Published online by Cambridge University Press: 08 January 2010
Kant takes autonomy to be recognizably valuable. In claiming that non-Kantian views of morality treat the morally good will as heteronomous, he intends to present an objection to these views. He expects proponents of these views to recognize that the implication of heteronomy is a serious objection; his task is not to convince them that heteronomy is bad, but to convince them that their views imply heteronomy.
1 ‘The freedom of the Willkür is this independence from sensuous impulse in the determination of the Willkür. This is the negative concept of freedom. The positive concept of freedom is that of the power of pure reason to be of itself practical. But pure reason can be practical only if the maxim of every action is subjected to the condition that it qualifies as a universal law.’ (MdS 213–14)
2 Hereafter I will use ‘will’ or ‘rational will’ to indicate Wille, and ‘choice’ or ‘elective will’ for Willkür. Kant also introduces a negative and a positive conception of freedom at G 446. But here these seem to be two ways of conceiving the same condition, rather than two distinct conditions; at least Kant does not insist on their distinctness as sharply as he needs to.
3 I am assuming some controversial points that need further discussion. In particular, I assume that it is possible for a rational will (Wille) to be heteronomous, by failing to give itself for the law; in these cases the matter of volition is a determining ground of the Wille. See KpV 36, 39, 41. I doubt whether this view can be reconciled with Groundwork, ch. 3, but I take it to be the most plausible account of Kant's position in KpV and MdS. On this point, I generally agree with the position of Allison, Henry, Kant's Theory of Freedom (Cambridge: CUP, 1990), 94–106.CrossRefGoogle Scholar
4 … freedom of choice is of a wholly unique nature in that an incentive can determine choice to an action only so far as the individual has incorporated it into his maxim (has made it the general rule in accordance with which he will conduct himself); only thus can an incentive, whatever it may be, co-exist with the absolute spontaneity of choice (i.e. freedom). (Religion, 23f)
5 ‘Our age is the genuine age of criticism, to which everything must submit. Religion through its holiness and legislation through its majesty commonly seek to exempt themselves from it. But in this way they excite a just suspicion against themselves, and cannot lay claim to that unfeigned respect that reason grants only to that which has been able to withstand its free and public examination.’ (KrV Axi)
6 See Practical Philosophy (ed. Gregor), 17.
7 ‘If a man be in some cases obliged to prefer his own judgment, he is in all cases obliged to consult that judgment, before he can determine whether the matter in question be of the sort provided for or no. So that from this reasoning it ultimately appears, that the conviction of a man's individual understanding is the only legitimate principle imposing on him the duty of adopting any species of conduct.’ (Godwin, Enquiry concerning Political Justice, ii 6 (ed. Priestley, 181).)
8 Beck, L. W., Commentary on Kant's Critique of Practical Reason (University of Chicago Press, 1960)Google Scholar, 200 cites Rousseau, Social Contract, i 8: ‘… moral freedom, which alone makes a person truly master of himself. For the impulse of appetite alone is slavery, and obedience to the law that one has prescribed to oneself is freedom.’
9 It is convenient to call this position ‘Kantian constructivism’. To be more precise, I have in mind the moral constructivism that Rawls ascribes to Kant at Political Liberalism (NY: Columbia UP, 1993), 99–101.Google ScholarPubMed This is distinct from the political constructivism that Rawls defends.
10 In discussing Kantian constructivism as Rawls understands it, I am not taking a position on other views that have been called ‘constructivism’ and are inspired by Kant. One such view is described by O'Neill, Onora, Constructions of Reason (Cambridge: CUP, 1989)Google Scholar, 188n: ‘I use the term [sc. ‘constructivist’] to cover approaches that seek to justify ethical principles by reference to an account of agency and rationality, without relying on claims about desires or preferences.’ It is not clear, without further argument, whether a ‘constructivist’ in this sense must be a metaphysical constructivist (in the sense that excludes moral realism). Similarly, Hill, , Dignity and Practical Reason (Ithaca: Cornell UP, 1992), 231Google Scholar, has a doctrine of method in mind. He identifies Kantian constructivism with Rawls's use of the original position: ‘The choosers are not seen as seeking to discover a moral order, Platonic, natural, or divine, which exists independently of their reasoned choices; rather, we are to view principles as justifiable by virtue of their being what persons with the specified values would choose in the defined situation.’ This is a description of a constructivist method, without any explicit commitment to a constructivist metaphysics; one might agree with Hill that the people in the original position are not conceived as metaphysical realists about morality, while still maintaining that this constructive method is correct because it matches moral truths.
11 ‘Yet in Kant's moral constructivism it suffices for heteronomy that first principles obtain in virtue of relations among objects the nature of which is not affected or determined by our conception of ourselves as reasonable and rational persons (as possessing the powers of practical reason), and of the public role of moral principles in a society of such persons. Of particular importance is the conception of persons as reasonable and rational, and, therefore, as free and equal, and the basic units of agency and responsibility. Kant's idea of autonomy implies that there exists no moral order prior to and independent of these conceptions that is to determine the form of the procedure that specifies the content of first principles of right and justice among free and equal persons. Heteronomy obtains not only when these first principles are fixed by the special psychological constitution of human nature, as in Hume, but also when they are fixed by an order of universals, or of moral values grasped by rational intuition, as in Plato's realm of forms or in Leibniz's hierarchy of perfections’. (Rawls, ‘Themes in Kant's moral philosophy’, in Collected Papers, ed. Freeman, S. [Cambridge, Mass.: Harvard UP, 1999], 512.Google Scholar) A similar statement appears in ‘Kantian constructivism’, in Papers 345, and in Lectures on the History of Moral Philosophy (Cambridge, Mass.: Harvard UP, 2000), 236.Google Scholar It is clarified at Lectures 72, 229.
12 Further comments by Rawls clarify his view: ‘In contrast with rational intuitionism, constitutive autonomy says that the so-called independent order of values does not constitute itself but is constituted by the activity, actual or ideal, of practical (human) reason itself. I believe this, or something like it, is Kant's view.’ (Political Liberalism, 99.) ‘Rational intuitionism says: the procedure is correct because following it correctly usually gives the correct (independently given) result. Constructivism says: the result is correct because it issues from the correct reasonable and rational procedure correctly followed.’ (Lectures 242.)
13 The argument from autonomy is not the only argument for taking Kant to be a constructivist. Hence, even if I am right about this argument, it does not follow that Kant is not a constructivist about morality. But it will be reasonable to re-examine other arguments for taking him to be a constructivist.
14 See Schneewind, J. B., ‘Autonomy, obligation, and virtue’, in Cambridge Companion to Kant, ed. Guyer, P. (Cambridge: CUP, 1992)Google Scholar, ch. 10, at 316: ‘Suppose that a kind of state of affairs is intrinsically good because of the nature of that state of affairs. Then the goodness occurs independently of the will of any finite moral agent, and if she must will to pursue it, she is not self-legislating. Suppose the goodness of states of affairs comes from some standard. Then the standard is either the outcome of someone's will—say God's—or it is self-subsistent and eternal. In either case conformity to it is not autonomy. Conformity would be what Kant calls heteronomy.’ At 315 Schneewind seems to impose less extreme anti-objectivist conditions for self-legislation.
In ‘The use of autonomy in ethical theory’ (in Reconstructing Individualism, Heller, T. C., Sosna, M., and Wellberry, D. E. (eds.) [Stanford: Stanford UP, 1986], 64–75.Google Scholar), 66, Schneewind contrasts Kant with Butler: ‘The defining feature of an autonomous agent, in Kant's view, is its ability to guide its own action by the choice of a will that is such that whatever it wills is good simply because it is willed by it. The point is not that the autonomous will unerringly hones in on what is independently and antecedently good, as Butler's conscience does. The point is rather that when something is chosen or pursued by such a will, that very fact makes the object of the will good. An agent so guided is not led by anything outside himself.’ Schneewind goes on to compare and contrast Kant with Pufendorf.
15 I use ‘objectivist’ to refer to the view that Rawls rejects on Kant's behalf. ‘Realist’ is often used in the same sense. Rawls notes (Lectures, 243–7) that the constructivism he ascribes to Kant also allows a conception of objectivity.
16 ‘In natural science the principles of what occurs … are at the same time laws of nature, for there the use of reason is theoretical and determined by the constitution of the object. In practical philosophy, which has to do only with the grounds of determination of the will, the principles that one makes for oneself are not yet laws by which one is inexorably bound, because reason, in practice, has to do with a subject and especially with its faculty of desire, the special character of which may occasion variety in the rule … This rule, … is an imperative for a being whose reason is not the sole determinant of the will. It is a rule characterized by ‘ought’, which expresses the objective necessitation of the act and indicates that, if reason completely determined the will, the action would without exception
17 ‘If reason infallibly determines the will, then the actions of such a being which are recognized as objectively necessary are subjectively necessary also, i.e., the will is a faculty to choose that only which reason independent of inclination recognizes as practically necessary, i.e., as good. But if reason of itself does not sufficiently determine the will, if the latter is subject also to subjective conditions (particular incentives) which do not always coincide with the objective conditions; in a word, if the will does not in itself completely accord with reason (which is actually the case with human beings), then the actions which objectively are recognized as necessary are subjectively contingent, and the determination of such a will according to objective laws is obligation, that is to say, the relation of the objective laws to a will that is not thoroughly good is conceived as the determination of the will of a rational being by principles of reason but which the will from its nature does not of necessity follow. The conception of an objective principle, in so far as it is obligatory for a will, is called a command (of reason), and the formula of the command is called an imperative. All imperatives are expressed by an ought, and thereby indicate the relation of an objective law of reason to a will that from its subjective constitution is not necessarily determined by it (an obligation). They say that something would be good to do or to forbear, but they say it to a will which does not always do a thing because it is conceived to be good to do it.’ (G 412f) ‘Now this principle of morality, … includes the Infinite Being as the supreme intelligence. [In finite beings], …the moral law is an imperative that commands categorically, because the law is unconditioned; the relation of such a will to this law is dependence under the name of obligation, which signifies a necessitation to an action, though only by reason and its objective law; and this action is called duty, because a choice subject to pathological affections (though not determined by them, and, therefore, still free), implies a wish that arises from subjective causes and therefore may often be opposed to the pure objective determining principle; whence it requires the moral necessitation of a resistance of the practical reason, which may be called an internal but intellectual, constraint. In the supreme intelligence choice (Willkür) is rightly conceived as incapable of any maxim which could not at the same time be objectively a law; and the notion of holiness, which on that account belongs to it, places it, not indeed above all practical laws, but above all practically restrictive laws, and consequently above obligation and duty. This holiness of will is, however, a practical idea, which must necessarily serve as a type to which finite rational beings can only approximate indefinitely, and which the pure moral law, which is itself on this account called holy, constantly and rightly holds before their eyes.’ (KpV 32–4) ‘That will whose maxims are necessarily in accord with the laws of autonomy is a holy, or absolutely good, will. The dependence upon the principle of autonomy of a will that is not absolutely good (i.e., moral necessitation) is obligation, which cannot therefore be applied to a holy being. The objective necessity of an action from obligation is called duty.’ (G 439) ‘An imperative differs from a practical law in that a law indeed represents an action as necessary but takes no account of whether this action already inheres by an inner necessity in the acting subject (as in a holy being) or whether it is contingent (as in the human being); for where the former is the case there is no imperative. … A categorical imperative, because it asserts an obligation with respect to certain actions, is a morally practical law. But since obligation involves not merely practical necessity (such as a law in general asserts) but also necessitation, a categorical imperative is a law that either commands or prohibits …’ (MdS 222f)
18 The use of ‘without’ over-simplifies Suarez's position. A few passages indicating Suarez's position: ‘This will of God, prohibition or prescription, is not the whole character of the goodness and badness that is present in the observance or transgression of natural law, but it assumes in the actions themselves some necessary rightness or wrongness, and joins to them a special obligation of divine law.’ (De Legibus, ii 6.11) ‘In this opinion, I take to be true the teaching that it assumes in its foundation about the intrinsic rightness or wrongness of actions, by which they fall under the natural law that forbids or prescribes …’ (ii 5.5). I hope to discuss Suarez's position more fully elsewhere.
19 ‘… this legislation of its own by pure and thus practical reason is freedom in the positive sense.’ (KpV 33).
20 Andrews Reath discusses many of the relevant issues helpfully in ‘Legislating the moral law’ Nous 28 (1994), 435–64Google Scholar, and in ‘Legislating for a realm of ends’, in Reath, A., Herman, B., and Korsgaard, C, eds.. Reclaiming the History of Ethics. (Cambridge: CUP, 1997), 214–39.Google Scholar At some places he seems to suggest that one gives the law to oneself simply in so far as the law is the law of the will. See ‘Legislating’ 456: ‘The fundamental law regulating moral deliberation is a principle derived from the nature of rational volition; it is thus the law which the rational will gives to itself.’ In that case, the will need not choose what the content of the law will be.
21 ‘A law (a morally-practical one) is a proposition which contains a categorical imperative (a command). He who commands (imperans) through a law is the lawgiver (legislator). He is the author (auctor) of the obligation (Verbindlichkeit) in accordance with the law, but he is not always the author of the law. If he were so, the law would be positive (contingent) and arbitrary (willkürlich).’ (MdS 227)
22 These resolutions are preferable to those that would abandon the division between holy and imperfect wills; for Kant insists that the moral law is not only relevant to the distinctively human aspects of human rational wills. We would do serious damage to his whole position by abandoning this element in it. The preferable resolutions do not involve such serious damage.
23 Rousseau draws a somewhat analogous contrast between the legislator (corresponding to Kant's ‘author of the law’) and the person or body who adopts the laws for a given state. See Rousseau, , Social Contract ii 7. Cf. R. Grimsley's edition (Oxford: OUR 1972), 30.Google Scholar
24 As I mentioned earlier, this is not how Suarez puts it, since he disagrees with Kant about necessary conditions for law. I mean only that Kant's division captures the substance of Suarez's distinction in other terms.
25 ‘The law which obliges (verbindet) us a priori and unconditionally through our own reason can also be expressed as proceeding from the will of a supreme lawgiver, i.e. of one who has only rights and no duties (accordingly, from the Divine Will). But this only signifies the idea of a moral being whose will is law for all, without his being conceived as the author of the law.’ (MdS 227.)
26 Wood, Allen, Kant's Ethical Theory (Cambridge: CUP 1999)CrossRefGoogle Scholar, 161, takes this passage to show that the rational will must be the author of the moral law: ‘He distinguishes the legislator of a law, the one who issues a command and may attach positive or negative sanctions to it, from the law's author, the one whose will imposes the obligation to obey it. In these terms Kant has no objection to regarding God's will as the legislator of the moral law, but thinks only the rational will of the person obligated can be its author.’ The passage, however, offers no support for ‘but thinks only …’. If Wood were correct, it would imply that the law by which we are obliged is ‘positive and arbitrary’.
As further evidence of Kant's holding the view he ascribes to him, Wood cites two other passages: (1) G 448: ‘Reason must regard itself as the author (Urheberin) of its principles, independently of alien influences.’ It is not clear that being the author of its principles implies being the author of the moral law that its principles embody. (2) Rel. 99: ‘But neither can ethical laws be thought of as proceeding originally merely from the will of this superior (as statutes that would not be binding without his prior sanction), for then they would not be ethical laws, and the duty commensurate to them would not be a free virtue, but an entirely enforceable legal duty. Therefore only such a one can be thought of as the supreme lawgiver of an ethical community, with respect to whom all true duties, hence also the ethical, must be represented as at the same time his commands.’ This passage says nothing about the will of the person obliged being the author of the law. On the contrary, it suggests that ethical laws do not proceed from anyone's will. Kant's remark about the supreme lawgiver suggests that something's already being a true duty is presupposed by its being one of his commands.
27 ‘The will is thus not merely subject to the law, but is subject to the law in such a way that it must be regarded also as legislating for itself and only on this account as being subject to the law (of which it can regard itself as the author).’ (G 431)
28 ‘Pure reason is of itself alone practical, and gives (to the human being) a universal law, which we call the moral law. … Now this principle of morality, just on account of the universality of the lawgiving that makes it the formal supreme determining ground of the will regardless of all subjective differences, reason declares (erklärt) to be at the same time a law for all rational beings in so far as they have a will, i.e., faculty of determining their causality through the conception of a rule, and consequently in so far as they are competent to determine their actions according to principles and thus to act according to practical a priori principles, which alone have the necessity which reason demands in a principle. It is thus not limited to human beings, but extends to all finite beings having reason and will; indeed it includes the Infinite Being as the supreme intelligence. In the former case, however, the law has the form of an imperative.’ (KpV 32)
29 ‘Those external laws whose obligation can be recognized a priori by reason even without external legislation are natural laws; those, on the other hand, which without actual external legislation do not bind (and so without it would not be laws) are called positive laws. Hence it is possible to conceive of an external legislation which contains only positive laws; but then it would have to be preceded by a natural law providing the ground of the authority of the legislator (i.e., his authorization to obligate others by his mere choice).’ (MdS 224)
30 See Wood, intro. to Practical Philosophy, xxii: ‘The principle of autonomy is also a principle of enlightenment, because it locates the source of moral legislation in the reason of human individuals who think for themselves, locating the ultimate criterion of morality in "the moral judgment of every human being in so far as he makes the effort to think [the moral law] clearly" (A807/B835).’
31 See §2 above.
32 ‘… The concept of duty cannot be derived from it (sc. a feeling of satisfaction), for we would have to presuppose a feeling for law as such and regard as an object of sensation what can only be thought by reason. If this did not end up in the flattest contradiction, it would destroy every concept of duty and fill its place with a merely mechanical play of refined inclinations, sometimes contending with the coarser.’ (KpV 38–9) In the last sentence, the apodosis (‘it would destroy …’) seems to refer to an attempt to offer a non-circular sentimentalist account.
33 ‘… the appeal to the principle of moral feeling is superficial, since men who cannot think believe that they will be helped out by feeling, even when the question is solely one of universal law. They do so even though feelings naturally differ from one another by an infinity of degrees, so that feelings are not capable of providing a uniform measure of good and evil; furthermore, they do so even though one man cannot by his feeling judge validly at all for other men.’ (G 442)
34 See Rawls, Lectures 228–30, 235–7.
35 Rawls mentions Clarke at ‘Kantian constructivism’, in Papers, 343, and discusses him in Lectures, 77–81.
36 See Clarke's, Discourse of Natural Religion, in British Moralists, ed. Raphael, D. D. (2 vols, Oxford: OUP, 1969)Google Scholar, § 225f.
37 Quoted more fully above at n. 29.
38 An abbreviated quotation from the passage quoted in full at n. 12 above.
39 I will set aside the difficult issues that arise in deciding whether Rawls's own theory meets this test for being genuinely constructivist