is the argument that institutions and policies should not take into account membership in a group but merely deal with individuals, in order that they be treated equally. Ironically, this argument is usually raised in opposition to policies intended to increase equality. Most liberals disagree with it, acknowledging that such policies are designed to counter ways that group membership already affects the distribution of social goods. Third, and perhaps most importantly for our purposes, there is the form of individualism that claims that people are self-interested. This often involves the suggestion that people are self-interested in a selfish way—that is, that our motivations are less than entirely moral. While much liberal theory is premissed on a conception of individual rationality that assumes individuals wi l l take their own best interests into account and can judge possibilities on their potential benefit to their own good, it never claims that people do this to the exclusion of considering others' goods, nor does it assume the good of others is not often among an individual's own strongest desires. Most importantly, liberalism is premissed on the desire of individuals to treat each other as justly as possible, 46Harry Brighouse, School Choice and Social Justice (New York: Oxford University Press, 2000), 5-6.1 am here using his terms for the three forms of individualism. Liberalism: Theory and Criticisms 56 equally, and with respect, and to counter injustice and the limiting of others' liberty, even when such instances of injustice benefit oneself in selfish ways. There is at least one more way that critics of liberalism have claimed it to be individualistic. Some think that the liberal conception is of a universal and impersonal just society peopled by abstract individuals, "unalloyed with particular identifications." 4 7 In the next section I wi l l review what are generally taken to be the most powerful examples of such criticisms. Most liberals would argue that one might think occasionally in these terms, but "such a conception of persons detracts from a good deal that is important about persons, and makes for theory that is not only unrealistic but also highly undesirable. That liberalism can accommodate real people with their own attachments is, I hope, one outcome of the remainder of this thesis. C R I T I C I S M S O F L I B E R A L T H E O R Y Liberalism is not unchallenged; there are several strong criticisms of it and of either its underlying concern with justice, or else with its interpretation of justice and its other commitments. In this section I wi l l describe what I take to be the strongest of such criticisms. I wi l l point out some of the fundamental ways that I think these critics fail in their attempts to criticize the core of liberal theory. But I do think that valid concerns are raised, i f not with the actual intent of liberal theory then perhaps with its interpretation. I canvass these critics here in order to point out these legitimate concerns. 4 7 Charles Taylor, "Cross-Purposes: The Liberal-Communitarian Debate," in Liberalism and the Moral,Life, ed. Nancy L . Rosenblum (Cambridge, Mass.: Harvard University Press, 1 9 8 9 ) , 175. Liberalism: Theory and Criticisms 57 COMMUNITARIAN CRITICISMS According to Michael Sandel, the liberal account, as exemplified by Rawls ' and Dworkin 's conceptions, requires a strong sense of community in order to redeem the notion of common assets that underlies proposals for just systems of distribution like the difference principle or an equality of resources. But Sandel believes the liberal actor, as exemplified by the actor in the original position, has a sense of self which does not allow for the 'constitutive sense of community' necessary. This is because For a subject such as Rawls ' the paradigmatic moral question is not 'Who am I?', for the answer to this question is regarded as self-evident, but rather 'What ends shall I choose?', and this is a question addressed to the wi l l . Rawls ' subject would thus appear epistemologically impoverished where the self is concerned, conceptually ill-equipped to engage in the sort of self-reflection capable of going beyond an attention to its preferences and desires to contemplate, and so to re-describe, the subject that contains t h e m 4 8 Sandel believes the truly empowered actor is one "whose identity is constituted in the light of ends already before it, [for whom] agency consists less in summoning the wil l than in seeking self-understanding"; such an actor "is empowered to participate in the constitution of its identity." 4 9 But how are we to understand such an actor's 'participation' in defining her own identity? Sandel believes that by recognizing one's identity as formed by the communities one is a part of, and seeking to come to an understanding of one's identity as shaped by one's community, one is playing a role in shaping one's own identity. 5 0 4 8 M i c h a e l Sandel, Liberalism and the Limits of Justice (Cambridge: Cambridge University Press, 1982), 153. 4 9 Ib id . , 152-53. 5 0 F o r Sandel, "what marks such a community is not merely a spirit of benevolence, or the prevalence of communitarian values, or even certain 'shared final ends' alone, but a common vocabulary of discourse and a background of implicit practices and understandings within which the opacity of the participants is reduced i f never finally dissolved." Ibid., 172-73. Liberalism: Theory and Criticisms 5 8 Let me use a personal example to illustrate Sandel's claim. The community I am most a product of is my immediate family, consisting of my parents and sisters. A great deal of my own sense of myself—much of my language, my values, my personal history—comes from my growing up in and being a part of that particular family. Some of this I could have gained had I grown up in another family, but some of it—notably much of my personal history—I could not have. I would be a very different person in a very real, tangible way i f I was not a product of, and a constituent of, this community. I am also a product of a variety of other communities, such as the various circles of friends I belong to, the nation that I grew up in, and so on. A s Sandel says, my being a part of these communities is "not just a feeling but a mode of self-understanding partly constitutive of [my] identity," and this seems to be an apt description of our existence in communities. 5 1 Sandel does allow that the individual can distinguish herself from others. He says that The bounds between the self and (some) others are . . . relaxed on the intersubjective account, but not so completely relaxed as to give way to a radically situated subject. The bounds that remain are not given by the physical, bodily differences between individual human beings, but by the capacity of the self through reflection to participate in the constitution of its identity, and where circumstances permit, to arrive at an expansive self-under-standing. 5 2 But this is, for him, bounds of "an enlarged self, conceived as community," or, I take it, bounds between community understandings, and not to be understood as bounds between individuals defined by their 'physical, bodily differences.' Sandel also claims that members of communities are not just constituted by their relationships, but that they also 'discover' these relationships. What does 5lIbid., 150. 52lbid., 144. Liberalism: Theory and Criticisms 59 he mean by this? If many of our values and beliefs are formed by the communities we are a part of, and formed without our complete awareness, then we may come to recognize a value we hold, and along with this we may recognize its origins in one of the communities we are part of. This might happen as someone who realizes he or she holds a value that he or she was not conscious of before, and which he or she can identify the origins of within a community he or she is or has been a part of, or it could happen as a process of discovery as a community examines itself and its practices and is able to identify some of its practices as reflecting a value in a way its members had not been aware of before. So, for example, a community might come to realize that some of its practices and institutions actually reflect sexist biases they were previously unaware of. For myself, the more I become aware of myself the more I see some of my values and beliefs to have their origins in my family. But note that we do not only discover ourselves to be part of communities, we also choose some of our communities. We are free to, and do, choose the circle of friends we are a part of. We are also able to leave communities—for myself, while a child growing up I came to realize I was part of a family, I remain part of that family today by choice. A l l of the shared activities of spending time together, remembering, discussing, sharing values and aspirations, making decisions together and so forth are now entered into by choice. N o doubt, were I to disassociate myself from my family and refuse ever to communicate with them again I would continue for some time, perhaps my whole life, to recognize elements of myself that had their origins in my history with my family. But Sandel does not seem to want to acknowledge the fact that people do choose between communities, and between ways of life, or at least this fact cannot be explained by a conception of the empowered actor whose aim is self-understanding but not reconsidering and revising one's ends, especially when those may be at the cost of embeddedness in one's community. Liberalism: Theory and Criticisms 60 On his interpretation of Rawls Sandel would seem to be confusing the hypothetical actor in the original position with actual individuals in society. It is not entirely true that the Rawlsian actor is impoverished because her primary moral question is 'What are the appropriate ends?' This may be the question that organizes just institutions, but it does not need to be the one that motivates actual individuals in the world. Indeed, Rawls ' principles of justice are chosen to allow the individual to be motivated by a range of understandings about her situation. To some extent Sandel's characterization of how understanding our identity as shaped by our communities itself shapes our identity must be right: how we understand something changes its nature for us, how we understand ourselves and our embeddedness in community changes who we are and the range of options open to us. But even this would seem to go further than Sandel wants, for it implies some freedom of choice among possible interpretations, and the need for criteria as to what would count as a better understanding, or a worse one. Suppose a better understanding of ourselves and our community is one of inheriting a society of grossly immoral traditions. How could we face up to that? Sandel's individual would seek to understand her inheritance, but even i f she did have criteria on which to judge her inheritance as immoral (and it is not clear she could ever find it truly repulsive) it is not clear why she would be motivated to act to change it. She would have to find elsewhere a desire for justice to prevail, and this may not be any part of her community's tradition. Nor is it clear why, in Sandel's account of the bounds between the self and (some) others, the bounds cannot go deeper—why cannot this capacity of the self to reflect on its identity allow for distinction between individuals even within a shared community of traditions? Why cannot it allow that we all are going to interpret our inheritance a little differently, even though substantial parts of that inheritance are common? Liberalism: Theory and Criticisms 61 Even more importantly how are we, together or individually, to understand choice, and to have the power to make moral decisions about what aims are worthwhile? How are we to decide whether our institutions are just, and, i f not, how we are to refashion them? Sandel's discussion of the role communities have in shaping identities might be expected to lead to a concern for the justness of this relationship, and for the appropriate bounds on the control that communities can exert over individuals. But his account does not give any answer to this, other than to point out that just institutions must take account of our membership in community—indeed, for him, justice ultimately takes a back seat to our self-knowledge of ourselves as members of a community. Sandel's analysis involves sliding between two senses of 'prior'—two senses that it is important to distinguish between. The temporal sense of 'prior' refers to the timing of one thing's existence in comparison to another; the moral sense refers to which is taken to be morally more important than, or to form the criteria on which to judge, the other. In the temporal sense, Sandel is absolutely right that communities come before, and therefore shape, our conception of ourselves. Of what sorts of things is our conception of ourself constituted? Among other things we see ourselves as having aims and goals, desires, values, a personal history, relationships with other persons, a physical sense of our bodily selves, the ability to evaluate and reconsider and change all of these things, and a language. A l l of these are learned, and therefore learned from others. Presumably, most of these things are learned subconsciously, the way a child learns a language—the child may, at some point, become conscious of the fact that he is learning a language, but he likely remains unaware of many of the lessons about that language that he is learning, while he is learning it. But the fact that we learn these things from the communities we are part of does not lift them beyond moral evaluation; indeed, as I have said, if anything it places a greater demand on us to do what we can to ensure that our families and communities are morally just. Liberalism: Theory and Criticisms 62 Alasdair Maclntyre believes that the conception of the person underlying liberalism is too abstract and universal. 5 3 In contrast, he believes that we understand ourselves through narrative, that what gives intelligibility to ourselves and our actions is to see these as existing within our life described in a narrative framework, and this in connection to the narrative frameworks of other lives around us, and within larger narratives. This account gives us some sense of how one might endeavour to seek the self-understanding that Sandel refers to. But Maclntyre also intends such an understanding of ourselves to be normative; in his terms, "the key question for men is not about their own authorship; I can only answer the question "What am I to do?' if I can answer the prior question 'Of what story or stories do I find myself a part? '" 5 4 For Maclntyre our context is not prescriptive in the sense that we all must act the same, or all members of a particular group or class must meet the same demands. Rather, "I am someone's son or daughter, someone else's cousin or uncle; I am a citizen of this or that city, a member of this or that guild or profession; I belong to this clan, that tribe, this nation. Hence what is good for me has to be good for one who inhabits these roles." 5 5 As W i l l Kymlicka has pointed out, the difficulty with this explanation is that it is not clear how normative Maclntyre intends it to be. 5 6 On the one hand he says that we are to understand individuals as the subject of their narratives, and that "I am never able to seek for the good or exercise, the virtues only qua individual. . . . I inherit from the past of my family, my city, my tribe, my nation, a 5 3Alasdair Maclntyre, After Virtue: A Study in Moral Theory, 2" ed. (Notre Dame, Indiana: University of Notre Dame Press, 1984). See, in particular, chapter 15. 5 4Ibid.,216. 5 5Ibid., 220. 5 6 W i l l Kymlicka, Liberalism, Community, and Culture (Oxford: Clarendon Press, 1989), 56-57. Liberalism: Theory and Criticisms 6 3 variety of debts, inheritances, rightful expectations and obligations." 5 7 From this we might think Maclntyre intends that our choices are already so embedded in our traditions that they are prescribed. But this cannot be entirely true—while it may be necessary to know my personal history and context to have a satisfying understanding of my actions and motivations, we often evaluate our own actions on other criteria. A n d we expect that others wi l l take some of these criteria into account in evaluating their actions—and we have every reason to believe that people do: people do evaluate their actions and their traditions on moral grounds, even when these contradict important elements of their inheritance. A n d they often decide that much of their inheritance is immoral, or unjust, and act to change this. On the other hand, Maclntyre says that "the fact that the self has to find its moral identity in and through its membership in communities such as those of the family, the neighborhood, the city and the tribe does not entail that the self has to accept the moral limitations of the particularity of those forms of community." 5 8 Here Maclntyre seems to be saying that we can step outside of our inherited traditions in some way to judge these traditions. But i f this is what he is saying then it is not clear how he intends this to be different from the liberal position. Maclntyre is right that what is important to people is often contained in traditions that have existed over time, and that in order to understand the motivations of people and how they judge what is important to them we have to know their personal history and context. 5 9 A n d liberals would agree with him that "Mac ln ty re , 217, 220. 5 8 Ib id . , 221 . 5 0 It seems to me an open question as to whether we do always understand ourselves and our context in terms of narratives. D o we not sometimes give reasons for our actions in terms of rational choices, in which we use reasons abstracted from our own personal situations? Maclntyre might respond that to do so we must see ourselves as inheritors of a tradition of reason-giving, but then his description of how we understand ourselves becomes meaningless, as narrative becomes a synonym for explanation. Liberalism: Theory and Criticisms 64 "all reasoning takes place within the context of some traditional mode of thought, transcending through criticism and invention the limitations of what had hitherto been reasoned in that tradition; this is as true of modem physics as of medieval logic ." 0 0 But whereas liberals assume that people can make choices about their inherited traditions and the way they live these out, and believe that this choosing is done on criteria and that these criteria are themselves debatable, Maclntyre does not make it clear whether he believes people can do these things, or to what extent they can. Liberals must believe that individuals can evaluate differing traditions against one another, and consider what criteria should or could be used to evaluate traditions, but it is not clear that Maclntyre considers this possible. Taylor has attempted to explain how he believes our sense of identity and our values are embedded in what he calls frameworks. For him, "a framework incorporates a crucial set of qualitative distinctions. To think, feel, judge within such a framework is to function with the sense that some action, or mode of life, or mode of feeling is incomparably higher than the others which are more readily available to us." 6 1 Our frameworks are worthy or desirable to us, but not "on the same scale as our ordinary ends, goods, desirabilia. They are not just more desirable, in the same sense though to a greater degree, than some of our ordinary goods are." 6 2 Rather, these frameworks frame our lives, our sense of identity, and our choices. Frameworks provide the background, explicit or implicit, for our moral judgements, intuitions, or reactions in [our moral life]. To articulate a framework is to explicate what makes sense of our moral responses. That is, when we try to spell out what it is that we presuppose when we judge that a certain form of life is truly worthwhile, or place our dignity in a certain achievement or status, 0 U Ib id . , 222. 6 l Char les Taylor, Sources of the Self: The Making of the Modern Identity (Cambridge, Mass.: Harvard University Press, 1989), 19. 6 2 Ib id . , 20. Liberalism: Theory and Criticisms 65 or define our moral obligations in a certain manner, we find ourselves articulating inter alia [our] 'frameworks.' 6 3 A n example of such a framework is a warrior ethic, which has guided lives and civilizations and from within which lives exemplifying bravery command awe and fame; another is the "large family of views which see the good life as a mastery of self which consists in the dominance of reason over desire." 6 4 A third, and historically much more recent, example is the affirmation of ordinary life as commanding dignity. Some of these frameworks, such as the warrior ethic, are not 'articulated theoretically,' while others, such as an ethic of self-mastery through reason, require a "theory, a reasoned account of what human life is about." 6 5 Taylor is not entirely clear on this point, but he seems to believe that we cannot fully understand these frameworks; they provide the scale against which we measure our other goods and desires, but they themselves are beyond rational analysis. At the same time, doing without frameworks is utterly impossible for us; otherwise p u t . . . the horizons within which we live our lives and which make sense of them have to include these strong qualitative discriminations . . . l iving within such strongly qualified horizons is constitutive of human agency,. . . stepping outside these limits would be tantamount to stepping outside what we would recognize as integral, that is, undamaged human personhood. 6 6 He therefore considers these frameworks 'inescapable.' 6 7 For Taylor, these frames come from our cultural backgrounds, or from our religious beliefs. "What this brings to light is the essential link between identity and a kind of orientation. To know who you are is to be oriented in moral space, a 'Ibid., 26. 'ibid., 21. Ibid., 20. 'Ibid., 27. 'ibid., 31. Liberalism: Theory and Criticisms 66 space in which questions arise about what is good or bad, what is worth doing and what not, what has meaning and importance for you and what is trivial and secondary." 6 8 Thus, for him, to say who we are is to identify these commitments and valuations. Taylor believes that the modem, liberal conception of the free individual depends upon notions of self-dependence and re-inventing oneself entirely from scratch. It is thus, for him, empty. Complete freedom [is] a void in which nothing [is] worth doing, nothing . . . deserve[s] to count for anything. The self which has arrived at freedom by setting aside all external obstacles and impingements is characterless, and hence without defined purpose, however much this is hidden by such seemingly positive terms as 'rationality' or 'creativity.' These are ultimately quite indeterminate as criteria for human action or mode of life. They cannot specify any content to our action outside of a situation which sets goals for us, which thus imparts a shape to rationality and provides an inspiration for creativity. 6 9 Taylor terms the belief in the possibility of viewing the world from outside any frameworks the 'naturalist' view. But it is not clear why liberals must be committed to it—indeed, the liberal is the inheritor of the rationalist framework or tradition which Taylor ascribes in large part to Plato. Where the liberal is likely to disagree with Taylor is on our ability to rationally analyze these frameworks, and it is this that Taylor seems to be claiming implies an ability to step outside of our frameworks altogether. But it is not clear why this must be our only choice. As Jenold Coombs has pointed out, one can accept a notion of frame-works without being committed to the view that the frames themselves must be beyond rational analysis. There is room for evaluation within frames themselves, because "there are always diverse strands of thought and competing values within 68Ibid., 28. 6 9Charles Taylor, Hegel and Modern Society (Cambridge: Cambridge University Press, 1979), 157. Liberalism: Theory and Criticisms 67 any 'frame'. Moreover 'frames' are always evolving as the result of the experiences of the persons who have them . . . Also frames are not static, nor 70 immune to outside influences." There is-also room for evaluation between frames, as many persons participate in more than one community—so they experience more than one 'frame.' . . . [Also], any two groups of human beings, no matter how diverse their 'frames' wi l l share many common values, concepts and commitments. This is because as humans they need to eat, protect themselves from harm, procreate, etc. Any view of frames that sees them as enclosing solipsistic worlds is just [plain] false on the facts. 7 1 But this is not to suggest that we evaluate frames against each other, rather, "What 72 we evaluate are things like practices, policies, theories and institutions." Taylor does not say so explicitly, but for the most part we must inherit our frameworks; for him, i f we can make any choice at all within and between frames it must be as a leap of faith, because we are unable to make the relevant evaluation. It must be for this reason that he advocates a form of republicanism in which values like justice are valued through a form of patriotism—a collective valuing which he believes increases the value precisely because it is collectively felt.7"' But Taylor is left unable to argue for any particular value. Why would the citizens of Taylor's republic not value their shared hatred for another country or culture? Or take stubborn pride in an ignorance about the rest of the world? Why could not such values help shape a community's republican world-view? The arguments against such commitments are pragmatic and liberal, and therefore not available to Taylor on his model. °Jerrold Coombs, personal communication, October 2002. 7 1Ib.d. 7 2Ibid. 3Taylor, "Cross-Purposes," 165-70. Liberalism: Theory and Criticisms 68 It may well be that we do have some kind of framework, or frameworks, as Taylor has described them—assumptions, beliefs and commitments, closely allied with our culture and which, for the most part, we inherit. It is probable also that these often go unquestioned and unanalyzed, and that elements of them may even always exist beyond scrutiny. But I think, as Coombs has suggested, we believe we can analyze them to some extent, and to that extent should. Are they just? Do they harm people? What are the outcomes of having such beliefs? Taylor might argue that I am just viewing the world from within my own rationalist, liberal framework, and unable to see the entirely different values I would have from another. But I think this is false—we do talk to one another, we do have sympathy for one another's views, we do put a lot of effort into understanding one another, and we do often achieve this to the apparent satisfaction of the other. But even i f Taylor were right the liberal can respond that her framework at least allows us to consider that we may have a responsibility to go beyond simply accepting our assumptions about the world; Taylor would seem to be denying us this line of examination. F E M I N I S T C R I T I C I S M S . Feminism's original aim was ending the oppression of women. But the base of feminist theory has widened in an attempt to understand and respond to different contexts of oppression because of challenges raised by women belonging to other oppressed groups, such as black women and lesbians, against feminist theory whenever it attempted to draw broad generalizations about the status or situation of women. As a result feminist theory has become closely allied with the struggle against oppression by any group over others. Feminism raises questions of liberal theory at several levels. A t one level it has challenged liberal theorists by challenging the gender-bias inherent in Liberalism: Theory and Criticisms 69 existing institutions. A n important example of this is its challenge of the private-public distinction, and in particular the alignment of the private sphere with the family. Since liberalism puts such an emphasis on the right of each individual to form and pursue her or his own conception of the good, along with such rights as the right to free association, the right to solitude, and the recognition of the time and privacy required to foster and sustain intimate relationships, it places a high value on privacy. But, traditionally, along with recognizing men as heads of the household and as the primary actors in the public/political realm, many liberal theorists have regarded the family and household as his private realm. As feminism has challenged each of these assumptions—that males are the head of households, and the primary actors in civi l society—it has also challenged the assumption that what goes on in the family lies within the private realm. For the demarcation of the family unit as one requiring privacy has often resulted in protecting conditions of inequality and abuse from public scrutiny and legal action. But feminism's challenge goes both ways; feminists have not just challenged the right to keep acts and relations that occur in the family out of public scrutiny, they have also challenged the lack of recognition of the influence of domestic life on the public sphere. As Jane Mansbridge and Susan Moller Okin have written, "Challenging the public/private dichotomy means insisting on the non-triviality, the non-exclusion from central public debate, of intimate, domestic concerns. It means insisting that what goes on between a man and a woman in their home, even in their bedroom, is created by and in turn creates what goes on in legislatures and on battlefields." 7 4 This is to argue that the family gets treated as any important association within society would—appropriate recognition for one of the most common and important associations most of us share. 74Jane Mansbridge and Susan Moller Okin, "Feminism," in A Companion to Contemporary Political Philosophy, eds. Robert E. Goodin and Philip Pettit (Oxford: Blackwell Publishers, 1995), 274. Liberalism: Theory and Criticisms 70 Note that this is not to imply that privacy is not important to feminists. Feminists recognize the need for solitude and quiet reflection, the need to develop and nurture intimate relationships, the need for a space from which one can critically consider and raise questions if necessary of the dominant institutions and prevailing assumptions. In this they agree with liberals who claim that [The] boundary between spheres does not imply that private life is radically apolitical or antisocial. Private life means life in civil society, not some presocial state of nature or antisocial condition of isolation and detachment. Liberation from traditional ascriptive attachments does not indicate naked individualism or anomie. . . . Private liberty provides escape from the surveillance • and interference of public officials, multiplying possibilities for private associations and combinations. . . . Far from inviting apathy, private liberty is supposed to encourage public discussion and the formation of groups that give individuals access to wider social contexts and to government. 7 5 This suggests that liberal principles can be used both to challenge some of the extent and the effects of the demand for privacy, and to bolster the arguments for privacy. Thus liberalism may demand a redrawing of the traditional line between the public and the private, without offering a firm answer as to where a redrawn line should be. Liberalism may inform the arguments here without fully resolving them. It is important to recognize that feminists have often argued against the private-public dichotomy through using liberal values. Thus, often they are arguing against the work of individual liberal theorists, rather than liberalism itself. For instance, when feminists argue against treating what goes on in the family as totally private they are arguing that families are not co-extensive with individuals, but with associations-—albeit close, intimate associations—between individuals. Thus they are calling for a better understanding of the existence and 7 5 N a n c y L . Rosenblum, Another Liberalism: Romanticism and the Reconstruction of Liberal Thought (Cambridge, Mass.: Harvard University Press, 1987), 61. Liberalism: Theory and Criticisms 71 context of oppression, and an application of the principles of justice and equality where before these were lacking. As Alison Jaggar has pointed out, as feminism "challenges the dominant tradition of western ethics, [it] simultaneously contributes to that tradition." 7 6 Quoting Margaret Walker and invoking Audre Lorde, she says that The values [feminism] invokes 'are of specifically democratic, participatory, and emancipatory kinds, squarely founded on moral and political ideals of modern Western social thought.' Rather than scrapping the master's tools, many feminist philosophers are working to transform them so that we may build a moral household 77 that has no head nor master. Thus significant dimensions of feminist criticism can be seen as contributing to liberal theory and a better understanding of its own values, rather than attempting to dismantle and replace it. There are other examples of how feminism has challenged the gender-bias present in many contemporary institutions, institutions that are understood to either arise from or be protected by liberal theory. But the public-private distinction is an important one because it overlaps with so many other examples, such as the assumption that males form the head of households. Carole Pateman goes so far as to claim that "The dichotomy between the private and the public . . . is, ultimately, what the feminist movement is about." 7 9 7 6 A l i s o n Jaggar, "Feminism in Ethics: Mora l Justification," in The Cambridge Companion to Feminism in Philosophy, eds. Miranda Fricker and Jennifer Hornsby (Cambridge: Cambridge University Press, 2000), 242. 7 7 lbid. 7 8 Susan Mol l e r Ok in has analyzed the assumption made by Rawls in A Theory of Justice that the 'representative man' in the original position is the head of a household, and argues that families must be given the same just consideration as the rest of society. Justice, Gender and the Family (New York : Basic Books, 1989). 7 9 Caro le Pateman, "Feminist Critiques of the Public/Private Dichotomy," in Public and Private in Social Life, eds. Stanley I. Benn and Gerald Gauss (New York : St. Mart in 's Press, 1983), 281. Liberalism: Theory and Criticisms 72 At a deeper theoretical level feminism challenges some of the underlying assumptions of liberalism, such as those concerning human nature and the nature of justice. For example, Jaggar argues that liberalism depends on the "assumption that human individuals are essentially solitary, with needs and interests that are separate from if not in opposition to other individuals." Jaggar is not alone in believing this about liberal theory; this reflects a theme common to much feminist thought. For example, Evelyn Fox Keller writes "that autonomy can be bought only at the price of unrelatedness," and Seyla Benhabib says that feminist and other critics of Enlightenment universalism "have questioned the abstract and disembodied, distorting and nostalgic ideal of the autonomous male ego which the l 81 universalist tradition privileges." This assumption gives rise, Jaggar believes, to a contract theory that attempts to answer how essentially solitary and conflict-prone individuals might agree to enter into a civi l society together, and how conflict could be prevented in such a situation. Thus liberalism is often concerned 82 foremost with "the protection of life, c ivi l liberties and property." But this goes against the actual facts of human existence, and the need humans have for one another for their survival. Individual self-sufficiency . . . is an unrealistic assumption even i f one conceives of all human beings as healthy adults, which most social contract theorists have done. As soon as one takes into account the facts of human biology, especially reproductive biology, it becomes obvious that the assumption of individual self-sufficiency is impossible. . . . [I]n order to raise enough children to continue the species, humans must live in social groups where individuals share resources with the young and the temporarily disabled. Human interdependence is thus necessitated by human 8 0 A l i s o n Jaggar, Feminist Politics and Human Nature (Sussex: Rowman and Allanheld, 1983), 40. 8 l E v e l y n Fox Keller , Reflections on Gender and Science (New Haven, Conn.: Ya le University Press, 1985), 72. Seyla Benhabib, Situating the Self: Gender, Community and Postmodernism in Contemporary Ethics (Cambridge: Poli ty Press, 1992), 3. 8 2Jaggar, Feminist Politics, 40. Liberalism: Theory and Criticisms 73 biology, and the assumption of individual self-sufficiency is C O plausible only i f one ignores human biology. Jaggar believes that, i f liberals took seriously our need for one another, and treated this as a fundamental aspect of human nature rather than a tendency to conflict and solitude, then, "Instead of community and cooperation being taken as phenomena whose existence and even possibility is puzzling, and sometimes even regarded as impossible, the existence of egoism, competitiveness and conflict, phenomena which liberalism takes as endemic to the human condition, would 84 themselves become puzzling and problematic." But it is not clear what different results such a starting point for a conception of social union would lead to. If liberal theory is not always clearly premissed on laying out the groundwork for community and co-operation, then at the very least it should always make room for these. But such things as egoism, competitiveness and conflict, while perhaps not necessarily parts of human biology, do seem to be regular elements of human existence and social union. Any theory which attempts to explain the terms on which people can come together to co-operate has to take into account the fact that people wil l at times attempt to deceive each other, wi l l desire an unequal share of resources, and wil l fail to respect each other's rights, and wil l have to make clear that these are not workable grounds for successful long-term social union and the flourishing of co-operation. Indeed, the assumption that people are individually motivated to acts of altruism and co-operation would seem to vitiate the necessity of a formal structure for social union (except perhaps a need for limits on acts of selflessness to the detriment of one's own long-term good), since people would simply get along with and help one another. Tbid., 40-41. 4Ibid.,41. Liberalism: Theory and Criticisms 74 It seems that any useable description of how people could come together that would adequately meet Jaggar's concerns would make as much room as possible for people to co-operate with one another, help one another, form the kind of deep, meaningful attachments they want to form and require in order to survive and reproduce, while at the same time minimizing the presence or effect of such harmful forces as serious disagreement over relative rights to resources, bigotry, lack of respect, and so on. But this is just what liberalism attempts to do; Jaggar does not make clear what different results she believes her feminist starting point would yield. Another example of feminist criticism of liberal theory, and an important one in contemporary ethical theory, is the contention that liberalism's emphasis on justice is inappropriate. In the 1980s Carol Gill igan criticized Lawrence Kohlberg's cognitive theory of moral development as failing to represent women's morality. Kohlberg utilized Rawls ' conception of justice in developing his stage theory; in particular, Kohlberg's ultimate stage reflected a 'universal-ethical-principle orientation' toward moral situations, where "Right is defined by the decision of conscience in accord with self-chosen ethical principles appealing to logical comprehensiveness, universality, and consistency. . . . At heart, these are universal principles of justice, of the reciprocity and equality of human rights, and of respect for the dignity of human beings as individual persons."*5 Gilligan argued that the moral ideal of being concerned with "separation, autonomy, individuation, and natural rights" was not a description of moral maturity per se, but a description of a masculine moral maturity. 8 6 What 8: ,Lawrence Kohlberg, "The Claim to Moral Adequacy of a Highest Stage of Moral Judgment" Journal of Philosophy 70 (1973): 632. 8 6Gilligan, 23. It remains contentious whether Gilligan was talking about a women's perspective or just a different moral perspective. While she starts by saying that "The different voice I describe is characterized not by gender but by theme," she goes on to discuss women's moral development in contrast to men's (2). Liberalism: Theory and Criticisms 75 psychological theory lacked, she said, was an understanding of what it meant to be a morally mature woman, because "the very traits that traditionally have defined the 'goodness' of women, their care and sensitivity to the needs of others, are those that mark them as deficient in [the traditional study of] moral development." Gilligan juxtaposed a caring perspective with the traditional (or traditionally -/male) perspective of being concerned with rights, autonomy, individuality and justice. Authors such as Nel Noddings have developed on Gill igan's work, attempting to explain what a moral orientation focused on care oo rather than justice and rights would look like. While Noddings described the mother-child relationship as the prototypical caring relationship, she does not on believe that a caring orientation is limited to women. One simple way to divide possible responses to these two views on morality is into the positions that care is somehow morally inferior to a principled perspective, that the two views are in some way equal or complementary, or that care is a more important or 'moral' consideration than reasoning via some set of universal ethical principles. Several variations can be advanced for each of these positions, but they broadly define the responses taken to the two views. 9 0 Liberals have commonly responded by occupying one of the two former positions. 9 1 It can seem that the dictates of care are not always clear, and that in general some form of principle-based ethics wi l l be clearer in its prescriptions. 8 / I b i d „ 18. 8 S See, in particular, Ne l Noddings, Caring: A Feminist Approach to Ethics and Moral Education (Berkeley: University of California Press, 1984). 8 < J Nel Noddings, " A n Ethic of Caring and Its Implications for Instructional Arrangements," American Journal of Education 96 (February 1988): 219. 9 0 F o r instance, Lawrence B l u m has considered eight alternative positions that defenders of a universalist ethics might take in responding to Gil l igan 's work. Seven of these are different versions of the claim that a care perspective is somehow inferior to a principled one. Lawrence B l u m , "Gi l l igan and Kohlberg: Implications for Mora l Theory," Ethics 98 (1988): 472-91. Liberalism: Theory and Criticisms 76 For instance, the fact that we care deeply about someone who is in conflict with another may lead us to not want to have to treat that third person fairly. Noddings deals with such possible difficulties by saying that any interaction with others is an opportunity to care, and that a truly caring action is one that is recognized as 92 such by the cared-for. It can also seem clearer to claim that a particular situation is to be considered as moral, or as demanding moral reasoning, when it can be seen to involve moral principles. For instance, i f a situation is one in which respect might be involved, then we know it demands moral deliberation. But this may be to start from a principled perspective. A care theorist might claim that any situation in which care for another is or could be involved demands such consideration. Without resolving the conflict as to whether one perspective is a more important moral consideration, it seems reasonable to say that both perspectives can be seen to inform the other, and are necessary parts of a complete moral picture—whether morality is to be seen as at root dependent upon principled reasoning or a caring orientation toward the world, or indeed any other basis. That is, a desirable moral stance in the world from any perspective requires both an understanding of moral principles and good judgment in applying them, and the care required to be attuned to the needs of others and to desire to undertake the appropriate response. A n important point is that "liberalism was not meant to be a complete account of morality. It was Kohlberg, not liberals in general, who elevated a concern with justice to be the central concern of morality."9"1 So feminist charges against justice-based reasoning as a description of our moral reasoning may fail to 9 lFor instance, Kohlberg is said to have defended variously both of theses positions. Ibid., 482-83. 92Nel Noddings, Starting at Home: Caring and Social Policy (Berkeley: University of California Press, 2002), chapter l. 93Coombs. Liberalism: Theorv and Criticisms 77 effectively challenge liberalism; what we need to be clear on is how feminist criticisms challenge or change the underlying theory of social union. One result of seeing care as an alternative or additional moral perspective is to emphasize the moral importance of relationships, and possibly to understand that these have traditionally been perceived as more important to women. Thus it may be that men, including almost all of the liberal theorists considered above, "are more likely to retain an implausibly individualistic outlook and to seek an impersonal, impartial, universalistic stance for moral reasoning." 9 4 So, for instance, Benhabib has laid the charge against Rawls that "the construction of the 'original position' [is] an implausibly restricted process of individual deliberation rather than . . . an open-ended process of collective moral argumentation." 9 5 Benhabib believes that there is an internal inconsistency between the veil of ignorance required by the original position and the requirement of the moral point of view that it be able to reasonably consider the 'concrete other. ' 9 6 To see the concrete other, she believes, "requires us to view each and every rational being as an individual with a concrete history, identity and affective-emotional constitution. In assuming this standpoint, we abstract from what constitutes our commonality, and focus on individuality." 9 7 Thus such concepts as "autonomy, impartiality, rights, liberty and social contract can be interpreted in ways that presuppose highly separate individual selves who lack mutual concern and deep no interpersonal attachment." Her argument, therefore, is a procedural one: she 9 4 M a r i l y n Friedman, "Feminism in Ethics: Conceptions of Autonomy," in The Cambridge Companion to Feminism in Philosophy, eds. Miranda Fl icker and Jennifer Hornsby (Cambridge: Cambridge University Press, 2000), 209. ' ; 5 Seyla Benhabib, "The Generalized and the Concrete Other: The Kohlberg-Gi l l igan Controversy and M o r a l Theory," in Situating the Self, 169. % I b i d „ 164-70. '"ibid., 159. ""Friedman, 222, n. 16. Liberalism: Theory and Criticisms 78 believes that the Rawlsian construction of the original position is flawed because it is a process of individual deliberation which does not allow us to fully appreciate what she calls the 'alterity of the other,' the other in her or his concrete situation and with irreducible differences. She believes it should be replaced with a discourse ethics in which: an actual dialogue between moral agents takes place; there is full knowledge of the agents and their situation, their history, their values and so on, and that more knowledge of this sort is seen to improve the dialogue and likely outcome rather than detract from it; nothing is restricted from the conversation and, in particular, not only is the conversation about the goods that are desired, it is also about such things as the desires themselves; and, finally, the participants can talk about the actual conditions and restraints under which the dialogue takes place." Benhabib sees at least two conclusions that would arise from such a procedure. She believes that moral discourse would encompass questions of the good life as well as issues of public justice, and that the assumptions about needs and interests which she believes sustain rights claims would be made clear. Benhabib raises a very important point about the use of the original position. It is always possible that those who are actually engaged in the debate about the shape of institutions of justice wi l l make assumptions based on their own identity in that society, which would be hidden under a 'veil of ignorance' in a discussion emulating the one in the original position. Of course, this is not how Rawls intends his construction; for him, the actors in the original position do not pretend not to be aware of the positions they occupy in society, they truly are ignorant of these. For this reason we can never really have an original position, it remains a rhetorical device intended to help us reflect on our discourse about justice. Of course, Benhabib is attempting to level a stronger charge at Rawls ' Benhabib, 169. Liberalism: Theory and Criticisms 79 work, that his conception is brought about by his own assumptions about what constitutes a good life. But to make this point Benhabib has to say in what way Rawls ' conception actually privileges some people. A more important point that Benhabib raises, I believe, is that liberals must not forget to include real people with real backgrounds in the discussion of what constitutes a just society. That is, the result of a discussion among a very few about the best theoretical underpinnings for social union may indeed privilege a few members of that society; the ones most likely to be privileged, even unintentionally, w i l l , of course, be those people engaged in such a discussion. We cannot be certain that a particular conception privileges some if all members of society do not have a voice in the discussion. (It is important to add that including all in the discussion is not a sufficient criterion for guaranteeing that some wi l l not be oppressed, but it is an important step.) A n important result of having such voices should be the recognition of the real history and context that the different members of the polity actually have. But this is not to deny the power of Rawls ' original position as a rhetorical device. As Rawls points out, an abstract discussion generates an ideal theory, which we then compare against our intuitions and our actual considered judgments about justice. If these do not agree, then we allow them to inform each other, reconsider what assumptions might be informing our intuitions or the starting place for our theory, and start again. In this way both our theory and our practice continually inform the other and we are able to come to a deeper understanding and justification of both our practice and our theory. 1 0 0 The original position by itself cannot be expected to produce a final conception of justice, Rawls, Theory of Justice, § 9. Liberalism: Theory and Criticisms 80 because it "is to be seen as a device of representation and hence any agreement reached by the parties must be regarded as . . . hypothetical." 1 0 1 At the same time, the notion of a veil of ignorance reminds us that certain personal considerations and values should not be allowed to inform the creation of a just theory. Just which considerations and values should be left out, as Benhabib points out, should be debated, and debated by all the members of society who stand to be affected by its institutions. While in general we would not want to privilege some groups of people over others, characteristics of groups which are oppressed may indeed demand our attention—but they demand our attention because oppression is wrong, not because a particularly vocal member of the discussion is oppressed in that way. In this chapter I have summarized how some of the major theorists have described liberalism, and I have canvassed what I take to be the strongest criticisms of liberal theory. If a conceptualization of autonomy intended to have an impact on our educational practices is to be defensible, it must both meet the demands of liberalism and not succumb to the legitimate criticisms raised of it. I have shown that I do not believe all the criticisms of liberalism to be justified. Nonetheless I think that, taken together, these criticisms add up to a general charge that we must take into account, that liberalism is not seen to accommodate much of what is important to us as people. If autonomy is not to be seen as describing an abstract or atomistic conception of the person devoid of any commitments to others it must be made clear how it can accommodate a conception of the person as embedded in communities and contexts. I wi l l attempt to respond to this challenge in chapter 5. It is now possible to summarize the criteria necessary for an adequate conceptualization of autonomy that arise from the demands of liberalism and the criticisms of it. These criteria wi l l allow us to 'John Rawls, Political Liberalism (New York: Columbia University Press, 1993), 24. Liberalism: Theory and Criticisms 81 examine existing conceptualizations of autonomy in educational discourse more deeply, and should give us a defensible basis on which to re-conceptualize it. This is the topic of the next chapter. C h a p t e r 3 : C r i t e r i a f o r a C o n c e p t u a l i z a t i o n o f A u t o n o m y In this chapter I will summarize the criteria we can derive from liberal theory and its critics that a conceptualization of autonomy must meet. A notion of autonomy that will be useful for guiding educational policy will also at least not contradict what we understand to be other worthwhile educational goals, and will preferably help us to understand the importance of these other goals. It is not my purpose to give a full description of the aims of education here, but it will be useful for this project to summarize some of the more widely accepted general aims of education. Let me first make a couple of preliminary comments about the requirements that any discussion of autonomy must meet. As we have seen, as a moral and political concept autonomy has drawn much criticism. I take it that a good conceptualization should be fairly obviously desirable, though I do not think it likely (or even possible) that any conceptualization would be above criticism. Autonomy is a term not without some history and an already established place in liberal theory; after Kant it has generally meant some form of self-legislation, invoking a person who obeys universal laws because of his or her own "recognition of [his or her own] obligations both to obey the moral law and to 82 Criteria for a Conceptualization of Autonomy 83 exclude any determining influence outside [his or her] own reason."1 For Kant, obeying the moral law was being moved to be governed by the demands of one's own reason. Thus Kant was unable to give any content for a moral law, because doing so would be dictating an external morality—exactly what Kant was trying to move away from. There are elements of this that I think we intuitively want to appeal to when we talk about autonomy, if we do not want to lose any semblance of a notion of 'self-rule.' Self-rule suggests rule by the individual of himself in accord with his own reasoning. But I think we need to add that his reasoning must be of a certain standard, or that it must, in some way, be good reasoning. In other words, the individual is not free, or acting in a way that we would call ruling himself, if he acts on his own reasoning but his reasoning is poor. The standards for good reasons are public ones, so in a sense the standards for good reasons are external to the individual, although in the sense that the individual has to understand and agree that these are good standards, and desire to meet them for him or herself, they are internalized. This is distinct from Kant's notion of the individual acting such that he could will his actions to be universal, but in which there is no external authority to which this must appeal. Among the relevant standards are moral ones, such as respect, honesty and fairness. These are important because they are universalizable standards for interacting and dealing with others. If we are to be able to discuss the demands of co-operating with one another, or to enter into agreements with each other, or understand the need to respect the bounds of another's privacy, then we must act in ways consistent with such standards. 'Roger J. Sull ivan, An Introduction to Kant's Ethics (Cambridge: Cambridge University Press, 1994), 47. According to Kant, the wi l l is subject to the law " in such a way that it must be viewed as also giving the law to itself and just because of this as . . . subject to the law." Immanuel Kant, Groundwork of the Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge University Press, 1997 [1785]), 431. Criteria for a Conceptualization of Autonomy 84 So one criterion for autonomy will be that the individual acts in accord with the demands of such standards as respect, honesty and fairness. This may appear to be a different use of the term than is currently intended by modern liberal philosophers. John Rawls, for example, makes clear that he wants to distinguish between his version of the "full autonomy of political life . . . from the ethical values of autonomy and individuality, which may apply to the whole of life, both social and individual, as expressed by the comprehensive liberalisms of Kant and M i l l . " 2 Rawls discusses two versions of autonomy. For him, rational autonomy "rests on persons' intellectual and moral powers."3 Thus it is demonstrated by persons choosing their own ends, their conception of the good, reasoning in accord with this, and also in their ability to enter into reasonable agreements with others. The second version of autonomy, full autonomy, "is realized by citizens when they act from principles of justice that specify the fair terms of cooperation they would give to themselves when fairly represented as free and equal persons."4 That is, full autonomy is evinced when persons act in ways that respect the terms of co-operation they would choose in the original position. It is full autonomy that Rawls intends to be political, not comprehensive. But Rawls' full autonomy requires moral principles. This is because it is represented by citizens acting and deliberating in a position of freedom and equality. And for Rawls, "citizens are equal in virtue of possessing, to the requisite minimum degree, the two moral powers and the other capacities that enable us to be normal and fully cooperating members of society."5 The two moral powers are "the capacity for a sense of justice and the capacity for a 2John Rawls, Political Liberalism (New York: Columbia University Press, 1993), 78. 3Ibid.,72. 4Ibid., 77. 5Ibid.,79. Criteria for a Conceptualization of Autonomy 85 conception of the good." 6 Rawls ' account of autonomy is thus one of citizens who are wil l ing to co-operate together as free and equal citizens, and who are will ing to be moved by the demands of a public morality—or the rules necessary for such fair co-operation as equal citizens. At the same time he wants to make room for individuals to have their own private comprehensive conceptions of the good, although he says these must be restricted by the demand that they be reasonable. I am claiming much the same thing when I say that the demands of acting in accord with good reasoning include acting with one another in ways that are sustainable, that are consistent and universalizable. 7 Thus persons in a liberal society would, for example, be bound by the demand to be honest in their dealings with one another, to be fair, and to treat one another with respect, that is, to treat what is important to others as important for that reason. There are some other basic criteria that I think a useful conceptualization of autonomy should meet. Some of these wil l be obvious, like the demand that one person's having or evincing autonomy must be consistent with others also having or acting in an autonomous manner. This means that autonomous actions cannot include actions that diminish another's capacity for autonomous action. This is a straightforward result of the demands of liberty and equality. I also think that autonomy is not as useful as it could be as a descriptor if it is thought of as an ideal that few people can attain. This may remove it too far from the realm of what is readily seen as important to most people. Similarly I think a conceptualization of autonomy that demands an overwhelming amount of self-°Ibid., 81. 7 By universalizable I merely mean that our actions should accord with principles that we all can publicly agree as universally applying. Such prescriptions and proscriptions often have a reciprocal nature and so we all depend upon their being followed by everyone. The proscription against lying is like this: our being believed when we need to be depends upon everyone's believing and following this rule; if people generally just said what they wanted, there would be no basis for trusting anyone. Criteria for a Conceptualization of Autonomy 86 reflection is too far removed from too many possible good lives to be very useful; it seems likely that autonomy must involve some knowledge of a certain kind, or a certain kind of reflection on one's choices, but if it requires of an individual that she spend a great deal of her life in a continuous state of self-reflection then it wi l l be hard to make a case for its appeal to most people, and, I think, rightly so. CRITERIA F R O M LIBERAL THEORY If our notion of autonomy is to be understood within liberal theory, then my discussion of this theory in chapter 2 suggests that, at the very least, autonomous citizens wil l act fairly and justly, wi l l have an understanding of equality, and wi l l be capable of responsibly exercising their own freedom. Additionally, i f autonomy is not to carry with it the threat of a loss of identity or the ability to form deep attachments, then it must be possible for the autonomous individual to belong in a meaningful way to communities or to care in a meaningful way for others. So autonomy must be seen to be compatible with these things. This suggests that the term might be best thought of as a descriptor of persons who are in some way successful as members of the liberal polity. What is required of citizens in the liberal state? The list of abilities and dispositions claimed by John Rawls and Eamonn Callan as required by the liberal virtues, that I cited in chapter 1, is a good summary of the demands of justice on liberal citizens that I discussed in chapter 2: Justice requires of citizens that they understand, and are able to reflect on and act within the norms prescribed by, a notion of public justice. Rawls ' use of the original position suggests that this requires that citizens can and are wil l ing to give equal consideration to the claims of all and take the welfare of all equally seriously, that they are able to reason abstractly, and that they can imagine what it would be like to be in someone else's situation and to view disagreements from their point of view. Citizens also have to treat what others think is important as Criteria for a Conceptualization of Autonomy 87 important, simply because these things are valued by others—that is, citizens are required to show respect for others. They have the 'power to form, revise, and rationally to pursue a conception o f their own good and they hold such a conception. They are wil l ing to engage in public discourse, are able and wil l ing to propose and discuss fair terms of co-operation with others, and they are wil l ing to abide by the terms of co-operation so long as these terms are fair and others are also wil l ing to abide by them. Disagreement wi l l arise in society over worthwhile conceptions of the good, the ends that are to be pursued, and the support that various potential goods are to receive from the state. Autonomous citizens are wil l ing to accept this disagreement and its sources, as well as the fact that the terms of co-operation must accommodate the various beliefs as comprehensively as possible. A n d they have the general powers of deliberation, reason and judgment required to conduct one's own affairs and to participate in the state. Citizens in a liberal society must also have an understanding of equality, what it means for persons to be valued and treated equally, and why this does not necessarily lead to people being treated the same. Elizabeth Anderson's analysis of equality suggests that citizens need to have an understanding of inequality, they must be sensitive to oppression, and they must be motivated to act against these in instances where they believe these are occurring. Citizens in a liberal society need, to have an understanding of rights, a desire to have their own rights respected and upheld and a respect for the rights of others, along with an understanding of the reciprocal nature of the relationship between the protection of our own individual rights and our meeting the demands of the shared responsibility with- all other individuals with whom we share these rights. Anderson's analysis of equality as the ability of each to participate equally in the state points out that we must value those things that this requires, like literacy and the right to vote, and that we must be will ing to work toward acquiring and protecting these. This requires that citizens value education to at least some minimal level. It also requires that we Criteria for a Conceptualization of Autonomy 88 can make reasoned and responsible decisions about how we wi l l exercise our rights; for instance, it may be that I should be free to choose not to exercise my right to housing, but in order for this to be a free choice I need to be able to both give good reasons for my choice and know that, i f I were to desire housing, I could attain it. Liberty demands that citizens have an understanding of the need for laws to protect freedom; that appropriate laws increase and protect liberty rather than infringe upon it. M i l l points out that liberty requires a 'maturity of faculties,' which must include the willingness to take responsibility for one's own choices. From M i l l we also get that, i f freedom is to be fully appreciated, individuals require the ability to make good choices, they should hold a variety of opinions, and they must have the ability to debate the truth. M i l l believed that persons who are free in this way are more likely to display spontaneity and originality and variety in their lives, necessary elements of what he believed to. be a free and advancing civilization. These abilities require openmindedness, an ability to learn from others, an ability and willingness to be moved by reasons, a willingness to be exposed to other ways of life and the ability to make judgments as to their worth, and the ability to give reasons in justifying one's own choices. And , as Charles Taylor points out, if freedom is to have real meaning we must have the ability to make worthwhile choices with regard to our aims, purposes, goals and values. These are the demands placed on persons by liberalism. A n autonomous person in a liberal society must be able to meet these demands, and so an account of autonomy should help us make sense of at least some of them. For instance, it would seem prima facie that a conceptualization of autonomy would include an account of a person's ability to exercise their freedom to their own best advantage. But it is not so obvious that a conceptualization of autonomy would necessarily explain all of the demands liberalism places on persons; for instance, Criteria for a Conceptualization of Autonomy ' 8 9 it might seem counterintuitive to think that autonomy would involve an account of equality between persons. The more a conceptualization of autonomy can explain the better, but of course it must not do this at the expense of other criteria, such as being clearly reasonable and attainable. C R I T E R I A F R O M C O M M U N I T A R I A N T H E O R Y While a conceptualization of autonomy must be consistent with an account of persons that is compatible with liberal theory, a conceptualization of autonomous persons that is to be successful must also take into account the legitimate criticisms raised of the liberal view of persons. As I showed in chapter 2 I do not think that all of the criticisms raised of liberalism are valid; nonetheless I do believe that there are important elements of what it means to be a person that liberal theory is seen to have neglected. Let me summarize these. Michael Sandel believes that a conception of justice requires a strong, constitutive sense of community. Whether justice requires it or not I do think that many people desire such a strong sense of community, one that helps them to define their values. A n d I think that he is right when he says that being a part of a community is, for many if not for all , more than a feeling, it is a mode of self-understanding partly constitutive of one's identity. A n d we are all embedded in communities and traditions, and so for many an important sense of self-understanding is coming to know these. These are complex, two-way relationships: how we interpret our communities changes how we understand ourselves. A conception of persons that is to apply to most people must therefore accommodate an understanding of the importance of communities to individual identity or sense of self, wi l l not create an artificial notion of the individual as an atomistic unit with a sense of self not tied to others, and wi l l allow communities to thrive and flourish while protecting the rights of the individual within them. It wil l also clearly not diminish the sense of community or solidarity or caring Criteria for a Conceptualization of Autonomy 90 between people required for them to be motivated to act justly toward one another. Alasdair Maclntyre believes that we must be able to see our particular selves within any conception of the person; he believes that the abstract and universal conception of the liberal person denies our individuality. The fact that many people believe they are not represented by the liberal conception should give liberals some pause; this is some evidence to suggest either that the liberal conception of the person might be mistaken and in some way fails to capture an important element of what it means to be a person, or that liberals have failed to communicate their conception very well. But there is nothing in the description I have given of liberalism to suggest it is predicated on a conception of the person as an atomistic unit with a sense of self not tied to others, or as abstract and universal in a way that denies our individuality. In fact, as I hope to demonstrate in chapter 5 , liberal theory as presented by its major theorists and as I have described it here is entirely compatible with a conception of the person as embedded in communities and traditions that serve to define the individual's identity, and her relationships. This suggests that the problem is rather that liberals have failed to communicate adequately that their theory is not predicated on such a conception of the person as either an atomistic unit or an abstract, universal entity. Maclntyre believes that we understand ourselves within narratives; I have argued that this is a debatable point, as is just how normative those narratives are understood to be. But I do think that a person's motivations and judgments of what is important to her or him exist—and are only comprehensible within—a personal history and context, and it may also be true that many people understand and explain these through the use of narratives. One important task, then, is to find a way to understand the person as embedded in a context and at the same time able to reflect on and make Criteria for a Conceptualization of Autonomy 91 responsible decisions about this context and her or his own course of action. Maclntyre's republicanism reminds us that individuals have very different histories and contexts that make them very different people with very different ways of understanding themselves and their relation to the world, and perhaps even with different ways of reasoning. On the one hand this leads to the concern that people have available worthwhile ways of reasoning that help them lead out their lives to their own satisfaction, on the other hand this should lead us to a concern that we display an adequate respect for the variety of goods to be found within the large variety of traditions that are lived by people in the world. 9 Charles Taylor's notion of frameworks seems to capture the general com-munitarian charge that there are some things that guide us that either cannot or should not be questioned. The prime argument for this is that there is knowledge, or standards, or criteria, or norms—be they described as our horizons, embedded contexts, traditions or frameworks—beyond our complete recognition and that we can never quite get a full handle on. The liberal response to this has to be, at least in part, that we are responsible for the decisions we make, which need to incorporate the knowledge we can and can reasonably be expected to have, along 8 I do not mean to suggest by this that there are incommensurable ways of reasoning. Rather I believe that the fact that we speak a common language—or common languages that we can meaningfully translate between—and can understand each other, suggests that we all reason through a fairly common and consistent logical structure, though this may be more or less developed. What I do wish to suggest is that the fact that we all start from personal values which may or may not be shared, and understand these through differing sets of experiences and beliefs, means that we wi l l often interpret situations differently, have a variety of opinions with regard to what features of a situation are important and what values are to be emphasized in our conclusions, and arrive at different,conclusions from similar sets of information. Thus, although we may, in a sense, reason very differently, it still makes sense to speak of considerations and principles that have the potential to improve reasoning across the board. I wi l l discuss some of these below and in chapter 5 when I talk about critical thinking. 9 One conclusion that we might derive from communitarian theorists is that any complete description of liberal theory may require a fully worked out conception of community, one that includes its various moral, social and epistemological dimensions. I wi l l not undertake to provide one here, my intention is to explain how we can understand the autonomous citizen in liberal theory. But a satisfying answer to this wi l l have important implications for how we are to view communities. Criteria for a Conceptualization of Autonomy 92 with a belief that together we wil l be able to probe deeper into and come to a better understanding of these traditions, the assumptions embedded in our communities, or the horizons that help to make sense of our lives. (Note that this too implies a community, though one defined epistemologically.) These may not always be fully transparent, but over time we can come to know them, and so ourselves, better, and therefore make even better decisions. To deny this position is to deny our ability to raise questions of justice about our deepest motivations. The communitarian may want to take such a stand, but I do not think the liberal wi l l want to share this stand. The liberal should, however, be concerned with the importance such defining elements have for people, and the relation between belief systems and communities. While there is no reason to think that liberal theory has ever denied or precluded these, it is obviously seen by at least some critics (like Taylor) as not accommodating a sufficient recognition of the importance of communities and their traditions to individual identities, and the ability and desire of people to make commitments to such epistemological and normative frameworks. A conceptualization of autonomy suitable for educational theorizing must adequately recognize the important role that the communities we are a part of have in teaching us such things as what counts as good, what is valuable, and what the appropriate criteria for value judgments are. What is required is to be able to see how individuals can be understood both as free and as existing in a context that provides us with our values, interests and morals. We need to know what would count as free and in what ways individuals can be helped to act freely without creating some kind of imaginary independent self, devoid of commitments. I believe this is a very important part of my thesis. I wi l l attempt to explain how we can understand autonomous individuals in this light through the description of autonomy, and the relationship between it and a conception of the person, that I give in chapter 5 . Criteria for a Conceptualization of Autonomy 93 CRITERIA F R O M FEMINIST THEORY Feminists have argued that liberals start with the assumption that individuals are solitary and 'conflict-prone.' In place of this they argue that a conception of persons must take into account that not all persons in society are, or are always, self-sufficient adults, and therefore persons have an empirical, biological need for one another for survival. Persons also have a real need and desire to live together and to look after one another, to form intimate relationships and close personal ties, and to share resources. A satisfactory conceptualization of autonomy, therefore, should be premissed on the assumption that people generally desire to co-operate with one another and to form the kind of deep, meaningful attachments they want and need in order to survive and reproduce. Feminists also argue that any theory of social union must emphasize the moral importance of relationships, and in particular show a concern for "substantive moral matters of care, personal relationships and avoiding hurt to others," and a focus on "contextual detail and emotional responsiveness." 1 0 In part this requires that persons are attuned to the needs of others, and have the desire to undertake the correct moral, or caring, response; more generally it means that our social theory must "recognize the moral importance of emotions, close personal relationships, social relationships generally, and the non-impartial nature of any actual ethical standpoint."1 1 Feminists have also argued that moral reasoning and reasoning about justice involve an attention to the particular, to the actual persons involved as opposed to abstract representations of persons, including their history and their context. The best person to speak on someone's behalf is the individual her or himself; to speak for another is always to risk abusing the power imbalance l 0 M a r i l y n Friedman, "Feminism in Ethics: Conceptions of Autonomy," in The Cambridge Companion to Feminism in Philosophy, eds. Miranda Fricker and Jennifer Hornsby (Cambridge: Cambridge University Press, 2000), 206. " i b i d . , 212. Criteria for a Conceptualization of Autonomy 94 this situation creates (although obviously there are times when such a situation is unavoidable). Thus participants in the social union must desire to participate in this discussion, and have the skills to do so in a productive and meaningful manner. A n d a model for social reasoning must involve all the actual participants in the discussion, not only about the social institutions that are a result of that discussion but also about the format of and constraints on the actual discussion itself. Let me repeat that I do not believe that liberal theory has precluded much of what it has been criticized for being unable to accommodate. I therefore believe that it is not that liberal theory must change in order to accommodate those things its critics point out as important to persons, their sense of identity and their moral motivation, so much as it must be seen to accommodate these things—that is, it must be made clear how it can and does accommodate them. A conceptualization of autonomy that is to be useful, therefore, wi l l not only not succumb to the criticisms raised here of liberalism, but wi l l make it clear as to how it accommodates a conception of persons that have a sense of community necessary for justice, that are able to form deep, caring relationships, and that derive their sense of identity at least in part from their community memberships and the traditions they see themselves a part of. CRITERIA F R O M T H E AIMS OF EDUCATION A notion of autonomy that is to guide education must be understood in relation to our other aims of education. It may be that a conceptualization of autonomy wil l provide an additional argument for a particular educational aim, or it may provide a clearer grounding for an argument we already make in support of an aim. A conceptualization of autonomy might also raise questions of a particular aim that we take for granted, in which case we would have reason to question that aim and the reasons we understand as supporting it. In order to consider how a Criteria for a Conceptualization of Autonomy 9 5 conceptualization of autonomy must be shaped if it is to be educationally relevant and useful, I summarize here what are generally taken to be the major aims of education. I do not intend this list to be exclusive, for people with different understandings of education wil l have in mind additional particular aims for it. But I think this list captures the aims generally understood by all as part of education. Education is practically synonymous with the teaching of knowledge, knowledge as organized in a particular way and as understood to have some value. Traditionally, the knowledge associated with education is organized into the scholarly disciplines of math, the sciences, languages, literature, philosophy and so on. R. S. Peters and Paul Hirst gave what is perhaps the best-known defence of education as being about the learning of knowledge, and of this kind of knowledge in particular. Peters believed that what is important to education, and what distinguishes it from similar activities such as training and discipline, has as 12 its aim the development of an educated person (in his words, an educated man). For him, this is associated with knowledge and an all-round development. Such an educated person does not view knowledge from a purely instrumental perspective, nor is he or she a narrow-minded specialist, rather he or she is "someone who is capable of delighting in a variety of pursuits and projects for their own sake and whose pursuit of .them and general conduct of his life are transformed by some degree of all-round understanding and sensitivity." 1 3 Knowledge thus helps to shape how the individual understands the world and acts within it, as "Being educated . . . involves a capacity for absorption and enjoyment 1 2 R . S. Peters, "Education and the Educated M a n , " in Education and the Development of Reason, eds. R. F. Dearden, P . H . Hirst and R. S. Peters (London: Routledge and Kegan Paul , 1972), 3-18. l 3 Ib id . , 11. Criteria for a Conceptualization of Autonomy 96 which is connected with sensitization to standards which structure activities and pursuits." 1 4 Hirst agreed, saying that "To acquire knowledge is to learn to see, to experience the world in a way otherwise unknown, and thereby come to have a mind in a fuller sense."1 5 He argued that the rational mind structures its experience under conceptual schemes provided by distinct disciplines or forms of knowledge, such as mathematics, the sciences, history and literature. These traditional disciplines are important, he believed, because they embodied four characteristics of worthwhile ways of organizing knowledge: each has concepts peculiar to itself, a distinct logical structure, is testable against experience and has "developed particular techniques and skills for exploring experience and testing [its] distinctive expressions." 1 6 Jane -Roland Martin characterizes the Peters-Hirst educated person as "objective, analytic, rational; they are interested in ideas and things." 1 7 Such a person is incomplete, she claims, because they do not necessarily have any feelings for others. That is, she believed that such a person is ill-equipped to care about others, to act morally, or to be concerned about social issues. In response to this ideal she argues that education must incorporate what she terms the reproductive processes, 'construed broadly.' She includes in this not simply biological reproduction of the species, but the whole process of reproduction from conception until the individual reaches more or less independence from the family. This process [she takes] to include not simply childcare and rearing, but the l 4 Ib id . , 13. 1 5 Paul Hirst, "Liberal Education and the Nature of Knowledge," in Knowledge and the Curriculum: A Collection of Philosophical Papers, (London: Routledge and Kegan Paul, 1974), 40. l 6 I b i d „ 44. 1 7Jane Roland Mart in, "The Ideal of the Educated Person," in Philosophy of Education 1981, ed. Daniel DeNico la (Normal, Illinois: Philosophy of Education Society, 1982), 9. Criteria for a Conceptualization of Autonomy 97 related activities of keeping house, running the household and 18 serving the needs and purposes of all the family members. To this we can add the skills and emotional responsiveness necessary for nurturing friendships, caring for others generally, and perhaps other care-oriented aspects of community life. Martin argues for these on the grounds that without them our notion of education is of the development of someone into having i certain kinds of knowledge and certain dispositions toward that knowledge, but who wil l not "care about [the] welfare [of others], let alone to act kindly toward them. That person wil l have some understanding of society, but wi l l not have been taught to feel its injustices or even to be concerned over its fate." 1 9 Critical thinking is commonly mentioned as an aim of education by teachers, academics, and those responsible for provincial and state curricula. But it is often discussed in vague terms, and without a good understanding of what is to be considered as critical thinking it is not clear just exactly what the reasons are for its use as an educational aim, or how it is to be achieved. In their analysis of the term Sharon Bai l in , Roland Case, Jerrold Coombs and LeRoi Daniels have argued that it is best not thought of as a skil l , a process, or a procedure, because none of these singly captures the range of achievements or capabilities that is meant by the term. Instead, they suggest that it is best thought of as a normative 20 term applied to judgments that are considered good. Good judgments, they say, generally reflect the use of five intellectual resources: background knowledge, a knowledge of the appropriate criteria, a knowledge of key critical concepts, various heuristics and good habits of mind. Given this conception of critical thinking it is easy to see the value in teaching critical thinking as teaching l s lbid. , 12. "ib.d.,n. 20Sharon Bailin, Roland Case, Jerrold Coombs and LeRoi Daniels, "Common Misconceptions of Critical Thinking," Journal of Curriculum Studies 31 (1999): 269-83, and "Conceptualizing Critical Thinking," Journal of Curriculum Studies 31 (1999): 285-302. Criteria for a Conceptualization of Autonomy 98 students the valuable but complex ability to make judgments for themselves, where what counts as a good judgment varies from field to field, or from one area of life to another, and in which there is no simple set of steps to follow to come up with a right answer, but rather a complex set of considerations and possibilities, including the ability to evaluate the very criteria for what is to count as a good solution. A different sort of educational aim is the promotion of justice in society. As Jerrold Coombs has pointed out, because public educational programmes have such an impact on one's later achievements, and in particular on one's ability to access economic and social goods, they are an important means by which society demonstrates its desire to act justly. 2 1 The promotion of justice is reflected in the demand to give students as equal an opportunity in life as possible through the learning of such necessary skills as literacy, or through exposure to worthwhile experiences they would otherwise not get. It is also achieved through educating people about important justice issues such as racism or poverty. This gives us a summary of our most prominent educational aims. A conceptualization of autonomy wil l be useful to the extent that it has a very clear relationship to at least some of these. For instance, a conceptualization of autonomy may also give us additional reasons why certain bodies of knowledge —be they traditional academic subjects, Martin's reproductive processes 'construed broadly,' or others still—should be taught, or, conversely, it may cause us to question the necessity of teaching some subjects currently considered a part of general knowledge. I assume that a conceptualization of autonomy wi l l give us additional reasons in support of the teaching of critical thinking, and that it may shed additional light on other educational debates, such as the appropriate way to 2 lJerrold Coombs, "Equal Access to Education: The Ideal and the Issues," Journal of Curriculum Studies 26 (1994): 281. Criteria for a Conceptualization of Autonomy 99 promote social justice through education. In particular, we expect that a good understanding of autonomy wil l help to clarify those aims that have their grounding in liberalism. As well, a worthwhile conceptualization of autonomy wi l l help us achieve clarity in our reasoning about some of those aims of education which are contentious. For instance, an aim of education with much currency in the current economic and political climate is the ability to find and do productive work, and to contribute to the economy defined on some level, be it the local, the national or, in our age, the global. This would seem to stem from a variety of concerns, including a concern for the student's welfare and the assumption that this depends on her or his ability to get a job and support at least her or himself, and the belief that our collective well-being is dependent to a large extent on the national economy. There are a variety of expressions of this aim, one of the most common of which is the demand that prescribed sets of skills are taught in schools in order to make students more employable, another is the demand that school systems respond to the market place and teach people the skills necessary for specific jobs currently considered to be important, or at least in demand. There is no doubt that, in our world, individuals should have either the skills necessary for employment or the skills, habits and dispositions necessary to get the skills necessary for employment (I use the term skills here very broadly), and that, as presently constituted, at least part of the health of our local and national communities depends upon the economy and the ability of people to work within a capitalist system. On the other hand, as Peters has pointed out, "there is something inappropriate about this way of speaking; for we would normally use the word 'train' when we had such a specifiable extrinsic objective 22 in mind." There are two general objections to be made here: one is that there is !R. S. Peters, Ethics and Education (London: George Allen and Unwin, 1966), 27. Criteria for a Conceptualization of Autonomy 100 something intrinsic to education that is missing from such a description of its aims, the other is that it is morally problematic to have such ends in mind for others. A clear notion of autonomy should help us think about such educationally and morally problematic issues. Additionally, a conceptualization of autonomy that is to be useful to educators should help to clarify what is involved in teaching for autonomy, or the means for helping students to attain autonomy, and it should make it clear why such an aim is desirable. SUMMARY OF T H E CRITERIA FOR A CONCEPTUALIZATION OF A U T O N O M Y Let me summarize the criteria we have arrived at. To begin with there are general constraints on the form of a conceptualization of autonomy. Thus, a conceptualization should: • be obviously desirable. Where this is not the case, there should be clear reasons for its acceptance. • be attainable. • not demand an overwhelming amount of self-reflection. • reflect some defensible version of self-rule in accord with publicly acceptable moral principles. A conceptualization of autonomy must be consistent with the demands of liberalism. This means that: • autonomous citizens wil l act fairly and justly, wi l l have an understanding of equality, and wi l l be capable of responsibly exercising their own freedom. This includes: o an understanding of the demands of justice, equality and liberty, and a desire to act within the norms prescribed by these. o the ability to reason abstractly and the willingness to respect Criteria for a Conceptualization of Autonomy 101 others. o the willingness to engage in a discussion of the fair terms of co-operation, and to abide by these terms so long as they are fair and others are also wil l ing to abide by them. o valuing and being will ing to work toward those abilities necessary for equal participation in the state, and the ability to make reasoned and responsible decisions as to the exercise of one's rights. o abilities and traits such as openmindedness, an ability to learn from others, an ability and willingness to be moved by reasons, a willingness to be exposed to other ways of life and the ability to make judgments as to their worth, the ability to give reasons in justifying one's own choices, and the ability to make worthwhile choices with regard to our aims, purposes, goals and values. • the more a conceptualization of autonomy can reasonably explain with regard to other liberal commitments, the better. A conceptualization should also accommodate the reasonable criticisms of liberal theory. Thus, a conceptualization should: • be compatible with a strong sense of community. This includes allowing us to understand people as defining their identity through their membership in a community or communities. It wi l l also be consistent with allowing communities to thrive and flourish. • be compatible with understanding persons as embedded in social and epistemological contexts, including traditions, as well as able to reflect on and make responsible decisions about these contexts, and able to freely make commitments within and to them. • be compatible with the assumption that people generally desire to co-operate with one another and to form deep, caring and meaningful Criteria for a Conceptualization of Autonomy 102 attachments with one another. • be seen to accommodate the requirements for consistency with a sense of community, with understanding persons as embedded in contexts, and with the desire and ability to co-operate with one another and to form deep and meaningful attachments. Finally, a conceptualization that is embedded in the theory that sustains our educational beliefs and practices, and that is useful to that practice, should: • be consistent with our other accepted aims of education, in particular those whose grounds for defence are found in our liberal principles. • help us to resolve, or at least re-examine in a clearer light, problematic issues in education, and in particular those issues that seem to offend our liberal values and grounds for education. • help to clarify what is involved in teaching for autonomy, and • make it clear why such an aim is desirable. This list of the criteria that a conceptualization of autonomy must meet gives us something against which we can appraise existing conceptualizations, and provides us with a guide for attempting to reconceptualize autonomy. In chapter 5 I wi l l attempt to provide a conceptualization that meets these criteria in a satisfactory manner. First, though, I wi l l reconsider the accounts of autonomy I introduced in chapter 1, to see how they fare against these criteria, and I wi l l introduce a couple of other versions of autonomy that feature in current discussions of education, to see how they do. C h a p t e r 4: A n A n a l y s i s o f C u r r e n t C o n c e p t u a l i z a t i o n s o f A u t o n o m y In this chapter I wi l l examine several current conceptualizations of autonomy. To start with I wi l l take a closer look at the conceptualizations I introduced in chapter 1; in particular, I want to see. how they fare when evaluated against the criteria for a conceptualization of autonomy that I outlined in the last chapter. At the same time I wi l l discuss other conceptualizations of autonomy that have currency among liberal educational theorists. DEARDEN REVISITED As I discussed in chapter 1, R. F. Dearden describes autonomy as independent-mindedness, which, he says, exists to the extent that an individual's choices and actions "cannot be explained without reference to his own activity of mind." B y 'activity of mind' Dearden means that one could not explain one's choices or actions without referring to one's own decisions, reflections, thoughts, judgments, and so on. We might think of this in this way: If a person explains a choice he made by giving iterations of reasons—I chose this for these reasons, and I think these reasons are good ones for these further reasons, and so on—then the more obviously a particular reason is a person's own, that is, given in his own terms (or 'with reference to his own activity of mind'), or the further down the chain of reasons that displays the fact that the reasons are his own, then the more 103 Analysis of Current Conceptualizations of Autonomy 104 autonomous the person, or the particular choice he made, is. Remember that autonomy is a matter of degree for Dearden. From this a few things are clear: autonomy is a normative claim for Dearden—it is a desirable description of one's choices, actions and so on. But, at the same time, neither the choices one makes nor the reasons one gives must be good ones by Dearden's account. What is important for autonomy is that a strong chain of personal reasoning exists for each of a person's important choices, but not that the reasons themselves must be good. In conceptualizing autonomy as existing to the extent that one can provide one's own reasons for one's actions, he does allow that somewhere along the line we have learned such things as how to reason, the criteria for good reasons, as well as much of the knowledge upon which we base our decisions, from others. Insofar as we have to refer to what we have learned from others in the past in order to exercise independent decision-making, reasoning and planning, autonomous persons cannot be imagined to exist without a shared history with others. But Dearden's conceptualization requires little reference to a social context. We can see this in the following way. Suppose we imagined a person stranded alone on an otherwise deserted island—could we call such a person autonomous? According to Dearden, inasmuch as she is still able to enunciate her own reasons for why she did as she did: how she organized her day, perhaps, or why she located her shelter here rather than there, or why,she undertook the particular attempts at getting herself off the island that she did—so long as her reasoning and decisions are describable in her own terms, with regard to her particular beliefs and desires and plans, by Dearden's definition we could consider such a person autonomous to a high degree. This is significant, because it means that we are not adding much to a description of an isolated individual as capable of reasoning for herself at a fairly high level, or exhibiting a particular kind of rationality, by labelling her Analysis of Current Conceptualizations of Autonomy 105 autonomous. Indeed, we have to wonder just how much distinction there is to draw between the autonomous and the non-autonomous individual. If what defines the autonomous person is that her thoughts and actions require for their explanation reference to the person's own choices, deliberations and so on, as Dearden would have it, then we would expect that any person cut off from others, and especially cut off from others for any length of time, would be autonomous to some degree. But what does this tell us about the person, other than that she appears to be someone who has been alone for a while? Consider some of the desirable characteristics such a person may be lacking: she may not be cognizant of the extent to which her own thinking is distinctive from the reasoning she has learned from others; she may have no desire to have her thoughts and actions a product of her own queries, ideas and so on (beyond her probable unhappiness at having this forced on her by dint of isolation); and her thinking and decision-making skills may be of quite a low quality—as measured by her success at survival in her situation. That is, her ability and desire to think critically and independently—or her ability and desire to responsibly exercise her own freedom, one of the criteria for a conceptualization of autonomy—may be lacking. On the other hand, it might also be the case that the person who can enunciate his own reasons for his decisions fares poorly over the long haul in his situation, and the person stuck with a set of unexamined assumptions from her upbringing does much better. So autonomy, on Dearden's account, may have little or no connection with one's actual ability to survive in one's environment. Dearden's account of autonomy does not rule out the autonomous individual's membership in a community. But Charles Taylor might argue that Dearden's version of reference to the person's own activity of mind ignores the ongoing importance of communities and traditions in forming an individual's conception of her or himself and her or his role in society. It is fair to say that ' Analysis of Current Conceptualizations of Autonomy 106 Dearden's account does not depend on imagining the individual as seeing him or herself as a current member of a social or an epistemological context. Thus, what Dearden's conceptualization of autonomy does not do is explain the relationship between autonomy and one's role in a liberal society. I take it that the iteration of one's own reasons for one's beliefs and choices is meant to indicate a version of the individual's liberty. But it is not clear that this is successful. For instance, what would distinguish the thinking of someone who freely chose a life of religious devotion from someone who also led such a life, but of whom we would be less inclined to say he freely chose it—perhaps he was born into it and was never able to imagine any reasonable alternatives, or perhaps he was indoctrinated into his belief system? I think it is possible to imagine either person capable of giving reasons that are clearly in his or her own terms, and to which he or she is committed. But since we could not say of the person born or indoctrinated into his life of devotion that he had freely chosen his commitments, we have to conclude that Dearden's description of autonomy does not necessarily entail any notion of free choice in one's life or beliefs. This is not to suggest that we could not differentiate between an indoctrinated person and someone who is not indoctrinated by discussing with him the reasons for his beliefs. Rather, the fact that we believe we could do this, even though the indoctrinated person may be able to offer several iterations of —indoctrinated—beliefs in his own terms, shows that it is more than simply reference to the person's own thoughts, choices and so on that we are concerned about. It suggests that the judgment of an individual's liberty is a more complex matter than checking his iteration of his own reasons; among other things we would want to know the kind and validity of the reasons he- has to offer. We would want to know, for instance, that the person knows what would count as good reasons against the course of action he is taking, that these are indeed good reasons and that the individual would be likely to act on them if they held sway; Analysis of Current Conceptualizations of Autonomy 107 that the person would be willing to entertain such reasons for other courses of action; that the person has a sense of justice and respect and is acting in accord with these; and that the person has a sense of himself as demanding respect and the appropriate freedom, and the right to choose for himself, and some idea of what this means. That is, we are concerned about a qualitative assessment of his beliefs as well. So Dearden's version of autonomy gives us no indication as to whether the individual has any sense of such things as equality or the demand for respect for others, and therefore is not clearly consistent with the demands of liberalism. C A L L A N REVISITED, AND BRIGHOUSE Recall that in his earlier book Eamonn Callan described autonomy as regulation of the wi l l , which plays out for the most part as regulation of our interests, and requires such things as realism and independence-of-mind. We must grant that almost all our interests are shared, in one form or another, with others. Indeed, some of our most important interests are others and our relationships with them. Even if we imagine an individual pursuing his interests on his own, in a fairly secluded manner, there is much dependence on others: an individual could not really be said to pursue an interest in philosophy of education, say, in any way we understand, without there existing a community of philosophers who together decide what counts as philosophy of education, what the criteria for good work in the discipline are, and so on. And, of course, our devotion of time and energy to any of these pursuits has implications for the others involved and is therefore bound by moral principles. Callan's conceptualization is different from Dearden's in an important way. Dearden's description requires that a person's reasons just be demonstrably his or her own, and I said in chapter 1 that this implies having an important form of direction and control over one's own life. But Callan is clear that his Analysis of Current Conceptualizations of Autonomy 108 conceptualization requires that a person's reasoning display further characteristics, such as realism and independence-of-mind. That is, that we would evaluate autonomous reasons as good in certain ways, notably in comparison to reasoning that did not demonstrate such characteristics. A n d the fact that one exercises autonomy through regulation of the wi l l shows how Callan believes autonomous actors exercise power over themselves.1 But we should ask whether we can imagine an individual alone on a deserted island, and autonomous by Callan's description. Imagine that such a person shows much realism with regard to her situation, and independence-of-mind; that she does not carry any prejudices or superstitions, poorly examined beliefs about being stranded on an island, and so on, with her; and that she devotes her time and energy to practical things that benefit her situation, to trying to discover things that make her happy and help her keep a positive frame of mind about her situation—perhaps she develops an interest in some of the natural world around her. Over time her identity would increasingly become bound up in her situation as resident of this island, and this would be fostered in large part by the interests she develops and nurtures. We could take this image further, but I think this allows us to imagine such a person cut off from social interaction—at least at present—yet autonomous by Callan's definition. While it would seem that Callan's autonomous person cannot have autonomy thrust upon her or him in quite the same way as Dearden's, it does seem as though she or he can almost stumble upon it—through being isolated and then discovering that she or he is forced to develop certain interests or suffer. This person, too, may not have the desire to be autonomous in the way that she is, nor ' i do not mean to invoke a problematic mind-body distinction here, or some kind of mind-mind variant. By this I simply mean that the autonomous individual has some power over some of their own characteristics—for Callan, what they are interested in, but also possibly their emotional reactions to events and experiences, etcetera. Analysis of Current Conceptualizations of Autonomy 109 might she be aware of her rational independence from the social world around her and the knowledge she has gained from it. That is, while her thinking may start to differ from that of the culture she was a product of before becoming stranded on an island, she may not be aware of this change and her new-found independence from her old ways of thinking, and nor might she desire such a state of affairs if confronted with it. Nor does Callan's conceptualization require any knowledge of moral principles. For Callan autonomy is related to such things as moral virtue and freedom. This is because moral virtue requires realism and independence-of-mind, components of autonomy for Callan; this does not mean that autonomy necessarily leads one to moral virtue. In the same way we do not know of Callan's autonomous individual whether she or he wil l be led to act justly, or in a way that respects the equality of others. We can, for instance, imagine a tyrant who displays a marked ability to regulate her wi l l , who successfully exercises control over her motivational structure, and displays much realism and independence-of-mind in making her tyrannical plans. Thus, calling someone autonomous by Callan's description gives no indication of his or her desire and ability to act justly, or demonstrate an understanding of equality. Callan's autonomous individual has certain cognitive capacities, but we know nothing of her or his success as a member of the liberal polity. I said in chapter 3 that a successful conceptualization of autonomy wil l capture some defensible version of the Kantian notion of self-legislation in accord with moral principles. I suggested there that a defensible version of this would be a liberal version of self-rule. In order to understand individuals as ruling themselves in a way that makes sense, is consistent with others doing the same, and that can occur in a liberal society, they must be seen to do this in a way that accords with such principles underlying the liberal democratic ideal as respect, honesty and fairness. Both Dearden's and Callan's descriptions, as descriptions of Analysis of Current Conceptualizations of Autonomy 110 cognitive characteristics and abilities, leave any version of this out. The individual alone on an island is not bound by any such rules since there is no one else present to whom she has to justify her decisions, or be honest to, or ensure her actions do not harm. Therefore we have no way of knowing whether she is able to 'legislate for [herself]' in accordance with moral rules. A display of morality may not be necessary for autonomy to exist, but an important component of autonomy is the ability to operate in the moral sphere. The fact that we can so easily ascribe autonomy to the individual alone on the island, without knowing how she might navigate the moral world, emphasizes that these conceptualizations leave out this important component of autonomy. I have not singled out the conceptualizations of autonomy put forth by Dearden and Callan because they uniquely fail in this regard, rather I believe they merit discussion because they represent the way in which many conceptualizations of autonomy intended to motivate educational policies are limited. For example, Hairy Brighouse has recently described autonomy in a similar manner. Brighouse argues that appropriate educational aims for schools do . not preclude school choice. To show this he describes the important educational aims, and he believes that justice requires that "The first fundamental value that should guide the design of educational policy is the ideal that all children should have a realistic opportunity to become autonomous adults."2 His argument for this is that justice requires that everyone have an equal right to live a good life, or a life which is good for them. Liv ing a good life has two requirements: one, that the way of life chosen actually is good and worthwhile and, two, that "the person living it must endorse it 'from the inside.'" For Brighouse, this requires "epistemically reliable ways of 2Harry Brighouse, School Choice and Social Justice (New York: Oxford University Press, 2000), 65. Analysis of Current Conceptualizations of Autonomy 111 evaluating different ways of life"; for him, autonomy is synonymous with "the basic methods of rational evaluation" required for deciding how to live wel l . 3 These include the ability to evaluate information, investigate truth claims, and judge the relative merit and probable truth or falsehood of various claims. They also include the ways in which people sometimes adapt their beliefs to accommodate various, and sometimes unjust, circumstances. These are all cognitive skills and characteristics; while Brighouse describes autonomy in terms of living with others—such as the ability to counter or re-assess beliefs one has been inculcated with as a child—we have no knowledge of whether the individual is led to respect others, what his or her view of equality is or whether he or she is motivated to act justly. So while this is a worthwhile set of cognitive resources or abilities for the individual to have, and perhaps necessary for freely choosing and committing to a good life, Brighouse's conceptualization is limited in the same way that Dearden's and Callan's are. STRIKE AND CALLAN/RAWLS REVISITED, AND GUTMANN The version of autonomy offered by Kenneth Strike is very different from the previous ones, as his is an explicitly political conceptualization. Strike describes autonomy as the opportunity to make decisions for oneself. Remember that, for Strike, this opportunity entails both the negative liberty to choose one's own beliefs and lifestyle, and the ability to make good choices. Strike's description of autonomy is clearly situated in liberal theory, and the rights he believes it entails are liberal rights. Thus Strike explicitly includes elements that other descriptions have left out but that I claimed are fundamental to our intuitive understanding of the concept, and so has captured much more of what I suggested a useful and liberal conceptualization of autonomy would be like. 3Ibid., 69-70. Analysis of Current Conceptualizations of Autonomy 112 Strike claims that the ability to make good choices—a component of autonomy—entails such intellectual virtues "as a regard for and commitment to truth, honesty and fairness."4 This suggests that by his account autonomy is not just concerned with choices that are identifiably an individual's own, but that would be normatively evaluated as 'good.' But note that Strike is using the term to describe individuals who are capable of making good choices and committed to doing so, rather than individuals who do make such good choices. That is, on Strike's account, we do not have to evaluate someone's actions and accomplishments as actually being good before we call him or her autonomous. His description of autonomy entails the attributes required to make good choices, rather than an assessment of the choices themselves as demonstrating certain characteristics. There are, I believe, good reasons to use the term to demarcate either actions that we would consider 'good' in certain ways—such as being just, fair, or promoting one's own ends in concert with the ends of others—or persons who we do not just judge as capable of such actions or decisions but who also demonstrate this ability. Consider the individual who has all of the attributes Strike would ascribe to him or her, as well as the various freedoms and rights he deems necessary for making decisions for oneself, but who constantly makes poor choices—perhaps chooses short-term pleasures that seriously hinder her or his long-term opportunities or health, or to pursue personal gain in a way that threatens the well-being or freedom of others. Such a person would be autonomous by Strike's account. But at least some of her or his choices are not consistent with others making the same choices, or exercising their autonomy in the same way, an important criterion for a conceptualization of autonomy. A n d this does not seem to capture in any way a notion of self-rule in accord with such 4Kenneth Strike, Liberty and Learning (Oxford: Robertson, 1982), 44. Analysis of Current Conceptualizations of Autonomy 113 principles as respect, honesty and fairness. Indeed, it is not entirely clear how we could judge someone as capable and even desiring of making good choices i f she or he constantly made poor ones. What counts as evidence for the ability to make good choices? To be able to say this would seem to require as evidence her or his having made good choices in the past. Certain features of the conceptualization of autonomy used by Callan in his later book also depend for their meaning on imagining the individual existing within society. Important features of what I wi l l call the Callan/Rawls autonomous individual include the ability to engage in rational deliberation with others, and the desire to be fair. This version of autonomy also makes explicit its intimate connection with features of the social sphere. The Callan/Rawls autonomous individual participates in a liberal society which is defined, at least in part, by a shared notion of public justice. He or she is wil l ing to be fair and co-operate so long as others are also wil l ing to abide by the terms of co-operation, and he or she believes that the state must find ways to accommodate a presumably wide range of disagreement over conceptions of the good. If autonomy can only exist within a just, liberal sphere then we have an additional, and important, constraint on it. But even i f we remove these requirements on the social structure this conceptualization still requires that individuals are wil l ing and able to pursue their relationships with others in certain ways; that is, other important elements of the Callan/Rawls conceptualization do not make sense in the context of the solitary individual. Thus, in a different way the Callan/Rawls conceptualization of autonomy also circumscribes elements fundamental to our intuitive understanding of the concept, although its dependence on a liberal society that is just in a particular way is an additional constraint on its use that must be considered separately. Describing autonomy in the manner that Strike has, or as reflected in parts of the Callan/Rawls conceptualization, is to make clear that a very important part of Analysis of Current Conceptualizations of Autonomy 114 autonomy is about how we relate to our community, and not just about some kind of freedom—perhaps imposed—from it. But the Callan/Rawls conceptualization ultimately raises the same concern that Strike's does. The Callan/Rawls conceptualization has the additional criteria of showing a willingness for such things as fairness, engaging in public discourse, and abiding by the terms of fair co-operation so long as others are similarly bound. Requiring such a willingness suggests that the autonomous individual is likely to act fairly and so on, but this still leaves open the possibility of making mistakes. Do we say of someone who has the abilities and dispositions that the Callan/Rawls description of autonomy requires, but who does something that is clearly outside the bounds of liberal, rational agreement through a mistake in judgment on her or his part, that she or he was acting autonomously at this time? We may be able to say that in such a situation a person was acting independently, and even that she or he was will ing to be bound by the terms of fair co-operation, but we cannot say that in this case she or he was acting in a manner consistent with others exercising their autonomy in this way, or acting in a way we would judge as fair or just, and so on. This suggests that it might be clearer to concern ourselves with judging certain actions or decisions as autonomous, or reflecting the characteristics of autonomous behaviour, than judging individuals as having autonomy. Switching our focus in this way makes the success of our decisions and actions in reflecting autonomous standards an important component of a judgment of autonomy. For instance, an action intended to be to fair, and to abide by the terms of fair co-operation, but which failed to be and do so, would not be judged autonomous on this account. I believe that switching our focus in this way makes it easier to see the relevant dispositions and abilities, too. Consider the individual who desired his whole life to act in a manner that was fair and just toward others, but who consistently failed to do so. Such an individual might have the Analysis of Current Conceptualizations of Autonomy 115 appropriate desire to act autonomously, but clearly lacks some necessary ability or abilities to follow through on this desire. Thus, switching our focus to a judgment of actions or decisions as meeting the standards of autonomy should not only make our judgment of autonomy easier but ultimately prove more profitable in clarifying the relevant dispositions and abilities: the relevant dispositions and abilities are those that result in autonomous actions and decisions. 5 Nor are Strike's and the Callan/Rawls conceptualizations of autonomy unique; they reflect how others have written about autonomy. For instance, A m y Gutmann, in her Democratic Education, discusses moral autonomy in a manner that she too models on Rawls and the work of Lawrence Kohlberg. She takes this to mean having "the desire and capacity to make moral choices based on principles that are generalizable among all persons."6 While this is an obviously desirable state of affairs, and captures much of what we want to mean by autonomy, it still leaves us with the problem of whether a person who has these characteristics is acting autonomously when he or she fails to act in ways consistent with them. This is not to say that this is not a desirable set of characteristics, nor one that we would not want to aim for in our students (although Gutmann discusses it here as too high a goal and one that education has proven incapable of achieving). The complexity and specificity of the knowledge needed to choose responsibly as required by conceptualizations like Strike's or Callan/Rawls' also makes clear the necessity, for their realization, of an education, and this particular '^ Such a switch in our focus should also make it easier to plan and evaluate the necessary educational tasks. If we desire to educate toward the development of autonomous citizens, and so to' teach a variety of relevant dispositions and abilities, but find that our students are unable to successfully carry out tasks that we think demand autonomy, we will know we are failing our students in this regard. 6 Amy Gutmann, Democratic Education (Princeton, New Jersey: Princeton University Press, 1987), 59. Analysis of Current Conceptualizations of Autonomy 116 characteristic of their definitions is, I believe, a further attribute in their favour. The condition of autonomy implies some kind of critical distance from the assumptions and possible constraints of the knowledge of one's upbringing and surrounding social sphere. The possibility that we could gain this by simple separation from others, or by happening upon it, would trivialize this critical distance by suggesting that it is not a very demanding or particularly well-defined kind of distance that is required. But this would not give us much cause to value autonomy. It is more likely that what we value is a particular sort of critical distance, or at least a perspective that includes certain kinds of knowledge and dispositions. Indeed, I think this is precisely how we view autonomy, and what makes it valuable. Among the attributes of this critical distance that I believe we value are the knowledge of our rational independence from our social world and the disposition to value this autonomy. I have examined several conceptualizations of autonomy, and discussed the ways that I believe they fail to meet the criteria that I set out in chapter 3 . In particular, I have said that a conceptualization of autonomy as a set of cognitive skills and abilities does not capture the criteria that a useful conceptualization wil l include an ability to act congruently with the demands of justice and equality. As well, conceptualizations of autonomy that discuss it in terms of being able and wil l ing to make good decisions, or the kinds of decisions necessary of citizens in liberal democracies, are unclear on whether an action or decision must itself be seen as good in order to be considered autonomous. A useful conceptualization, wi l l , I have argued, be clear on this point. In the next chapter I wi l l present a conceptualization that, I hope, achieves this. Chapter 5: Re-Conceptualizing Autonomy So far we have seen how autonomy is central to liberal theory, and I have discussed the ways in which I believe attempts to conceptualize autonomy have proven inadequate. In this chapter I wi l l attempt to conceptualize autonomy in a way that meets the criteria for an adequate conceptualization that I laid out in chapter 3. I wi l l start by discussing a conception of the person in which one's existence in social contexts is clearly central to one's own identity; a desirable conceptualization of autonomy wil l take such a conception of the person seriously. T H E LIBERAL CONCEPTION OF THE PERSON I want to start by examining what the criticisms raised of liberalism suggest is a commonly held understanding of the liberal conception of the person. Bruce Ackerman's characterization of liberal education provides us with a useful image for identifying this understanding: The entire educational system wi l l , if you like, resemble a great sphere. Children land upon the sphere at different points, depending upon their primary culture; the task is to help them explore the globe in a way that permits them to glimpse the deeper meanings of the life dramas passing on around them. At the end of the journey, however, the now mature citizen has every right to locate himself at the very point from which he began—just as he may also strike out to discover an unoccupied portion of the sphere. . . . The liberality of an education is to be measured not by 117 Re-Conceptualizing Autonomy 118 outcomes but by the extent that the growing child's question of legitimacy is taken seriously. The ideal liberal education is one that permits the child to move from his initial resistances to an ability to define his own objectives in the light of the universal culture defined by all humankind. 1 I am going to treat the assumptions of human nature that underlie this picture as a caricature of the liberal position; I do not mean by this to criticize Ackerman's model, because he has drawn this metaphor for different ends. Nor do I mean to suggest that any liberal has ever claimed the conception of persons that I draw here, and the implications for how we are to understand the relationship between persons and their realm of possibilities. But the criticisms of liberalism that I recounted in chapter 2 suggest that many of its critics believe that liberals must hold such a picture of the person and her possibilities. I think, therefore, that this caricature is useful, if not for seeing what is wrong with the liberal position, then for understanding what some believe to be wrong with the liberal position. How is the individual characterized in this picture? The individual occupies a point in a range of opportunities. This point represents the things that are important to an individual's way of life: such things as her beliefs, her com-mitments, and the range of choices that are currently available to her because of her experiences and her beliefs. Through the appropriate experience and education the individual wi l l understand the world more deeply, and her choices wi l l gain legitimacy. She is freed to choose her original point on the sphere, only now her choice wil l be made more legitimate. A n d she is 'free' to explore and choose another point on the sphere. The image of the person occupying a point on the sphere and then moving from one point to another suggests that, with the appropriate liberal education, the person can in some sense adopt or take on a new set of beliefs and commitments 'Bruce Ackerman, Social Justice in the Liberal State (New Haven: Yale University Press, 1980), 159-60. Re-Conceptualizing Autonomy 119 and, similarly, that we can leave behind or discard our old set. Ackerman does not say whether we are to understand the 'great sphere' as universal, but i f it is like the 'universal culture defined by all humankind,' it may be that he intends this to be understood as a single, universal culture equally available to all . It is pretty easy to imagine the relevant criticisms of such a picture. The picture suggests that the liberal view of the person is of one who can adopt and discard their primary commitments, based merely on evaluating the relevant evidence and options. And it may suggest that we all have access to the same sphere, and that all options are equally available to everyone. What would an alternative picture of the person look like? It seems to me that it would be better to imagine the person's primary values, beliefs and commitments as occupying an area rather than a point on the sphere. Early in a person's life this area would be primarily inherited from their family, culture, ethnicity, and social location. Imagining a person's primary values, commitments, beliefs and so on as occupying an area allows us to see that two persons, perhaps of very similar background and experiences, would occupy overlapping beliefs and commitments, rather than completely distinct points. And when an individual changes his commitments he would be free to change some of his commitments without changing all ; thus, we can imagine the area he occupies on the sphere changing shape to cover a slightly new area, but perhaps substantially the same set of points. In such an image we can imagine that all the members of a particular community would occupy overlapping areas on the sphere, representing their shared beliefs and commitments. In fact, we would expect that their (our) shared beliefs and commitments would be tightly entangled or interwoven with each other, so that we could see that these are not just 2I include in this such social and cultural characteristics as gender, sexuality, income level and membership in a visible minority, as well as the way such characteristics are described within the dominant culture. Re-Conceptualizing Autonomy 120 overlapping in the sense that they are shared, but also because they arise through group membership, are in many ways dependent on the beliefs and experiences they (we) have shared with each other, as well as on their (our) communal understandings and practices, and so on. This is how we might imagine the person and her or his beliefs to look; what about the 'sphere,' or range O f possibilities? To start with, we might consider that the possibilities available to each person are going to vary according to her starting area. One person with one set of commitments, firmly held, is going to have a different set of possibilities open to her, even i f she is wil l ing to take a critical look at her current beliefs and commitments, than another, with different and perhaps less firmly held commitments to start with. Now one could argue with Ackerman that this is what a liberal education is all about—it is about opening up and expanding each person's range of possibilities, so that the largest possible range of worthwhile choices is available to all. In his discussion of Ackerman's image Eamonn Callan suggests a caveat that helps make this possibility more accommodating and attractive. He says we should remember that Children are hardly well-placed even to glimpse the 'deeper meanings of the dramas unfolding around them' until they have some secure grasp of the meaning of their own culture of birth. Nor are they likely to profit from a nomadic curriculum if that wrenches them abruptly away from all that gives meaning to their lives outside the school/ Bearing this in mind does not mean we would be forced to reject the picture of education that Ackerman has drawn, rather it would mean being sensitive in our educational practices. As Callan points out, education can and should be provided with a 'certain gradualism and sensitivity,' so that the student is not pulled sharply 3 Eamonn Callan, Creating Citizens: Political Education and Liberal Democracy (Oxford: Clarendon Press, 1997), 132. Re-Conceptualizing Autonomy 121 away from her or his original value commitments, and the community and life they find this within. But even with this caveat the argument cannot be entirely right: we may agree that a major aim of education should be expanding the student's possibilities, his or her knowledge of the world and possible good lives within it, along with the development of a critical eye toward his or her current beliefs, but this does not force us to adopt an epistemology that claims it is possible to expand every student's view of the world so that he or she can see the whole of the sphere, nor one that implies the sphere might be available in the same way to all. Indeed, it is not only that the range of possibilities is different depending on one's starting point—as if, standing tall at one point on the sphere we can see a different part of the sphere as it curves away from us than the perspective of someone standing at even a slightly different point allows—but that the whole sphere itself differs depending on one's beliefs and one's current commitments. A n d it is not just what we see but what is available to us. How we know the world and how we understand our place in it changes the world—in a sense it changes our picture of the world, but since that is all any one of us has to work with, and since there is no single picture we can claim as accurate and to which all other perceptions should strive to resemble, it is not incorrect to say that it changes the world. A better picture might therefore be of a person occupying an area of beliefs and commitments, an area that overlaps or has elements in common with the areas of many others, in particular those close to her. A n d the area occupied by a person can be seen to sit on a shape—not some universal sphere, but a blob of one shape or another. This blob represents the range of possibilities open to the person. It is unique—the blob of possibilities that the next person occupies, no matter how close the person, is different, though, like the area of beliefs and commitments, the closer the next person is in terms of his current beliefs and commitments the closer wi l l be the shape of the blob his area occupies. Re-Conceptualizing Autonomy 122 Now what happens as the person learns about new possibilities available to her in the world—say she comes to learn about a community with a very different faith than is shared by the members of the person's own community? The person's beliefs about the world wi l l be changed to a greater or lesser degree, depending on how unusual this bit of information is; on whether she simply acknowledges that there exist people with beliefs different from her own; or whether she can go further and imagine that there may be other people still with yet different beliefs again; or whether she can sympathetically identify with the members of the new faith and see how their faith is important to them in the same way that the person's own faith is important to her; or the extent to which she can actually imagine herself adopting the new faith, and finding worthwhile and rewarding commitments within it. This is a range of possible responses and new knowledge available from learning about another possibility in the world. The important point is that by incorporating this new knowledge the person's current beliefs wi l l change to a greater or lesser degree, and because her beliefs have changed her possibilities wi l l change. In our revised picture of the person the shape of the area she occupies wi l l change to a similar degree, and as this changes so wi l l the shape of her blob be changed. I want to draw a fine line here. For I believe that each person's blob, or range of possibilities, is qualitatively different, and that we cannot know exactly what another person's blob is like, but at the same time I do not want to fall into the relativistic trap of either suggesting that we cannot describe our personal beliefs to another or meaningfully discuss how we perceive the range of options available to us; or of suggesting that we cannot work toward some standard of 'goodness' in the pursuit of available options (a standard as applied to the decision-making process, not as applied to the particular choices made by an individual—although it is clear that an appropriate standard for the process wi l l result in better choices), which might include such things as the ability to consider Re-Conceptualizing Autonomy 123 one's own commitments in light of a sympathetic engagement with other possibilities. In this case one might still wish to hold with the liberal educator that if indeed the range of a person's possibilities changes with the knowledge he or she has of the world, then our job as educators is to help everyone attain a certain amount or kind of knowledge to ensure that everyone has similar possibilities available to them. And I think this is almost right: I think we can help individuals to strive for defensible versions of their own beliefs, but doing so must respect the individual differences arising from differences in background, culture, and so on. But this does not leave us with either no, or an impossible, task. Because we can discuss our own situations and how we perceive the world, we can talk together about such things as what the consequences are, both for ourselves and others, of our personal beliefs and commitments. We can talk about the range of available alternatives to our beliefs and commitments, both by examining how others live out similar commitments in different ways, and by considering the vastly different commitments that others in the world hold, and why they hold these and find them rewarding and valuable. By considering the consequences of and alternatives to our beliefs and commitments, we can make our choices better informed. Together we can also examine and question our values, beliefs and commitments, either by looking for ways others have answered concerns we have, or by questioning things we are uncertain about in each other. Thus we can work to make our own commitments more consistent, have more power in explaining our changing understanding of the world, and be more useful in helping us to understand our own role within it and to predict the consequences of our actions. We can also discuss such things as what our responsibilities to others are, in what ways fairness might or should limit our conceptions of what is valuable, and so on, and in this way we can together examine the kinds of external constraints we Re-Conceptualizing Autonomy 124 have to live with i f we are to live together, enjoy personal security, and so on. This is to say that there are ways in which one's values, beliefs and commitments, and the way one lives one's life as a result of these, might be made better. In a sense the standards for better are universal, in that we might all agree that, all else being equal, we desire for our own beliefs to be as consistent as possible, to be true in some sense of reasonably reflecting the world we see around us, and to be useful in interpreting and allowing us to flourish within that world. But in an important sense the standards are individual, in that we each have to apply our own understanding of them to our own lives. For example, I hold certain personal environmental values. These values have changed over time as others have challenged me about them, as I have had to question what evidence I had for certain beliefs about the environment I held, and as I have been forced to consider the values of those who hold other values as more important, and been forced to work to some agreement over more practical matters with such people. This has caused me to rethink my values and how I am committed to them, and how I express that commitment. At a deeper level it has at times led me to question my commitments in this area altogether, as well as question other aspects of myself, like why I often fail to meet what I understand to be the demands of such commitments. Thus I have been forced to rethink the breadth and depths of my beliefs and commitments in one area of my life as I have been brought into situations that caused me to reflect on what empirical evidence I had for some of my beliefs, the extent to which my own preferences impinged on others and their right to express their own values, and so on. So these universal standards for better beliefs: such things as usefulness, fairness, consistency with my other beliefs and values, have been brought to bear, i f fitfully, on my personal commitments. I have outlined here a very simple conception of the person; I do not intend my picture as a complex conception of what it means to be a person, or to Re-Conceptualizing Autonomy 125 capture all of what is important to personal identity. Rather my outline is intended to show that it is possible to conceive of the person in ways entirely compatible with liberal theory, that should also satisfy some of the important demands of the critics of that theory, even in very broad strokes. A CONCEPTUALIZATION OF AUTONOMY The discussion so far suggests fertile ground for a definition of autonomy.4 We can refine this further by considering the kinds of actions or decisions that would not normally be termed autonomous. For instance, someone who reflects on how his or her religious beliefs compare to those of others, but who has only a biased and impoverished view of any other belief system and lacks the tools and desire necessary for a sympathetic and critical look at other beliefs, could not be said to autonomously choose her own religious beliefs. This suggests that particular decisions with an important bearing on one's choice of what life is worth leading, but that do not meet certain standards of adequacy for such decision-making, are not autonomous. Similarly, consider a person who, in making an important decision regarding the course of her life, tried to figure out what would best meet her community's standards. Suppose she pursued this without any reflection on these standards and their desirability, but in doing so accidentally met some of the standards for good decision-making, perhaps by critically reflecting on her current life and her options. We would not normally say of such a person that she had reasoned autonomously about her choices, for she had not intended to do so. Indeed, in this case she had consciously intended to meet standards that we would 4 M y definition of autonomy as presented here, and much of the discussion surrounding it, is modelled on the discussion of critical thinking presented by Sharon Ba i l in , Roland Case, Jerrold Coombs and L e R o i Daniels in their "Conceptualizing Crit ical Thinking," Journal of Curriculum Studies 31 (1999): 285-302. Re-Conceptualizing Autonomy 126 not describe as adequate for such decision-making—compliance with an arbitrary set of external standards. Or consider someone who had his own internal, but poor or mistaken, standards in mind. Suppose that in attempting to meet them he accidentally met the standards for autonomous decision-making; we would not usually judge such a person to have performed autonomously, either. This suggests that someone must have in mind the intent of fill ing standards that are appropriate to autonomy. Finally, as R. F. Dearden has pointed out, we would not normally count as autonomous the decision involved in what route to take to work, or such other choices generally irrelevant to how one chooses and lives a good life. This suggests that to describe an action or decision as autonomous is to say that: • it is a decision or action that has a bearing on an individual's l iving out what is for her or him a worthwhile life, • in making the decision the individual consciously wants to meet the standards relevant to the making of such decisions, and • the decision or action goes a long way toward meeting those standards. If this is correct, then to describe someone as having autonomy is less clear, but by it we would mean that they are generally capable of meeting the relevant standards, and that they have undertaken at least some decisions with regard to the conduct of their own lives and have met these standards in doing so. This conceptualization leaves us with the task of delineating the realm of decisions that would be considered 'autonomous' i f they met the relevant standards. The image of the person I drew helps to clarify Dearden's reference to decisions about important areas of one's life. We can say that autonomy refers to those decisions and actions about how the individual lives out a worthwhile life for him or herself. In the image of the individual as occupying an area of value commitments and choices as to her or his way of life, these decisions and actions Re-Conceptualizing Autonomy 127 are of two kinds: the ones, such as affirming their current choices and commitments, that shape the area of the person's current life, and the ones that carve out the blob of possibilities that the person perceives for her or himself. I do not think that these are exclusive categories; rather I think it may be the case that all decisions wil l ultimately have some impact on how the individual perceives their 'blob,' or range of legitimate possibilities in life. Decisions and actions of both these kinds together wi l l include decisions about one's faith, the basis of one's moral principles and how one lives these within their community, the choice and acceptance of one's community, family and major relationships, the kind of career one pursues, and choices about what activities to put one's energy into, including such things as hobbies, interests and religious devotion. It is true that we might say of many people that they did not consciously choose at least some of these things. But I think we should include not reflecting on the current course of one's life, on one's community and relationships, and on the assumptions underlying what one believes, as a decision about how one is living and going to live the good life, although obviously not one that wi l l meet the appropriate standards for good decisions. This conceptualization also requires that we describe the standards such decisions must meet in order to be considered autonomous. The standards apply to both the decision-making process itself and to the outcomes of that process, or the choices made and actions taken. The appropriate decision-making wi l l , of course, produce outcomes that meet the appropriate standards, but since it is clearer and more intelligible to consider outcomes than the mental processes undertaken to produce those outcomes it is worthwhile to consider and be aware of the appropriate standards for outcomes. This should give us some indication of the process necessary, and the education necessary, to enable the individual to attain such ends. Re-Conceptualizing Autonomy 128 The standards or principles relevant to guiding decision-making include such things as: • Choosing according to principles that would be reasonably acceptable to all. • Consideration of the impact of one's choices on others. Such principles are a direct result of the demands of justice on the autonomous citizen. Justice also demands that persons are rational, and can reason with a knowledge of their own goals and plans, and the kinds of choices that would be efficacious with regard to these. • Judging from a recognition of one's own right to choose for oneself. • Consciously endorsing one's choice for oneself. Such principles reflect the demands of equality on how we are to view citizens in the liberal state. Simply making good choices also requires other things, such as knowledge of the context within which one chooses, and a reasonable knowledge of the options. If a person does not know that there are plausible options she cannot make an informed decision even to maintain the present course of action. Therefore, we have such additional standards as: • Judging from a knowledge of other plausible, worthwhile choices. • Judging in accord with one's own sense of oneself and one's desires and plans, Thus the standards that the outcomes of autonomous decision-making wi l l meet wil l include respecting the principles of justice and rationality, being morally acceptable, being reasonable and realistic, and being coherent with one's own values and conception of the good life. These are only meant to be examples of the standards that wi l l apply to the process of making autonomous decisions, and to the kinds of choices and actions that wi l l count as autonomous. As well, at times the demands of different Re-Conceptualizing Autonomy 129 standards may contradict one another. For instance, judging in accord with one's own life goals may at times contradict the demand to choose according to principles that would be reasonably acceptable to all. In such cases one would have to weigh the demands of different principles; doing so, and the outcomes of such weighings, should themselves reflect the principles of autonomous decision-making. DISCUSSION It is worth repeating that the description of an individual's choice as autonomous is a recognition of, among other things, the fundamental right of the individual to make choices for her or himself with regard to what is important to him or her, the equal status of each of us within society, and the equal claim of each of us to having respect shown by the state and each other for ourselves and our values. This formed the cornerstone of the liberal principles of justice, liberty and equality, and thus this conceptualization of autonomy is clearly predicated on liberal theory. Further, the elements of liberalism provide many of the standards for what is to count as autonomous behaviour, such as being just, or chosen according to principles that would be reasonably acceptable to all, and so the justification for these standards comes from liberalism. A n d some of the requirements for making autonomous choices that I wi l l discuss in chapter 6, such as the self-confidence to think and judge for oneself and the ability to participate in collective decision-making, are among the requirements for equality and liberty enunciated by such people as J. S. M i l l and Elizabeth Anderson. So the term has a clear place within and relationship to liberal theory. At the same time it would be expected that among an individual's prime values wil l be the relationships she shares with others and the traditions and communities she has inherited or adopted. So the individual and her personal choices need not be seen as abstract, Re-Conceptualizing Autonomy 130 disembodied, unrelated nor without a shared sense of value in order to be considered autonomous. One task remaining in fleshing out this conceptualization of autonomy is to describe the types of intellectual resources that a person attempting to come to an autonomous decision wil l require. I wil l leave this task for the next chapter, where I wi l l describe what an education for autonomy wi l l look like, but so far we can see that, among other things, a person wil l need to have a knowledge of the standards for autonomous behaviour, and the desire to meet them. Let me discuss a couple of examples of individuals exhibiting or attempting to exhibit auto-nomous behaviour. These examples are intended to help make clearer what is implied by my conceptualization, and to demonstrate how the standards are brought to bear in guiding autonomous actions or decision-making, but they should also suggest the kinds of resources a person wil l need to act autonomously. As a first example consider the kinds of thinking that a person reflecting on her religious beliefs might undertake. Since we know that many people have changed religions, or gone from considering themselves as holding a religious belief to becoming an atheist, or vice versa, we know/that it is possible to know more than one option with regard to religious beliefs from the inside, or to have a fairly good personal knowledge of more than one point of view, and to endorse one of them as one's own. We would not want to say that only individuals who have changed religions are to be considered autonomous with regard to their religious choices. But it does seem reasonable that, in order for a person to be understood as having autonomously chosen her religious commitments, she should have to know that there exist other possibilities, and she should in some sense understand at least some of these as viable options. Someone thinking about her own faith might consider what sorts of things about her beliefs are important to her, what sorts of things about other faiths are important to their adherents, and what reasons she has for believing what she Re-Conceptualizing Autonomy 131 does. In this way she would be starting to examine the reasons she has for what she believes. Upon examining these kinds of things she might decide that she has no better reasons for believing what she does than others have, or probably have, for believing what they believe. This could lead her into a deeper examination of what other people believe and why, or it might lead her to probe deeper into her own beliefs, and the reasons available for believing the claims she already does. She might also consider how her beliefs complement or conflict with other beliefs she holds. For instance, if she is a Roman Catholic who also believes that women are the equal of men, she may question why women are not allowed to serve as priests in her church. This might lead her away from her religious beliefs if she concludes that this means that there is a better moral position than is reflected by her church, and that her church must not, therefore, represent the wil l of an infallible god, or it may lead her to think that the institution of her religion is fallible, and that she should therefore work toward improving it. On the other hand, given such reflection she might instead decide that she cannot face where her questioning is taking her, and resolve to live with the inconsistency in her beliefs rather than face the personal and perhaps social disruption involved in attempting to sort out her inconsistent beliefs. This might be a practical conclusion, but we may or may not judge it an autonomous one. It may reflect an appropriate judgment of one's own sense of oneself and one's plans, and it may be a true judgment of the hardship such an inquiry would cause, and it may be that in some situations one has to weigh these standards as of more importance than others. On the other hand such a decision could reflect an unwillingness to do the work involved in examining the basis for one's own beliefs, or a lack of recognition of one's own right to choose for oneself. Or the individual may herself see it as an attempt to avoid the consequences of applying the appropriate standards for judgment, rather than an attempt to apply the Re-Conceptualizing Autonomy 132 appropriate standards. In such a case we could not call her choice an autonomous one. What i f her reflection led her to decide that she could not face the disruption involved in fully examining her beliefs and their veracity, but that she would search out her own way of finding meaning within the traditions she had grown up with, with minimal disruption to them? 5 Depending how she went about this we might consider this course of action to be a partially autonomous one. Suppose she attempted to reconcile her own beliefs about gender equality with her religion's official position of gender inequality by coming to the conclusion that the official institutional version of her religion was flawed, and that she was partly responsible for working to improve its embodiment of a better morality. Then she would be demonstrating the principle of not accepting an external moral authority by questioning the one presented to her, and believing that it is only legitimate i f it adheres to certain stronger moral principles. This would be a demonstration of some measure of autonomy. On the other hand, if her attempts to find meaning within the traditions she knows led her to decide simply to live a life others had desired for her to live, and to accept their interpretation of its meaning and the principles by which it should be conducted, we would be hard-pressed to consider her choice autonomous. This would seem to demonstrate a desire simply to avoid conflict or a challenging course of action, for no reason that we can judge to meet the standards for autonomous action. I use religious choice as an example of when we may or may not consider a set of choices as autonomous because it is a problematic area with regard to autonomy; it is not always clear to what extent we can understand a person's religious choice as autonomous. Clarifying how we would apply the standards for 5Pamela Courtenay-Hall suggested this possibility to me. Re-Conceptualizing Autonomy 133 autonomous decision-making to harder cases like religious choice should make the general application of those standards easier. I have attempted to show that, although indeed religious choice may not always qualify as autonomous, there may often be good reasons for us to consider it as such, and it should be possible for us to imagine what an autonomous religious choice would look like. As a second example, consider the account of Socrates as described in the Crito. Here, Socrates has accepted a guilty verdict from the Athenian jury, and is refusing his friends' attempts to convince him to escape from the imposed penalty of death. In the Crito, Socrates listens to one of his friends plead with him, and then inquires into the matter with his friend. 6 He argues that he must act according to whatever is the right thing to do, and that the right thing to do is whatever is required by the principles of justice. Socrates maintains that he must be bound by the laws which he had argued in support of throughout his life, and which, by continuing to live in Athens, he has shown his consent to. In this story Socrates embodies the ideal of being motivated by the best reasons available, and of inquiring into what those reasons might be. He is guided by a notion of justice, and an attempt to come to the best possible understanding of what the demands of justice are on his actions. It is interesting that here we have an example of someone autonomously choosing death as the best way to live out a worthwhile life. Socrates says that he has spent his life arguing for justice and virtue, and that for him to break the laws of the state by attempting to escape would be for him to deny the value of those laws, and a life of virtue. One interesting result of this is the conclusion that it is possible to autonomously choose death, although this should not be interpreted as implying that anyone who dies for what they believe in, or because they believed that for 6Plato, Five Dialogues, trans. G. M . A. Grube (Indianapolis: Hackett Publishing, 1981 [399 BC]), 45-56. Re-Conceptualizing Autonomy . 134 some reason they could no longer live what was for them a good life, is to be assessed as having made an autonomous decision. But this decision may demonstrate some of the standards necessary for autonomy, such as being in accord with one's own conception of the good and of what is valuable to oneself, and of taking responsibility for one's own choices and actions. We would assess such a decision as autonomous to the degree that the decision fulfills all the relevant standards, including the demands of justice. Nor do I mean to imply by the use of these two examples that autonomy is only to be evinced in major life decisions, or during the kinds of crises that might only arise a few times in a lifetime, if at all. Rather I think it is the case that autonomy is reflected in many of our actions and the way we live our life day-to-day. Nor is it the case that we only evince autonomy when we rehearse all of the relevant thinking and decision-making. Rather, as Dearden has suggested, our autonomous actions today may reflect thoughts and decisions taken in the past, even years in the past, "but without losing its force as the explanation" for our autonomy.7 In what way is my conceptualization of autonomy different from other conceptualizations? First of all I am claiming that it is best thought of as a descriptor of actions, choices and judgments; that it denotes that particular actions meet the three criteria I have given. This is different from conceptualizations such as those provided by Dearden and Eamonn Callan in his earlier work, in which the term is used either to denote a characteristic or characteristics of a person, or as a general description of a person's reasoning. A n d it is different from a conceptualization like Kenneth Strike's, in which autonomy denotes an opportunity, or the state of affairs necessary to allow for good choices. 7 R. F. Dearden, "Autonomy and Education," in Education and the Development of Reason, eds. R. F. Dearden, P.H. Hirst and R. S. Peters (London: Routledge and Kegan Paul, 1972), 454. Re-Conceptualizing Autonomy 135 Importantly, my conceptualization makes explicit the normative component of the term: that it involves a judgment as to the adequacy with which such actions and decisions undertaken meet the relevant standards. In this way it also differs from conceptualizations like Strike's, or the Callan/Rawls conceptualization, in which the autonomous person may have the ability and/or the motivation to make good choices, but in which it is not clear that the choices themselves have to be good, or meet relevant standards. Second, my conceptualization is clearly premissed on a conception of the person as embedded in communities and traditions, and interdependent with others. Nothing I have said about autonomy should detract from understanding persons as in large measure constituted by the communities and traditions they are a part of, nor from their ability to participate in those communities, families, traditions and so on. In this way my conceptualization clearly meets the criteria that I drew from communitarian and feminist theory. What my conceptualization does is point out the relevant standards that should be brought to bear in considering who one might want to share one's life with, in reflecting on the communities one is a part of, or on reflecting on how one's life manifests one's ideals about relationships. On the other hand my conceptualization wi l l clearly be antithetical to some forms of community, and to forms of relationships and interactions in otherwise liberal communities. M y conceptualization helps to make it clear just what sorts of communities and forms of interaction are to be considered unfriendly to autonomy—those that discourage or impede people from making decisions that meet the appropriate standards. Thus my conceptualization is antithetical to certain forms of community. But that should not be a problem, as I believe the arguments mounted for autonomy are sufficient arguments against such forms of relationships. Nor should we interpret the fact that autonomy so described is antithetical to certain types of community as meaning that it is Re-Conceptualizing Autonomy 136 antithetical to all communities, or even to most, or to the fact that we all need and live in and desire communities and relationships. Rather, this conceptualization should help us to be clear on just what kinds of relationships, communities and so on actually foster autonomy. A third way in which my conceptualization is different from others is that it makes clear how the tenets of liberalism serve as a foundation for autonomy: liberalism provides the relevant standards against which we judge a person's choices and actions as autonomous or not. In chapter 4 I pointed out that this was lacking in conceptualizations like Dearden's or Callan's earlier one. I argued in chapter 3 that a good conceptualization of autonomy should also reflect what most people mean by the term, as some version of self-rule according to one's own reasoning. This meaning is captured in my conceptualization by the ascription of 'autonomous' to choices that the individual makes that meet the appropriate standards, including the standards for moral behaviour. I think my description of autonomy is clear on this. Thus my conceptualization is explicitly moral, in that one of the standards to be applied in judging a decision or action as autonomous is whether it is moral. That is, moral in the sense that the action does not harm others, and is consistent with such moral principles as honesty, fairness and respect for others. Many people hold for themselves rules or standards of behaviour they consider 'moral' and which go far beyond these principles. M y conceptualization of autonomy does not demand such standards of morality—indeed, it cannot, for it is meant to be a public conception and any standards it adopts must be readily agreed upon by all, as well as universalizable, and not contradict such other principles as liberty and equality. But this conceptualization does make clear how we can understand the individual as 'autonomously' choosing to live by such moral principles as they choose for themselves. At the same time it makes clear that the exercise of an Re-Conceptualizing Autonomy 137 individual's personal morality cannot contradict the public standards for morality I have mentioned—like honesty and respect for others. There is perhaps a criticism that my conceptualization fails to capture what we commonly mean by the term that might still be leveled. This is the claim that, by using the term to denote acts which conform to liberal principles, I have lost the primary meaning of the term 'autonomy' as self-rule. B y this argument there is something about self-rule as choosing for oneself that a focus on what might be understood as external principles leaves out. On the other hand conceptualizations like Dearden's, which attempt to explain how we can understand this idea of 'self-rule,' would be seen to be preferable. But I think this criticism fails. I think the aim of such a criticism is to suggest some motivation or processes internal to the person that has not been made clear. But I think this equates internal with an inability to describe; that is, by this suggestion i f we say a person acts for these reasons we must have failed to describe what reasoning is really taking place internal to the person, because such reasoning is by definition indescribable. I think this is false. The fact that we often cannot describe our thinking around our choices, preferences and actions does not mean that we can never do so, or that such things are at heart indescribable. Rather, it suggests that we happen not to be clear in our thinking in this particular case—in other words, that we have not arrived at clarity in this case yet. The fact that we can often describe and debate our reasoning is evidence of our ability to, at least sometimes, describe our reasoning, it is not simply an incomplete, superficial representation of processes buried within us. I am suggesting that, in order for our reasoning to count as self-rule, our reasons must conform to certain standards, standards for epistemic validity, for power and usefulness to ourselves, and for moral action in the public sphere. That is, there are important criteria that such an explanation has to be consistent with. M y argument, therefore, is that there are such criteria, and that an ability to act Re-Conceptualizing Autonomy 138 autonomously means having an awareness of these criteria and the desire and ability to meet them. Furthermore, inherent in these criteria is the demand that the individual is the only one able to decide for her or himself whether she or he wil l attempt to act in line with the appropriate principles. One implication of my definition that is congruent with all of the conceptualizations I examined in previous chapters is that it can be seen to exist along a continuum—in my definition, actions and decisions can evince more or less autonomy depending on the degree to which they meet the standards necessary for autonomous behaviour. Similarly, an individual wi l l be judged more or less autonomous to the extent that he or she is capable of meeting the relevant standards in a wide range of contexts, and does so. Another way in which this definition is congruent with what others have claimed for autonomy is that it denotes both intrinsic and instrumental value: Autonomy is a normative term that says certain things are considered good in certain ways. These things—autono-mous actions and choices—have intrinsic value because the standards for what is to count as good affirm the importance of our choosing them, and the course of our lives, for ourselves. The term denotes an instrumental value because in meeting the standards for autonomy our actions and choices are made better— more coherent, just, and so on. One might criticize my conceptualization by saying that I have denied the image I drew of unique individuals interdependent with others and deriving their beliefs and values from shared traditions. Such a criticism might proceed by claiming that in imposing the standards necessary for autonomy many current ways of life considered worthwhile by their adherents, and the value commitments they entail, would be seriously hindered. I do not think this is entirely false; I think the claim that some values that people hold are not fair, just, moral, and so on, is correct, and I think that the desire of some in passing on these values is for others not to be free to decide for themselves whether to adopt these values as Re-Conceptualizing Autonomy 139 their own or not, but rather to have others believe these claims outright. Such values and intentions are illiberal, and, as A m y Gutmann argues, liberalism is not neutral with regard to such values, and nor should it be. 8 It is true, for example, that liberalism implies that everyone has a right to be the final judge of their own value commitments, and this itself condemns some opposing values. But this is appropriate—such values are undesirable precisely because they contradict this central value. But this is not to say that such a conceptualization of autonomy has nothing to say to such traditions and values. I think one result of my concept-ualization is that we are reminded that only the individual can move herself from one set of values to another—in this case from a set with some illiberal claims to one that the individual can autonomously endorse. We can help the individual to consider the standards for better beliefs, but we cannot choose those beliefs for her. A n d in moving from one set of values to another the individual wi l l be perceiving the new values from her old set, testing out possible new beliefs, trying some beliefs against others. The individual wil l be perceiving and understanding the new values on her own account—that is, she wi l l be testing them out and adopting them from her own knowledge and values. Thus, while an individual might adopt some values that we take to be universally preferable—equality between the genders, for instance—she has to adopt them for herself, from her own understanding of the issues. In arriving at one's own value commitments one wil l inevitably choose them or interpret them in a way that reflects values that one started with, or experiences one had with one's earlier values. Part of this process for any individual wi l l involve adapting the beliefs and traditions she has inherited from her culture and community, whether these are originally illiberal or not. In 8See, for example, A m y Gutmann, "Challenges of Multiculturalism in Democratic Education," in Philosophy of Education 1995, ed. Alven Neiman (Normal, Illinois: Philosophy of Education Society, 1996), 96. Re-Conceptualizing Autonomy 140 doing so she wil l change the range of possible conceptions of the good for herself, but it wi l l still be identifiably her range, and still unique. Thus I think that while the claim is correct that some values—notably illiberal ones—stand to lose under a demand for autonomy, I do not think the implication that many worthwhile ways of life wi l l be lost, or that we all wi l l end up ultimately much the same, needs to follow. Rather I think it more likely that there wil l remain, much as there is now, a broad variety of individuals with i notably different values, commitments and lives, and in many different cultures and communities. In this chapter I have laid out a conceptualization of autonomy as a descriptor of actions that affect a person's conception of the good, that are intended to meet certain standards and are successful in this regard, and I have shown how this meets the criteria I claimed necessary for a worthwhile understanding of autonomy. I hope I have also succeeded in showing that it is possible to have an explicitly liberal conceptualization of autonomy that is also consistent with understanding the individual and her sense of identity as deeply embedded in a community and a tradition, and an integral part of close, nurturing relationships. It is now possible to show how this description of autonomy is useful in current discussions of education. This is the topic for the next and final chapter. Chapter 6: Autonomy and Education In order for a conceptualization of autonomy to be useful in guiding our educational practices it must meet several criteria specific to those practices. In chapter 3 I claimed that such a conceptualization should: • be consistent with our other accepted aims of education, particularly those whose justifications are found in our liberal principles; • help us to resolve, or at least re-examine in a clearer light, problematic issues in education, particularly those issues that seem to offend our liberal values and grounds for education; • help to clarify what is involved in teaching for autonomy; and • make clear why autonomy is desirable as an educational aim. In this chapter I wi l l explain how the way of thinking about autonomy that I have proposed meets these criteria. TEACHING FOR AUTONOMY I wi l l start with the penultimate criterion I identified, that a conceptualization should help to clarify what teachers must do in order to foster autonomy in their students. According to the picture I have suggested, to teach autonomy is to help students gain the necessary skills, knowledge, abilities and dispositions to make autonomous choices. Thus, thinking of autonomy in this way—as the ability to 141 Autonomy and Education 142 make certain kinds of decisions, and regularly doing so—means that it is fairly easy to outline what we have to do in order to teach for autonomy. A n important point to be made here is that one's exercise of autonomy is intimately connected with understanding oneself as having the right to make important decisions, judge the veracity of evidence, choose between plausible options, participate in society and be heard, for oneself. It is implausible to think that we could foster or instil this in our students through means that do not themselves show a respect for our students' autonomy; that is, this conceptualization gives us reason to believe that we cannot instil autonomy in others by such means as beating it into them, drilling them, or indoctrinating them. Rather, the nature of autonomy suggests that in order to teach it we have to presume our students are capable of either making or learning how to make good judgments for themselves, that they have the right both to make such judgments and to the education necessary to prepare them to do so, and we have to treat them accordingly. That is, we have to treat them in a way that respects both their personhood, and their ability and right to make autonomous decisions, or their developing ability to do so. One important result of this is that the task of teachers is to attempt to gain their students' assent for what they are teaching. Kenneth Strike has argued that seeing the student as a moral agent necessitates giving them reasons.' I think this is correct; seeing even the youngest student as a moral equal, and as someone whose ability to reason morally is developing, entails treating them as though it is up to them whether or not they wil l finally come to believe what you are teaching them. Obviously it is usually in our students' interest to believe much of what they are being taught, nonetheless the only justifiable way to get our students to 'Kenneth Strike, Liberty and Learning (Oxford: Robertson, 1982), 46. Autonomy and Education 143 believe something is to give them the best reasons we can and in doing so transparently appeal to their own judgment. One reason for this is epistemological; we are teaching students how to assess the validity of reasons and truth-claims, and we do this by giving them reasons and demonstrating how to assess reasons. But another is moral; we are teaching them that they either have the ultimate right to make all important decisions for themselves, or wi l l have this right one day. There is an epistemological truth to this, too, of course—ultimately only they can determine whether they wil l believe something or not. But the history of indoctrination makes it clear that people in authority can abuse that authority in such a way as to make it very hard not to believe certain claims. The ability to make autonomous decisions requires certain intellectual resources. These include the following: • the resources necessary for making good judgments generally, including good habits of mind; • a knowledge of the standards of autonomous decisions and actions; • a desire to meet those standards in one's own actions, and in particular a commitment to such principles as equality, freedom and a respect for rights; and • an awareness of a variety of worthwhile ways of life and of the variety of commitments held by different people, and a deep understanding and appreciation of at least some of these. In addition to these intellectual resources autonomy would seem to require certain psychological conditions, such as the self-confidence required to act on one's own judgments when appropriate. As I showed in chapter 2, Elizabeth Anderson has also suggested such attributes. I do not think it is important to add to her discussion here, except to say that i f my arguments for the importance of autonomy are successful then we have an important reason to be concerned with Autonomy and Education 144 how successfully schools foster such things as self-confidence. In what follows I wi l l discuss each of the intellectual resources I claimed are necessary for autonomy. The Resources Necessary for Making Good Judgments If people are to make good judgments with regard to those choices and actions that have a bearing on living out worthwhile ways of life, planning a course of action and the means to pursue these, then they require those resources necessary to the making of good judgments in general. In their work on critical thinking Sharon Bai l in , Roland Case, Jerrold Coombs and LeRoi Daniels suggest that the making of good judgments requires the following five intellectual resources: a knowledge of the standards of good thinking and good performance, a background knowledge of the context in which one's judgment is taking place, a knowledge of critical concepts and vocabulary, a knowledge of a variety of strategies and heuristics for examining information and making decisions, and good habits of mind. Bai l in et al argue that the nature of each of these resources, with the possible exception of habits of mind, tend to be specific to the field within which the judgment is being made. (For example, the criteria for judging a good performance by a mathematician—that it be consistent with the logical structure of mathematics, that it be clear, that it have important mathematical results or implications—are very different from those used to judge the decisions made by a basketball player at play—that they help to move the ball up the court or get the ball in the basket, that they be made quickly, that they help to evade the defence— or someone attempting to resolve a personal crisis.) This suggests that we wi l l not only have to teach students what is involved in the making of good judgments generally, and an understanding of such concepts as criteria and weighing 2 Sharon Ba i l in , Roland Case, Jerrold Coombs and L e R o i Daniels, "Conceptualizing Crit ical Thinking," Journal of Curriculum Studies 31 (1999): 285-302. Autonomy and Education 145 evidence, we wil l also need to help students gain a facility with making judgments about those things that affect how one chooses to live out one's life specifically. That is, not only do students need to explore the world around them and the options it presents to them, they wil l also have to learn how to make decisions about those options, and about the relevant criteria for those decisions. Some of these tools are reflected in the other resources I have claimed necessary for making judgments about one's life. I have emphasized good habits of mind among the resources necessary for making good judgments. This is because there is reason to believe that such habits are more generalizable than the other resources, and that such things as open-mindedness are generally good to have in all situations. Practising such things as listening to others, trying to figure out what reasons others have for what they are trying to get you to believe, and being will ing to be moved by reasons, may be transferable, at least as habits, from one context to the next. Such worthwhile habits of mind are just as important for making good choices about one's values, the course of one's life, planning a course of action and the means to pursue these, and so on. Among the many possible worthwhile habits of mind would be open-mindedness, a willingness to withhold making a decision until enough relevant information is known, and independent-mindedness. Bai l in et al have had success teaching this conceptualization of critical thinking to teachers and student-teachers, and in improving students' mastery of these intellectual resources.3 This suggests that one important aim with regard to the development of autonomy that we can expect public schools to be able to achieve is an understanding of what is necessary for the making of good 3 Ibid. , 298-99. See also the series of teacher resources on critical challenges in the classroom, edited by Roland Case and Leroi Daniels and published by the Fie ld Relations and Teacher In-Service Education Unit of the Faculty of Education at Simon Fraser University, and written by public school teachers who have learned this conceptualization of critical thinking from Case and Daniels. Autonomy and Education 146 judgments. While not all students at all age and developmental levels wi l l be able to master this in relation to the more complex and involved kinds of choices related to living out a worthwhile life, there is no reason to believe that even young students cannot start to develop some of the necessary skills. Knowledge of, and Desire to Meet, the Relevant Standards for Autonomous Actions If making an autonomous decision requires that one has in mind the relevant standards for autonomous decisions and desires to meet them, then an important component of an education for autonomy wil l be teaching these standards, and their value. In the last chapter I suggested the following as examples of the relevant standards: • Choosing according to principles that would be reasonably acceptable to all. • Consideration of the impact of one's choices on others. • Judging from the recognition of one's own right to choose for oneself. • Consciously endorsing one's choice for oneself. • Judging from the knowledge of other plausible, worthwhile choices. • Judging in accord with one's own sense of oneself and one's desires and plans. These standards come from the principles that underlie a liberal society, from the demands of living in communities and traditions, of being reasonable, realistic, and coherent with one's own values and conception of the good life. The need to endorse one's choices for oneself comes from the demand to recognize one's own rights to make decisions for oneself, thus students need to learn both that they have this right, and how to make good decisions. The standards also reflect the respect and trust necessary for co-operation with one another, for helping one another, and for forming the kind of deep, meaningful attachments that we desire, and require in order to survive and reproduce. Autonomy and Education 147 There are a few comments that can be made about the teaching of such standards. The first is that an important part of being able to make good judgments with regard to a particular standard is having a very good conceptual understanding of the standard. This sort of knowledge includes such things as recognizing cases to which a standard applies, and knowing when a particular choice or action successfully meets the standard. In the above section I talked about learning to make decisions, and that this involves such things as practice at making and evaluating such decisions, which in turn requires situations that call for important decisions to be made. I want to emphasize that it also very much includes learning about the related concepts. These two wi l l , of course, often be intertwined, but I want to emphasize that one cannot successfully apply particular standards to one's decision-making until one has some knowledge of the standard itself. For example, consider the child who grows up assuming she wil l follow in her father's footsteps and become a farmer because that is what she knows, and that is what her family has always done. Learning about choices, and that good choices are usually made from a knowledge of other plausible, worthwhile choices, are necessary components of recognizing that becoming a farmer is a choice, and a very important one to the course of her life. (They are also necessary components of recognizing that such choices are not irreversible.) Practice at finding relevant information about options may lead her to investigate other possible choices that she may want to pursue for a career. But the important point is that simply learning what choices are and how to recognize them is itself a very important part of being free and able to act on them in one's life. Second, coming to an understanding of these standards involves learning some very complicated concepts, like the demands of justice. But this does not mean that even the youngest students cannot learn about elements of these Autonomy and Education 148 concepts in meaningful ways. For example, consider what Rawls ' discussion of justice requires of the autonomous citizen. Rawls argues that justice requires that the individual has a sense of how she would like to lead her life, and is able to reason toward achieving this. This requires that individuals are capable of imagining and choosing what it is they want and value, and the ability to assess the worth of these as goals and to reason toward achieving them. A n education intended to foster autonomy, therefore, wi l l have to assist people in exploring, discovering, assessing and revising their values. It wi l l also require teaching them how to work to achieve their goals, how to reason in accord with their own aims and values, and how to make strategic decisions toward the realization of these. But an education at even a very elementary level can involve a discussion of the sorts of things that students find important and why these are important to them. It could also involve discussing the sorts of things that are important to other people, and why, and coming to some understanding of the different kinds of things that very different people value, part of understanding the demands of respect. In this way even very young students can work, in a worthwhile manner, toward an understanding of justice and its demands. Exposure to a Range of Possible Good Lives If one is to make informed decisions with regard to living out a worthwhile life, then one should have knowledge of a variety of ways that people have chosen to exercise this. This might, for example, include learning about the different kinds of values that people hold, the different kinds of family, social and political arrangements that exist and the different kinds of religious beliefs people have. In order to have a fair appreciation of these one would have to know the reasons why people value each of their own lived expressions of these different things, and the kinds of values that people who subscribe to different kinds of goods and values believe they get out of them. Autonomy and Education 149 The picture I drew of people starting from an area on the blob suggests that learning about the expanse and depth of one's own culture, and the commitments entailed by it, may be the only place one can start to learn about such things. That is, the range of beliefs and values most readily available for consideration are those that are shared in one's own community. Studying these helps us learn more about our own community and relationships, and also helps us come to an appreciation of the sorts of values and choices that people hold and make. But at the same time the variety of values and choices and beliefs that people consider a part of a worthwhile life will be greatly restricted if we are unable to consider lives lived outside the communities immediately surrounding us. At the same time, we are likely able to come to a much richer understanding of our own values if we can explore ones very different from the ones we are used to, and if we can leam to appreciate how others might see us and the lives and values that seem normal to us. One thing this suggests is that some familiarity with different ways of life as gained through time spent with people with very different values and commitments from one's own seems necessary in order to appreciate both the range of commitments available and the value that very different commitments can hold for people. This may well be a two-way street, as it might be that the deeper one's appreciation of one's own culture, and the more conscious one is of the assumptions behind one's own culture, the better one is placed to appreciate other cultures. My argument is that these resources are necessary in a liberal society, because they are required for autonomous decision-making. But if my argument in chapter 2 is successful, then these attributes are necessary not just for the attainment of autonomy in our actions, but also for the flourishing of the individual in communities and relationships, and for the flourishing of healthy relationships. These attributes and resources are complex abilities, they include Autonomy and Education 150 such things as abstract and principled reasoning, and an ability and disposition to seek out other points of view. Such things are not innate, therefore they require an education if they are to be attained. Liberal democracies, therefore, are required to provide such an education. APPLICATION TO CONTROVERSIAL ISSUES IN EDUCATION I said that one way a conceptualization of autonomy would prove useful would be if it helped us to examine in a clearer light, or even to resolve, some of the controversies and debatable ideas that currently face us in education. I wi l l discuss here a couple of examples by way of illustrating how I think my description of autonomy is useful in this way. One such debatable idea is the call to teach skills that make students more employable. It is clear that having employment-related skills increases the range of options open to students, including the kinds of goals they can pursue in their lives. Even pursuits that do not themselves require such skills often require that a person be able to support him or herself in order that he or she can pursue other interests. So teaching students such skills may indeed increase, in a practical way, their ability to act on the choices they make for themselves about how to live out a worthwhile life. But it should also be clear that what is important is that teaching these skills should increase students' ability to make such decisions for themselves. A n d these decisions must include the ability to critique choices premissed on a variety of values. For instance, teaching students the ability to type and use a variety of computer programs might both increase their employability and give them a range of skills that they might also find useful in their pursuits and activities outside of employment. But teaching someone these skills does not preclude her or his personal decision never to type again, i f she or he felt so inclined. Autonomy and Education 151 On the other hand, there have been suggestions in recent years that schools should not only teach students a particular range of employment-related skills, but should promote certain values related to the consumer market-place as well. For instance, a couple of years ago a group in the province of Alberta proposed a charter school that would teach the principles of business. The school was planned for students from the age of eight on, and had, according to news reports, the goal of turning "out entrepreneurs capable of bypassing time-consuming university years and vaulting straight from high school into the high-paying global economy." 4 The proposal included requiring students to dress in business attire, and to leam subjects such as reading and writing by reading "resumes, memos and business reports . . . [and learning] mortgage tables, real-estate deals and stock-market movements . . . Children would sit at computer work stations and receive lessons through headsets. It would look like an office rather than a school." 5 It was reported that the provincial Education Minister 'loved the idea.' While elements of such an education have the aim of giving students the same sorts of skills that I suggested above could be useful to the student in the right context, other elements appear to have the aim of decreasing the child's future ability to make choices for herself. (To say nothing of the paltry nature of such an education as exemplified by the proposed reading and writing curriculum.) The child who dropped out of such a program to pursue her dreams of being an artist, or a philosopher, could not be viewed as a success when measured against the aims of this program. While individual teachers in such a school might very well encourage students to pursue their dreams, the program as a whole is set up to foster the pursuit of a very narrow range of goals. 4AIanna Mitchell, "Calgary School for Businesskids Stirs Up a Fuss," The Globe and Mail, 4 May 1998, A6. 5Ibid., A l , A6. Autonomy and Education 152 It is precisely the sorts of intellectual resources required by the conceptualization of autonomy I have suggested that are lacking in the report of this proposal. While no educational program can give students experience with every possible worthwhile set of choices they might make, one could prepare them to make reasonable decisions about their own preferences and choices; there is nothing in the description of this proposal that suggests its proponents are interested in fostering such an ability in its students. A serious controversy in education surrounds the issue of the rights of parents to control the content of their children's education. Many of the arguments about the control over the content of that education crystallize around Mozert, an American court case. It is an important case because it is a clear example of the demands of the liberal state in conflict with the beliefs of some of its citizens, and the pressures this can put on educational institutions. The facts of the case are widely known. In 1983 the Hawkins County, Tennessee Board of Education adopted a new reading series, the Holt, Rinehart and Winston basic reading series, for use in teaching reading in grades 1-8 in the county. Shortly after the start of the school year one parent noticed that some of the material in the series offended her religious beliefs, and started to organize other parents in protest. Ultimately several families came to believe their religious beliefs were being offended, and demanded that the schools make arrangements to teach their children reading without this series. At least one of the schools in the district complied with these requests, and alternative arrangements were made for the students involved. But in November the Board of Education unanimously adopted a resolution requiring teachers to, in effect, use only the Holt reading series. This resulted in several suspensions as students refused to attend classes in which the series was used. In December 1983 several children and their parents brought a suit against the Board alleging that the Board had violated their First Amendment right to free exercise of religion. The Autonomy and Education 153 parents lost the case, but appealed. The Court of Appeal reversed the decision. This in turn was appealed by the Board to a Federal Court, leading to the final decision that the requirement to use the basic reader series did not create an unconstitutional burden on the plaintiffs. 6 The parents had complained that various items contained in the reading series had images or ideas that offended their beliefs. Among these was a poem that presented the idea that by using imagination a child can become part of anything and thus understand it better; one of the parents said that it was wrong for children to use their imagination beyond the "limitation of scriptural authority." The story of a child's visit to Mars was also an inappropriate use of the imagination. Several passages contained reference to evolution, which was interpreted as a teaching that there is no God. Stories in which women were recognized for their achievements outside the home, or where boys enjoyed roles traditionally seen as women's, were seen as inappropriate. And the stories of children who belonged to other religions or who had values that contradicted the parents' were also objectionable, in part because they did not come with a statement claiming these views to be incorrect and the parents' views the correct ones.7 Clearly this case is not just about teaching students to read, for all of the parties involved wanted to ensure the children learned how to read, and the parents were demanding alternate reading programs, not exclusion from learning how to read altogether. Rather this is about a clash between an education in civic virtues, and the ability of parents to pass on views inimical to those virtues. Stephen Macedo claims that it is the liberal value of tolerance that is at stake here, 6Mozert v. Hawkins County Public Schools 647 F.Supp. 1194 (F.D. Tenn. 1986) and Mozert v. Hawkins County Bd. of Educ. 827 F.2d 1058 (6 t h Cir. 1987). 1 Mozert 827 F.2d at 1062. ' Autonomy and Education 154 and argues that the families cannot be accommodated because the promotion of a Q basic public purpose like tolerance trumps religious complaints. But to look at it in terms of tolerance is too narrow. The children do need to leant tolerance, but they also need to learn to be able to make important choices for themselves, and this wil l one day include making decisions about any faith they might subscribe to. As a standard for good choices in a liberal state, and an important virtue for living in a state in which all are free to make such choices for themselves, tolerance is an important part of such an ability. But it is not all of it. Making choices for themselves also, for instance, involves their own awareness of their right to make such decisions. It is instructive to examine this case in terms of the children's right to an education that increases their ability to make such choices for themselves. It is precisely this right that is at risk in this case; the parents' position amounted to attempting to deny their children this right, and with it their prospects of being able to make such choices well in the future.9 Notice that, notwithstanding at least some of the children's participation as plaintiffs in the original suit (and the unclear wording in the final decision as to just who the requirement to use the reading series was deemed not to create a burden on), most of the judges' reasoning addressed the parents' right to withdraw their children from this program versus the demand on the state to provide an education. Similarly, most of those commenting on this case have also discussed whether the parents had any right to withdraw their children from the reading 8Stephen Macedo, "Liberal C i v i c Education and Religious Fundamentalism: The Case of G o d v. John Rawls ," Ethics 105 (Apr i l 1995), 470-91. 9 A developmental argument could be made here, that the children, particularly the younger ones, are not yet able to adequately consider views that challenge the beliefs promoted by their parents. This would be an empirical argument, and would be an important one i f there were evidence to support it. But it does not seem to be the parents' concern. Autonomy and Education 155 program; no one assumes the children had the right to withdraw themselves. 1 0 The fact that everybody addresses the parents' rights suggests that everyone assumes the children are not capable of making such decisions for themselves—that is, the children's autonomy, or lack of it, is treated as centrally important, even i f it is not openly recognized as such. If the children were regarded as fully capable of making autonomous decisions then there would be no question of their right to withdraw from forced exposure to a program with a message they did not accept. No one, for example, would suggest that the parents should be forced to sit through a program that expounded views they did not believe. The children in Mozert, like all children, require schooling to give them the best chance.of gaining—and the state the best chance of ensuring that they gain—at least a minimal ability at making autonomous decisions. Eamonn Callan suggests that the strongest argument in support of the parents' position would be the argument that the parents' beliefs were discriminated against in the reading series." But notice that this would amount to an argument by the parents that their children would not be able to make a reasonable judgment as to the legitimacy of their parents' beliefs, because they would already have been encouraged to form a biased and unfavourable opinion of those beliefs. That is, if the parents had argued this they would be claiming their children were not being prepared for making the best decisions possible, and thus they would also be arguing on the grounds of their children's right to an education that prepared them to make autonomous decisions. 10See, for example, Eamonn Callan, Creating Citizens: Political Education and Liberal Democracy (Oxford: Clarendon Press, 1997), 157-61; Amy Gutmann, "Undemocratic Education," in Liberalism and the Moral Life, ed. Nancy L . Rosenblum (Cambridge, Mass.: Harvard University Press, 1989), 71-88; and Macedo. "There is some suggestion of this position among the parents' testimony, but it is not the focus of the parents' complaints, and is not adequately replied to in the Court's decision. Callan, 159-60. Autonomy and Education 156 One of the parents suggested this line of argument, but otherwise there is not much evidence to suggest that the parents in general saw their own argument in this light. Rather it is pretty clear that the parents wanted their children to uncritically adopt a particular stance toward the parents' religious beliefs. For instance, Judge Lively observed that "Although it is not clear that the plaintiffs object to all critical reading, Mrs. Frost [one of the parents] did testify that she did not want her children to make critical judgments and exercise choices in areas where the Bible provides the answer." 1 2 The parents in Mozert are, of course, entitled to their beliefs. But, as Macedo has pointed out, this right does not extend to the freedom to circumscribe their children's beliefs. 1 3 Everyone should be happy that these children are being brought up in what are presumably loving and caring families. A n d the children are entitled to a deep understanding of their parents' beliefs and to come to an appreciation of their parents' way of life as having considerable value for their parents, among others. But the students are also entitled to an education that enables them to make decisions for themselves, including decisions with regard to their religious beliefs, and such an education includes such things as exposure to other beliefs and ways of life and a probing examination of the grounds for the type of fundamentalist beliefs held by their parents. Ultimately, the students must be regarded as either having the right to make such decisions for themselves, or the eventual right. I am arguing that, in a liberal democracy, the need to teach the ability to make autonomous decisions is comparable to the need to teach literacy: there may have been a time when a parent's desire that his son or daughter not be taught how to read, because it would interfere with how he or she saw his or her place in l2Mozert&21 F.2d at 1069. 13Macedo, 486. Autonomy and Education 157 the world, would be respected. But we no longer believe that such a request should be met, because doing so unfairly restricts the child's future development in order to honour the parent's wishes. It is exactly the same with regard to the right of the child to an education that develops her ability to make autonomous decisions: to allow parents to deny their children this right is to allow parents to restrict the future development of their children into adults who can make important decisions for themselves. That is, this would allow the wishes of one person (or group of persons) to restrict the current and future rights, and potential for development, of another. As A m y Gutmann has argued with regard to a similar case, given the practical constraints in the Mozert situation it may be best for the school or school district to provide the alternative classes, and exempt the students from the reading program. 1 4 But this would be as a means of ensuring the students remain in schools and of giving them the widest exposure to an autonomy-facilitating education possible in the circumstances. No one should be defending a right of the i parents to deny their children such an education, or to hinder their children's growth into adults capable of making the best decisions possible for themselves. Examining Mozert as an example of the right of people in a liberal society to an education that increases their ability to make autonomous decisions in conflict with beliefs that might be threatened by this ability does not produce results that differ from the analysis of other liberal commentators, or from the court's ultimate decision. But it does give us different reasons for our conclusions, and it makes that reasoning clear. That the results of such reasoning are sometimes different is brought out by examining another well-known American case, Yoder. l 4 A m y Gutmann, "Challenges of Multiculturalism in Democratic Education," in Philosophy of Education 1995, ed. Alven Neiman (Normal, Illinois: Philosophy of Education Society, 1996), 96-97. Autonomy and Education 158 In 1972 the U.S. Supreme Court ruled that a group of Amish living in Wisconsin had the right to withdraw their children from compulsory school attendance, after grade 8 but before age sixteen, the age required by Wisconsin's compulsory school-attendance law. 1 5 The Amish supported an elementary educa-tion for their children because they felt it was necessary for reading the Bible and being good farmers, and for dealing adequately with the non-Amish people they might come in contact with. But beyond grade 8 they felt education taught values that were in opposition to their beliefs, values such as competitiveness, worldly success, and the value of intellectual and scientific accomplishments. These are distinctly in opposition to the Amish values of a simple life of hard work, devotion to the soil and the community's welfare, and separation from contemporary worldly society. The Amish believe that the adolescent period is a crucial, formative one during which a person "must acquire Amish attitudes favoring manual work and self-reliance and the specific skills needed to perform the adult role of an Amish farmer or housewife. They must learn to enjoy physical labor." 1 6 The Court agreed in this case that the right of the Amish to practice their religion was more important than two more years of compulsory schooling. Yoder is also an important case for liberal educational theorists, but ultimately its resolution has not proven to be as contentious as the debate over Mozert. Perhaps this is because, as Macedo observes, "being Amish is not a growth industry . . . Protestant fundamentalists are far more numerous and power-17 ful and are often highly politicized and hostile to at least some liberal values." If Macedo is correct in this then it is a black mark on liberals that their motivation is a concern for the existing liberal polity, and not for the various individuals both 1 5Wisconsin v. Yoder 406 U.S. 205 (1971). '"ibid, at 211. l 7Macedo, 472. Macedo for one says that although "this remains a difficult matter... [he] would lean toward deciding" in favour of the Amish. Ibid., 489. Autonomy and Education 159 within and ostensibly without that polity that might be affected by its decisions. The state has a responsibility to the children of the Amish that is not revoked simply because the children's ancestors wanted to separate themselves from its reach. It is clear that one concern in this case is that the Amish way of life is threatened by a public education; it is widely recognized that, given an education that takes the children beyond the teachings of the immediate community, there is a good chance that many of the children may choose to leave the Amish faith and way of life. So this is a case in which not just a few parents' views but a way of life that has existed for several centuries, that threatens no one else, and that puts few demands on the surrounding state, may be at risk. The State argued in Yoder that exempting the Amish children from compulsory schooling "fails to recognize the substantive right of the Amish child to a secondary education, and fails to give due regard to the power of the State as parens patriae to extend the benefit of secondary education to children regardless of the wishes of their parents."1 8 The Court responded in terms of the danger to the children's mental or physical health that might occur by missing two years of schooling, or the danger of their future "inability to be self-supporting or to discharge the duties and responsibilities of citizenship, or in any other way materially detract from the welfare of society." 1 9 But this response fails to recognize that the benefit of an education should be measured, at least in part, by its benefit to the student as a person, and not just by its benefit to society through not detracting from the society's welfare. The parents in Yoder may be right that the schools promote values that are at odds with the Amish's legitimate values, and in a way that biases students l*Yoderat 229. "ibid, at 234. Autonomy and Education 160 against these values. If so, this would be a serious criticism of the public educational system. The parents, however, cannot be allowed to respond by withdrawing their children from school and promoting their own values at the expense of other legitimate values. It is not wrong to teach children the value of hard work, or simplicity, just as it is not wrong to teach them the value of intellectual study. But what is at stake here is the children's ability to choose values for themselves, and teaching the children that one set of values is the only possible and desirable one is to unfairly attempt to bias their future selection. This is the important line of argument that the State's claim about the substantive right of the children to an education at least suggests. Admittedly there is a value, or set of values, that is being promoted here at the expense of others. This is the liberal principles, which includes the right of people to choose their own good for themselves, along with such other values as equality and justice. Understandably, this set of values clashes with the Amish 's belief that their very salvation may be at stake if their children are not brought up to adopt Amish values and the Amish way of life. This is unfortunate for the Amish. But simply because the Amish adults believe something does not mean they can illiberally impose their wil l over other persons, even when those persons are their own children. The decision in Yoder should be given in terms of the children's ability to make such important decisions for themselves. It may be that after grade 8 the children are capable of making such decisions, and making them well. But this is not proven by the plaintiffs, and does not seem likely. If not, the children have a right to an education that cannot be denied by their parents, or even by themselves before they are capable of adequately making such decisions. It is interesting that in this case the one judge who dissented in part noted that the Court had weighed the interests of the parents against the interest of the State, but had not considered the interests of the children involved. But he agreed Autonomy and Education _ 161 with the Court about the rights of the one student who had testified that her own views were opposed to high school education; he dissented with regard to the parents whose children had not testified, arguing that the children's views may indeed be opposed to those presented by their parents. He also noted that what was important was whether the child "is mature enough to have that desire respected." 2 0 Thus, the decision in Yoder should be made in terms of whether the children have received, by the end of grade 8, an education thought to have sufficiently prepared them to make such decisions for themselves, which would have to include the kinds of resources I discussed earlier in this chapter. I have not discussed in this thesis what we might judge to be a minimum threshold that we could agree to be necessary, and that we would aim to achieve within the limits of compulsory schooling, but my own suspicion is that this is likely not achieved in what currently constitutes an elementary education. The fact that the parents in Yoder are happy with the education that their children are receiving before the end of grade 8 suggests that, in this case, their education is failing to prepare them in a substantive way to make autonomous decisions. If this is the case it seems to me a major failing, as this should be an important part of, and an important legitimation for, any compulsory schooling. If we believe that people have the right to make good decisions for themselves, this must extend to the Amish children: we must view them as either having this right, or having it in the future and therefore currently having the need to develop the ability to exercise this right. 2 1 A decision in favour of the survival of the Amish way of life is one that privileges that way of life over the children's z uIbid. at 242. 2 1 We should say that this extends to the adult members of the Amish community, too, but in their case we say that they are exercising their right to make decisions about how to lead their Autonomy and Education 1 6 2 22 right to learn how to make the best decisions possible for themselves. Now it may be that most of the children, perhaps even more than currently, given a reasonable knowledge of other ways of life and the ability to make good decisions, would opt to remain Amish the rest of their lives. This would be a result we could all rejoice in, liberals because the children were learning how to make decisions for themselves, and exercising this, and Amish who presumably want to see their way of life survive and be passed on. But the Amish obviously fear that this wi l l not be the case, and we have no reason to question their doubts. At the very least some of the Amish's beliefs and values wil l be challenged and, one might expect, have to change over time. That a way of life, which obviously has and has had much value for many people, could end, or at least be severely changed, is a strong possibility, and obviously one that could cause much unhappiness for many people. This should make the state be very careful in its deliberations, but it is not reason enough to deny the children their ability to make autonomous decisions. Macedo might call my resolution of this case an example of an autonomy-centred comprehensive liberalism. He believes that political liberalism offers us more promising grounds for a resolution of such cases than does a comprehensive liberalism that asserts "the political authority of a vision of the good life as a whole informed by autonomy." 2 j The problem with this is that, as Macedo says, i lives by focusing on the life they live, and by reducing their ability to choose otherwise. This may not be the. best decision, but it is not for us to say. 2 2 It may be that giving the children the skills and knowledge to make good decisions for themselves can only be gained at the expense of the break-up of the Amish way of life, and it could be that such a break-up could be traumatizing to the community's members to such an extent as to ruin all the children's future lives. This would be an argument against delivering an autonomy-facilitating education that we would want to consider, since it is one that considers the children's rights versus their own well-being, but it is an empirical argument, and as such would require strong empirical evidence before it should be weighed against the rights o f the children. 2 3 Macedo , "Liberal C iv i c Education and Religious Fundamentalism," 476. Autonomy and Education 163 political or not liberalism "does not claim to be non-partisan." 2 4 Among the liberal values we must be partisan about is the right of everyone to make decisions for themselves and, I have claimed, the right of everyone to an education that enables them to do this as well as possible. Macedo suggests as much himself; when he says that "While the state has no business promoting broad ideals like personal autonomy . . . to allow Amish parents to withdraw their children from high school could thwart the children's ability to make adequately informed decisions about how to live their lives f the ability to make good choices that he is concerned with protecting is remarkably similar to the version of autonomy for which I have argued. 2 5 AUTONOMY'S RELATIONSHIP TO O T H E R A C C E P T E D A I M S OF EDUCATION The first criterion I gave for a conceptualization of autonomy that is to be useful in guiding our educational practices is that it should be consistent with our other accepted aims of education, or else it should make clear the grounds for rejecting or changing those currently accepted aims. In chapter 3 I cited R. S. Peters' argument that education has as its aim the development of an educated person, and thus a conception of the educated person helps to clarify our educational practices. A conceptualization of autonomy as a normative claim about choices suggests that an educated person is one who is able to make good decisions about how to live out what is for her or him a worthwhile life, where good decisions are ones that meet the standards relevant to such decisions to a high degree. I have argued that this is a necessary aim of education in liberal democracies, and I believe this is intuitively attractive as a criterion for an educated person. As such, Z 4Ibid„ 482. 2 5 Ibid„ 488. Emphasis added. Autonomy and Education 164 it gives us an additional reason for the importance of knowledge, as knowledge is an important ingredient in the ability to make decisions for oneself. Further, Peters' conception of an educated person in terms of her relationship to knowledge—as someone who takes delight in pursuits for their own sake and who enjoys knowledge in a particular way—suggests a specific definition of the good for people. But this is a very narrow aim for education in a liberal society which must include a variety of goods and allow individuals to choose and pursue their own conception of the good for themselves. Describing the educated person as, in part, one who is capable of making autonomous decisions for her or himself to a high degree helps us to recognize that what we want to achieve through education is the development of individuals who can make just such choices for themselves, but not what sorts of good lives we want people to choose and value for themselves. Peters' conception also suggests that all sorts of esoteric information might be taught in schools, along with an appreciation for such information as interesting in its own right. But autonomy as a criterion for the educated person provides us with different and important criteria: knowledge wil l be judged important for a public education to the extent that it increases the individual's ability to make important choices for her or himself. This complements the argument that Jane Roland Martin has given in support of what she termed the reproductive processes, or what we might call (also broadly) the private sphere, including such things as caring for others, and the related skills such as keeping house. These activities and their related dispositions are an important part of almost every life; to ignore such a large and important aspect of life is to fail to prepare people adequately to make fully-informed decisions about the world they wil l inhabit, and to ignore an important realm of what gives many people value. Autonomy and Education 165 At the same time, focusing on enhancing the ability of people to make important choices for themselves may cause us to rethink the place of what are currently common parts of the curriculum; some subjects taught currently in our schools are there because of a historical judgment of their value, where much value was based on academic knowledge. It is not clear, for instance, why calculus should be widely valued as part of an advanced high school education for most. Calculus is useful for those people who intend to go on to study other subjects that require it, like engineering and science, and is also of value to those who particularly enjoy studying mathematics. But otherwise encouraging or even requiring students to study it and similarly specialized subjects might be seen as a bit like requiring the study of chess: it certainly gives pleasure to many, but it is hard to justify as necessary for most. I said that the promotion of social justice through education is partly achieved through attempting to give students as equal an opportunity in life as possible through the learning of such necessary skills and abilities as literacy or critical thinking, and through an exposure to a range of worthwhile experiences. Learning to make good decisions about one's life, about the kinds of values one holds, about how to plan to achieve one's goals, and about such things as how to get along with others who hold very different values from oneself, obviously increases one's range of opportunities. As well, learning about the demands of justice, equality and respect on one's actions, and about one's own rights with regard to these increases the ability of the individual to demand justice for herself. M y description of autonomy as a descriptor of good decisions also adds weight to the arguments for teaching critical thinking, if critical thinking is viewed as the making of good judgments. The final criterion I stated for a conceptualization of autonomy that is to be useful in guiding our educational practices is that it should help to make clear why Autonomy and Education 166 autonomy is desirable as an educational aim. In this thesis I have argued that liberalism demands the ability to make good choices for oneself, and in chapter 5 I presented a conceptualization of autonomy as the making of such choices. If I have been successful in my argument then the reader should by now have good reasons for valuing autonomy as an educational aim. In this final chapter I have shown how such a description of autonomy meets the other criteria I gave for a conceptualization of autonomy that is to be useful to educators. I have argued that the conceptualization I drew helps us to clarify what is involved in teaching for autonomy, and helps us to resolve or at least clarify some currently problematic issues for us in our educational practices. Finally, I have argued that this conceptualization helps to clarify and buttress some of our current educational practices and aims, and so meets the final criteria that I gave for a such a conceptualization. This gives us further reasons in its favour and, I hope, suggests lines along which it might be found useful. B i b l i o g r a p h y Ackerman, Bruce. Social Justice in the Liberal State. New Haven: Yale University Press, 1980. Anderson, Elizabeth. "What Is the Point of Equality?" Ethics 109 (January 1999): 287-337. Bai l in , Sharon, Roland Case, Jerrold Coombs and LeRoi Daniels. "Common Misconceptions of Critical Thinking." Journal of Curriculum Studies 31 (1999): 269-83. . "Conceptualizing Critical Thinking." Journal of Curriculum Studies 31 (1999): 285-302. Benhabib, Seyla. Situating the Self: Gender, Community and Postmodernism in Contemporary Ethics. Cambridge: Polity Press, 1992. . "The Generalized and the Concrete Other: The Kohlberg-Gill igan Controversy and Moral Theory." In Benhabib (1992), 148-77. Benn, Stanley I., and Gerald Gauss, eds. Public and Private in Social Life. New York: St. Martin's Press, 1983. Berlin, Isaiah. Four Essays on Liberty. Oxford: Oxford University Press, 1969. Blum, Lawrence. "Gill igan and Kohlberg: Implications for Moral Theory." Ethics 98 (April 1988): 472-91. Brighouse, Harry. School Choice and Social Justice. New York: Oxford University Press, 2000. Callan, Eamonn. Autonomy and Schooling. Montreal: McGil l -Queen 's University Press, 1988. . Creating Citizens: Political Education and Liberal Democracy. Oxford: Clarendon Press, 1997. Coombs, Jerrold. "Equal Access to Education: The Ideal and the Issues." Journal of Curriculum Studies 26 (1994): 281-95. 167 Bibliography 168 Dearden, R. F. "Autonomy and Education." In Dearden, Hirst and Peters (1972), 448-65. Dearden, R. F. , Paul H . Hirst and Richard S. Peters, eds. Education and the Development of Reason. London: Routledge and Kegan Paul, 1972. DeNicola, Daniel, ed. Philosophy of Education 1981. Normal, Illinois: Philosophy of Education Society, 1982. Dworkin, Gerald. The Theory and Practice of Autonomy. New York: Cambridge University Press, 1988. Dworkin, Ronald. "Liberalism." In Hampshire (1978), 113-43. . " W i l l Clinton's Plan Be Fair?" New York Review of Books, 13 January 1994, 20-25. . Sovereign Virtue: The Theory and Practice of Equality. Cambridge, Mass.: Harvard University Press, 2000. Fein berg, Joel. Rights, Justice, and the Bounds of Liberty: Essays in Social Philosophy. Princeton, New Jersey: Princeton University Press, 1980. Flicker, Miranda and Jennifer Hornsby, eds. The Cambridge Companion to Feminism in Philosophy. Cambridge: Cambridge University Press, 2000. Friedman, Mari lyn. "Feminism in Ethics:, Conceptions of Autonomy." In Fricker and Hornsby (2000), 205-24. Gil l igan, Carol. In a Different Voice: Psychological Theory and Women's Development. Cambridge, Mass.: Harvard University Press, 1982. Goodin, Robert E . and Philip Pettit, eds. A Companion to Contemporary Political Philosophy. Oxford: Blackwell Publishers, 1995. Gutmann, Amy. Democratic Education. Princeton, New Jersey: Princeton University Press, 1987. . "Undemocratic Education." In Rosenblum (1989), 71-88. . "Challenges of Multiculturalism in Democratic Education." In Neiman (1996), 86-105. Hampshire, Stuart, ed. Public and Private Morality. Cambridge: Cambridge University Press, 1978. Bibliography 169 Hirst, Paul. Knowledge and the Curriculum: A Collection of Philosophical Papers. London: Routledge and Kegan Paul, 1974. _ . "Liberal Education and the Nature of Knowledge." In Hirst (1974), 30-53. Jaggar, Alison. Feminist Politics and Human Nature. Sussex: Rowman and Allanheld, 1983. . "Feminism in Ethics: Moral Justification." In Fricker and Hornsby (2000), 225^14. Kant, Immanuel. Groundwork of the Metaphysics of Morals. Translated by Mary Gregor. Cambridge: Cambridge University Press, 1997 [1785]. Keller, Evelyn Fox. Reflections on Gender and Science. New Haven, Conn.: Yale University Press, 1985. K e i T , Donald. "Devoid of Community: Examining Conceptions of Autonomy in Education." Educational Theory 52 (Winter 2002): 13-25. Kohlberg, Lawrence. "The Cla im to Moral Adequacy of a Highest Stage of Moral Judgment." Journal of Philosophy 70 (25 October 1973): 630-46. Kymlicka , W i l l . Liberalism, Community, and Culture. Oxford: Clarendon Press, 1989. . Contemporary Political Philosophy: An Introduction. Oxford: Clarendon Press, 1990. . "Community." In Goodin and Pettit (1995), 366-78. Macedo, Stephen. "Liberal Civic Education and Religious Fundamentalism: The Case of God v. John Rawls." Ethics 105 (April 1995): 468-96. Maclntyre, Alasdair. After Virtue: A Study in Moral Theory, 2 n d ed. Notre Dame, Indiana: University of Notre Dame Press, 1984. Mansbridge, Jane and Susan Moller Okin. "Feminism." In Goodin and Pettit (1995), 269-90. Martin, Jane Roland. "The Ideal of the Educated Person." In DeNicola (1982), 3-20. Bibliography 170 M i l l , John Stuart. On Liberty. Edited by Gertrude Himmelfarb. London: Penguin Books, 1974(1859). Mitchel l , Alanna. "Calgary School for Businesskids Stirs Up a Fuss." The Globe and Mail, 4 May 1998: A l , A6 . Morse, Jane Fowler. "Fostering Autonomy." Educational Theory 47 (Winter 1997): 31-50. Mozert v. Hawkins County Bd . of Educ. 827 F.2d 1058 (6 t h C i n 1987). Mozert v. Hawkins County Public Schools 647 F.Supp. 1194 (F.D. Tenn. 1986). Neiman, Alven, ed. Philosophy of Education 1995. Normal, Illinois: Philosophy of Education Society, 1996. Noddings, Nel . Caring: A Feminist Approach to Ethics and Moral Education. Berkeley: University of California Press, 1984. . " A n Ethic of Caring and Its Implications for Instructional Arrangements." American Journal of Education 96 (February 1988): 215-30. . Starting at Home: Caring and Social Policy. Berkeley: University of California Press, 2002. Norman, Richard. Free and Equal: A Philosophical Examination of Political Values. Oxford: Oxford University Press, 1987. Nozick, Robert. Anarchy, State, and Utopia. Oxford: Basil Blackwell , 1974. Pateman, Carole. "Feminist Critiques of the Public/Private Dichotomy." In Benn and Gauss (1983), 281-303. Peters, R. S. Ethics and Education. London: George Al len and Unwin, 1966. . "Education and the Educated Man . " In Dearden, Hirst and Peters (1972), 3-18. Plato. Five Dialogues. Translated by G . M . A . Grube. Indianapolis: Hackett Publishing, 1981 [399 B C ] . Okin, Susan Moller. Justice, Gender and the Family. New York: Basic Books, 1989. Bibliog raphy 171 Rawls, John. "Justice as Fairness." Philosophical Review 67 (1958): 164-94. . A Theory of Justice. Cambridge, Mass.: Belknap Press, 1971. . Political Liberalism. New York: Columbia University Press, 1993. Rosenblum, Nancy L . Another Liberalism: Romanticism and the Reconstruction of Liberal Thought. Cambridge, Mass.: Harvard University Press, 1987. , ed. Liberalism and the Moral Life. Cambridge, Mass.: Harvard University Press, 1989. Ryan, Alan, ed. The Idea of Freedom. Oxford: Oxford University Press, 1979. Sandel, Michael. Liberalism and the Limits of Justice. Cambridge: Cambridge University Press, 1982. Strike, Kenneth. Educational Policy and the Just Society. Urbana, Illinois: University of Illinois Press, 1982. . Liberty and Learning. Oxford: Robertson, 1982. Sullivan, Roger J. Immanuel Kant's Moral Theory. Cambridge: Cambridge University Press, 1989. . An Introduction to Kant's Ethics. Cambridge: Cambridge University Press, 1994. Taylor, Charles. Hegel and Modern Society. Cambridge: Cambridge University Press, 1979. . "What's Wrong with Negative Liberty?" In Ryan (1979), 175-93. . "Cross-Purposes: The Liberal-Communitarian Debate." In Rosenblum (1989), 159-82. . Sources of the Self: The Making of the Modern Identity. Cambridge, Mass.: Harvard University Press, 1989. Wisconsin v. Yoder 406 U.S. 205 (1971). Wolff, Robert Paul. In Defense of Anarchism. New York: Harper & Row, 1970.