I thank my three colleagues for their comments; in this response I will try to address the themes common to all of them.
In general, the comments helped me see I should have been more circumspect in how I presented my thesis about Kant as the “ideal of classical liberalism.” Instead of promoting Kant as a broad figure, with all of his personal and philosophical complications, I should have limited my thesis to the essence of his ethical and political theory (as I summarized them), which is where I think the value lies, especially in terms of linking him to classical liberalism.
When I think about Kant, I think in generalities, such as the basic ideas of autonomy, dignity, and duty, and the implications thereof—in particular, the judgment made necessary by the vagueness of duties and the inevitable conflicts between them, which leads to the flexibility I claim for Kant’s ethics. One can see this in the casuistical questions scatted throughout the Doctrine of Virtue, which represent a series of “but what if” questions that I can easily imagine him tossing out at his students to show how difficult moral decisionmaking is in the real world. I also focus on his essential comments about the role of government, particular the Formula of Justice that holds government to the same ethical standards to which we hold individuals, particularly to treat them with respect as persons with dignity.
I regard this general framework for ethical thinking as valuable in and of itself. As such, I don’t concern myself much about the phenomenal/noumenal distinction in Kant’s metaphysics, which Hicks emphasized in his commentary. To be sure, I completely agree with him that it’s contentious (at best). I don’t accept it, and I don’t think many modern Kantian ethicists do either, but we all believe nonetheless that Kant’s ethics are important and valuable, regardless of where freedom actually resides. As I said above, his basic positon starting with autonomy and dignity generates a system of ethics that governs the basic interactions between persons as well as between citizens and state, which can be applied to practical problems without being overly concerned about metaphysical complications such as the timelessness of the noumenal self.
However, the internal inconsistencies in Kant’s moral philosophy that Hicks and Long mention are potentially very troubling in that they cast the integrity of the core of Kant’s ethics into doubt. Both commenters acknowledge that specific ethical statements must be interpreted in light of general ones, and I would add that we must also be willing to reject specific claims if they are not consistent with general ones. The fact that Kant expressed racist and sexist ideas that conflicted directly with his writings about universal dignity for all persons should not, in my opinion, diminish the value of those profound words any more than should his metaphysical complications. Of course, we must be mindful not to cherry-pick, endorsing as integral the things we happen to like and dismissing as irrelevant those we don’t. We must hold Kant’s specific claims to the standard he established through his core principles—as Kant recommends we do with our own ethical decisionmaking—and we must also be willing to hold some claims to be mistaken in light of those principles.
In addition to the obvious examples of racism and sexism, I believe this applies to the notorious paragraph from the Doctrine of Right cited by Long (and an online commenter) that appears to endorse a system of coerced redistribution. Its tone and content are in striking contrast to Kant’s more general statements regarding the narrowly circumscribed domain of state coercion (to limit coercion by others), yet some Kant scholars have tried to interpret it within his general political philosophy, generating a provisional defense of a limited welfare state. Kant’s denial of a right of revolution also seems to be at cross-purposes with the rest of his political philosophy; it does reflect his strong belief that a state was necessary to protect rights (with which many classical liberals would agree), but cedes too much power in a tyrannical government by tying the hands of its citizens. As Roger Sullivan writes, “Kant’s discussions of civil disobedience and of the right to revolution are obviously inadequate, both in terms of his own moral theory and in terms of the judgments of ordinary moral reasoning” (emphasis mine), concluding that the only explanation can be the despotic state Kant lived in at the time.
I also believe that this identification of specific mistakes judged against general principles applies to the infamous case of the murderer-at-the-door that Salmieri mentions, in which Kant maintains (quite passionately) that you must tell the inquiring prospective murderer that his prey, your best friend, has hidden in your closet. Much ink has been spilled and many trees felled by philosophers trying to square this circle, but I think the clearer path is to judge that he applied his basic principles incorrectly in his verve to promote honesty and effective communication as a value above most others. In general, perfect duties do not automatically take precedence over imperfect duties; rather, their relative grounds of obligation, the substance behind the duties, determine which is more important. Because the specific ground of the duty of beneficence in this case is an innocent life, and the duty of not lying is owed to a person who wrongfully threatens that life, I think it is clear that, in weighing the two grounds, the former obligation prevails.
Consistency is an ideal goal, and certainly the initial presumption should be that all of a philosopher’s claims fit together. But this goal is rarely achieved, especially with systematic philosophers like Kant, and at some point we must be willing to consider that particular claims are mistaken, especially when we can defend such a judgment on the basis on the majority of his other work. While inconsistencies and contradictions are worthy subjects for philosophical inquiry and can sometimes be illuminating, there is also the risk that the bulk of a thinker’s writing that is consistent and focused can get lost (in the spirit of Oliver Wendell Holmes’ famous saying, “hard cases make bad law”). I maintain that once you separate the wheat from the chaff, the distillation of Kant’s ethical and political system is a fine statement of classical liberalism that sits well among other, more widely acknowledged advocates.
But not all of the criticisms from my colleagues hinge on inconsistencies and complications; some questioned the core of Kantian ethics that I defend as an exemplar of classical liberalism. In particular, Long and Salmieri questioned Kant’s emphasis on duty and morality to the detriment of happiness and flourishing. This is one case (among several) in which Kant’s rhetoric works against him, for in fact Kant did prize human happiness, holding it up as the ultimate good (albeit proportionate to a person’s virtue). On the level of specific acts, he stressed self-regarding duties such as to cultivate one’s talents as well as to practice beneficence to oneself; on the latter, he wrote that “lawgiving reason … permits you to be benevolent to yourself on the condition of your being benevolent to every other as well.” Furthermore, this duty must be balanced with all others using our judgment, which may favor our own happiness, even to the exclusion of specific (and perfect) duties to others, depending on the specific choice circumstances.
Because persons can generally be trusted to pursue their own happiness, it was Kant’s intent to point out the limitations of that pursuit out of respect for the autonomy and dignity of other persons. In other words, Kant did not want us to sacrifice our happiness; he just wanted to make sure we didn’t gain it wrongfully at someone else’s expense. This corresponds well to ideas such as Mill’s harm principle and the principle of noninterference, providing further support of Kant’s core ethical ideas as consistent with classical liberalism.
 For this, I draw from Ronald Dworkin’s theory of finding mistakes in judicial precedent from Taking Rights Seriously (Cambridge, MA: Harvard University Press, 1977), pp. 118-23.
 Kant, The Metaphysics of Morals, 326.
 For instance, Jeffrie Murphy discusses this passage, acknowledging its inconsistency but nonetheless developing (what he calls) a “provisional account” of a Kantian basis for positive welfare payments (Kant: The Philosophy of Right, Macon, GA: Mercer University Press, 1970, pp. 123–5). Furthermore, Arthur Ripstein and Allen Wood argue for a welfare state on based on a conception equal opportunity drawn from Kantian principles (Ripsten, Force and Freedom Kant’s Legal and Political Philosophy, Cambridge, MA: Harvard University Press, 2009, ch. 9; Wood, Kantian Ethics, Cambridge: Cambridge University Press, 2008, 196).
 Kant, Metaphysics of Morals, 320. This problem has been noted by many Kant scholars, such as Peter Nicholson, “Kant on the Duty Never to Resist the Sovereign,” Ethics 86(1975): 214-30, and Sarah Holtman, “Revolution, Contradiction, and Kantian Citizenship,” in Mark Timmons (ed.), Kant’s Metaphysics of Morals: Interpretive Essays (Oxford: Oxford University Press, 2002), pp. 209-31.
 Roger J. Sullivan, Immanuel Kant’s Moral Theory (Cambridge: Cambridge University Press, 1989), p. 245.
 Kant, “On a Supposed Right to Lie Because of Philanthropic Concerns,” included in the Hackett edition of Grounding for the Metaphysics of Morals, translated by James W. Ellington (Indianapolis, IN: Hackett Publishing Company, 1993), pp. 63–7, and originally published in the Berlin Press in 1799.
 Kant, Metaphysics of Morals, 224.
 For an investigation of this, see Stephen Engstrom, “The Concept of the Highest Good in Kant’s Moral Theory,” Philosophy and Phenomenological Research 52(1992): 747-80.
 Kant, Metaphysics of Morals, 451. As Barbara Herman writes, “the Kantian charge against self-interest is not that it is inherently contrary to morality; it need not be. The problem lies in its tendency to be presumptive” (“Making Room for Character,” in Stephen Engstrom and Jennifer Whiting (eds), Aristotle, Kant, and the Stoics: Rethinking Happiness and Duty, Cambridge: Cambridge University Press, 1996, pp. 36–60, at p. 52).