The main claim of this essay is that an appropriately developed, nuanced concept of autonomy can help overcome a supposed dichotomy (often exaggerated) between liberal and communitarian thought, thereby contributing to uniting them in a constructive way relevant to moral education. The paper includes illustrations of the role of higher education in moral education.

First, discussions of liberalism and communitarianism are not, as some have thought, a debate between two clearly distinguishable competing orientations. These discussions do display a renewed interest in community, a topic of educational importance, as John Dewey recognized, for example, in Democracy and Education. Dewey is arguably a liberal and a communitarian both, and consistent in this. For purposes of educational theory, as Dewey noticed, "community" is used both as a descriptive or explanatory concept, and a normative concept. This essay is particularly concerned with certain normative issues about community and moral education.

Second, autonomy, or free, rational self-determination, especially on an individual level, is a needed complementary value to community (both considered in their uses as normative moral concepts). A suitably reinterpreted notion of individual autonomy could help express and unify some of what is educationally valuable in both liberalism and communitarianism.

Third, concepts of individual and group selfhood or identity should be central in philosophizing about moral education. Focus on autonomy and community should also involve a moral educational focus on self and identity. Understanding the individual self and the identity of a community (what makes that person the individual self he or she is, or what gives the community the identity it has) is valuable. It matters for understanding individual and collective self-determination or autonomy.

Fourth, the emotional sensibility connected with blame and punishment, how an individual, or a community, reacts emotionally in blame and punishment, is a crucial aspect of identity. Critical examination of and pragmatic intervention about that emotional sensibility and its alterability matters a great deal in education for autonomy. Education for autonomy, in this and other ways, is the work of many institutions, among them public universities.

Fifth, educationally framed description and evaluation of selfhood or identity, and the education for autonomy of individuals and groups should replace "the liberalism-communitarianism dichotomy." That dichotomy is in part an evanescent episode in university culture and history, not a manifestation of a deep, unavoidable political and educational conflict.

Some philosophers continue to cultivate, while some would wish to bury, the idea of a dispute between liberals and communitarians. The philosopher-political theorist Charles Taylor argues that there are genuine differences between two sides, but that it is a multiple-stranded, complex debate.¹

A main issue in "the liberalism-communitarianism debate" has been a tension between what some call neutral liberalism (not all liberalisms are neutral) and communitarianism. Neutral liberalism holds some version of the view that a society or (more usually) a state should be neutral about conceptions of good character and the good life. Typically neutral liberalism favors support for an ethical and political framework of basic (individual) rights within which, it is thought, individuals should, freely and equally, decide in their own lives and for themselves what good character and the good life should be. John Rawls, Ronald Dworkin, and others have supported such views.² It is notable that even such liberals, who tend to differentiate themselves from communitarians, may make use of the concept of community. Thus Dworkin argues for obligations of community in Law's Empire, and attempts to adjudicate conflicts about liberalism and conservativism (in "Liberalism") and about abortion and euthanasia (in Life's Dominion) by appealing to what he seems to regard as some shared communal principle such as equality or the sanctity of life, respectively.³ Rawls increasingly argues for his position as an interpretation of the shared intuitions of what he considers the shared constitutional democratic political culture of this (the United States) and similar countries. It is entirely possible that Rawls has been influenced by what he opposes, including the communitarian-style criticisms of his theory of justice by Michael Sandel.⁴ Will Kymlicka, in such works as Liberalism, Community, and Culture and Multicultural Citizenship, finds himself extending liberal ideas about normative commitment to equality among individuals to less well-understood ideas about the equality of cultures.⁵ Thus even self-styled anti-communitarian liberals use ideas such as community and culture prominently. Moreover, though their educational commitments vary, neutral liberals tend (whether wittingly or not) to hold views that commit them to political support of public education for character traits such as autonomy, rationality, openmindedness, respect for scientific method, and many other features of character. Neutral liberal views and writings have been among those taught extensively in U.S. universities, including public (state and federally governed) universities, as well as "private" universities supported in part by public resources. In classes in normative ethics and political philosophy, for example, such views and writings (among others) have been influential. Such views and writings have not been mandated from above by political authority, but typically are chosen for study by university and college faculty. While not typically plainly coercive, the teaching of these views in these circumstances often seems to move beyond liberal neutrality as usually meant.

Now, consider some of the communitarians, whether clearer cases such as Alasdair MacIntyre in After Virtue, or Taylor in various writings, including Sources of the Self, The Ethics of Authenticity, and "The Politics of Recognition"; or less clear cases of communitarians such as Michael Walzer or the legal and political theorist Roberto Unger.⁶ MacIntyre attacks liberal individualism as incoherent, a condemnation with an educational dimension. In the clearer cases, the author explicitly rejects neutral liberalism, advocating that under some circumstances, at least, society or the state should eschew neutrality and take sides on good character and the good life. Taylor, the clearest case, directly addresses educational issues in some of his writings, for example, multiculturalist criticisms of educational canons. Walzer studies the educational sphere as one sphere in which concepts of distributive justice operate. The matter is complicated since even some communitarian-leaning but self-avowed liberal writers, such as Joseph Raz in The Morality of Freedom, abandon or attack the ideal of neutrality, while communitarians such as Taylor and Walzer would declare themselves liberals, and in the case of Unger, he writes as an advocate of a "superliberalism!"⁷ Unger has important and controversial views concerning education about the law, especially in law schools (presumably private but also public law schools which are part of universities). Communitarians, however, unlike neutral liberals, concede or even flaunt their non-neutrality about education for character, about the good life, and so on. Communitarians also tend to stress more the connections between community and education.

Any adequate education for individual autonomy (as free, rational self-determination) requires the efforts of a community which has collectively reached some self-understanding about the nature of autonomy and the conditions for autonomy. These conditions include the available social options, personal traits, and skills needed to be an autonomous person. (Later, I shall explore more of the relevance of autonomy to all this.) Such a community must make a commitment of educational resources (for example, in its universities and other schools) in light of its values. When a community affirms the value of some plausible moral ideal of autonomy in that community's educational practices, the situation is arguably one which both the better sorts of liberals and the better sorts of communitarians could approve.

One crucial example: in the public school system of an autonomy supporting community (including its public, politically governed universities) there should be an anti-authoritarian effort to educate morally, including character education. One way in which this would show up would be in the existence of certain philosophy courses at public universities. These are courses about normative ethics and politics, in which the objective should include teaching the sort of critical thinking that is an essential part of autonomy. Such autonomy can and should be an ideal of character that leaves room for lots of variability in one's self-chosen character traits and skills. Such courses are only the most obvious manifestation of communication by the public university of ideals about character and the good life. For better or worse, public (and other state-supported though not necessarily state-governed) universities are in the character-building enterprise, not solely in certain philosophy courses, but in the furthest reaches of their technical areas, public service activities, and so on.

In seeking to further public education for autonomy among its individual citizens, the state can be acting non-neutrally, and presumably building on some community consensus (however limited) about the nature of, and conditions for autonomy. This is "communitarian," broadly speaking. But it can also be "liberal," in leaving room for considerable variability about how individuals should develop self-chosen aspects of their own characters (compare J.S. Mill on individuality in On Liberty) and "liberal" in its emphasis on autonomy as free, rational self-determination. Preferably such university courses should be encouraged, not required. Such normative and character-targeted courses should be contrasted with required courses, preferably of a more factual and informational nature, about American history and government. But there is no way to avoid the implications for character education of the operations of public universities, even in the "factual" domain. Whether normative ethics courses are required or not, the distinction between courses with normative ethical content that shapes character, and courses that do not shape character is a blurry distinction at best in the university curriculum. Moreover, the total university environment cannot escape functioning purposively in moral character education, and not solely on a course-by-course basis.

I have rejected dichotomies between liberals and communitarians. Rather than a dichotomy, there is a renewed development of a type of normative ethics and political philosophy which takes community seriously. What are some major issues in this renewed development? These are issues which are about huge social and political problems, but at universities the issues are typically most explicitly debated, ironically, within the comparatively marginalized context of parts of philosophy departments and a few related areas such as parts of education schools or political science departments. This narrowness in the explicit university debate is unfortunate. The university environment would be far better educationally were the issues critically discussed much more broadly.

An examination of the "philosophical" literature shows not only that Kantian-style neutral liberals such as Rawls, and others such as Dworkin and Kymlicka, make autonomy central to their normative political theories, but that a communitarian liberal such as Taylor, or a communitarian-leaning liberal such as Raz, also emphasize the importance of the idea of self-determining freedom, both for the individual and for the community.

I am obviously emphasizing the idea of autonomy, its central importance in some philosophy that takes community seriously, and its presupposed idea of the self (or the corresponding presumably more holist notion of the identity of a group which can have or lack self-determining freedom).

A suitably enriched and modified notion of autonomy could help us unite much of what is good in both liberalism and communitarianism. As one concept of individual freedom, autonomy lends itself to taking individuals seriously, while interpreting autonomy as requiring a background of educational institutions and a community lends itself to taking community seriously (and not simply the government of a community). Autonomy, properly understood, also helps check the authority of collectives, which looms large in some communitarianism.

In English-speaking academic normative ethics and political philosophy, there has been a notable, relatively recent shift from utilitarianism to Kantian contractarianism (as typified by Rawls) to the rebirth of more "communitarian" modes of thought. From the earlier Rawls of A Theory of Justice to the later Rawls of Political Liberalism, and related writings, one can discern the latter shift. There is a greater emphasis on interpreting the intuitions and institutions of constitutional democratic cultures, and a greater defensiveness on the part of Rawls and Rawlsians about responding to communitarian criticisms.

This essay supports a third way, after utilitarianism and Kantianism, of doing the normative ethics needed in normative political philosophy. Academic ethics in the English speaking world in recent years has been dominated by utilitarianism and Kantianism. The third approach alluded to here takes the notion of community seriously, but need not be exclusively communitarian, so the word will occasionally be put in scare quotes. What is suggested here is that a type of "communitarian" ethics is worth trying out as a replacement for the two previous paradigms. Such an ethics need not abandon all the intuitions that drive utilitarian and Kantian thought. It can benefit from, but need not merely repeat some older types of virtue ethics.

All normative ethical theory should be developed with the aim of guiding moral educational practice. The question, "What is ethical theory for?" has a ready answer. Ethical theory is primarily for the pragmatic guidance of moral education in its many manifestations across a broad range of institutions, including but not only schools. That philosophical ethics has recently typically been academic ethics, in a university context, has obscured the pragmatic function of ethical theory. It is especially the comparative isolation of much academic philosophy from other disciplines and from the extra-academic world that has obscured this. The past reign of philosophical ethics conceived as primarily meta-ethics, also, has not furthered clarity about philosophical ethics conceived as a pragmatic guide for moral education.

Taylor's version of communitarian ethics, as expounded, for example, in Sources of the Self, and elsewhere, has much to be said for it as a guide to moral education. While indebted to Aristotle, Taylor is a sophisticated social scientist as well as a political philosopher. He takes autonomy seriously, but does not interpret it at its best as an overly individualistic notion. Rather, autonomy is seen as presupposing a societal and communal context. Shared institutions and history are deemed by him as necessary conditions for selfhood to emerge at all. He values some of the motivations in utilitarianism and Kantianism, even as he criticizes them. Taylor advocates a rejection of one-basic-reason monistic normative ethics. Even autonomy is not so rich that it can substitute for a plurality of values. There is in this approach a plurality of reasons relevant to ethics. One can, moreover, only grasp ethics in dialogue. Dialogue is how we learn and decide about what reasons should count in moral deliberation. Taylor rejects monological, first-person singular philosophizing in favor of dialogue. Ethical discussion uses varying intuitions in dialogue with others, in particular intuitions that are connected by individuals with crucial personal and cultural narratives. The idea of identity on an individual and collective level clearly plays a large role in Taylor's ethics.

On my view, then, a number of things are to be regretted about the way in which discussions of community have arisen in the context of an alleged dispute between liberals and communitarians. Certain ideas such as autonomy on an individual level have been distorted, allowing some excessively individualist theorists to neglect the background culture, community, and educational institutions which are always necessary to give the idea of individual autonomy (as well as group autonomy or self-determination) its intelligibility.

In the remainder of this paper, I wish to focus mainly on issues about blame and punishment within a more community-centered third way in educational thought about normative ethics and politics. I wish to show how the emotional sensibility characteristic of moral blame and legal punishment is one important educationally disputed feature of individual and community identity. One reason this matters is that imposition of non-neutral ideas about character and the good life generates worries about social and political coercion and related phenomena, including blame and punishment. There is concern about coercive imposition of a morality of character and the good life. Another related reason this matters is that some influential liberals and communitarians have played up ideas about obligation, blame, and punishment.

Without being particularly communitarian or autonomy-centered, Bernard Williams has over the years interestingly polemicized against utilitarianism and Kantianism.⁸ Taking Kant as the paradigm expositor of "morality," Williams has criticized (among much else) what he regards as morality's obsessive concern with obligation, general principles, and certain types of blame. It is worth recalling Williams when we reflect on some of the implications of autonomy and community as these have often been conceived. Autonomy has often been used (among other things) as a concept denoting a capacity, a type of freedom, necessary for moral obligation, blame, and legal punishment to be justifiable. Community has sometimes been conceived as a valuable generative source of obligation and responsibility, in contrast to the supposedly more individualistic, egoistic-tending liberal ideologies. I have criticized this dualism, but it persists in influential places. Combining autonomy and community in an educational theory need not result in a theory centered excessively on general moral or legal principles, nor on blame and punishment, and corresponding emotions. A focus on autonomy and community can promote fruitful new ways to reflect on the critique and justification of blame and punishment, and corresponding emotions. One way to make the field more fruitful is to raise questions both about selfhood and identity and about emotional sensibility.

Identity can be critically examined both on an individual and a group level. On a group level, for example, we can ask, what does this community judge or feel to be right about blame and punishment? Asking this question can at least sometimes be a way of asking about the identity of a group. Given its history and current practices, what gives this group as a community the identity it has? It is not always obvious what this question means. However, it is not entirely unclear either. There are many ways to cope with the question, many educative venues for the debate. It has both factual and normative dimensions. One can discuss group history and current issues that seem salient to its educational practices. History departments at universities could be of assistance here. One can focus on a community's criminal law. Legal studies at universities in and out of law schools can be helpful at this. Some interesting recent discussions of punishment suggest that punishment is inherently both a communicative and educational concept. The criminal law aims to teach, for better or worse. One can study a community's dominant narratives, which are so important in a group's moral education, including its blame and punishment. English departments, comparative literature units, film and video studies, journalism and mass communications, and other areas at universities are potentially useful in this regard. One can study a community's characteristic emotional sensibility, which is itself to a large extent the result of the group's moral educational practices, including its education about the moral emotions. Educational psychology, and the social sciences generally can be of service about this at universities.

Blame and punishment are poorly understood or justified solely by reference to the consequences of these practices. Retributivism seems more promising in this respect when it frankly confesses its linkages with retributivist emotions.⁹ To some persons, this makes retributivism less attractive; but it does also seem to make it more realistic. Communitarians do not always help about criticism and justification of blame and punishment, especially those who appeal to actually existing shared emotional sensibility as if it were the bottom line.¹⁰ However, communitarians can be a help in elucidating how a community's education in a broad sense constructs individual and community identity, including moral emotions. If the debate attracts attention at universities, and fosters interdisciplinary inquiry and discussion, the factual/moral issues about criticism or justification of blame and punishment and the accompanying emotions could be enriched. Perhaps more important, pragmatic educational correctives for potentially excessive or misdirected coercion (and blame or punishment) might be devised.

In justifying retributive (or other blaming and punitive) emotions, one approach would be to attempt to describe, critically assess, and devise educational modes of regulation for an overall psychology of identity, whether of an individual or a community. One must approach this with an understanding of the educational habits of the individual or group. It is obviously important that we cannot successfully ethically justify legal punishment by pointing to the facts about individual or community psychology and identity. Rather, when we have a better grasp of the facts, interpretations, and explanations of how blame and punishment work on an individual and community level, and how they fit in with identity, we are in a better position to discuss critically what is ethically attractive or not about the identity. This could be, for example, its proneness to lack of individual self-knowledge, its unimaginative sticking to a limited picture of what practices could give expression to individual or community identity, or its causal linkages with varying better or worse effects. Self-righteousness and malice sometimes figure in identities (individual or collective) centered on blame, punishment, and related moral emotions. This is one morally unattractive feature of such identities. Educational remedies for such identities need to be devised for many institutional contexts, including universities. Sigmund Freud and Michel Foucault might help, whatever department uses them, because of their wary examination of the excesses of blame and punishment.

There is no obvious reason why the identity of an individual or a community must be primarily or in large part constituted by a retributivist emotional sensibility, or a blaming and punitive emotional sensibility in general. Any communitarianism that suggests or even hints at this does not understand its own potential for generating educational options. By a retributivist emotional sensibility, or more broadly a blaming and punitive emotional sensibility, what I mean is one version of what P.F. Strawson has called the "reactive attitudes."

In Strawson's sense, the reactive attitudes include much more than retributivist emotional sensibility. Among the reactive attitudes, however, are the blaming and punitive reactions with which we are now occupied. Strawson writes: "The concepts we are concerned with are those of responsibility and guilt, qualified as 'moral', on the one hand - together with that of membership of a moral community; of demand, indignation, disapprobation, and condemnation - together with that of punishment."¹¹

In his essay, "Responsibility, Reactive Attitudes, and Liberalism in Philosophy and Politics," Samuel Scheffler (a self-described liberal, though apparently not a neutral liberal) argues that in light of Strawson's insights about the reactive attitudes, contemporary liberals ought to re-think their philosophical and political strategy.¹² Distinguishing philosophical from political liberalism, Scheffler writes that many outstanding recent philosophical liberals (including Rawls) and even some prominent communitarians (including Alasdair MacIntyre and Michael Sandel) do not acknowledge the reactive attitudes and desert (what Scheffler, following T.M. Scanlon, calls "pre-institutional desert") at their full significance. Scheffler rejects the possibility of neutral liberalism, so long as naturalism and a scientific attitude is taught as a basic part of the society's framework. There is some plausibility in Scheffler's diagnostic argument, and in his all-too-brief critique of liberal neutralism, even if his remedies are less than satisfying. The reactive attitudes, though culturally very variable, are in some form widespread and strong, and any politics that does not deal with this is likely to be doomed to failure.

Nonetheless, at a deeper level, Scheffler's argument, like Strawson's, does not consider enough the variable emotional roots and the educationally constructed nature of much about individual and community identity. Nor do Scheffler and Strawson fully appreciate the dangers of punitive moralism, once it is understood that a community is and must be in the character-building business. Both authors neglect the full range of educational options that we must at least contemplate as possibilities. Too little is known in this area, but we can speculate that as empirical knowledge grows, and especially as systematic university-based study of political psychology and political education grows, more will be learned concerning the educational options for modifying individual and community identity via changes in basic emotional sensibility. The normative moral educational possibilities regarding identity, individual and communal, might thus be expanded.

This essay thus agrees with those who have criticized the sometimes dichotomized debate between liberals and communitarians. Well-formulated concepts of autonomy and community could provide some common ground, and a better account of selfhood and identity could be central to a fresh approach to ethics, politics, and education. This new approach could include a critical examination of the educational practices that teach an emotional sensibility too often dominated excessively by unexamined emotions connected with blame and punishment. The identities of individuals and communities, contrary to what some philosophers have thought, may well be malleable and subject to change by educational choices which could appropriately modify the centrality of blame and punishment in identity formation. One reason such changes may be desirable is exactly that the reactive attitudes connected with blame and punishment are so strong, and tend to stifle free reflection and the critique of varied educational practices. This is not to say that we should do without blame and punishment (and corresponding emotions) in moral education. But it is to say we (at universities and elsewhere) need to be careful to evaluate the educational shaping of moral emotions connected with responsibility, without assuming their defensibility in their current manifestations. Concern for autonomy demands no less.

For a response to this essay, see Jones.

12. Samuel Scheffler, "Responsibility, Reactive Attitudes, and Liberalism in Philosophy and Politics," Philosophy and Public Affairs 21, no. 4 (1992): 299-323.