¹ I. Kant, Groundwork of the Metaphysics of Morals in Practical Philosophy, ed. and trans. Gregor, M. J. (Cambridge: Cambridge University Press, [1785] 1996), pp. 41–108Google Scholar. Hereafter referred to as Groundwork. All citations of Kant's works will refer to the Akademie edition volume and page numbers.

² See, for example, Kant, Groundwork, 4: 447.

³ As British philosopher Henry Sidgwick explains, arguing for this connection between freedom and morality forces Kant to deny that any free being would ever act contrary to what morality demands. That is to say, Kant's attempt to uphold the idea that all free beings are necessarily bound to the moral law forces him to characterize human freedom as an exclusively ‘good freedom’ and ultimately to maintain that a free will can do nothing but act morally. This unintentional endorsement of ‘good freedom’ thus has the consequence of prohibiting the possibility that any free agent might commit a genuine immoral act — and if evil actions cannot be attributed to a free choice of the will, then it seems there is no alternative other than that such actions must instead result from our sensuous nature having overwhelmed our capacity to act according to the moral law. See Sidgwick, , Methods of Ethics (London: Macmillan Press, [1874] 1962), pp. 512–15Google Scholar.

⁴ Kant, I., Religion within the Boundaries of Mere Reason, ed. and trans. Wood, A., Giovanni, G. Di (Cambridge: Cambridge University Press, [1792/1793] 1998)Google Scholar. Hereafter referred to as Religion. The term ‘incorporation thesis’ was first coined by Henry Allison. See Allison, , Kant's Theory of Freedom (Cambridge: Cambridge University Press, 1990), pp. 5–6CrossRefGoogle Scholar and Religion, 6: 23–4.

⁵ Kant, I., Anthropology from a Pragmatic Point of View, trans. Dowdell, V L. (Carbondale, IL: Southern Illinois University Press, [1800], 1978Google Scholar. Hereafter referred to AS Anthropology.

⁶ Hereafter referred to as DV

⁷ Kant, I., Metaphysics of Morals in Practical Philosophy ed. and trans. Gregor, M. J. (Cambridge: Cambridge University Press, [1797] 1996), pp. 353–603.Google Scholar

⁸ A discussion of Schiller's frequently cited and sarcastic concern that if one were to experience sensible enjoyment when coming to the aid of friends then he would no longer be a virtuous person can be found in Guyer, Paul, Kant and the Experience of Freedom (Cambridge: Cambridge University Press, 1993), p. 336.CrossRefGoogle Scholar

⁹ See Henson, Richard, ‘What Kant Might Have Said: Moral Worth and the Overdetermination of Dutiful Action’, The Philosophical Review 88.1 (1979), pp. 39–54Google Scholar; see especially pp. 49–51. See also Allison, Kant's Theory of Freedom, pp. 113–20, for an extensive critique of Henson's arguments.

¹⁰ My concern is to show that the moral worth of an action is not determined by the presence of countervailing inclinations; I will not discuss whether there is some other category of moral evaluation for which the existence of such inclinations might matter. In his article ‘Kant's Conception of Merit’ (Pacific Philosophical Quarterly 77.4 (1996), pp. 310–34), Robert N. Johnson argues that there is indeed such a category, namely that of moral merit. Moral worth, he claims, is determined solely by whether an agent has acted from the motive of duty, while merit is due to an agent who does both more than is expected and more than he or she can be rightfully coerced to do. Johnson argues that there are several different factors — each independent of one another and each important to varying degrees — that determine the amount of moral merit that an action deserves: (1) the presence of ‘natural obstacles’, i.e. opposing inclinations, (2) if there is a ‘moral obstacle’ to the action, i.e. a significant incentive to fulfil the duty, such as when the action is legally coercible, and (3) whether the action has been performed from the motive of duty. Whether there is, as Johnson argues, a conception of moral merit in Kant along the lines he sketches need not be investigated here, as we both agree on those points relevant to this paper, namely that the moral worth of an action is determined only by whether it is done from the motive of duty and that the presence of countervailing inclinations in an agent has no bearing on this. For Johnson's own thoughts on this last point, see especially ‘Kant's Conception of Merit’, pp. 324–5.

¹¹ Kant, DY 6: 399^*00.

¹² I will discuss in greater details Kant's definitions of the terms, instinct, inclination and desire on pages 134–5 above.

¹³ For an interpretation of sensibility as a ‘product’, see Guyer, Paul, Kant on Freedom, Law, and Happiness (Cambridge: Cambridge University Press, 2000), pp. 297–303CrossRefGoogle Scholar, and Guyer, Kant and the Experience of Freedom, pp. 361–5.

¹⁴ Kant, Religion, 6: 29n.

¹⁵ Kant, Anthropology, 7: 265.

¹⁶ Since I take the terms ‘instinct’, ‘inclination’ and ‘passion’ to refer to states of desire which differ only in degree (and not in kind), I will use them interchangeably and as synonymous with desire in general.

¹⁷ Although Kant sometimes appears to suggest in both the Religion and the Anthropology (e.g. 7: 265–6) that our passions could grow so extreme that they might end up overriding our attempts to act morally, any such interpretation would obviously contradict Kant's incorporation thesis. We should instead understand such claims as mere hyperboles which are meant only to demonstrate that passions can indeed become quite entrenched.

¹⁸ I will address the question concerning to what degree we can influence the content of our desires in section 2 of this paper.

¹⁹ Kant, Religion, 6: 58.

²⁰ Kant, Anthropology, 1: 266.

²¹ Kant, DV, 6: 392.

²² Ibid., 6: 386–7.

²³ Kant, DV, 6: 387. On the point that Kant's concept of self-perfection necessarily includes perfection of our sensible capacities, see Paul Guyer, ‘Moral Anthropology in Kant's Aesthetics and Ethics: A Reply to Ameriks and Sherman’, Philosophy and Phenomenological Research, 55.2 (1995), 379–92; see especially pp. 386–7.

²⁴ On this point, see Guyer, ‘Moral Anthropology in Kant's Aesthetics and Ethics: A Reply to Ameriks and Sherman’, p. 390.

²⁵ This general argument can be found in Guyer, Kant and the Experience of Freedom, pp. 378–80.

²⁶ See, for example, Religion, 6: 21.

²⁷ Thanks to Professor Guyer for comments on an earlier version of this paper that pointed out to me the relevance of Kant's claim concerning the inscrutability of the decision to take the moral law as one's motive.

²⁸ Although it should be noted that Kant's statement that this duty is narrow conflicts with the claim made in the Introduction to DV that all duties of virtue are of wide obligation (i.e. indeterminate in so far as they simply identify a range of permissible actions and never oblige us to act in any specific instance), the distinction between narrow and wide duties and how it holds up in DV lies outside the scope of this paper.

²⁹ Kant, , Lecture on Ethics, ed. and trans, p. Heath and Schneewind, J. B. (Cambridge: Cambridge University Press, 1997)CrossRefGoogle Scholar; quote from Moral Collins [1794–5], 27: 351.

³⁰ Kant, DV, 6:401.

³¹ Kant, DV, 6: 452.

³² See Guyer, Kant and the Experience of Freedom, pp. 382–9; Allison, Kant's Theory of Freedom, pp. 118–20; Sherman, Nancy, Making a Necessity of Virtue: Aristotle and Kant on Virtue (Cambridge: Cambridge University Press, 1997), pp. 145–9CrossRefGoogle Scholar; and Baron, Marcia, Kantian Ethics Almost without Apology (Ithaca: Cornell University Press, 1995), p. 220.Google Scholar

³³ In stressing the merely instrumental value of feelings such as love and sympathy, I am here trying to distance my view from any interpretation of Kant whereby it is morally permissible to be motivated by one's desires or inclinations. Such an interpretation is offered by Barbara Herman, who argues that inclinations or feelings can serve as one's ‘primary motive’ for action so long as the moral law functions as a ‘secondary motive’ that puts a check on acting from feelings in those cases where they fail to conform to the demands of morality. As Guyer notes, however, the difficulty here is that on Herman's view inclinations are understood not as ‘products’ but instead as passive affections on which neither one's overall respect for the moral law nor the manner in which one engages her predispositions have any bearing. For this point and a thorough critique of Herman's position, see Guyer, Kant on Freedom, Law, and Happiness, pp. 291–303, especially pp. 298–9. Herman's distinction between primary and secondary motives is in ‘On the Value of Acting from the Motive of Duty’, Philosophical Review 90 (1981), pp. 359–82CrossRefGoogle Scholar, which also appears in revised form as chapter 1 of The Practice of Moral Judgment (Cambridge: Harvard University Press, 1993).Google Scholar

³⁴ Kant, DV, 6: 394.

³⁵ See footnote 10.

³⁶ See page 132 above.

³⁷ Thanks to the two anonymous referees from Kantian Review, the audience at the 2nd Annual UK Kant Society Graduate Conference, and Professors Paul Guyer and Kenneth Westphal for helpful comments on earlier versions of this paper.