20th WCP: Kant's Moral Constructivism and his Conception of Legislation

Introduction

The contemporary Kantianism of John Rawls and some of his followers includes a commitment to "moral constructivism", a metaethical conception that is proposed as an interpretation of Kant and as a substantive position in its own right. Today I will discuss an important strand in Kant’s thought that challenges this constructivist interpretation of his moral philosophy. First I will provide a brief overview of the constructivist interpretation of Kant, starting with Rawls's important "Themes in Kant's Moral Philosophy". (1) Then I challenge this interpretation with a discussion of Kant's conception of legislation, with particular attention to the way this conception is developed in his lectures and Reflexionen.

I. Summary: the Meaning of "Kant's Moral Constructivism"

A. John Rawls

In A Theory of Justice, although Rawls's method of generating principles of justice from a choice in the Original Position is described as "constructive", in the sense of "helpful to settle disputes", the idea of "constructivism" is hardly present. Constructivism, in the sense that interests us here, first plays a major role in Rawls's 1980 Dewey Lectures, "Kantian Constructivism in Moral Theory", where Rawls clarifies his own project as a limited programme in political theory, not in moral philosophy in general, a programme he has also described as something "political, not metaphysical". (2) There Rawls proposes a procedure of construction that connects a particular conception of the person with first principles of justice. In his article "Themes" Rawls emphasizes a similar idea in his interpretation of Kant moral philosophy, leading him to speak of what he calls "Kant's moral constructivism".

In "Themes" Rawls begins with an outline of the "CI-Procedure" (where CI is an abbreviation for "categorical imperative"), which he sees as something given or laid out, based on the conception of free and equal persons as "reasonable" and "rational". The procedure specifies the first principles of right and justice, and through the procedure the content of the doctrine is constructed. Kant, Rawls had explained in Theory,

begins with the idea that moral principles are the object of rational choice. They define the moral law that men can rationally will to govern their conduct in an ethical commonwealth. Moral philosophy becomes the study of the conception and outcome of a suitably defined rational decision. (3)

In Rawls’s reading, autonomy is given a procedural interpretation according to which it is choice via the procedure which specifies the content of the moral law. Rawls contrasts this constructivism with "rational intuitionism" which maintains that moral principles are true or false in virtue of an order of values, known by rational intuition, that is prior to and independent of our conception of the person. Rawls insists that Kant's conception of "autonomy" rules out the existence of such an order. On Rawls's interpretation, the content of the moral law is constructed by the procedure rather than discovered by it.

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. . . . The intuitionist’s independently given order of values is part of the transcendental realism Kant takes his transcendental idealism to oppose. (4)

Along similar lines, Stephen Darwall has spoken of Kant’s "practical or ethical idealism," (5) suggesting that "the idea of an independent order of normative fact is inconsistent with the autonomy of the will." (6)

This constructivist conception is intended to possess both epistemological and metaphysical significance: before the use of the procedure, we can not know any moral facts, furthermore, before the use of the procedure, there are no moral facts to be known. The metaethical significance of this position is rather straightforward: there are moral facts but they depend upon our conceptions and our constructive activity. In the classical terminology: this constructivism is a cognitivist theory (there are moral facts that are true or false), but it is anti-realist (because the moral facts are dependent upon us and our activities). (7)

B. J. B. Schneewind and Christine Korsgaard

J.B. Schneewind seems to concur in the ascription of such a view to Kant when he writes that, "[Kant's] theory of agency as freedom showing itself through rational self-legislation is meant to show how we as moral agents can be independent of divine legislation. It enables us to understand morality, with its due weight, as a human creation." (8) Schneewind’s studies of early modern theories of obligation contend that Kant's advance consists in his discovery that morality is something we impose on ourselves, that the moral law is something we make. (9) Christine Korsgaard suggests that Kant's "idea is that rational choice has a value-conferring status." (10) Her important study of the sources of normativity in the modern period aligns Kant with the view that, "there are answers to moral questions because there are correct procedures for arriving at them," not because there are independent moral truths or facts that the procedure tracks. (11) She insists that, "... the Kantian laws of Autonomy are positive laws: moral laws exist because we legislate them." (12)

C. A motivation for the constructivist reading: the principle of autonomy

This constructivist interpretation has been proposed for a number of different reasons. (13) Perhaps the most widespread suggestion, and the thesis we shall examine today, is that constructivism is a straightforward implication of Kant’s analysis of obligation: Kant’s analysis reveals that moral obligation involves autonomous self-legislation, and constructivism seems to be a natural way to interpret this contention.

II. Kant’s Derivation of the Principle of Autonomy

Let us review Kant’s deduction of the principle of autonomy. In the first section of the Grundlegung, Kant begins with the concept of a good will as the only unqualified good. A good will acts not out of inclination, but from objective practical laws. For a being with needs this means that a good will acts from duty, according to an "ought". Because duties or moral obligations are categorically rather than hypothetically commanded, this means that a will that acts from duty is not determined by material motives but by pure practical reason, out of respect for the moral law. Thus Kant writes,

For since, other than the law, the imperative contains only the necessity that the maxim be in accord with this law, but the law contains no limiting conditions, nothing remains for the maxim of action to accord with other than the universality of law itself. The only thing the imperative actually represents as necessary is this conformity. (14)

The essential thing is the law-like, properly legislative form. Pure practical reason can determine our will to act only on maxims that are appropriate for universal legislation. A will that is so determined is a good will. And that is the only way a good will can be specified. Kant infers a principle: "the idea of the will of every rational being as a universally legislative will." (15) He explains:

"According to this principle, all maxims are rejected that can't co-exist with the will's own legislation. Thus, the will is not merely subjected to the law, rather it is subjected in such a way that it must also be regarded as self-legislating and precisely on this account, above all, as subjected to the law (of which he can consider himself author)." (16)

Moral necessity, Kant argues, must be based on the self-legislation of a will, because, otherwise, this necessity would have to be based on an interest, which would contradict the requirement that moral necessity be a categorical necessity. This analysis generates the third formulation of the categorical imperative: "Act only such that the will can at the same time regard itself, through its maxim, as universally legislating" and is supposed to be the basis for the dignity of humanity. (17)

III. Kant’s Conception of Legislation

A. Legislation and Authorship in der Metaphysik der Sitten

I assume that this derivation of the principle of autonomy is more or less familiar. The present question, of course, focuses on the metaphysical significance of the requirement that agents regard the moral law as self-legislated. Does this imply that the moral law is a positive law or a human creation? That it is the object of a procedure of construction rather than discovery? While our contemporary conception of legislation might suggest such conclusions, there has been remarkably little discussion of Kant’s conception of legislation. (18) One explanation for this is that in the Grundlegung and the Kritik der praktischen Vernunft, Kant never elaborates his conception of legislation. But there is some interesting material in the Metaphysik der Sitten. In the introduction Kant wrote:

A (morally practical) law is a proposition that contains a categorical imperative (a command). The one who commands (imperans) through a law is the legislator (legislator). He is the author (autor) of the obligation in accordance with the law, but not always the author of the law. In the latter case, the law would be a positive (contingent) and arbitrary [willkürlich] law. The law which obligates us a priori and unconditionally by our own reason can also be expressed as proceeding from the will of a supreme legislator, i.e., one that has only rights and no duties (hence from the divine will), but this only signifies the idea of a moral being whose will is a law for everyone, without his being thought of as the author of it." (19)

Here Kant explains his conception of legislation in connection with his distinction between the author of obligation in accordance with a law and the author of a law. Unfortunately, the point and significance of the distinction seems obscure. As is often in the case in Kant’s writings, especially the Metaphysik der Sitten, a passage absent its context may remain cryptic. But further study of this distinction reveals an anti-constructivist strand in Kant’s thought which lies at the core of his conception of legislation.

B. Kant's reaction to Baumgarten in his early lectures

We know that, since the beginning of the 1760s Kant had an ongoing and wide-ranging engagement with Baumgarten's Wolffian textbook; Kant used Baumgarten's Initia philosophia practicae primae as the text for his lectures on moral philosophy. An examination of Kant's reactions to Baumgarten provides a starting point for the reconstruction of his conception of moral legislation.

In §100, Baumgarten asserts that God is the legislator and author of all natural obligations (simultaneously legislator and autor), because he is the author of the world. (20) This idea that God is the author of the moral law, concerned Kant for several reasons. First, Kant needed to defend his distinction between moral and pragmatic laws, between categorical and hypothetical imperatives. According to Kant, morality is not only action in accordance with law, but also a good character, something categorically required. But according to Baumgarten, God is the author of law because he is the author of nature, i.e., because he has connected certain actions with natural consequences that relate them to our inclinations. But if the moral law were binding only through this sort of connection, it could only bind us through our inclinations, in which case it wouldn't be categorical. In response, Kant insisted that a categorical command must lie in the nature of things, i.e. neither in our inclination, nor in God's will, nor in a combination of the two. Kant said in his lectures during the 1770s:

The one who declares that a law, which is in accord with his will, obligates another, he gives [legislates] a law. The legislator is not simultaneously an author of the law, only when the law is contingent. When the laws are necessarily practical and he only declares that they are in accord with his will, he is the legislator. Thus no one, including God is the author of the moral laws, since they do not spring from the will [choice], but are practically necessary. Were they not necessary, it would be possible that lying be a virtue. The moral laws can only stand under a legislator; it can be a being that has the power and authority to execute the laws and to declare that the moral law is simultaneously a law of his will, and obligate everyone to act in accordance. Thus, this being is a legislator, but not author. Precisely as God is not the author of the fact that triangles have three angles..." (21)

I assume that the distinction between legislator and author is relatively clear here: an author makes a law through his will; whereas a legislator simply declares that a law is in accord with his will, whether or not it is something he makes. The significant thing here is the reason why Kant distinguishes the legislator and author of the law, and why he rejects theological moral positivism: he rejects it because he rejects all kinds of moral positivism. It is as clear to Kant that lying is a vice (22) as it is that triangles have three angles. (The echoes of classical Leibnizianism rationalism opposing Cartesian voluntarism, as well as its attack on moral positivism in specific, are no accident here.) (23) If either proposition had an author, then it would be true only due to her will, i.e., they wouldn't be necessarily true. But since they are necessarily true, they can not have an author. Indeed, God could not be the author of the moral law, Kant concludes, not due to any of his special properties, but rather because it can't have any author whatsoever.

In this passage from Kant’s lectures, the essence of legislation is the declaration that someone is obliged, sometimes conjoined with the fact that sanctions are connected to the law. Natural, non-positive laws can have a legislator, but they don't have an author.

C. Legislation and Authorship in the Reflexionen and later lectures

Of course, there is reason to proceed cautiously with the notes of Kant’s lectures, since the sources vary and are certainly not Kant’s manuscripts (not in the least because Kant almost certainly did not lecture from a manuscript). (24) It is important to correlate ideas in the lectures with evidence from Kant’s own hand, whether published or unpublished. Here the notes Kant wrote in the margins of his personal copy of Baumgarten's Initia suggest that Kant was reflecting on this point as early as 1762 and that he was developing this line of thought throughout the late 1760s and 1770s. (25)

Up to this point though, we haven't yet found Kant's distinction between the author of the law and the author of obligation in accordance with the law (which we saw in the Metaphysik der Sitten). When we continue reading later lecture notes, this distinction does emerge. It seems to appear first around the time of the composition of the Grundlegung, in the fragment of lecture notes referred to as "Moral Mrongovius II". (26) This is significant since it suggests not only that Kant developed this line of thought in the 1760s and 1770s and continued to hold it, but also that in fact he was elaborating this anti-constructivist conception of legislation at the same time he was formulating his conception of autonomy. (27) I’ll cite from the "Metaphysik der Sitten Vigilantius" notes of 1793-94, since the distinction we saw in the Metaphysik der Sitten is more clearly elaborated there (and since we also have reason to believe, these are the most reliable text of any of Kant's lectures on ethics). Here we find Kant saying:

Were one to understand under 'legislator' an author of the law, this would merely concern statutory laws. But for laws that are recognized through reason from the nature of things, when one lays an author alongside them, he could only be the author of the obligation which is contained in the law. ... as author of a natural law only the author of the obligation in accordance with the imperative of the law can be thought of." (28)

Thus the legislator or author of obligation declares a correspondence of the law with his will and may connect sanctions with the law. Kant has found a way to concede to Baumgarten some sense in which God is author of the moral law (he is author of the obligation in accordance with it) without reducing morality to positive divine law. Kant insists that while the moral law can have an author of obligation in accord with it, the moral law can have no true author.

IV. Conclusion: Legislation and Autonomy in Kant

What we have then is a variety of sources that clarify Kant's discussion of legislation in the Metaphysik der Sitten and suggest that, from the early 1760s through at least the mid-1790s, Kant worked with an anti-positivistic and anti-constructivist conception of moral legislation. As Kant told his students in the mid-1780s, moral laws "are not positive laws." (29) This position follows directly from Kant's understanding of the necessity of the moral law.

This presents a serious problem for the constructivist interpretation and suggests an alternative interpretive hypothesis. According to Kant moral obligation requires autonomous, but non-positive, self-legislation. That implies that the self-legislating will recognizes moral obligation as something that coexists with and must harmonize with its own freedom. Obligation comes into being through the agent’s own will and she is not the subject of a foreign will. But, on this understanding, the content of the moral law does not depend upon her or her activities.

It remains for this non-constructivist interpretation to come to grips with the variety of uses Kant makes of the concept of autonomy and engage the variety of motivations for the constructivist reading. It remains possible that there is an inconsistency between Kant's anti-constructivist conception of legislation and other aspects of his conception of autonomy, which are also supposed to follow from his analysis of moral necessity. It could very well be that there is a serious tension in Kant's thought here. In any event, Kant’s conception of legislation represents a persistent and significant anti-constructivist strand in his thought that must be taken seriously. (30)

Notes

(1) John Rawls, "Themes in Kant's Moral Philosophy", in Kant's Transcendental Deductions: The Three 'Critiques' and the 'Opus postumum'. ed. Eckart Förster. (Stanford: Stanford University Press, 1989.)

(2) John Rawls, A Theory of Justice. (Cambridge: Harvard, 1971.) "Kantian Constructivism in Moral Theory". Journal of Philosophy 77: 515-572. (1980) "Justice as Fairness: Political not Metaphysical". Philosophy and Public Affairs 14: 223-251. (1985)

(3) Rawls, A Theory of Justice, p. 251.

(4) Rawls, Political Liberalism. (New York: Columbia UP, 1993.) pp. 99-100; cf. pp. 112, 125. It is important to note that here Rawls’s own "political constructivism" no longer denies this thesis of rational intuitionism, in contrast to his reading of Kant and earlier formulations of his own view. cf. "Kantian Constructivism", p.568.

(5) Stephen Darwall, The British Moralists and the Internal "Ought": 1640-1740. (Cambridge: Cambridge UP, 1995) p. 126.

(6) Darwall, "Reasons, Motives, and the Demands of Morality: An Introduction." Moral Discourse and Practice: Some Philosophical Approaches. ed. Darwall, Gibbard, and Railton. (Oxford, 1997) pp. 310.

(7) For the sake of clarity, it must be pointed out that while Rawls denies that facts are constructed, he does maintain that it is the constructive activity that makes facts morally relevant, or turns ordinary non-moral facts into moral facts. cf. "Themes", pp. 99,101.

(8) J.B. Schneewind, "Natural Law, Skepticism, and Methods of Ethics". Journal of the History of Ideas 52: 289-308. (1991) citation at p. 307. Karl Ameriks has recently challenged several aspects of Schneewind’s constructivist interpretation. "On Schneewind and Kant’s Method in Ethics", Ideas y Valores 102:28-53. (1996)

(9) Schneewind, "Autonomy, Obligation, and Virtue: an overview of Kant's moral philosophy", in The Cambridge Companion to Kant. ed. Paul Guyer. (Cambridge, 1992), see pp. 309; 312. For simplicity, I omit consideration of Schneewind’s important The Invention of Autonomy which may signal some deviation from his earlier interpretation of Kant on this point. (Cambridge: Cambridge University Press, 1998).

(10) Christine Korsgaard, "Kant's Formula of Humanity." Kant-Studien 77:183-202. (1986). citation at 196.

(11) Korsgaard, "The Sources of Normativity." The Tanner Lectures on Human Values. vol. 15. (1994) citation at 38.

(12) Korsgaard, "Kant's Analysis of Obligation: The Argument of Foundations I", Monist 73: 311-40. (1989) citation at 331.

(13) It is sometimes suggested, for example, that moral constructivism is part of Kant's overall answer to skepticism, both theoretical and practical. cf., for example, Korsgaard, "The Sources of Normativity"; Schneewind, "Modern Moral Philosophy: From Beginning to End?", in Philosophical Imagination and Cultural Memory: Appropriating Historical Traditions, ed. Patricia Cook. (Duke, 1993) p. 100; Schneewind, "Natural Law, Skepticism, and Methods of Ethics". Some also maintain that motivational internalism or the nature of practical, as opposed to theoretical reason, implies moral constructivism.

(14) "Denn da der Imperativ außer dem Gesetze nur die Notwendigkeit der Maxime enthält, diesem Gesetze gemäß zu sein, das Gesetz aber keine Bedingung enthält, auf die es eingeschränkt war, so bleibt nichts, als die Allgemeinheit eines Gesetzes überhaupt übrig, welchem die Maxime der Handlung gemäß sein soll, und welche Gemäßheit allein der Imperativ eigentlich als notwendig vorstellt." (KGS) IV, 420-421.

(15) "die Idee des Willens jedes vernünftigen Wesens als eines allgemein gesetzgebenden Willens." (KGS) IV, 431. For a good discussion on the translation of this passage, see H.B. Acton, Kant’s Moral Philosophy. (London: Macmillan, 1970) pp. 37.

(16) "Alle Maximen werden nach diesem Prinzip verworfen, die mit der eigenen Gesetzgebung des Willens nicht zusammen bestehen können. Der Wille wird also nicht lediglich dem Gesetze unterworfen, sondern so unterworfen, daß er auch als selbstgesetzgebend und eben um deswillen allererst dem Gesetze (davon er selbst sich als Urheber betrachten kann) unterworfen, angesehen werden muß." (KGS) IV, 431.

(17) "tun ... nur so, daß der Wille durch seine Maxime sich selbst zugleich als allgemein gesetzgebend betrachten könne."(KGS) IV, 434, 440.

(18) One apparent exception is Andrews Reath’s "Legislating the Moral Law", Nous 28:435-464. (1994) But Reath seems to simply assume, given Rousseau’s influence, that Kant’s conception of legislation must be that of giving positive law.

(19) "Gesetz (ein moralisch praktisches) ist ein Satz, der einen kategorischen Imperativ (Gebot) enthält. Der Gebietende (imperans) durch ein Gesetz ist der Gesetzgeber (legislator). Er ist Urheber (autor) der Verbindlichkeit nach dem Gesetz, aber nicht immer Urheber des Gesetzes. Im letzteren Fall würde das Gesetz positiv (zufällig) und willkürlich sein. Das Gesetz, was uns a priori und unbedingt durch unsere eigene Vernunft verbindet, kann auch als aus dem Willen eines höchsten Gesetzgebers, d. i. eines solchen, der lauter Rechte und keine Pflichten hat, (mithin dem göttlichen Willen) hervorgehend ausgedrückt werden, welches aber nur die Idee von einem moralischen Wesen bedeutet, dessen Wille für alle Gesetz ist, ohne ihn doch als Urheber desselben zu denken." (KGS) VI, 227.

(20) Baumgarten, Initia philosophiae practicae primae. (1760). Reprinted in (KGS) XIX. §100. This section of my talk is greatly indebted to Josef Schmucker's important Die Ursprünge der Ethik Kants (Meisenheim: Verlag Anton Hain, 1961) esp. pp. 278ff.

(21) "Derjenige, der deklariert, daß ein Gesetz, welches seinem Willen gemäß ist, den anderen dazu obligiert, der gibt ein Gesetz. Der Gesetzgeber ist nicht zugleich ein Urheber des Gesetzes, sondern nur dann, wenn die Gesetze zufällig sind. Wenn aber die Gesetze notwendig praktisch sind, und er sie nur deklariert, daß sie seinem Willen gemäß sind, der ist ein Gesetzgeber. Von moralischen Gesetzen ist also kein Wesen, auch das göttliche nicht ein Urheber, denn sie sind nicht aus der Willkür entsprungen, sondern sind praktisch notwendig. Wären sie nicht notwendig, so könnte es auch sein, daß die Lüge eine Tugend wäre. Allein die moralischen Gesetze können doch unter einem Gesetzgeber stehen; es kann ein Wesen sein, welches die Macht und Gewalt hat, diese Gesetze zu exekutieren und zu deklarieren, daß dieses moralische Gesetz zugleich ein Gesetz seines Willens sei, und alle zu obligieren danach zu handeln. Alsdann ist dieses Wesen ein Gesetzgeber, aber kein Urheber. Ebenso, wie Gott kein Urheber ist, daß der Triangel drei Winkel hat. ... Allein Gott will die Gesinnung, die Moralität ist seinem Willen gemäß, und als solche Gesetze obligieren sie schon vollkommen. ... Wir haben also Gott nicht als einen pragmatischen Gesetzgeber, sondern als einen moralischen Gesetzgeber anzusehen." Paul Menzer, Eine Vorlesung Kants Über Ethik. (Berlin: Pan Verlag Rolf Heise, 1924) p. 61-63; cf. "Moralphilosophie Collins" KGS XXVII, 282-3.

(22) cf. "Moralphilosophie Collins" KGS 262/16-21; 277/23.

(23) cf. Leibniz, Codex Juris Gentium §xiii-xiv. trans. Patrick Riley, The Political Writings of Leibniz. (Cambridge: Cambridge University Press, 1972) pp. 174. Kant is recorded early in his career making this same point. See "Praktische Philosophie Herder" KGS XXVII, 9-10.

(24) In addition, the extensive philological work of Wilhelm Kraus shows that the extant notes from Kant’s lectures on ethics from the 1770s (both Menzer’s sources and the Moralphilosophie Collins which is the foundation of the Akademie edition) were the least reliable copies of the original source or sources. Wilhelm Kraus, "Untersuchung zu Kants moralphilosophischen Vorlesungen." Dissertation. Tübingen, 1926. esp. pp. 66. I am greatly indebted to Dr. Werner Stark of the Marburger Kant-Archiv for instructive discussion about this and related matters.

(25) For example, see R.6680, 6769, 7089, 7090, 7092. KGS XIX. By 1762-63 Kant had already begun to write marginalia on this subject. This is interesting because Schmucker persuasively argues that Kant's critical interaction with Rousseau's writings first began between October 1763 and February 1764. Thus, it appears that Kant’s reflections on this issue began before his encounters with Rousseau. cf. Schmucker, pp. 142.

(26) "Moral Mrongovius II". KGS XXIX, 633.

(27) I believe this fact is also crucial to our understanding Kant’s remark in the Grundlegung, quoted earlier that in some sense, the will "can consider itself as author" of the moral law. KGS IV, 431.

(28) "Würde man unter dem Legislator einen autorem legis verstehen, so würde dieses bloß statutarische Gesetze betreffen. Gesetzen aber, die aus der Natur der Sache durch die Vernunft erkannt werden, wenn man denen einen auctorem beilegt, so kann er nur autor der Verbindlichkeit sein, die im Gesetz enthalten ist. ... unter dem autore eines natürlichen Gesetzes nur der Urheber der Verbindlichkeit nach dem Imperativ des Gesetzes gedacht werden kann...""Metaphysik der Sitten Vigilantius", KGS XXVII, 544-45.

(29) "...sind nicht positive Gesetze." "Moral Mrongovius II", KGS XXIX, 634.

(30) My thanks to Profs. Otfried Höffe, Karl Ameriks, David Solomon, and audiences at the University of Tübingen and the 1998 APA Pacific Division Meetings and my commentator there, Mark LeBar, for helpful comments on earlier versions of this paper.