¹ I am using Gregor's, Mary translations of the Groundwork of the Metaphysics of Morals (Cambridge: Cambridge University Press, 1997)Google Scholar, the Critique of Practical Reason (Cambridge: Cambridge University Press, 1997)Google Scholar and the Metaphysics of Morals (Cambridge: Cambridge University Press, 1996)Google Scholar, the second half of which is the Doctrine of Virtue. References will be to the Prussian Academy pagination, and inserted into the text. I will abbreviate the Groundwork as G, the Critique of Practical Reason as C2, and the Metaphysics of Morals as MdS.

² For example, H. J. Paton argues that FHE ‘may be said to enjoin respect for personality as such’, and that ‘[s]trictly speaking, this formula, like all others, should cover rational beings as such; but since the only rational beings with whom we are acquainted are men, we are bidden to respect men as men, or men as rational beings’. , Paton, The Categorical Imperative (New York: Harper & Row, 1965), p. 165Google Scholar .

³ Cranor, Carl F. stops just short of drawing this conclusion in ‘Kant's respect-for-persons principle’, International Studies in Philosophy, 12 (1980), 19–40CrossRefGoogle Scholar, holding that ‘[sjomeone might be able to argue for this claim’ – that is, that FHE can be understood, on Kant's terms, as a principle enjoining respect for persons – ‘but an argument is needed’ (p. 31). I attempt such an argument below.

⁴ On which see Hill, Thomas E. Jr, ‘Must respect be earned?’, in Respect, Pluralism, and Justice: Kantian Perspectives (Oxford: Oxford University Press, 2000), pp. 87–118CrossRefGoogle Scholar .

⁵ The distinction implicit in these passages tracks the distinction Darwall, Stephen draws between recognition-respect and appraisal-respect in ‘Two kinds of respect’, Ethics, 88 (1977), 36–49CrossRefGoogle Scholar in many but not all respects, as Baron, Marcia notes in ‘Love and respect in the Doctrine of Virtue’, in Timmons, Mark (ed.), Kant's Metaphysics of Morals: Interpretive Essays (Oxford: Oxford University Press, 2002), pp. 397–8Google Scholar . Rather than risk confusion, I have adopted the (admittedly awkward) vocabulary above. Hill draws a similar distinction, between what he calls ‘respect for a person's position’ and ‘respect for a person's merit’, in Hill, ‘Must respect be earned?’.

⁶ Kant implicitly invokes one more distinction in the taxonomy of respect for persons in the Doctrine of Virtue. At the end of his treatment of the duties of respect he tells us that

[t]he different forms of respect to be shown to others in accordance with differences in their qualities or contingent relations – differences of age, sex, birth, strength, or weakness, or even rank and dignity, which depend in part on arbitrary arrangements – cannot be set forth in detail and classified in the metaphysical first principles of a doctrine of virtue, since this has to do only with its pure rational principles. (MdS 468)

This describes several instances of a kind of respect akin to what I called respect-as. The respect owed someone owing to her age or rank, is, like respect-as, respect owed her owing to her status, rather than what she has revealed about herself through her actions. The respect owed a person owing to her age or rank is, however, only akin to respect-as because the relevant status to which such respect attaches is not her being a person, but rather her being a particular kind of person owing to her ‘qualities or contingent relations’. I will set this aside because Kant did not regard the discriminations relevant to this kind of respect to be part of a metaphysics of morals. They are, we might say, a matter of etiquette rather than of morality proper.

⁷ See, in particular, Book I, Chapter III of the Critique of Practical Reason, On the Incentives of Pure Practical Reason, C2,71-89.

⁸ Thus Kant says in the second Critique that ‘respect for the law is not the incentive to morality; instead it is morality itself subjectively considered as an incentive’ (C2,76).

⁹ I borrow this characterization from Reath's, Andrews very helpful discussion in ‘Kant's theory of moral sensibility’, Kant-Studien, 80 (1989), 293CrossRefGoogle Scholar .

¹⁰ It is of course a matter of controversy just what feature or features of humanity confer upon it absolute value, and just how its having this value entails that we may not use humanity in ourselves or in others as a mere means. But I will set this to the side here. The answer requires reconstructing the argument for FHE, a task beyond the scope of this article.

¹¹ I will consider what it means to respect the humanity in a person below. For now I only want to show how we can get to the view that FHE enjoins us to respect persons.

¹² See also MdS, 434-5.

¹³ As O'Neill, Onora emphasizes in ‘Universal laws and ends-in-themselves’, Constructions of Reason (Cambridge: Cambridge University Press, 1989), pp. 138–9Google Scholar .

¹⁴ See Herman, Barbara, ‘Mutual aid and respect for persons’, Ethics, 94 (1984), 577–602CrossRefGoogle Scholar, reprinted in , Herman, The Practice of Moral Judgment (Cambridge: Harvard University Press, 1993), pp. 45–72Google Scholar .

¹⁵ The word rendered ‘agreement’ by Gregor (Übereinstimmung) is sometimes rendered ‘harmony’. Robert Binldey suggested to me that this nicely captures the negative/positive contrast and its mapping onto the distinction between perfect and imperfect duties, as follows. We might say I am in negative harmony with a melody if I do not sound a dissonant note – something I may do by keeping silent – whereas I am in positive harmony with it if I sound a note that supports it, that is, that harmonizes with it in the sense in which we usually use the word in the musical context.

¹⁶ , Reath, ‘Kant's theory’, 293Google Scholar .

¹⁷ See the contrast Kant draws between temperament and character in Anthropology from a Pragmatic Point of View, trans. Dowdell, V. L. and Rudnick, H. H. (Carbondale and Edwardsville, IL: Southern Illinois University Press, 1996), part 2, section AGoogle Scholar .

¹⁸ Baron, Marcia, Kantian Ethics almost without Apology (Ithaca, NY: Cornell University Press, 1995), pp. 44–5Google Scholar .

¹⁹ See C2,159, and Religion within the Limits of Mere Reason, trans. , Wood and , Giovanni (Cambridge: Cambridge University Press, 1998), 6: 50Google Scholar .

²⁰ Kant does not quite say ‘ought’ here. In this passage he says instead that reason represents the ambiguous cases as products of cultivation (C2,78). Earlier he suggests that the benefit of the doubt is granted just because I know my own weaknesses better than others (C2, 77). But it seems to me that he means to say that morality requires us to adopt this disposition.

²¹ See too G, 401n., where Kant argues that ‘[bjecause we also regard enlarging our talents as a duty, we represent a person of talents also as, so to speak, an example of the law (to become like him in this by practice), and this is what constitutes our respect’.

²² See Guyer, Paul, Kant on Freedom, Law, and Happiness (Cambridge: Cambridge University Press, 2000), pp. 303–23CrossRefGoogle Scholar, for a very fine-grained treatment of these puzzles.

²³ This on the assumption that all duties are either duties of right or duties of virtue. Kant makes only one exception explicitly, namely the duty to adopt the maxim ‘that an action in conformity with duty must also be done from duty’ (MdS, 383). But this is curious, because this is just the duty to increase one's own moral perfection which later appears as a duty of virtue to oneself (MdS, 446–7). The more plausible candidate exceptional category is found within the perfect duties owed to others. The enforcement of many such duties is outside the state's purview (for example, broken promises not actionable under the law of contract), yet they are not a matter of ‘wide’ obligation, as Kant at one point characterizes all duties of virtue (MdS, 390). On this question, and on the taxonomy of duties in Kant's practical philosophy generally, see Gregor, Mary, Laws of Freedom (Oxford: Basil Blackwell, 1963), pp. 95–112Google Scholar ; Nell, Onora, Acting on Principle (New York: Columbia University Press, 1975), pp. 43–58Google Scholar ; and Schaller, Walter E., ‘From the Groundwork to the Metaphysics of Morals: what happened to morality in Kant's theory of justice?’, History of Philosophy Quarterly, 12 (1995), 333–5Google Scholar .

²⁴ There is more still contained in the claim that duties of right are duties for which external constraint is possible, but we need not attend to the further details here. I attempt a more comprehensive unpacking of the nuances of ‘possibility’ here in ‘Necessity, deterrence and standing’, Legal Theory, 8 (2002), 339–58Google Scholar .

²⁵ In the Introduction to the Doctrine of Virtue Kant tells us that all duties of virtue are duties to adopt one or the other of the ends that are also duties, namely our own perfection and the happiness of others. But it is not at all obvious that the duties of respect can be reckoned instances of either. The latter duty to aim at the happiness of others seems the more likely. But I can be said to promote your happiness by, for example, refraining from ridiculing you only in the attenuated sense that in so refraining I do not detract from your happiness. So perhaps the point is that because self-constraint is the only kind of constraint possible for duties of respect, in discharging those duties I contribute to my moral perfection. But then this makes all duties of virtue into duties to oneself. I will set this puzzle to the side here.

²⁶ Conversely, ‘no one is wronged if duties of love are neglected; but a failure in the duty of respect infringes upon one's lawful claim’ (MdS, 464).

²⁷ It is obvious, then, why duties of love are duties of virtue. Duties of love issue from the duty to adopt others' happiness as an end. Thus they are duties for which external compulsion is impossible in the first sense of possibility. They might also fail the second condition of possibility, but the question need not be raised. Now, duties of respect are plainly duties for which external compulsion is impossible in the second sense: it is not n i the state's purview, for example, to rein in my arrogance. But it is not clear whether duties of respect are duties for which external compulsion is impossible only in the second sense. If so, one may discharge a duty of respect (simply) by acting in the right way, regardless of one's motivations. I might refrain from ridiculing others, for example, just because I want to keep my reputation intact. We might take Kant to be suggesting as much when he says that ‘[a]t times one cannot, it is true, help inwardly looking down on some in comparison with others … but the outward manifestation of this is, nevertheless, an offence’ {MdS, 463). This ‘offence’, it seems to me, is the breach of a duty of respect, which, then, seems to be satisfied by a (mere) omission. But if this is right, we have the odd consequence that one might discharge a duty of virtue with base or at least amoral motives. On this question and other issues in Kant's treatment of the duties of love and respect I have not raised here see , Baron, ‘Love and respect in the Doctrine of Virtue’, pp. 391–407Google Scholar .

²⁸ Another example is punishment. The state has a duty to punish criminal wrongdoers {MdS, 331-2), but respect for persons forbids ‘disgraceful punishments’ such as quartering (MdS, 463), (or so I read the passage at MdS, 463, which is rather ambiguous).

²⁹ Noggle, Robert, ‘Kantian respect and particular persons’, Canadian Journal of Philosophy, 29 (1999), 455CrossRefGoogle Scholar .

³⁰ Neuman, Michael, ‘Did Kant respect persons?’, Res Publica, 6 (2000), 299Google Scholar .

³¹ This is my disaggregation, not Noggle's or Neuman's. I think it fairly respects their common concerns. Each raise other worries, which I will not consider here.

³² , Noggle, ‘Kantian respect’, pp. 454–5Google Scholar .

³³ Neuman similarly allows that the duty of beneficence requires us to consider the desires of others – and so that respect for persons is at least i n part for persons as we ordinarily understand ourselves – but dismisses the significance of this owing to the fact that the duty of beneficence is, among other things, imperfect, and so ‘not even a true moral obligation’ ( , Neuman, ‘Did Kant respect persons?’, p. 298)Google Scholar . It is not even a true moral obligation, Neuman holds, because Kant allows that the failure to discharge it registers only as a failure to do what is praiseworthy. But Kant means this to apply to a particular given case of a foregone beneficent action. ‘Fulfillment of [imperfect duties] is merit (meritum) = +a; but failure to fulfill them is not in itself culpability (demeritum) = -a, but rather mere deficiency in moral worth – 0', Kant tells us, ‘unless the subject should make it his principle not to comply with such duties’ (MdS, 390; final emphasis mine). The obligation to adopt the end of others' happiness is categorical, no less ‘true’ than the obligation to discharge a perfect duty.

³⁴ Kant means this without qualification, I think. While I may not promote others' impermissible ends {MdS, 388, 450), Kant holds that nonetheless I cannot deny all respect to even a vicious man as a human being; ‘I cannot withdraw at least the respect that belongs to his quality as a human being, even though by his deeds he makes himself unworthy of it’ (MdS, 463). This raises important and unanswered questions, for example, how, exactly, an agent can make herself unworthy of respect and why it nonetheless cannot be withheld from her. I will set these aside here.

³⁵ I add the qualification ‘political’ to distinguish this sense of ‘the priority of the right to the good’ from the doctrine, sometimes also referred to under that name, that ‘the concept of good and evil must not be determined before the moral law (for which, as it would seem, this conception would have to be made the basis) but only … after it and by means of it’ (C2, 63; italics removed), from which Kant derives the idea of the highest good.