Content area
William Galston and Ronald Dworkin have tried to articulate a foundational conception of the good from which liberal politics can be derived. In an analysis of their competing conceptions, Lund assumes that any reasonable liberalism must be consistent with individual agency and the idea that genuine goodness is a product of inner persuasion and belief.
According to a number of recent critics, various efforts to justify liberalism in a neutral or merely "political" fashion have failed badly In response, William Galston and Ronald Dworkin, as well as other theorists, have tried to articulate a foundational conception of the good from which liberal politics can be derived. In analyzing their competing conceptions of ethical liberalism, I assume that any reasonable liberalism must be consistent with individual agency and the idea that genuine goodness is a product of inner persuasion and belief. Galston's "purposive liberalism" rests on a teleological structure that compromises his expressed concern for those value considerations and opens the door to public coercion and manipulation as a way of maximizing his liberal good. Since Dworkin's approach manages to avoid such problems, I argue that his "challenge model" of the good life provides a superior ethical foundation for liberalism.
While much of the world struggles to achieve liberal democracy, communitarians and other critics have been challenging the practical viability and even the ethical desirability of liberalism itself. In responding to these external critics, liberal theorists have been driven to a vigorous internal debate regarding the best way to characterize and ground the familiar principles and institutional arrangements they favor. As a result, we now confront a range of liberalisms, each of which offers a different and competing justification of liberal politics. In one recent analysis, Patrick Neal has identified three competing "'models' of a liberal order." The first, which he defends and I will leave aside, treats liberal institutions and principles as a "vulgar," neo-Hobbesian modus vivendi between citizens with irreconcilable conflicts of interest, but a shared strategic interest in avoiding suffering and violence. The second model, familiar from the work of John Rawls and others, treats liberalism as a "neutral" and merely "political" response to the presence of plural and incommensurable views of the good amidst an underlying commitment to equality. Without much evaluation, Neal also mentions recent efforts to develop an "ideal-based" model that specifies "a theory of the good life generally as the foundation of a political theory" justifying liberal institutions and practices (1993: 623, 626-27).
In what follows, I hope to contribute to the evaluation of this third model by explicating and analyzing two competing versions of what I will call ethical liberalism. Both William Galston and Ronald Dworkin lay out a conception of the liberal good from which liberal politics can be derived, and both take up this task in response to perceived shortcomings in the more dominant Rawlsian approach that eschews appeals to foundational goods. While there are other important versions of ethical or perfectionist liberalism (e.g., Raz 1986; Macedo 1991; and Kateb 1992), Galston and Dworkin each combine a well-developed ethical framework with concrete prescriptions on contemporary policy debates. Moreover, they disagree sharply on the character of the liberal conception of the good and on those policy prescriptions. Thus, their arguments may be a particularly apt subject for analysis not only for liberals interested in an in-house dispute, but also for those external critics whose worries have prompted this increased emphasis on the good in liberal theory.
On the other hand, the criteria I use in evaluating their work may well seem question begging to those outside the liberal tradition. However, both theorists claim to be defending a version of liberalism; hence, whatever other merits their particular theories might have, it does seem reasonable to expect them to support and promote several substantive desiderata that have been widely shared in liberal theory and practice. We can begin with "agency," or the idea that individuals are both authors of and actors in a life that features the pursuit of projects and goals reflecting their values and beliefs. As J. Donald Moon points out, the attribution of value to agency is historically contingent, but it is now both widely shared in liberal societies and crucial to our ideas of responsibility and the practice of justifying or excusing our conduct (1993: 108-09; see also Johnston 1994: 22-24). A closely related concern lies in the claim that both the goodness of fulfilling particular goals or projects and genuine virtue are internally related to the agent's values and beliefs. That is, any liberal theory must take very seriously the idea that, like religious faith, neither virtue nor ethical value can be genuine if it is the product of force, fraud, or manipulation.1 When combined, these considerations tend to support and explain other standard liberal commitments, including those to equality and individual rights, that limit public coercion in the name of agency and the search for a genuinely good life.
The argument that emerges from applying these criteria to Galston and Dworkin is that the less ground an ethical liberalism yields to substantive and independent ideals, the better off (or at least the more liberal) we are. In simple terms, Dworkin offers a superior approach to uncovering ethical foundations for liberal politics because he is able to incorporate the strengths of neutralist liberalism while also avoiding its major problems.2 Despite Galston's powerful critique of ethically thin liberalism, his constructive argument is ultimately vitiated by structural weaknesses that open a door for potential abuse. Generally, his theory moves too far from liberalism in the room it leaves for overriding agency and the constitutive relationship between choice and ethical value in the name of a contestable conception of the good. With that in mind, we should perhaps begin with a brief discussion of neutralist liberalism and its supposed shortcomings, shortcomings against which both theorists are reacting.
1. "POLITICAL" LIBERALISM AND PUBLIC NEUTRALITY
While the notion that legitimate political institutions must eschew the goals of religious salvation or secular ethical perfection has its roots in Locke and Kant, its recent popularity stems from Rawls's criticism of the tendency of various perfectionisms to maximize the achievement of ethical goals at the expense of individual freedom (1971: esp. 325-32). Shortly after A Theory of Justice was published, Dworkin, Bruce Ackerman, and others began to work out the implications of that critique by describing and justifying liberal politics through reference to a demand that public institutions and policies should be "neutral on what might be called the question of the good life" (Dworkin 1985: 191; see also Ackerman 1980). For proponents of this view, public neutrality is the morally best response to the fact that we confront plural and competing conceptions of ethical flourishing. They argue that a theory that supports the public enforcement of a substantive good must make one of two errors. Either it assumes the infallibility of claims for the ultimate value and truth of a particular way of life such as Aristotelian virtue or Marxian conscious and cooperative labor, or it proceeds by handing over to a particular group, including simply "the majority," the right to make and implement such decisions on the basis of their supposed wisdom or other signs of special status. In either case, the result will disrespect the equality of those whose conceptions of the good or whose group status are disfavored.
The alternative is to restrict legitimate public institutions and rules to providing an impartial framework within which citizens are free of public coercion in matters involving decisions on intrinsic value and the good life. Here, a liberal state should simply enforce those rights and duties that guarantee a moral floor on which citizens have the opportunity to make more exalted ethical choices in a private sphere of families, churches, and other voluntary associations. A sound political framework does not reflect judgments about the relative superiority of particular ways of life or particular groups. Instead, it strives for an ethically thin conception of justice and rights in which the principles, institutions, and policies required by that conception can "gain the support of an overlapping consensus of reasonable religious, philosophical, and moral doctrines" available in a society (Rawls 1993: 10, 15).
In developing and qualifying their earlier statements, Rawls and others have recently begun to argue that liberalism should be understood as a merely "political" doctrine. This means, in the first place, that the scope of liberalism is limited to "a specific kind of subject, namely, for political, social, and economic institutions" rather than being treated as a detailed picture of the ends and goods of individual lives. Contrary to some earlier arguments, however, political liberals deny that the neutrality constraint is itself foundational or that it can be defended in a purely neutral way Political liberalism is, after all, a "moral conception" that rests on certain ideals and principles, including especially the idea of equality, while still rejecting a priori standards of human excellence. In the second place, political liberalism and the moral basis on which it rests are seen as "freestanding" views which may be supported by, but are not derived from, more "comprehensive" conceptions of value and virtue. Finally, political liberalism draws on and reflects ideals of fairness, equality, and freedom that are "implicit in the public political culture of a democratic society" rather than being offered as a timeless and universally rational response to public dilemmas (Rawls 1993: 11-13).3
In modifying some of the more inflated claims of early neutralists, political liberals have argued that their doctrine will not be completely neutral in its effect or impact on the citizens and ways of life that exist within an appropriately neutral state. Public neutrality is a criterion that is used less to evaluate political outcomes than to judge and constrain political inputs, or the reasons and justifications that can be offered for particular policies. There are no public choices without gains and losses for adherents of particular ways of life, but if those choices "can be justified without appealing to the presumed intrinsic superiority of any particular conception of the good life," neutrality will have been honored (Larmore 1987: 44; see also Rawls 1993: 192-93; and Kymlicka 1989).
Political liberalism thus describes legitimate public institutions and policies as those which can be justified without settling on a choice of "the good life" that will be applied coercively to all citizens, and it justifies that approach by appealing to the ways in which it respects ideals of freedom and equality that are held by citizens who cling to any number of those more particular ethical or religious views. For its proponents, it is intended to be capacious and inclusive; but it is neither suicidal in terms of being unable to defend itself against fanatics and others whose "unreasonable" conception of the good requires them to use public authority to enforce it on the unwilling, nor sociologically blind to the fact that even neutrally justified policies can have a negative impact on reasonable and worthy conceptions of the good (Rawls 1993: 193,197; and Larmore 1987: 66).
While powerful and appealing, political liberalism has been subject to serious criticisms. One line of attack focuses on the substantive ethical components of the theory and, simply put, argues that it is either too thin or too thick. For some critics, political liberalism is too thin and vacuous either to provide the kind of shared commitments that make a society capable of honoring goods that go beyond cold and austere justice or to legitimize the sacrifices necessary to its own viability over time (see, e.g., Sandel 1982; and Taylor 1989). Alternatively, other critics like Michael Perry (1991: ch. 1) and John Kekes (1993: ch. 11) argue that political liberalism's demand for public neutrality is actually a covert way of smuggling into play a particular set of values, and that its appearance of fairness and capaciousness masks a politics that excludes and undermines those whose interests cannot be framed in an appropriately secular or empirical fashion.
Beyond criticisms of its substance, political liberalism has also been criticized for a weakness in its motivational capacity In seeking "overlapping consensus," it claims to secure support for liberal politics from those who may not begin with particularly liberal beliefs. However, the price of this search is a rigid separation between our public and private identities reflecting what Nagel calls a "moral division of labor" between the sort of arguments appropriate to public and private spheres (1991: ch. 6; see also Nagel 1987). In their private relations, political liberalism leaves individuals free to act on partial and particular conceptions of the good. There, as family members, coreligionists, members of clubs, or magazine and journal subscribers, we are free to take up and act on commitments whose basis may not be widely shared or neutral. But as "citizens," where we are both authors and subjects of coercive rules, we must leave those substantive attachments behind. Political liberalism requires citizens to seek impartial and neutral ground when they make public claims. Thus, they must either frame their arguments in terms that could not be rejected by other free and reasonable persons or give up the search for public support of their partial view of the good life.
Communitarians have argued that such a view reflects a deep philosophical mistake regarding the very possibility of differentiating our personal ends and our public projects. Here, however, am more interested in an immanent criticism that has come from other liberals who accept the possibility of this differentiation but are concerned about the motivational plausibility of bracketing private commitments when we move to the political realm. The problem can be put fairly simply: in seeking support from those who are not already committed to liberal views, political liberalism requires them to leave their deepest and most meaningful attachments behind when they move to the public arena. For Brian Barry and others, the only justification for this lies in an appeal to neutrality, and "there is no way in which nonliberals can be sold the principle of neutrality without first injecting a large dose of liberalism into their outlook" (Barry 1990a: 54; see also Waldron 1987: esp. 145-46).
Both Galston and Dworkin accept this argument. For Dworkin, political liberals have adopted a strategy that requires a "discontinuity" between public and private. The result is "a politics of ethical and moral schizophrenia" in that they ask us "to ignore instincts and attitudes on political occasions that are central to the rest of our lives." His goal is to "reconcile" the public and private perspectives by laying out a liberal ethic that can ground liberal politics in "the central or most important part of the ethics most of us embrace" (1990: 15-17). For Galston, the search for neutrality and the differentiation of public and private selves leaves political liberalism unable to honor "the lived experience (and highest possibility) of liberal life," and he blames it for a discomforting moral gap between contemporary American elites and ordinary citizens. In trying to lay out the goods, virtues, and purposes appropriate to a liberal society, he seeks "a more morally and humanly attractive account of liberalism, an account that can relieve many thoughtful individuals of the need they now feel to choose between liberal principles and their own moral experience" (1991a: 76, 17-18).
In what follows, I will emphasize their differences on the character of the liberal ethic, on the relationship between choice and goodness, and on the role and stringency of rights in a properly liberal politics. Here, the point is to note that both Galston and Dworkin offer their ethical liberalisms as a way of remedying perceived shortcomings in the political liberal project. Both seek to preserve and strengthen the politics of toleration, limits on governmental power, and the twin commitments to liberty and equality by rooting our attachments to liberal politics in a view of the good life that is both ethically thicker than neutralists will permit and, at the same time, widely shared in our society
2. TELEOLOGY WITH A LIBERAL FACE: GALSTON'S PURPOSIVE LIBERALISM
In a series of publications culminating in his Liberal Purposes, Galston has articulated an ethical liberalism that is intended to stand between neutralist liberalism and the restrictive and thicker perfectionisms of antiliberal and communitarian alternatives (1991a: 9). His effort is motivated not only by theoretical puzzles, but also by a practical concern over troubling trends in crime, divorce, and other negative "civic experiences." He roots these trends in the failure of political liberals to "attend to the dependence of sound politics on sound culture" and in their disregard of the need for a liberal state to reproduce the cultural "conditions necessary to its own health and perpetuation" (1991a: 6). To overcome these practical and theoretical problems, we must renew our acquaintance with "the Aristotelian tradition of the virtues" and recognize that, while underappreciated by elites and liberal theorists, traditional "virtues are by no means dead" in the lives of ordinary citizens (1991a: 66, 71-72). For Galston, then, we simply need to get clear about both the substantive goods that are implicit in liberal society and the individual virtues and public purposes that support and reproduce those goods.
His account of a liberal conception of the human good (or what he also calls human well-being) begins with what is presumably a conceptual truth: "Every society-and liberal society is no exception-rests on some shared perception of the good to be achieved through collective endeavor" (1991a: 70). Neutralist liberals have been both overly coy and arbitrary in spelling out their "perception of the good." While they promise to eschew substantive conceptions of the good, they all wind up covertly relying "on the same triadic theory of the good, which assumes the worth of human existence, the worth of human purposiveness and the fulfillment of human purposes, and the worth of rationality as the chief constraint on social principles and social actions" (199 la: 92). Galston, then, aims to present a fuller and more explicit account of the particular human telos that is displayed in (and presupposed by) liberal societies.
To that end, he lays out a chapter-length analysis of liberal goods, and the plural form signals an account that is complex, heavily qualified, and comprised of goods that are both distinctively liberal as well as those that would be required by any reasonable society He begins with a list of "background conditions" that any liberal theory of the good must satisfy These include the claim that its conception of well-being must be compatible with a secular outlook; that it must combine a sense of "unity and objectivity" with considerable room for diversity; and that it must refer to "conditions, capacities, or functionings, not just internal states of feeling." For my purposes, however, the most important of his "background conditions" reflects a deep and doubled pluralism. For Galston, not only is the liberal good itself made up of different and potentially incompatible elements, but it must also compete with other basic but distinct moral notions such as equality and agency, notions which are not always consistent with the individual's good. In rejecting monistic efforts to provide an overriding hierarchy of goods, he argues that any liberal theory of the good must be sufficiently pluralistic to allow for the possibility of value trade-offs, trade-offs both between the competing values that make up individual well-being and between that good and other basic moral considerations such as agency (1991a: 166-73).4
With those conditions in hand, we can move to the "key dimensions" of what he takes to be "a liberal conception of the individual human good." He begins by accepting the neutralists' commitment to life, individual purposiveness and its satisfaction, and sufficient rationality to be a functioning member of both the socioeconomic order and a moral order of equals who must offer and hear justificatory reasons rather than relying on force and manipulation. To that list, he adds several other goods. These include a concern for the "normal development of basic capacities" of speech, motion, and sociability; freedom as a key means to satisfying specific interests, asserting our personality, and achieving the integrity that comes with identifying with and being responsible for our ends; a "network of significant relations" with others; and finally, the ability to take subjective satisfaction in our "real personal accomplishments" (1991a: 173-77).
The liberal purpose, then, is to use the polity to provide the conditions under which individuals are able to flourish by achieving some satisfactory mix of those goods. The real load-bearing portion of his constructive argument, however, is that we will be unable to meet that purpose without paying more attention to individual virtue than political liberalism allows. While diversity and individual freedom are part of the background conditions and the goods of liberalism, "it cannot be wholly indifferent to the character of its citizens" or wholly indifferent to the fact that "at some point, the attenuation of individual virtue will create pathologies with which liberal political contrivances . . . simply cannot cope" (1991a: 216-17). Thus, while assuming that liberal virtues will differ in substance from those of classical Athens, he largely accepts Aristotle's account of the formal relationship between political communities and virtue. As Galston acknowledges, this requires us to negotiate certain ambiguities since Aristotle treats the virtues in three different ways: as means to the flourishing of particular human beings, as intrinsically valuable ends that politics ought to promote, and as instrumental and indispensable means to the health and stability of any given regime (1991a: 218).
In mapping these ambiguities on to liberal society, Galston mentions theorists like Locke, Kant, and more romantic individualists who have posited "overlapping yet distinct" conceptions of "intrinsic individual excellence" and liberal virtue that are often in conflict with each other and with the requirements of good citizenship (1991a: 229-32). He also notes cases in which liberalism yields a portrait of the virtues as means to individual flourishing; for example, the virtue of tolerance is instrumental to the good of purposiveness and the virtue of moderation is instrumental to the good of rationality (1991a: 227-28). However, his main emphasis, and the real need he posits for our current situation, lies in understanding the way in which certain virtues are instrumental to the health and stability of a liberal polity. He argues that the relevant virtues include those traits and habits of character that can be hypothesized to have an empirical connection to the health of liberal institutions. These virtues must be held by "most citizen" if a liberal society is to function well, and they are to be understood as challenging and at least occasionally inconsistent with self-interest (1991a: 220).
The essential virtues include some, such as courage and law abidingness, that are required by all political systems. Galston's focus, however, is on those traits and habits that are especially relevant to the health of a liberal society and which, contrary to neutralism, the state must take an interest in reproducing. If a diverse liberal society is to survive, citizens must have "independence," "tolerance," and the "virtues of family solidarity" required for passing on those other traits. A liberal market economy requires citizens with either the imagination and initiative to function well as entrepreneurs or the reliability and punctuality required of good workers. More generally, a liberal economy cannot survive without citizens who display the "work ethic," a capacity for "moderate delay of gratification," and "adaptability" A liberal politics of representative democracy and individual rights requires moderate and self-disciplined citizens as well as leaders who have sufficient patience and restraint to avoid "pandering to immoderate public demands" (1991a: 222-26). Mounting evidence of our currently "tattered social fabric" means that we must take these virtues and the common good to which they contribute more seriously, and he argues that we can do so only by attending "to the moral requirements of liberal public life and by doing what is possible and proper to reinforce them" (1991a: 237).
Many of those "moral requirements" have been staples of recent appeals by political conservatives for public support of traditional views of the family and the socializing role of religion. Historically, of course, liberalism has often sought to substitute critical reason for unanalyzed custom and habit, but Galston argues that political liberalism has set its face too resolutely against the very idea of tradition. A sound liberal polity need not go so far as strict public neutrality on, and juridically enforced rights against, the imposition of traditional values on individual lives. For Galston, liberals should continue to be wary of what he calls "intrinsic" traditionalism, or efforts to justify public coercion by appealing to standards of intrinsic value grounded in personal ethical or religious sentiments. However, where the justification for coercion is an empirical claim that the conduct it enforces is "functional" for a liberal society, we have a different case: "If a plausible (not necessarily conclusive) link can be forged between some aspect of traditional morality . . . and some feature of liberal public purposes, institutions, or conduct, then a rational basis exists" for policies that promote or enforce those aspects of tradition (1991a: 280-81). This approach, he argues, is by definition supportive of (functional for) liberalism, responsive to empirical evidence about the dysfunctional results of excessively individualistic and demoralized public policy, and beneficial in creating potential common ground between currently irreconcilable voices in our public debates (1991a: 287-88).
In all this, Galston presents a very powerful challenge to political liberals, and it is obviously difficult to criticize his account of the substantive components of a liberal good. Indeed, he modestly notes that he is neither breaking new ground in, nor claiming to have demonstrated "the correctness or completeness" of, his list of goods (1991a: 166, 173). He intends only to prompt us to a "minimal perfectionism" that is both more "open ended" and inclusive than Platonic or Aristotelian versions and heavily qualified in terms of the support it provides for public coercion (1991a: 177). However, despite his apparent concern for protecting individual freedom and diversity, his rejection of neutrality in favor of the state flatly taking sides in support of substantive goods and the "functional" virtues that promote them is troubling for a supposedly liberal theory of public purposes. The problems stem from the way in which he fits value pluralism into the basic teleological structure of his argument. That in turn leads to ambiguities or inconsistencies in his treatment of agency and the constitutive relationship between choice and goodness and, at the most concrete level, a relaxed conception of the extent and stringency of individual rights with an opening to more compulsion than Galston imagines or admits.5
We can begin with his liberal qualifications. For Galston, all political theories rest on and promote ideas of the good, but the distinctive aspect of liberalism is that it also contains principled and prudential "reasons for restricting the movement from the good to public coercion" of citizens. He notes three basic concerns that limit efforts to force citizens to the liberal goods: "liberalism embodies a respect for individual agency" that competes with efforts to maximize the good; like religious faith, some virtues and goods "do not respond to coercion" and cannot be genuine in the absence of "inner persuasion"; and even where coercion might produce the good, it may prove "so intrusive and potentially tyrannical as to be, on balance, destructive rather than productive of the human good" (1991a: 179-80, also 89). Elsewhere, he has offered parallel reasons as warrants for using rights to protect liberty: rights are legitimate not only when there are no clear and compelling standards of goodness and virtue, but also when coercive restrictions either contradict the good sought or create enough "obnoxious externalities" to outweigh that good (Galston 1983: 323). Thus, Galston argues, his perfectionism is sufficiently hedged to warrant the liberal label and to satisfy all but the most extreme defenders of free choice: "For the most part, a liberal society will understand the good as opportunity rather than as coercive command . . . [and it] will typically not try to enforce its conception on resisters" (1991a: 182).
Note the qualifications in the qualification and just how much hangs on ambiguities in "for the most part" and "typically" Those ambiguities stem in part from his conception of ethical pluralism. Recall that not only is his liberal good made up of heterogeneous and often incommensurable elements, but also that the good itself is held in tension with the separate and independent values of agency and equality. Following Brian Barry and others, Galston rejects the possibility of a monistic or "lexical" ordering of these diverse values and argues that a consistent and "intelligible structure of trade-offs" among the various values that come into conflict is the best that we can hope for (1991a: 181, also 41). Many political liberals (as well as Dworkin) also begin from ethical pluralism, but they then go on to treat equality or agency as principles that ought to be respected in and against political efforts at value maximization. In this, neutrality and stringent rights serve as constraints on public balancing designed to insure that value conflicts will typically be resolved by individuals and their voluntary associates rather than by public authorities.6 Galston, however, treats those principles as simply other values that have to be balanced against the good, and in practice he leaves considerable room for an unconstrained public balancing of the liberal good and those other values.
As a result, Galston's expressed concern for agency and the internal relationship between persuasion and value is seriously attenuated by his refusal to grant them any special weight against his substantive telos. The independence of the latter makes it an "agent-neutral value." This means that "anyone has reason to want it to happen" and that anyone has a reason to promote it even against an "agent-relative" reason that includes "essential reference to the person who has it" and yields deontological restraints on what we are allowed to do in the service of impersonal goods (Nagel 1986: 152-53, 175-80).
This tends to vitiate the principled reasons liberalism has "for restricting the movement from the good to public coercion," and it leaves only prudential checks on a community and its government treating a particular conception of the good as weightier than either agency or equality For Galston, after all, "the real good and the apparent good are not identical," and those liberals who emphasize agency or equal respect in order to require that we provide those we coerce with justificatory reasons they could accept ignore two facts: individuals can be mistaken about the good, and "the real good is what we really want" (1991a: 86). Thus, if public balancing tells us that we ought to coerce particular citizens to their "real good," we will not violate any fundamental principles if we offer them "what we take to be our true and best reasons for acting as we do" (1991a: 109). The fact that they may not accept or understand those reasons is of less concern than that we are promoting their good.
With this, Galston retreats from the practical weight that other liberals have granted to the constitutive relationship between inner persuasion and value. On his view, moralistic or paternalistic invasions of liberty can be wrong in principle only if public balancing of values is constrained by the idea that "the mere fact that the impetus toward the good is external somehow negates the worth of the good end so achieved," and this move toward making free choice a "necessary condition" of the good is one that he firmly declines (1991a: 86). To support this case, he notes that we often interfere with the freedom of heroin addicts or school age children to act on their immediate inclinations without worrying overmuch about whether they affirm the reasons behind such coercion. While reasonable, these examples may not be as telling as Galston seems to think. Any number of liberals might also accept them on the grounds that the interference is temporary and has a high likelihood that the individuals will in time come to see the worth of the conduct they are forced into, thus reconnecting choice and goodness. But, while he notes the expectation of "retrospective" affirmations in these cases, Galston says little about how that concern might limit or qualify his generalized rejection of the claim that "the noncoerced pursuit of the bad enjoys priority in principle-that is, in every case-over the coerced pursuit of the good" (1991a: 87).
Since he treats the good as independent of, and potentially weightier than, agency or the constitutive relationship between value and choice, Galston's conception of individual rights is also narrower and less stringent than that of many other liberals. He has argued in other places for a teleological justification of rights that rests on a what he calls a "modest naturalism." Here, rights are justified as instrumental means to the promotion and protection of the good defined as the realization of "quasi-moral facts: widely shared empirical features of human individuals (their nature, reason, needs, and purposes) that are invested with moral significance" (199 lb: 254-55). Since he takes liberalism to be "about the protection of diversity, not the valorization of choice" (1991a: 329, n. 12), rights cannot be used to protect choices that lead individuals away from goodness and the shared "quasi-moral facts" of human existence. Thus, Galston flatly denies that there is any "right to do wrong," and he argues that duties stemming from the "virtue and right conduct" which lead to the good are prior to the "integrity and self-constitution" that such a right might promote. Even when we grant rights because coercion will contradict the good we are seeking, our real aim is "promoting right conduct," and such rights are about "the ground of moral action" itself rather than agency or integrity (1983: 323-24).
To conclude this section then, despite Galston's claim to be explicating liberal purposes and the qualifications he attaches to the coercive enforcement of his liberal good, the structure of his theory must give some pause to those concerned with agency and the constitutive relationship between choice and goodness. Having subordinated rights to rightness, Galston himself sanctions any number of state actions that would violate those principles in the name of either "functional" virtues or a paternalistic concern for the individual's opportunity to achieve the liberal telos. In Liberal Purposes, he suggests or endorses criminalizing the sacramental use of drugs (263), compelling bigots to mouth tolerance against their beliefs (104), and removing children from the care of drug-addicted parents (182). He also suggests that any consciencebased exemptions from general duties such as saluting the flag or being subject to a military draft rest on "a prudential determination" of social costs rather than on any "general principle" (250-51).7 Elsewhere, he has also argued in favor of mandatory waiting periods and enforced counseling for parents seeking divorce and, perhaps most tellingly has left open the question of whether there is a right of privacy in the Constitution that can protect individuals against coercive interferences based on conceptions of the good which they reject (1992: 46; and 1991b: 260).
While his own list of restrictions may be troubling, the greater problem lies in the extent to which his theory leaves an open door for those with less liberal inclinations. He lays a lot of weight on the distinction between legitimately enforced "functional" traditionalism and illegitimately enforced "intrinsic" traditionalism. However, since all that is required to justify public coercion is a "plausible (not necessarily conclusive)" argument that some piece of behavior is healthy for our common purposes and institutions, and since no extra weight is given to equality or agency in conflicts between individual freedom and those traditions, he leaves us with little in the way of a principled presumption against public coercion. Thus, his rejection of neutrality in favor of public support for liberal purposes may turn out to yield results that are a lot less liberal than he leads us to believe. He claims that much public coercion will turn out to be "retrospectively" accepted by those whose freedom is momentarily limited and that, over time, the coerced behavior will "evoke a higher level of normative internalization and voluntary compliance" (1990/91: 15). For a supposedly liberal theory, the point should be that such "internalization" remains a future hope while the compulsion that denies agency and creates gaps in the relationship between choice and goodness is certain and immediate.
3. DWORKIN AND THE CHALLENGE" MODEL OF ETHICAL VALUE
In "Foundations of Liberal Equality," Dworkin seeks to transcend political liberalism and his earlier conception of public neutrality while preserving the rigorously liberal substance of his well-known political and constitutional arguments. He now accepts the criticism that political liberalism's effort to distinguish "the personal from the political perspective" overemphasizes "detachment and impartiality" and leaves us with an "austere, gray" politics that is unable to capture the richness and commitment of our private ethical lives. He seeks to meet that criticism with an account of ethical foundations that will reconnect public and private by grounding liberal politics in "a plausible and attractive account of how people should think and act in their private lives" (1990: 12-13).
To meet that goal, he eschews the "discontinuity" strategy adopted by Rawls and others who treat justice and principles of right as independent of (and constraints on) private conceptions of the good. Instead, he argues for a "continuity" strategy in which liberal politics is derived from "liberal ethicsinstincts and convictions about the character and ends of human life" that are especially conducive to such politics. Since any liberal politics must be tolerant with regard to the diverse ways of life its citizens choose, this liberal ethic must be quite abstract: "It cannot consist in some detailed description of the good life that is controversial within the political community," and it must therefore have a "structural and philosophical rather than a substantive character" (1990: 17, 20).
Contrary to Galston, then, Dworkin's ethical liberalism is not an argument for the superiority of particular substantive commitments regarding the good life or a defense of the good of particular virtues which aid in realizing that life. Rather, it is an account of a general conception of what it is to have a good life. He has argued elsewhere that we are equal in terms of a fundamental interest in the intrinsic value of having as good a life as possible (1983: 2627). Here, that goodness is explicated in terms of the extent to which a life contains within it the satisfaction of what he calls "critical" interests. Contrary to "volitional" interests, where something is thought good simply because we desire it, our critical interests involve elements of life that we ought to want, elements that are thought to be good independently of our current preferences and about which we may be mistaken. Contrasting a desire to avoid trips to the dentist with the desire to have a close relationship with one's children, Dworkin argues that the latter illustrates critical interests and that one's "critical well-being is improved by his having or achieving what he should want, that is, the achievements or experiences that it would make his life a worse one not to want" (1990: 43-44; see also Dworkin 1989: 484-85). With that, however, we still need some account of the general features of "critical well-being" that might be shared across the diverse commitments characteristic of liberal societies.
To that end, Dworkin begins by distinguishing two different "models of value." Each of these models will have some grip on our ethical imaginations, but they offer competing "ways of conceiving the source and nature of the value a life can have" for the person concerned with her critical interests. The first is the "model of impact." Here, "the value of a good life consists in its product," or the way in which it yields greater happiness, more artistic achievement, more religious devotion, or whatever other agent-neutral state of affairs is taken to be independently valuable. This model lies behind utilitarian and many communitarian theories and, I think, captures much of the flavor of Galston's ethical liberalism. The second, which Dworkin argues ought to be more prominent in a properly liberal ethic, is "a model of challenge" in which "the value of a life lies in the inherent value of a skillful performance of living," or in the relative skill with which an individual applies her judgment to the ethical and economic situation she confronts in order to achieve critically valuable events and experiences (1990: 53-55).8
So stated, these are very abstract conceptions. But, for Dworkin, the necessary flesh can be attached by way of the different answers the two models provide to a range of questions and anxieties that may puzzle anyone reflecting on their ethical convictions and challenges. These include concerns about how much ultimate significance ethical choices can have in the vastness of human history or the universe; about whether ethical standards are to be understood as universal and perennial or rooted in cultural particulars; and about the relationship between self-interest and morality. For my purposes, however, the most important and revealing of the ethical anxieties he mentions concerns the question of whether or not an individual can be said to have had a good life independently of her being persuaded of that fact (1990: 47-53).
On this last question, Dworkin argues, the "impact" model yields a positive answer. It leads us to what he calls an "additive" view in which we can judge a life good or bad without necessarily consulting the beliefs of the person whose life it is. If the life "has the components of a good life, then it is good for that reason," and those components provide agent-neutral reasons for securing that good in the world independently of whether or not other individuals have "endorsed" those components. If they do, it may add value to our calculations, but even if not, the "impact" will be to increase goodness, and "the ethical value of the components" will remain despite the rupture of the relationship between goodness and inner persuasion (1990: 50; also 76). The political point here is that the "impact-additive" view weakens principled opposition to public moralism or paternalism. There may be prudential limits to consider, but if we can increase, say, the stock of utilitarian happiness or Galston's "functional" goods by coercing behavior, then there will be a point in doing so whether or not the person behaving so as to enact those goods values them. For Galston, remember, the "consciously willed pursuit" of the good is not a "necessary condition of the value of attaining it" (1991a: 86).
The alternative is a "constitutive" view of the relationship between inner persuasion and the goodness of a life. This view reflects the "challenge" model's portrait of a good life as a skillful performance and the recognition that our judgment of how well someone meets such challenges is intimately connected to their having the correct intentions and motivations. On a constitutive view, the production of independently good features can not be judged to "so much as contribute to the value of a person's life without his endorsement" of their value. For Dworkin, as for Galston, the real good is what we want and we may be mistaken about what is of "critical" value. But he combines those disclaimers against subjectivism with a distinction between questions about the discovery of intrinsic goodness and questions about how political authorities ought to act to enable citizens to have good lives. With regard to the latter, Dworkin argues, "no event or achievement can make a person's life better against his opinion that it does not" (1990: 50). When combined with the assumption of an equal interest in achieving critical well-being, the constitutive fusion of persuasion and goodness yields agent-relative reasons against restrictions of liberty grounded on communal or human goods. Contrary to Galston, then, Dworkin highlights and gives priority to "ethical integrity," or the individual's ability "to live out of the conviction that his life . . . is an appropriate one for him" rather than acting on externally imposed, but unendorsed, values (1990: 80).
Thus, the "challenge-constitutive" view puts agency and the impossibility of coerced goodness at the core of liberal ethics rather than treating them as separate values to be balanced against the good. With that, it provides a principled limit on policies that restrict liberty in the name of unendorsed appeals to paternalistic or moralistic goods. Dworkin does allow for some forms of paternalism: "volitional" paternalisms, such as seat belt requirements, simply help people achieve "volitional interests" and preferences they are assumed to have. Other limits on acting from internal conviction may be justified if there is a high probability that they will ultimately be "endorsed" by the individual in a way that is genuine and free of fear or manipulation, or if they are structured so that liberty is not seriously and permanently restricted if the endorsement never comes (1990: 77-79). Beyond that, however, the "challenge-constitutive" model rules out public efforts to enforce and maximize the good whether they take the simple form of coercing particular behavior or the more sophisticated form of using the law to restrict options and awareness of lives thought to be bad on some ground.9 No performance can be a skillful one if the good it achieves comes at the expense of integrity, and efforts to override conviction and inner persuasion in the name of an unendorsed good are, for Dworkin, simply self-defeating.
Those restrictions, and the underlying distinction between maximizing agent-neutral goodness and providing the necessary conditions for agents to have a good life, must find their way into political principles and institutions if Dworkin is to succeed in harmonizing public and private. Thus, he asks us to imagine a group of people reflecting on the proper principles to govern their public relationships. They are each concerned to advance their critical interests, but each of them, while holding diverse conceptions of what makes a good life, understands "their different ethical convictions in the fashion of the challenge conception" rather than in terms of the "impact" model of value (1990: 89-90). He argues that with full and unveiled information about the structure and substance of their ethical convictions, these ethical liberals would choose the politics of "liberal equality" as the best way of harmonizing their public lives and private commitments.
Liberal equality reflects and summarizes Dworkin's earlier and more detailed arguments (1981, 1987) regarding the best distribution of income, opportunities, and rights and liberties. Here, he argues that ethical liberals would choose a conception of justice that is built "on the space of resources rather than well-being or welfare"; they would insist that justice demands that those resources follow an egalitarian distribution; and they would require such distributions to be sensitive to the difference between inequalities rooted in handicaps and other undeserved circumstances, and those that stem from expensive "tastes, ambitions, and preferences," the responsibility for which can be assigned to individuals. The final and, for my purposes, most important aspect of liberal equality is that it would be tolerant and in some sense "neutral on the good life" (1990: 90).
Given his commitment to the "continuity" strategy, however, Dworkin is now led to reject his earlier view that public neutrality is an "axiom" or a foundational constraint on ethics and politics. Instead, it should be seen as a political "theorem" that flows from ethical liberalism and that could be accepted by most people "not in spite of their own passionate ethical commitments, but because of these" (1990: 7, n. 2, and 118). In other words, rather than imposing neutrality at the front end of political discourse, those who view the good life in terms of the "challenge" model and its "constitutive" relationship between persuasion and value will be led naturally to the idea of public neutrality as a constraint on the day-to-day workings of their political system.
At this point, of course, critics might argue that Dworkin falsely assumes an existing consensus on the "challenge-constitutive" model, and he concedes that many people do not currently accept that view or its neutralist implications. However, he goes on to argue that liberal equality does offer significant "consensual promise" of achieving the liberal goal of principles that can be supported across a range of particular ethical commitments, and he claims that public reflection and argument will add to the number of those who see their particular views in "challenge" terms. Some support for that optimism stems from his distinction between "first-person" and "third-person" ethical views. The former are simply a person's "beliefs about the life right or best for him," while the latter include "convictions, about the best life for others."
Clearly, the politics derived from the "challenge-constitutive" model will constrain (and not be neutral toward) those "third person" ethical commitments that require the use of public authority to coerce other individuals to a life they cannot endorse. However, his optimism regarding the "consensual promise" of the "challenge" model lies in the claim that, with the exception of certain "fanatics" whose "first person" views require them to harm us, almost anyone can "occupy the position of an ethical liberal without abandoning the heart of his ethical convictions understood in the first person, that is, as convictions about how he should live well" (1990:112-13).
In operation then, liberal equality will allow (and require) restrictions of liberty based on claims that particular conduct is unjust or directly harmful to others, but it will be neutral in the sense that it rules out coercive denials of liberty rooted in "third person" or agent-neutral convictions regarding the components of a good life. Unlike Galston, Dworkin regards the law less as an ethical tutor and more as part of the "circumstances" in which individuals struggle with the "challenge" of securing the best life possible. Since equality of resources and circumstances is a central criterion of Dworkin's view of liberal justice, when choosing political principles we have to remember that things are "plainly unequal when the law forbids some to lead the lives they think best for them only because others disagree." For Dworkin's ethical liberals, then, both reasons of justice and reasons drawn from their acceptance of the "constitutive" structure of value rule in favor of political tolerance and the neutrality that tells against efforts to improve someone's life "against his steady conviction that it has not been" (1990: 115-17).
Against critics of a caricatured neutrality, Dworkin notes several qualifications here. Liberal equality is not expected to be (could not be) completely "neutral in consequence" or without differential effects on various ways of life; it cannot be neutral with regard to ethical views that reject its theory of justice; and it will not be neutral or give equal respect to the sort of "third person ethics" that require public efforts to override the "constitutive" relationship between choice and value in other citizens' efforts to secure a good life. Finally, and most importantly in the context of Galston's and others' concerns about liberal theory's capacity to address our decaying social fabric, the neutrality requirement does not rule out non-coercive efforts by government to promote or encourage particular conceptions of the good. Like other liberals, Dworkin assumes that ethical tutoring is best accomplished in the context of families and private associations, but his view of neutrality does not prohibit governmental policies aimed at "short-term educational paternalism" with a high probability of unmanipulated "endorsement" (1990:117).
More specifically, he has argued in other places that it is perfectly consistent with neutrality, and with the underlying commitments to agency and the "constitutive" view of a good life, for government to adopt policies that restrict or regulate the exercise of individual rights if their goal is the promotion of "responsibility" and a reflective examination of citizens' current convictions. Thus, policies that qualify abortion rights, regulate the availability (or suggest doubts about the value) of pornography, or support "high" culture and art will not fail the neutrality test if their justification is the promotion of ethical reflection and the preservation of valuable ethical options rather than that of simple "conformity"1o Despite these disclaimers, however, Dworkin's general view is that we can achieve a legitimate liberal "continuity" between public principles and private ethical views only if the latter still yield a strong and principled tolerance. His "challenge" model of value secures support for that tolerance from a variety of more particular ethical commitments, and in operation it will require that government coercion be "neutral about firstperson" ethics if they do "not embody antiliberal political principles" (1990: 118).
In seeking practical instruments for honoring that neutrality, we are led to the idea of individual rights. While Dworkin says very little about rights in his Tanner lecture, his ethical liberalism is fully consistent with his earlier demand that we take a wide array of stringent individual rights very seriously. However, his argument for that appears to have been modified and deepened. As is well known, his earlier view treated the "trumping" power of rights as "parasitic" on the utilitarian idea of a collective or communal goal, and he argued that the justification for trumping such goals lay in a largely ungrounded postulate that governments ought to treat citizens with "equal concern and respect" (Dworkin 1978: xi, 272-73). His more recent view seems to ground the political norm of equality in our equal and ethically fundamental interest in having as good a life as possible. Moreover, it now seems that the specific rights which embody the right to treatment as an equal are less "parasitic" on utilitarianism and more a reflection of the principles that would be chosen by those who see their particular ethical views along "challenge-constitutive" lines.
Whatever the derivation of rights may be, Dworkin has always shared Galston's view that they are not the same thing as rightness: "saying that someone has a right to do something" is not to say that it is either prudentially or ethically "the 'right' thing for him to do." However, unlike Galston's account, in which rights are purely instrumental to our interests in the liberal telos, Dworkin includes agency and integrity in the bundle of interests that rights protect. As a result, his version of ethical liberalism constrains public balancing of conflicts between rights and rightness by including those interests and the "constitutive" relationship between value and choice on the side of individual "endorsement." Rights are not rightness, and they may be overridden by concerns for justice and the rights of others, as well as by efforts to "prevent a catastrophe or even to obtain a clear and major public benefit" (Dworkin 1978: 188-91); but rights and even the right to do what others regard as wrong continue to trump public coercion rooted in concerns for "conformity" to the rightness of either a particular community's common good or agentneutral claims about ethical perfection.
Thus, Dworkin maintains a principled opposition to paternalistic or moralistic policies that deny equality and restrict liberty in the name of a substantive, but unendorsed, good. His view of a strong constitutional right of privacy, which reflects a pre-political constraint that he has variously called a right to "moral independence" (1985: ch. 17) or a right to judge "intrinsic value" (1993: ch. 1), helps to secure the public neutrality that is grounded in concerns for having a good life and the constitutive relationship between choice and value. His prior arguments against policies that ban or unduly regulate abortion, consensual adult homosexual relationships, euthanasia, and pornography should provide a roadmap regarding his views on the specific policy prescriptions that Galston and other less liberal theorists have proposed for promoting virtue and the good in our current circumstances.
4. CONCLUSION
We have before us several recent efforts to respond to the charge that liberal politics cannot succeed without greater attention and commitment to an underlying conception of the good. The charge itself may not ultimately hold up, and political liberals are still busily defending the idea of putting the right prior to the good. Fortunately that debate need not be settled here. All I have suggested is that if we do ultimately need an ethical liberalism, then we ought to be wary of throwing out the liberal baby with the neutralist bathwater. If we take agency and the impossibility of a coerced good life to be central elements of any defensible liberal theory, then a reasonable version of ethical liberalism must yield a politics that is compatible with and supportive of those concerns.
On those grounds, Dworkin's approach is more promising than Galston's because it provides greater weight for the idea of agency and because it explicitly rejects the idea that good lives and real virtue can be created without "endorsement." I have tried to demonstrate that greater promise by analyzing the ways in which the two theorists differ in their treatment of three common aspects in their ethical liberalisms. At the most abstract level, we can ask how the two views characterize their liberal ethics. Dworkin portrays his in structural and formal terms so that the "challenge" model is explicated in terms of the general properties and formal conditions of what it is to have a good life despite wide variation in the content of those lives. Galston, on the other hand, lays out an explicitly substantive account of what a good life is, and while that is meant to be capacious and only modestly perfectionist, it rests on a teleological and agent-neutral structure that invites the sort of publicly endorsed value maximizing that has always worried liberals.
At a less abstract level, both theories accept ethical pluralism and the problems created by the fact that we confront conflicts between incompatible and incommensurable values that may not have fully harmonious and rational resolutions. For Dworkin, finding the best resolution to conflicts of goals, obligations, and ways of life is a core aspect of the challenge individuals face in seeking a good life. However, he includes both equality and the "constitutive" relationship between choice and goodness as necessary conditions of any account of having a good life. Thus, within the constraints of justice, the "challenge" model of ethical liberalism aims to leave the resolution of value conflicts in the hands of individuals and their voluntary associates, and it provides principled resistance to efforts at publicly determining and enforcing the proper balance of conflicting values. Galston's pluralism, on the other hand, extends to the ideas of agency and equality themselves so that they are simply separate and distinct components of value that may well lose out when balanced against concerns for the good. Since he argues that inner persuasion is not necessary to achieve the good, it is always possible that individual resolutions of ethical conflict can and should be overridden by a publicly determined balance of value.
At the most concrete level, both theories go on to address the ideal of neutrality and the rights that give it its political punch. For Galston, it is conceptually impossible for a political theory to be neutral on the good life, and theories which mask or downplay their underlying ethical commitments in the name of a public purposelessness are ethically dubious and sociologically imprudent in stripping the polity of the means to promote and enforce those character traits and behaviors that enable it to endure over time. Further, he argues that attempting to implement a neutralist agenda by expanding the set of judicially enforced rights, including especially any idea of a "right to do wrong," simply confuses the instrumental relationship between rights and rightness. Since he sees neutrality and claims for rights that are independent of the good as primary causes of many of our current social and political maladies, he sets his face firmly against adding to our reliance on them.
For Dworkin, on the other hand, neutrality is an important "theorem" that flows out of the convictions of ethical liberals. It is not foundational, and it is clear that "liberalism takes sides" at the "philosophical level of ethics" where it favors, for example, the "challenge" over the "impact" models of value and the "constitutive" rather than the "additive" view of the relationship between choice and goodness. However, "the side it takes at the philosophical level dictates neutrality" at the operational level where we must consider the legitimacy of justifications for substantive policies (1990: 42). Rights, then, must still be taken very seriously as one mechanism for insuring that individual interests in integrity and having a good life by one's own lights will not be overridden just because others think that life to be ethically wrong. For Dworkin, we can and do fail in meeting our ethical challenges and securing our critical interests, but rights promote the idea of responsibility for those failures and insure that whatever goodness and virtue emerge from those challenges are genuine.
The debate over the explicit ethical foundations of liberal politics is an interesting and challenging addition to recent political theory Much more work needs to be done, both on the prior question of whether liberalism needs such foundations and then, if so, on what the shape and character of those foundations should be. In explicating and evaluating two ethical liberalisms, I have argued that Galston's purposive liberalism is motivated by the need to support and promote substantive values and virtues thought to be "functional" for restoring order and the common good in our current situation. Dworkin articulates an abstract view of ethics that is implicit in much of what many of us do believe and that can yield principles to be used in criticizing many of our current arrangements. We may leave open the question of whether liberalism is better seen as a prop or a critic of existing social arrangements. Here, I have simply suggested that Dworkin is truer and more faithful to the liberal tradition's emphasis on respecting individual agency and on securing real value through persuasion rather than coercion or manipulation.
NOTE: I would like to thank Greg Hill for some very helpful criticisms of an earlier version of this essay
1 On the constitutive relationship between inner persuasion and real value, see Kymlicka (1991: esp. 12-13) noting the importance liberals attach to being able to "lead our life from the inside," and Korsgaard (1993: 56) on "ethical individualism" in which "the
goodness of a life essentially depends on its being chosen and constructed by the person who lives it." Like Johnston (1994: 71-77) and others, I distinguish agency and the necessary connection between value and belief from the more demanding ideal of full Kantian autonomy That is, someone can be an agent and be inwardly persuaded of the value of their projects without having engaged in a self-conscious and critical evaluation of the beliefs and values that motivate those projects
2 It should be noted that Dworkin's well-known essay, "Liberalism" (1985: ch. 8), was one of the first and most important calls for requiring the liberal state to be neutral on the good life. As a reviewer points out, that makes his more recent view, on which I focus in this paper, an exercise in self-criticism. Dworkin's reasons for transcending his own earlier view of neutrality as a foundational aspect of liberalism, and the way it now figures in his account of liberal equality, will be discussed in Sections 1 and 3 below. It is, of course, still an open question whether he has moved any great distance from "political" and neutralist liberalism. Without pretending to resolve the issue here, I would suggest that the answer may depend on whether his abstract "challenge model" of the good life really meets the criteria of a Rawlsian "comprehensive conception of the good," or whether it just moves the distinction between "political" and private perspectives to a different stage of the argument.
3 For a marginally different explication of the meaning and point of political liberalism, see Larmore (1990). For Rawls, the point here appears to be to spell out his earlier argument (1971: 327) that a contractarian theory works with "an ideal of the person without invoking a prior standard of human excellence."
4 In arguing that concerns for goodness or well-being will have to be balanced against agency and equality, Galston claims to be following A.K. Sen's complex pluralism. At least some of Sen's remarks suggest that he might be more willing than Galston to grant the concern for "agency" special weight in at least certain kinds of questions: see Sen (1985: 186, 208) for the view that "in a person's own life" and in issues of "personal morality" we must treat agency as "especially important."
5 The concept of "compulsion" is, admittedly, rather general. For my purposes, I mean it to refer to a variety of possible state policies that would violate agency and break the necessary relationship between inner persuasion and the goodness of external behavior by coercive force, threats, or manipulation. Using the criminal law to punish particular "harmless" behaviors is the obvious and traditional liberal concern, but the less forceridden use of fiscal policies and general legal institutions can also run afoul of agency and inner persuasion if they have no other justification than to subsidize or raise to unacceptable levels the costs of conduct thought to reflect or deny a particular conception of the good life. There are obviously "fine line" questions regarding the difference between legitimate public persuasion and illegitimate manipulation to be considered, but to do so would require at least another paper. For some illuminating analysis of the issues in this area, see Waldron (1989: esp. 1138-52).
6 Even Brian Barry himself may be more sympathetic to efforts to constrain purely intuitionistic balances than Galston allows. See Barry's (1990b: Ix-lxxii) argument that the search for "lexicality" is implausible except in ideal circumstances, but that Rawls's effort to provide some constraint on public balancing by appealing "to general theoretical considerations," including principles "worthy of the assent of reasonable people," is much more defensible. For the problems with imposing on citizens an unconstrained public "balance" of pluralistic goods, and an analysis of the way in which Dworkin and more overtly political liberals draw compelling constraints on that process from the political weight of equality, see Lund (forthcoming).
7 In a recent essay, Galston (1995: 524-25) defends conscience-based exemptions from general duties and supports "maximum feasible space for the enactment of individual and group differences," especially if those differences are rooted in religious claims. The exemptions are, however, still constrained by the "more than minimal" requirements of liberal "public purposes," and they are warranted only as a means to the end of "diversity" rather than as a matter of autonomy or the constitutive relationship between choice and value.
8 Thomas Hurka (1995: 39, n. 11) argues that Dworkin's two models are "not exhaustive" and that there are "more attractive perfectionist" theories which do not have the shortcomings Dworkin attributes to the "impact" model. I'm not sure which theories Hurka has in mind, but I hope that the argument in section 2 indicates the extent to which Dworkin's "impact" model captures much of the flavor of Galston's "minimal perfectionism."
9 Wolfe (1994) offers a helpful survey of the various forms of paternalism that Dworkin discusses while arguing against any principled objections to paternalism. Like Galston, Wolfe emphasizes the role of the law as an ethical tutor and subordinates "ethical integrity" to the "common good" and "society's desire" to limit the environmental spillover caused by permitting conduct that is thought to be wrong even though it does not violate justice or second party rights (633). While I cannot go into it here, Wolfe appears to ignore completely the room which most liberals leave for private and familial efforts at habituating virtue and discouraging vice and does not consider the constitutive relationship between persuasion and value.
10 Hurka (1995: 36-37) shares Galston's view that contemporary neutralists deny authorities the opportunity to non-coercively encourage citizens to seek the good. Whatever may be true of other theorists, this charge seems wide of the mark with regard to Dworkin. For his distinction between policies that coerce or encourage in the name of "conformity" and those that promote "responsibility," see Dworkin (1993: 150-54). On the legitimacy of policies that regulate the exercise of the right to pornography or promote art, see Dworkin (1985: chs. 11,17).
REFERENCES
Ackerman, Bruce. 1980. Social Justice and the Liberal State. New Haven: Yale University Press.
Barry, Brian. 1990a. "How Not to Defend Liberal Institutions." In R. Bruce Douglas, Gerald Mara, and Henry Richardson, eds., Liberalism and the Good. New York: Routledge.
. 1990b. Political Argument: A Reissue With a New Introduction. Berkeley: University of California Press.
Dworkin, Ronald. 1978. Taking Rights Seriously. Cambridge, MA: Harvard University Press.
. 1981. "What Is Equality? Part I: Equality of Welfare," and "What Is Equality? Part II: Equality of Resources." Philosophy and Public Affairs 10: 185-246, 283-345.
.1983. "In Defense of Equality" Social Philosophy and Policy 1: 24-40.
.1985. A Matter of Principle. Cambridge, MA: Harvard University Press.
. 1987. "What Is Equality? Part III: The Place of Liberty." Iowa Law Review 73: 1-54.
.1989. "Liberal Community." California Law Review 77: 479-504.
.1990. "Foundations of Liberal Equality." In Grethe Peterson, ed., The Tanner Lectures on Human Values, Vol. 11. Salt Lake City: University of Utah Press.
. 1993. Life's Dominion: An Argument About Abortion, Euthanasia, and Individual Freedom. New York: Knopf.
Galston, William. 1983. "On the Alleged Right to Do Wrong: A Response to Waldron." Ethics 93: 320-24.
. 1990/91. "A Liberal-Democratic Case For the Two Parent Family" The Responsive Community 1: 14-26.
. 1991a. Liberal Purposes: Goods, Virtues, and Diversity in the Liberal State. Cambridge: Cambridge University Press.
. 1991b: "Practical Philosophy and the Bill of Rights: Perspectives on Some Contemporary Issues." In Michael Lacy and Knud Haakonssen, eds., A Culture of Rights: The Bill of Rights in Philosophy, Politics, and Law-1791 and 1991. Cambridge: Cambridge University Press.
.1992. "Putting Children First." American Educator 16: 8-13, 44-46.
.1995. "Two Concepts of Liberalism." Ethics 105: 516-34.
Hurka, Thomas. 1995. "Indirect Perfectionism: Kymlicka on Liberal Neutrality." Journal of Political Philosophy 3: 36-57.
Johnston, David. 1994. The Idea of a Liberal Theory: A Critique and Reconstruction. Princeton, NJ: Princeton University Press.
Kateb, George. 1992. The Inner Ocean. Ithaca, N.Y.: Cornell University Press.
Kekes, John.1993. The Morality of Pluralism. Princeton, NJ: Princeton University Press.
Korsgaard, Christine M. 1993. "Commentary on G. A. Cohen and Amartya Sen." In Martha Nussbaum and Amartya Sen, eds., The Quality of Life. Oxford: Clarendon Press.
Kymlicka, Will. 1989. "Liberal Individualism and Liberal Neutrality" Ethics 99: 883-905.
. 1991. Liberalism, Community and Culture. Oxford: Clarendon Press. Larmore, Charles.1987. Patterns of Moral Complexity. Cambridge: Cambridge University Press.
. 1990. "Political Liberalism." Political Theory 18: 339-60.
Lund, William R. Forthcoming. "Egalitarian Liberalism and the Fact of Pluralism." Journal of Social Philosophy.
Macedo, Stephen. 1991. Liberal Virtues: Citizenship, Virtue, and Community in Liberal Constitutionalism. Oxford: Clarendon Press.
Moon, J. Donald. 1993. Constructing Community: Moral Pluralism and Tragic Conflicts. Princeton, NJ: Princeton University Press.
Nagel, Thomas. 1986. The View From Nowhere. New York: Oxford University Press.
. 1987. "Moral Conflict and Political Legitimacy." Philosophy and Public Affairs 16: 215-40.
. 1991. Equality and Partiality. Oxford: Oxford University Press. Neal, Patrick. 1993. "Vulgar Liberalism." Political Theory 21: 623-42.
Perry, Michael.1991. Love and Power: The Role of Religion and Morality in American Politics. New York: Oxford University Press.
Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Belknap Press. . 1993. Political Liberalism. New York: Columbia University Press.
Raz, Joseph. 1986. The Morality of Freedom. Oxford: Clarendon Press. Sandel, Michael. 1982. Liberalism and the Limits of Justice. Cambridge: Cambridge University Press.
Sen, A. K.1985. "Well-Being, Agency and Freedom: The Dewey Lectures 1984." Journal of Philosophy 82: 169-221.
Taylor, Charles.1989. "Cross-Purposes: The Liberal-Communitarian Debate." In Nancy Rosenblum, ed., Liberalism and the Moral Life. Cambridge, MA: Harvard University Press.
Waldron, Jeremy. 1987. "Theoretical Foundations of Liberalism." Philosophical Quarterly 37: 127-50.
. 1989. "Autonomy and Perfectionism in Raz's Morality and Freedom." Southern California Law Review 62:1098-1152.
Wolfe, Christopher. 1994. "Liberalism and Paternalism: A Critique of Ronald Dworkin." Review of Politics 56: 615-39.
Copyright University of Utah, Political Science Dept. Sep 1996