KATERINA DELIGIORGI | The Scope of Autonomy: Kant and the Morality of Freedom | Oxford University Press 2012
By Jeppe von Platz
As with other work on key figures in the history of philosophy, work on Kant’s ethics comes in three genres. First, there are works that pursue the scholarly interest of understanding Kant’s ethics. Second, there are works that pursue the philosophical interest of finding the better ethical theory and look to Kant for inspiration. While such ‘Kantian ethics’ should stay true to some core principles of Kant’s philosophy, the defended position need not be the same as Kant’s ethics, and the defence can succeed without possessing all the scholarly virtues required by the first genre. Third is the genre of ‘interpretation and defence’, which combines scholarly and philosophical pursuits in the service of defending Kant’s ethics. It is hard to succeed in this genre of interpretation and defence. One’s interpretation must possess all the virtues of scholarship, and one must show that Kant’s ethics thus interpreted has advantages over other ethical theories. Yet, why study Kant, if not to understand morality? And why defend Kantian ethics, if Kant was not right? To succeed in this third venture would be a very fine achievement.
In The Scope of Autonomy Deligiorgi pursues the ambitious aim of interpretation and defence. Deligiorgi argues that Kant’s “autonomy-based ethics” (pp. 12, 26, 150) is less individualist and rationalist than its reputation would suggest and that it is superior to contemporary autonomist and Humean ethics.
Deligiorgi’s aim and method—to defend Kant’s ethics by bringing it into dialogue with contemporary alternatives—are ambitious and admirable. I was also impressed by how well Deligiorgi succeeds in bringing her interpretation of Kant into dialogue with contemporary theories. However, I was less impressed by Deligiorgi’s interpretation of Kant and thus I doubt that she succeeds in her aim. Deligiorgi makes a number of claims about Kant’s ethics, but the arguments she offers in support of these claims could (and should) be stronger: she rarely discusses passages that appear to contradict her interpretation; she does not explain how her interpretation answers relevant hard questions about how we should understand Kant’s ethics; and she rarely engages with relevant scholarship to show how her interpretation departs from other interpretations or why hers is better. So, in this comment I raise some interpretive questions with the hope that these questions will be useful for Deligiorgi to expand on the attractions of her interpretation of Kant.
Deligiorgi says that the view she defends “takes as its starting point” Kant’s “idea of the will of every rational being as a universally legislating will”, (p. 3, citing Groundwork, AA 4:431) and that this “formulation is the closest we come to a definition of autonomy” (p. 3). She then unfolds this idea of autonomy along several dimensions: normative ethics, moral epistemology, theory of action, and moral ontology. Bringing these dimensions of autonomy together is the idea that autonomy is the “practical ideal” (pp. 5, 103) of doing the “right thing, because it is the right thing to do” (p. 104), so that autonomy is realized in the “coincidence of normative and motivational reasons” (p. 107).
I’ll ask a couple of questions about Deligiorgi’s interpretation of Kant’s theory of autonomy for each of the four dimensions. I begin with questions about Kant’s normative ethics and theory of action and then move on to questions about moral epistemology and ontology.
In regard to Deligiorgi’s interpretation of Kant’s normative ethics, I’m especially interested in what Deligiorgi would say about some hard questions about how we should understand Kant’s exposition of the supreme principle of morality in the Groundwork—both because I’m very interested in these questions, but also because I fear that Deligiorgi’s silence about these questions indicates a general problem with her interpretation of Kant’s theory of autonomy.
Deligiorgi takes as her starting point the formula of autonomy from the Groundwork and her interpretation of the epistemology, motivation, and ontology of autonomy are ways to make sense of this formula. My worry is that Deligiorgi’s starting point, i.e. the formula of autonomy, is less central to Kant’s ethics than Deligiorgi assumes. Of course, Kant concludes section II of the Groundwork with saying that autonomy is the supreme principle of morality (GMS, AA 4:440), so there is no doubt that autonomy is a, or even the, central ordering idea of Kant’s ethics. Yet, we may doubt that the formula of autonomy that is the focus of Deligiorgi’s interpretation is the same as the principle of autonomy that is the supreme principle of morality. If so, then the problem is not merely that Deligiorgi mistakenly thinks that the principle of autonomy is the formula of autonomy (see e.g. p. 192). Rather, we may then worry that Deligiorgi’s interpretation misidentifies the basic principle of Kant’s ethics.
Kant states the supreme principle of morality, i.e. the principle of autonomy (PA), as the principle that we are “to choose only in such a way that the maxims of your choice are also included as universal law in the same volition” (GMS, AA 4:440). We have good reason to regard this principle of autonomy as in some sense the same principle as the categorical imperative known as the formula of universal law (FUL): ‘act only in accordance with that maxim through which you can at the same time will that it become a universal law’ (GMS, AA 4:421). In addition to their close similarity in wording, logic dictates that they are identical. According to Kant, there is one supreme principle of morality and Kant identifies both PA and FUL as this supreme principle: Kant says that PA is “the sole principle of morals” (GMS, AA 4:440); and that FUL is “the single categorical imperative” (GMS, AA 4:421) and “the supreme law” (GMS, AA 4:437); in notes from lectures he is reported to have called FUL “the objective principle” (V-Mo/Mron II, AA 29:621) and “the formula of the universal imperative” (V-MS/Vigil, AA 27:495–6); and in the second Critique Kant presents the “fundamental law of pure practical reason” as a variant of FUL: “So act that the maxim of your will could always hold at the same time as a principle in a giving of universal law” (KpV, AA 5:30). Therefore, both PA and FUL are the single supreme principle of morality, which means that FUL and PA are the same principle.
Of course, in addition to PA/FUL we find a number of formulae in section II of the Groundwork: the formula of law of nature (FLN), the formula of humanity as an end in itself (FH), the formula of autonomy (FA), and the formula of the realm of ends (FRE)—to mention the more prominent of these. Since PA/FUL is the “single categorical imperative” (GMS, AA 4:421), these other formulae must in some sense be variants of PA/FUL. As Kant writes, the formulae “are at bottom only so many formulae of the very same law” (GMS, AA 4:436). This claimed singularity in the face of plurality raises hard interpretive puzzles: How can we make sense of the singularity of the categorical imperative in light of this plurality of formulae? In what sense is there a single categorical imperative? In what sense are the different formulae variants of a single principle?
Unfortunately, Deligiorgi does not discuss these questions. And her interpretation should deal with at least two of these. First, Deligiorgi’s interpretation could have benefited from a discussion of the formula of humanity and its place in autonomy-based ethics. According to Deligiorgi, Kant’s ethics makes morality inter– rather than intra-subjective (pp. 16–17, 139–41, 172, 185). By inter-subjective Deligiorgi means, I take it, that morality concerns how we relate to each other as persons, not merely how we should relate to ourselves: “autonomy does not describe how an agent stands with respect to herself, her desires or plans of action, but rather how she stands with respect to other agents” (p. 5). I agree, but it seems to me that to make sense of how Kant’s ethics is inter- rather than intra-subjective, it would be helpful to consider FH which explicitly relates actors to other agents and also to consider the relation between this formula and PA/FUL. To fully make sense of Kant’s ethics as inter-subjective, then, Deligiorgi would do well to have a story to tell about the formula of humanity and how it is related to the principle of autonomy and the other formulas.
Second, since Deligiorgi takes FA as the central formula of Kant’s ethics, it would be nice to see an account of the relation between this formula, PA/FUL, and the other formulas. My underlying concern, again, is that FA is not the same as PA/FUL and so is not the supreme principle of morality, which means that Deligiorgi’s interpretation starts off on the wrong foot, so to speak.
Deligiorgi might argue that it does not matter whether you take FA or PA as starting point, since Kant’s singularity thesis implies that they are the same principle. Yet, the text belies the idea that FA and PA are the same principle. Kant initially presents FA as follows:
the ground of all practical lawgiving lies (in accordance with the first principle) objectively in the rule and the form of universality which makes it fit to be a law (perhaps even a law of nature); subjectively, however, it lies in the end; but the subject of all ends is every rational being as an end in itself (in accordance with the second principle); from this there follows now a third practical principle of the will, as supreme condition of its harmony with universal practical reason, the idea of the will of every rational being as a will giving universal law. (KpV, AA 4:431)
The implication is that FA is derived by means of the combination of FUL and FH. Now, if both FUL and FH are necessary to establish FA, then it follows that FA is not the same principle as either of these and so not the same as FUL. Moreover, since FUL and PA are the same principle, it follows that FA is not the same principle as PA. So, FA and PA are different principles.
If the preceding is correct, then the formula of autonomy, which is the starting point and focus of Deligiorgi’s interpretation, is not the supreme principle of morality of Kant’s ethics. So, we may doubt that Deligiorgi’s interpretation accurately captures the essence and structure of Kant’s ethics.
In sum, here are two questions that Deligiorgi might consider:
1. What is the role of FH in the autonomy-based ethics that Deligiorgi defends?
2. How should we understand the supreme principle of Kant’s ethics (the principle of autonomy) and what is the relation between this principle and the other formulas of the categorical imperative that Kant articulates in the Groundwork?
Theory of action (moral motivation)
Turning now to Kant’s theory of action, Deligiorgi defends the dual thesis that Kant’s theory is an “anti-Humeanism about practical reasons […] not anti-Humeanism about motivations” (p. 12). Kant’s theory is anti-Humean about practical reasons, since “the moral law reveals to us that we are capable of choosing ends, not means only” (p. 92, see also pp. 165, 171), which entails that reason is involved in the setting of ends and so is not a mere slave of the passions. Kant’s theory is not anti-Humean about motivation, because practical reason does not alone motivate action. Human beings can be habitually moral, can be motivated to be moral in the face of counter-moral interests, and can fail to be moral when imperatives of reason are too onerous (p. 89). Given this range of modes of motivation, Kant “is not committed to motivational anti-Humeanism” (p. 89). Deligiorgi thus finds fault with “the received view” of the practicality of pure reason, that “moral judgments […] necessarily motivate the agent” (p. 69), for misreading Kant as an internalist and being overly psychologistic. The received view is overly psychologistic, since it misses that Kantian autonomy is about our capacity to set ends. Kant is not an internalist about motivation, since we can know what we ought to do, yet not be motivated to do it.
I’m not persuaded by Deligiorgi’s argument that Kant is not an internalist. Kant, of course, recognizes the distinction between knowing what one ought to do and being motivated to do it, and it is possible to know what one ought to do and yet be insufficiently motivated to do it. So, we must agree with Deligiorgi’s assessment that Kant thinks that moral awareness can fail to result in moral action. But it does not follow that Kant is not an internalist about moral motivation (meaning that awareness of the moral law necessarily motivates), it merely follows that awareness of what one ought to do sometimes motivates insufficiently to determine one’s actions.
Moreover, Kant makes it clear that awareness of the moral law is both necessary and sufficient for moral motivation. The key concept here is respect (Achtung). Of course, Kant’s concept of respect is a complex topic, but at least it is clear that respect for the moral law (also) is an incentive to be moral produced by reason:
[R]espect for the law is […] morality itself subjectively considered as an incentive […] This feeling (under the name of moral feeling) is therefore produced solely by reason. (KpV, AA 5:76)
Awareness of the moral law is necessary for moral motivation, since respect is the only true moral incentive:
Respect for the moral law is therefore the sole and also the undoubted moral incentive.’ (KpV, AA 5:78)
Awareness of the moral law is sufficient for moral motivation, since such awareness necessarily involves respect:
Respect for the law, which in its subjective aspect is called moral feeling, is identical [ist einerlei] with consciousness of one’s duty. (MS, AA 6:464)
So, if by internalism we mean the position that awareness of what we ought to do is sufficient for being motivated to do it, then Deligiorgi’s argument falls short and we have good textual reasons to read Kant as an internalist. Deligiorgi might change the definition of internalism so that internalism is the position that moral reasons are sufficiently motivating to determine action. But that is not, I think, the sort of internalism that is attributed to Kant by ‘the received view’, and so not the target of Deligiorgi’s argument. It is virtually impossible to read Kant as defending that sort of internalism, so if she relies on this definition of internalism, Deligiorgi’s argument is of little consequence.
Another way in which Deligiorgi could support her interpretation of Kant’s theory of action would be by showing how she can make sense of the complex conceptual apparatus of moral psychology that Kant employs. Unfortunately, Deligiorgi’s discussion of agency and motivation bypasses questions about respect and self-respect, love and self-love, self-deception, self-knowledge, and self-conceit; about how Kant understands the relation between the faculty of desire (Begehrungsvermögen), (pure) practical reason, and the will; about Kant’s distinctions between incentives (Triebfedern), desires (Begierde), inclinations (Neigungen), feelings (Gefühle), affects (Affekte), and passions (Leidenschaften). Showing how her interpretation fits and makes sense of these and like elements in Kant’s ethics would support and deepen Deligiorgi’s interpretation.
In sum, here are two more questions Deligiorgi might consider in elaborating her interpretation:
3. How can Deligiorgi’s interpretation make sense of and be supported by what Kant says about the feeling of respect as the inescapable moral incentive?
4. How does Deligiorgi’s interpretation of Kant’s theory of action fit and make sense of Kant’s complex moral psychology?
Deligiorgi’s account of Kant’s moral epistemology answers three distinct questions:
How can we know what we ought to do?
What is the status of such knowledge of what we ought to do?
How can we know that there are moral truths – that duties truly exist?
Deligiorgi’s answer to the first question is that the categorical imperative, in the version known as the formula of universal law, presents an objective procedure for arriving at true ethical judgements. Thus, with respect to the second question, Deligiorgi argues that Kant’s moral epistemology is cognitivist, since moral judgements are truth-evaluable (p. 33); realist, since the truth of moral judgements does not depend on our attitudes or interests (p. 33n); non-naturalist, since the truth of moral judgements does not depend on external, non-moral facts (pp. 33n, 44–6); and objectivist, since we have an objective procedure for determining right and wrong (pp. 33n, 46–8). With regard to the third question, Deligiorgi rejects that Kant thinks that the moral law appears to us as a fact of reason. Instead, “the formula of universalizability represents the spontaneous order of reason” (p. 59). This order is spontaneous in that “reason frames an order for itself” (59), and orderly in that the universalizability test “is a rational pattern or an order that pure reason frames for itself in its practical employment” (p. 60).
I was puzzled by two features of Deligiorgi’s interpretation. First, in regard to Deligiorgi’s treatment of the first two questions, I’m unsure how much Deligiorgi’s interpretation moves us beyond what is clearly implied by the Groundwork and the second Critique. It goes without saying that Kant believes that there are duties and that the categorical imperative is, in some sense or other, the principle that generates these. So I am unsure that Deligiorgi’s claim that moral judgements are truth-evaluable adds anything further. If there are things that we ought to do, as Kant clearly argues there are, then I’m not quite sure what we learn by adding that there are truths about what we ought to do.
Second, I have more or less the opposite worry about Deligiorgi’s interpretation of Kant’s answer to the third question, namely by what authority we can affirm the validity of morality. Here, Deligiorgi asserts a controversial interpretation, but without much textual grounding and without engaging with relevant secondary literature. The Groundwork consists of an analytical part (sections I and II and some of section III), in which Kant articulates the categorical imperative as the supreme principle of morality, and a synthetical part (most of section III), in which Kant establishes the objective validity of this principle. Kant’s synthetical argument is notoriously obscure, and it may be that Kant changed the argument in the second Critique. In any case, any interpretation has to make sense of how Kant brings together the ideas of freedom, reason, and morality, the reciprocity thesis, and the role of the fact of reason in revealing the authority of morality. On Deligiorgi’s interpretation, Kant’s argument moves from the spontaneity of the order of reason to the validity of the formula of universalizability. I must confess that I didn’t quite understand how this move is supposed to go, how it draws upon and makes sense of the relevant texts (especially the reciprocity thesis, the fact of reason, and the link between autonomy and freedom), or how her interpretation differs from those offered by various Kant scholars. (In a later chapter, Deligiorgi argues that Kant’s “deductive argument” in the Groundwork fails and that he instead offers a “cumulative argument” [pp. 93–4], but it was unclear to me what this cumulative argument is or how it explains and supports Deilgiorgi’s interpretation.)
In sum, here are two more questions about Deligiorgi’s interpretation of Kant:
5. What is added by adding the notion of truth to Kant’s moral epistemology – apart from what is already implied by the notion of the objective validity of the categorical imperative?
6. How does Deligiorgi’s interpretation help us make sense of Kant’s argument for the validity of the categorical imperative in Groundwork III and the Analytic of KpV and the key moves in this argument (the reciprocity thesis, the fact of reason, the presupposition of freedom, etc.)?
An especially ambitious parts of Deligiorgi’s project is her interpretation of Kant’s theory of freedom. In a very dense section (Section 6.3) Deligiorgi argues that we can make sense of Kant’s theory of freedom, and therewith explain how agents can do the right thing, because it is the right thing to do, if we adequately distinguish between the different kinds of freedom at work in Kant’s philosophy, and if we interpret Kant’s claims about the causality of reason in terms of a “power of origination” (p. 193).
Crucial to Deligiorgi’s interpretation is the intriguing passage at AA 6:221 of the Metaphysics of Morals where Kant asserts the epistemic basis for affirming freedom on practical grounds. Here Kant writes:
[I]n reason’s practical use the concept of freedom proves its reality by practical principles, which are laws of a causality of pure reason for determining choice [Willkür] independently of any empirical conditions […] and prove a pure will [reinen Willen] in us, in which the moral concepts and laws have their origin. (MS, AA 6:221)
Deligiorgi suggests that we can make sense of this passage, and of Kant’s theory of freedom more generally, if we look closer at the distinction between Willkür, Wille, and the causality of pure reason that Kant here presents.
According to Deligiorgi, Wille is “the freedom that describes the moral content of the theory […] this type of willing gives us a proper, i.e. moral and rational, conception of our freedom” (p. 192). Wille is exercised whenever an action is “done from the commitment to do the right thing” (p. 196). Wille is thus a capacity for autonomy (for doing the right thing, because it is the right thing), but autonomy also presupposes two other concepts of freedom: Willkür and the causality of reason.
Following Kant’s definition at AA 6:213 of the Metaphysics of Morals, Deligiorgi defines Willkür as follows:
Willkür […] is the faculty “to do or to refrain from doing” (MS, AA 6:213) […] a power to choose between alternative possibilities. (p. 192)
Here I have a small quibble: the phrase that Deligiorgi quotes from the Metaphysics of Morals is not Kant’s definition of Willkür. Here is the full passage that Deligiorgi quotes from:
The faculty of desire in accordance with concepts, insofar as the ground determining it to action lies within itself and not in its object, is called a faculty [Vermögen] to do or to refrain from doing as one pleases. Insofar as it is joined with one’s consciousness of the ability [des Vermögens] to bring about its object by one’s action it is called choice [Willkür]; if it is not joined with this consciousness its act is called a wish. (MS, AA 6:213)
So the fragment Deligiorgi quotes as a definition of Willkür is not his definition of Willkür, but instead part of Kant’s definition of the faculty of desire, of which Willkür is one particular configuration. Willkür, according to this passage, is (the faculty of) desire, joined with consciousness of the ability to bring about the desired object. I do not, however, think that this is a serious quibble, since Deligiorgi’s further definition of Willkür as a power of choice seems correct enough—though perhaps a bit vague. Willkür, it seems, is an executive power: something like the power to pursue desires in accordance with concepts—including concepts (and principles) of right and wrong. So I don’t think that Deligiorgi is far from the truth, when she says that the exercise of Wille (the legislative function that issues principles interpreting the concepts of right and wrong, i.e. the moral laws) presupposes the executive capacity to act on the moral laws issued by Wille, i.e. presupposes Willkür.
And, of course, as Deligiorgi argues, all this (i.e. all this “acting for moral reasons” business) presupposes some sort of spontaneity or power of origination. Again, I’m not sure that I agree with Deligiorgi that the causality of pure reason is this power of origination—I’m inclined to think that the causality of pure reason is the whole story about how Wille determines actions, and thus that the causality of pure reason is a capacity that presupposes the power of origination without itself being this power. But again, this is a minor disagreement. Notwithstanding my quibble about Deligiorgi’s claims about particular concepts of Kant’s theory of freedom, it seems to me that she gets the theory right.
However, I would like to hear more about how Deligiorgi believes that Kant’s theory of freedom (as she interprets it) supports the autonomy-based ethics. One worry is that the sense she makes of Kant’s theory of freedom is an ill fit for the theory of autonomy she develops. Deligiorgi’s interpretation of Kant’s theory of autonomy is based on the Groundwork and the second Critique, but some scholars have argued that Kant’s distinction between Wille and Willkür (which only shows up in the 1790s) is meant to solve problems with the theory of autonomy (and its roots in Kant’s theory of freedom) that Kant defended in the Groundwork. If that is the case, then we at least need a story about how the theory of freedom we find in the later writings fits with the theory of autonomy of the earlier writings (and it would be nice if the story explains the apparent developments).
Moreover, Deligiorgi suggests that Kant’s theory of freedom can make sense of our ideas of personal responsibility, but it was not clear to me how that story went. The tight link between reason and morality in Kant’s ethics make him vulnerable to the criticism (made famous by Sidgwick, but raised already in the 1790s by Reinhold and Schulz) that, according to Kant’s ethics, the wicked are irrational and so are not truly free, which means that the wicked are not responsible for their wrongdoings. Deligiorgi mentions this problem and appears to think that her interpretation solves it, but it was not clear to me what her solution is.
So, in sum, here are my last two questions for Deligiorgi’s interpretation:
7. How does Kant’s theory of freedom fit and support his theory of autonomy?
8. How can Kant make sense of the responsibility of the wicked?
(Of course, one might follow Guyer and bring these two questions together with the thought that the developments in Kant’s theory of freedom and the distinction between Wille and Willkür are meant to solve problems with the theory of responsibility that beset the account of freedom and autonomy of the Groundwork.)
I learned a lot from The Scope of Autonomy, and I admire Deligiorgi’s attempt to defend Kant’s ethics by bringing it into dialogue with contemporary ethics. I do, however, worry that Deligiorgi bypasses many central interpretive questions and believe that her interpretation could be strengthened by considering at least some of these. So, in the hope that thinking about them will be useful, I have raised no less than eight such questions for Deligiorgi’s interpretation of Kant. Of course, I am not asking her to answer all of these questions (it would be unreasonable to ask any interpreter to cover this much ground), but present them with the hope that she’ll find inspiration in thinking about some of them.
© 2014, Jeppe von Platz.
Jeppe von Platz obtained his Ph.D. in philosophy from the University of Pennsylvania and is an Assistant Professor of Philosophy at Suffolk University, Boston. He specialises in social and political philosophy, ethics and its history, and Kant, and has published in various journals, such as Politics, Philosophy & Economics, Philosophy and Public Issues and Critical Review, as well as book contributions, such as to the forthcoming The Cambridge Companion to Liberalism.