This is a fine and well-crafted essay with which I substantially disagree. My response has more to do with the second fact than the first, but I would not have the reader ignorant of my appreciation of the excellence of the work.

Professor Hostetler argues that educational policy should be based on and aim toward achieving the best conceptions of a good life. These phrases conceal an important difference. No one denies that schools should aim toward helping students lead the best lives in some sense. But many deny that public policy should be based on any particular conception of such lives. They argue that people should be free to pursue their own conception of a good life, that governmental neutrality is a prerequisite of this freedom, and that neutral schools are more likely to abet good lives as those who lead them understand their lives than is perfectionism. To rebut this view, perfectionists need to explain how they can base education on perfectionist conceptions without oppressing those who hold other views.

Hostetler believes that perfectionism need not be monistic or maximizing. He argues this by saying that both teaching and performing may be among the best kinds of lives and that perfectionism does not requires us to decide which role is really best and to maximize it. But perfectionists do not characterize good lives at the level of roles. They describe good lives by asserting views about the criteria lives should meet to be considered good. Consider an analogy: utilitarians hold that happiness is the good, but also that people find their happiness in different things. Hence utilitarians are willing that some be teachers and others performers. But occupations are not conceptions of the good, and it has not been shown that perfectionism tolerates a diversity of conceptions of the good, which is, after all, the concern of the neutralists. Moreover, utilitarians desire role diversity for reasons that are, nonetheless, monistic and maximizing.

To put this differently, different conceptions of the good, different comprehensive doctrines, can sustain more than one rational life plan because more than one life plan can satisfy the criteria a conception of the good provides. If diversity of life plans is what is required by perfectionist pluralism who would disagree? And who would care? Most religions are consistent with pluralism of life plans. How does this help secure a pluralism of comprehensive doctrines? Religious people still reject the criteria for good lives put forth by utilitarians and utilitarians reject the conception of good lives put forth by religious people. That either view will let some of us be teachers and others performers hardly resolves their disagreement. Nor would we find the fact that Catholicism is pluralistic enough to recognize the worth of various occupations to be a reason in support of establishing it as the basis of public education. The pluralism that neutralists wish is a pluralism of comprehensive doctrines, not a pluralism of rational life plans.

Hostetler claims that perfectionism may have a coherent view of freedom and equality. Freedom may be served in that it may be that only autonomously chosen views of the good are genuinely experienced as worthwhile, equality because we might emphasize an equal development of the capacity for knowing what is best. This defense of freedom and equality works only if autonomy is an essential feature of a good life. Otherwise autonomy can be traded for other goods and there is no particular reason why we should define equality in terms of how we distribute the resources for autonomous choice.

What is to be sought in schools is the development of the capacity for rational decision-making. Two points: first, here Hostetler essentially reformulates the by- now familiar dispute between ethical and political liberalism opting for a curious perfectionist variant of ethical liberalism. Ethical liberalism claims that developing the capacity to choose wisely promotes autonomy and that goods are only good when autonomously chosen. Second, it is not clear that the emphasis on developing the capacity for rational and autonomous choice gets us very far in finding a suitable conception of equality. For John Rawls, the inequality at issue in the original position is the one that results when the moral foundation of the state is sought in conceptions of the good that are objects of disagreement among reasonable people. Of course it is true that perfectionists may have a conception of equality. However, the conception Professor Hostetler offers us is an answer to a question something like "How shall we distribute the resources on which the development of reason depends?" It might be viewed as a substitute for Rawls’s second principle of justice. However, the question asked in the original position is, "Shall our conceptions of the good be equal or shall one dominate?" Professor Hostetler’s answer is not an answer to the question asked.

Moreover, the strategy of focusing on the capacities for rational choice is consistent with the kind of pluralism neutralists seek only if we can know what the capacities for choosing the best lives are apart from assuming some particular view of the good. This is doubtful. Some may think that we learn how to choose wisely by reading the Bible, others that we need to learn how to do a hedonic calculus, and still others that we need some form of self-knowledge. In short, what is to count as "the capacities for choosing the best lives" is internal to different comprehensive doctrines.

It does not immediately follow that Hostetler’s view of perfectionism is wrong. What does, I think, follow is that his attempt to find a weak form of perfectionism, that grants much of what Rawlsians want without quite falling into neutrality, is not convincing because it does not address the neutralist’s concerns. Even the most benign perfectionists must show that the conception of the good they hold is worthy of public recognition and that the fact that there are those who disagree is not an adequate reason why the(ir) truth should not prevail. Among the obstacles to such a claim from a Rawlsian viewpoint are (1) his views on the burdens of judgment, and (2) the kind of equality the original position is intended to affirm.

The burdens of judgment assert a kind of fallibalism. Rawls claims that the evidence that can be adduced for particular conceptions of the good is not adequate to allow us to conclude that those who disagree can be dismissed as either stupid or malevolent. This is crucial to reasonable pluralism and to reciprocity. Neutrality between differing conceptions of the good is central to the kind of equality the original position seeks. Among other things, it is motivated by the conviction that our conceptions of the good are also likely to be our fundamental convictions. People in the original position are unlikely to agree to any principles of justice that put their religions or personal identities at risk. This is why Rawls believes they will not be perfectionists. Again, my point in noting this is not to claim that no view of perfectionism can be sustained. It is to call attention to the fact that Hostetler does not address the concerns that motivate the Rawlsian view. The original position is not about equality in the distribution of resources for reflective choice. It is about whether your view of the good can trump mine in the selection of basic social principles.

On to dialogue: Amy Gutmann and Dennis Thompson claim that those who deny the dignity and worth of others fail the test of reciprocity. In the example taken from Mozert v Hawkins, Hostetler objects that they attack a moral conclusion, not the reasoning that leads to it. The point of reciprocity, he claims, is to take an attitude towards others that makes argument possible. However, what Gutmann and Thompson are claiming is that to deny the dignity and worth of others is itself to undercut the basis of reciprocity. Nothing in their view prevents someone from attempting to persuade those who hold such views that they are wrong. The problem is that these others have no reason to listen. The issue is not about the incommensurability of views or whether we can or should communicate intelligibly across differences as Hostetler suggests. It is that to deny the dignity and worth of those who do not share one’s faith is to place such persons beyond the need for fair treatment, dialogue, and reciprocity.

Let us consider dialogue about the good from liberal and perfectionist positions. Liberals often insist on dialogical restraint. They want dialogue to be on the basis of premises that everyone can share. This, they claim, requires that we not view our distinctive religions or comprehensive doctrines as the basis of dialogue. It is important to be clear that dialogical restraint is intended as restraint on how we talk to one another when a decision involving the use of public power is envisioned. Dialogical restraint is best understood as conveying this message: Given that it is unfair for us to justify the exercise of public power by appealing to our distinctive religions or comprehensive doctrines, let us refrain from appealing to these doctrines as premises in arguments when decisions about the use of public power is envisioned. Dialogical restraint is not reasonably viewed as forbidding argument about the good, even in public spaces such as schools, when the use of public power is not at issue. Indeed, such an understanding of dialogical restraint would be inconsistent with standard liberal institutions such as free speech. To defend a perfectionist view of dialogue against this view, it is insufficient to be in favor of dialogue about good lives. One also needs to favor dialogue which permits non-neutral premises in discussions leading to non-neutral public decisions. It may be, as Hostetler suggests, that people such as the Mozert parents will feel affirmed if they are engaged in dialogue about their view of the good. But, at the end of the day, their view of the good will be officially judged by conceptions internal to another. One supposes they will object. Perfectionist dialogue is not just dialogue about the good. It is dialogue about the uses of power that presupposes a conception of the good.

I want to conclude with a (somewhat speculative) redescription of Hostetler’s project.

1. We should seek (dialogically) for a "thick, but vague" conception of the good (with a perfectionist slant) to function as the basis of educational policy. It should be thick else it will be useless, but vague in order that it can be big-tented.

2. Ethical liberalism (with a perfectionist slant) is the best candidate for a thick but vague conception of the good. It is thick in virtue of its commitment to autonomy and rationality, but vague because it is consistent with diverse life plans.

3. The form of equality appropriate to this picture emphasizes the equal and high distribution of the resources for rational decision-making.

Hostetler’s defense of this program amounts to the claim that while it may not be altogether neutral, it is neutral enough to avoid serious oppression. But, to the neutralist, this will be a bit like defending the establishment of a state church so long as it does not produce an inquisition. At the end of the day, after we have debated what is good, perfectionists will want a collective and reasonable decision about whose view of the good is to prevail. For liberals of a Rawlsian sort, that there are dissenters is a reason why we should avoid such decisions. Society is characterized by durable pluralism. No perfectionist view will be big-tented enough to accommodate this diversity and no amount of talk will produce consensus. Perfectionists will wish to decide what the moral basis of our fundamental institutions is to be despite this durable disagreement. Liberals of a Rawlsian stripe will demand neutrality. Hostetler’s arguments tend to skirt this issue.