A perennial theme in Immanuel Kant's ethics is the question of whether and to what extent one can welcome meaningful roles for our sensible natures in an a priori morality. Reading Anne Margaret Baxley's recent book Kant's Theory of Virtue: The Value of Autocracy Footnote 1 provided me with a new perspective, from which to think about this familiar problem. In this essay, I reflect on some claims Baxley makes about the relationship between reason and sensibility in Kant's theory of virtue. I then reflect on tensions I find in these claims as compared to the overall goal of her book: an account of Kant's conception of virtue as autocracy. Ultimately, I argue that interpreters like Baxley (and myself) who want to welcome a more robust role for feeling in Kantian ethics must, in order to achieve our purposes, move beyond the general account of the limits for the role of the moral feeling of respect in the grounding of Kant's ethics which Henry Allison established in his influential 1990 study of such things, Kant's Theory of Freedom.Footnote 2

I begin with a brief summary of Baxley's book. Baxley argues that Kantian virtue is, at its heart, what Kant calls ‘autocracy’, a state in which an agent has rational self-rule over herself, thereby ‘tak[ing] the right attitude toward her sensible nature and manag[ing] it appropriately’ (Baxley, 171). Although some would argue this is a state in which emotions must simply be constrained, Baxley argues to the contrary that, while Kantian virtue (because it is informed by radical evil) cannot be Aristotelian in nature, the Kantian virtuous person nonetheless is enjoined by duty to control, maintain and even cultivate certain emotional states. All of this results in a virtuous agent who is ‘at peace with herself, calm and tranquil, and yet always prepared to stand down the potential threat to good conduct that her propensity to evil could pose’ (81). Although one can, then, achieve virtue, this is a state within which at least the bare possibility (if not the actual presence) of attitudes and feelings opposed to the aims of morality remains. Virtue as autocracy is thus offered as a ‘distinctively modern, egalitarian conception of virtue’ (from book-jacket) which offers an alternative, non-Aristotelian way of appreciating the relationship of reason and sensibility in a virtuous life.

It is upon this last point that I want to dwell. Baxley's story of a more accommodating Kant – one whose conception of virtue welcomes our sensible selves as a part of us that can move us toward virtue even as we admit the ever-present possibility of our propensity towards evil undermining our efforts toward virtue – is one with which I have great sympathy. I have said more about Baxley's account of virtue elsewhere,Footnote 3 and I point readers to that work for my reflections on Baxley's account of virtue. Here, however, I want to stand back and raise a larger question. To defend her account of virtue, Baxley relies upon a story of the interrelationship of the rational and sensible aspects of the person that is more complex than traditional accounts. But is Baxley's story of the relationship between reason and sensibility adequate to the role she wants sensibility to play in this Kantian account of virtue? I argue that the conception of the relationship between reason and sensibility which emerges in this work is both at odds with itself, and ultimately inadequate for grounding the conception of virtue Baxley wants to defend.

To tell this story, I focus on Baxley's discussion of moral feeling as it arises in Kant's Doctrine of Virtue. I thus limit myself to only one part of Baxley's book (essentially, chapter 4), since Baxley considers how a whole range of feelings, from love to sympathy, are important in the life of virtue. But, as readers of Kant's texts will appreciate, making sense of moral feeling is a central concern for anyone wanting to make sense of the relationship of reason and sensibility in Kant's moral philosophy overall. Further, Baxley herself rightly dwells quite a bit on this central feeling; as such, we take it as the territory upon which to make sense of the larger question of the proper relationship of reason and sensibility in Kant's ethics.

One question about moral feeling is that of the relationship of this feeling to consciousness of the authority of the moral law: is moral feeling a consequence of a previous, non-felt consciousness of the authority of the moral law; or is it an aspect of that consciousness itself, something upon which a finite being relies as an epistemic tool for appreciating the authority of the moral law? On Henry Allison's standard reading of such things, we must take the previous route. According to Allison, the moral feeling of respect is something following from the fact of reason, not as an aspect or condition of it.Footnote 4 But moral feeling ‘also plays a crucial role in the presentation of the fact that is to be construed as a Fact of Reason’ (Allison, 237; emphasis added), particularly in ‘showing that we do in fact take an interest in morality’ (Allison, 237; emphasis added). For Allison, then, although the law as ‘a rule or principle of action’ (Allison, 233) is identified apart from any account of the moral feeling of respect, the law as ‘a motive to act or refrain from acting in ways specified by this principle’ (Allison, 233) relies upon moral feeling, for it is such feeling that ‘presents’ the fact of reason as a motive. For Allison, then, moral feeling is thus not an aspect of our founding consciousness of the moral law; it is, rather, a consequence of such consciousness, something to which we appeal in making sense of motivation to action in accordance with that already recognized authority of the moral law. What, though, is Baxley's take on such matters?

It is initially difficult to decide what Baxley's answer to this question is. She seems to suggest, initially, that she is going to take the non-Allisonian position that moral feeling is integral to consciousness of the moral law. (See, for example, Baxley, 146.) Ultimately, though, she takes the more traditional Allisonian point of view on this question. For example, while making a claim about the predisposition of love, she asserts that ‘love … seems not to follow immediately upon our consciousness of the moral law in the same way that moral feeling and conscience are directly affected by our awareness of the law’ (Baxley, 152; see also Baxley, 149–51). Here, it is clear that Baxley takes moral feeling to be a direct result of awareness of the law, not a component of it. Further, the moral import of this feeling, along straight Allisonian lines, is in its role for motivating agents to moral action: ‘Moral feeling … help[s] facilitate the strength we need to constrain our wills in accordance with duty and to fulfill our moral ends and obligations from duty’ (Baxley, 147).

One might have expected someone like Baxley, who is interested in defending the importance of feeling in Kant's conception of virtue, to push the envelope a bit on how to understand the role of moral feeling in a moral life. But Baxley instead takes the straight Allisonian line. Yet Baxley puts herself in a more problematic situation than Allison would in making similar claims. Baxley is, after all, providing an account of virtue that is centrally concerned to interpret the Doctrine of Virtue. But her claim that moral feeling is a consequence of our consciousness of the authority of the moral law, not a component thereof, is at odds with a claim central to the very section Baxley is interpreting closely. Indeed, Baxley's chapter 4 is almost entirely devoted to the section in the Doctrine of Virtue titled Concepts of What is Presupposed on the Part of Feeling by the Mind's Receptivity to Concepts of Duty as Such. But within this very section, Kant asserts pretty clearly that ‘consciousness of obligation [to the law of duty] depends upon moral feeling to make us aware of the constraint present in the thought of duty’ (6: 399/160). There is, perhaps, a way for Baxley to resolve this claim with her own suggestion that moral feeling is consequent upon such consciousness, not integral to it; but that she does not explicitly do so is perplexing and problematic.

In any event, we are left with the claim that moral feeling is the effect of our consciousness of the authority of the moral law. What, then, does that causal story look like in more detail? Baxley asserts that moral feeling is ‘a susceptibility to feel pleasure or displeasure based on our awareness that our actions conform to or conflict with duty’ (Baxley, 148). Furthermore, to say that this feeling is caused ‘by the moral law’ simply means that it is effected by ‘the weight of moral reasons that support the action’ in question (Baxley, 147–8): when moral reasons weigh in favour of me doing the action, I take pleasure in the thought of such action; when moral reasons weigh against the action, I feel pain at the thought of such action.

Baxley goes on, however, to make a couple of confusing claims about this otherwise straightforward account of the effect of moral reasons on sensibility. First, she asserts that ‘these aesthetic predispositions that make us receptive to duty’, including moral feeling, ‘present a model of the relation between reason and sensibility in which the two are so intimately intertwined that it is impossible to prize them apart’ (Baxley, 146; emphasis added). I take it that Baxley means here we cannot take apart aspects of moral feeling and say ‘this comes from reason’ or ‘that comes from sensibility’. Rather, we need to think of moral feeling, and of our predispositions generally, as ratio-sensible hybrids, or even what she later calls ‘volitional tendencies’ (Baxley, 146), a phrase which seems to try either to avoid or to efface the distinction between reason and sensibility entirely.

The second thing Baxley says is, however, in conflict with this first claim. She asserts that, because moral feeling has a rational cause, we should not consider it a sensible experience as such. This is a difficult claim to get one's hands on. Initially, Baxley asserts that ‘though none [of these predispositions, including moral feeling] is properly described as a merely sensible tendency, they are all phenomena that presuppose sensibility and in which sensibility is implicated’ (Baxley, 146). This seems straightforward enough. But then she claims not only that moral feeling ‘is not a merely sensible tendency’ (Baxley, 148; emphasis added), but also that because it ‘does not originate in sensibility’, Kant ‘does not locate this response in sensibility’ (Baxley, 148; emphases added). This is, in fact, a further articulation of what she means when she calls moral feeling a ‘volitional tendency’: because moral feeling has a non-sensible cause, we should not understand the feeling itself as a sensible thing. We should, rather, think of it as a volitional tendency, admittedly one that ‘presuppose[s] sensibility’ (Baxley, 146), but not one which is ‘locate[d]’ in sensibility. Moral feeling is thus best understood as a ‘volitional tendency’ which tracks our passive response to the authority of the moral law: ‘moral feeling is the way in which we are affected or moved by moral considerations, or the way in which we register their normative force’ (Baxley, 148–9).

I struggle to make sense of these claims. First, the claim that moral feeling is not sensible because it has a rational cause stands in tension with Baxley's earlier claim that reason and sensibility are so intimately connected in these predispositions that we cannot ‘prize apart’ the contributions of reason and sensibility. By asserting that the feeling has a rational cause, she has just prised apart what she said could not be prised apart!

It makes sense, though, that she should abandon this initial inseparability claim: the best way to prise apart the aspects of moral feeling is to say that the feeling has a rational cause, but a sensible effect. This fits very nicely with Kant's claims more generally, beyond the predispositions section in the Doctrine of Virtue, about moral feeling. From the Groundwork, through the second Critique and into the Metaphysics of Morals, there is a constant theme in Kant's understanding of the production of the feeling: reason is the cause of the feeling, itself inscrutable to us; and the feeling is the effect of this rational cause upon our sensibility.Footnote 5

But this is not what Baxley says. Rather, she asserts that because moral feeling has a rational cause, the feeling itself is not located within sensibility. But to say that we should not ‘locate this response [that is, feeling] in sensibility’ is just wrong, textually. Further, where would we find a capacity for passive receptivity to anything except in our sensible natures? Sensibility is, after all, that part of ourselves built to take things in! And yet, having asserted that moral feeling has a rational cause, Baxley does not go on to admit that this feeling itself is a sensible thing. Instead, we have the claim that moral feeling is not sensible as such, but is a volitional tendency located more broadly or pervasively within my person.

But why should we insist that something is not sensible because its cause is not sensible? Spinoza famously asserted that to know a thing was to know its ultimate cause, but Kant would want to avoid going down that route. Indeed, to do so would privilege the epistemic perspective of the active, causal force over the perspective of the caused phenomenon, that is, over the perspective of the finite being. But when Kant speaks of moral feeling, he wants to understand what it is like for a finite being to encounter moral demands. And, from a finite perspective, moral feeling is just a sensible feeling. It is, furthermore, one that is very familiar from the common, non-philosophical perspective. (See 5: 87–8/74 and 5: 91–2/77–8 for Kant's assertions of the commonality of the moral feeling of respect.) There is thus nothing unusual about the experience of moral feeling as such; it is only when we move to its causes that we begin to tell a distinctive story about its rational origin. All of that does make moral feeling a very special feeling, but it does absolutely nothing to remove the feeling from the realm of sensibility.

What is really interesting about moral feeling is that the realm of sensible feeling is itself expanded in virtue of this distinctive feeling with an admittedly mysterious rational cause. I take it, though, that Baxley would want to avoid such a story: although she removes moral feeling from the common, mundane realm of sensibility, she ironically also asserts a common, mundane location of those ‘reasons’ that are the cause of moral feeling. These reasons are for her, as we recall, simply the moral reasons that weigh in favour of some action as moral or immoral. Telling the story of how a truly rational, non-sensible cause can have a truly sensible effect, though, would involve a more metaphysical sense of the rational causes of moral feeling, a story more about our noumenal, rational natures than about the moral reasons we find in the course of conscious deliberation. Furthermore, in taking this route, we would welcome a certain mystery to our moral experience: moral feeling is very familiar, but we cannot explain how it is that a purely rational cause brings about so common and mundane a sensible feeling. Such a more complex story of the relationship of reason and sensibility is, I think, one more true to Kant's texts; it is, furthermore, my suspicion that telling such a story promises a more satisfying and coherent ground for any Kantian conception of virtue. But Baxley seems unwilling to go down this more metaphysical road.

Indeed, Baxley makes Kant's admonition to us to wonder at the source of the moral law a more mundane, less wondrous thing than it is. Consider her gloss on Kant's claim that we have a duty ‘to cultivate [moral feeling] and to strengthen it through wonder at its inscrutable source’ (MS6:399400; 528–9)’:

Here, Baxley fails to make a distinction between the source of moral feeling (‘the moral law within us’), and moral feeling itself. Our duty is to wonder at the source of moral feeling, but the ‘it’ in the above quotation takes ‘the source of moral feeling’ or ‘the moral law within us’ to be equivalent with ‘moral feeling’ (since that is the only thing that could be said to be ‘entirely different from all other merely pathological feelings’). But Kant isn't enjoining us with a duty to wondrously compare moral and pathological feeling. There is nothing particularly ‘inscrutable’ or wondrous about the feeling itself; it is, indeed, something he considers a common, familiar, even all too painfully obvious a part of our affective life. The duty to wonder involves wondering not at moral feeling, but at the source of moral feeling, which, despite being ‘within us’, is inscrutable. The inscrutability of our rational selves as the cause of a sensible effect is what inspires wonder here. Baxley, however, prefers a more mundane interpretation of such moments in Kant's texts.

There is a final curiosity to note in Baxley's interpretation of moral feeling. The moral feeling she articulates is only that moral feeling which Kant discusses in the section on the four predispositions in the Doctrine of Virtue. Baxley makes little or no effort to consider the extensive discussion of the moral feeling of respect in the second Critique, or to resolve the claims of the Doctrine of Virtue to these earlier discussions. Indeed, Baxley takes this limitation to consideration of the Doctrine of Virtue as a guiding principle in her interpretation.Footnote 6

Ironically, though, the account of moral feeling provided in the Doctrine of Virtue, on its own, does not prove most apt for application to her notion of virtue as autocracy. When Kant speaks of moral feeling in the second Critique, he claims that the first, negative feeling caused by the moral law's effect on our sensibility is one of constraint, even pain, since the moral law is acting to restrain our excessive prioritization of self-love. (See 5: 72–3/63.) It is only after this ‘negative’ moment of moral feeling that a positive feeling akin to pleasure, but not exactly pleasure, is caused. This is a feeling of self-approbation when we recognize ourselves as the source of that which constrained our inclinations. (See 5: 73/63 and 5:77–8/67.) But in the Doctrine of Virtue, we find a different description of ‘moral feeling’. There, one simply feels pleasure in reaction to potentially good actions and displeasure in relation to potentially bad ones.Footnote 7 All mention (at least in that passage) of having a negative or painful reaction to truly moral actions (or perhaps to the action of the moral law upon oneself) is absent.

But this taxonomy of the effects of the moral law on sensibility is problematic for Baxley's overall account. Central to Baxley's account of virtue as autocracy is that, contra Aristotle, even ‘full virtue’ involves the potential for temptation against the moral law. This is even the case when there is no actual temptation. (See Baxley, 133.) This distinctive account of virtue is grounded in Kant's, and Baxley's, commitment to understanding our propensity towards radical evil as invariably operative in the course of human agency. It is a claim at the very heart of both her account of virtue as autocracy and her claim that Kantian virtue is distinctively different from Aristotle's. It is curious, then, that Kant's Doctrine of Virtue account of the law's effect on our sensibility does not track this constant possibility of temptation against the law. That is, it is odd that moral feeling does not include a negative moment of displeasure as a response to moral (not immoral or bad) actions which, like in the second Critique account, would coincide with the constraint of one's excessive pretensions.

So, we have an interesting interpretative problem: if Kant's account of virtue is indeed autocracy, which involves self-governance and ‘control’ of sensibility by rationality (Baxley, 171), then why does his story of reason's effect on sensibility not reflect this? What has changed from the second Critique? Curiously, the second Critique account of moral feeling as involving both displeasure and something akin to pleasure (both in response to moral actions or the influence of the moral law) seems more appropriate to the account of virtue as autocracy.

But there might be an even more complex story to be told here: perhaps the four predispositions of which Kant speaks in the Doctrine of Virtue are simply four different aspects of what he had simply called one ‘moral feeling of respect’ in the second Critique. One could, for example, hope that the niggling voice of conscience in the Doctrine of Virtue might play the role of our negative response to moral demands which takes a more central role in the second Critique discussion of such things, or that the positive predisposition of respect might play the role of our positive response to morality's demands. Baxley does, briefly, consider this last connection between respect in the second Critique and the Doctrine of Virtue (see Baxley, 153–4),Footnote 8 but she cannot raise these interesting questions more broadly, because she has insisted on such a strong distinction between Kant's earlier works and the Metaphysics of Morals. The result, however, is not a convincing argument for how something new is going on in the Metaphysics of Morals, but instead an interpretation of his Metaphysics of Morals account of moral feeling not entirely appropriate to the overall goals of her project.

I thus observe regularly how Baxley's efforts to welcome a more feeling-friendly Kant come up against the epistemological constraints imposed by an Allisonian reading of Kant. Although I hesitate to pronounce firmly on Baxley's motivations in choosing this interpretative path for her understanding of moral feeling, I sense that, by desensibilizing moral feeling, Baxley hopes she is making that feeling more appropriate for integration into an a priori morality; and that by denying that feeling a constitutive role in consciousness of the moral law, she is respecting necessary, unavoidable constraints on the introduction of feeling to that a priori morality. But these efforts face strong textual evidence to the contrary. Ultimately, they reveal the twisting and turning in which Baxley has to engage to make her own feeling-friendly account of virtue fit with Allisonian constraints.

But as we continue down the road toward a more feeling-friendly conception of Kantian virtue, we need to leave the constraints Allison has introduced behind. One has to admit, of course, that Allison has some textual evidence for insisting that feeling can be involved only in motivation to moral action, and not in the justification of the moral law. Yet the general acceptance of this interpretative move by current interpreters can blind us to the fact that there is another story to be told about the involvement of feeling in an a priori morality. This is a story of the grounding of morality in common human judgement, a judgement which takes just this moral feeling as the means by which it comes to terms with the authority of the moral law. Just after the section in the second Critique where we find that passage most sympathetic to the Allisonian view on such matters,Footnote 9 we find the following claim:

Here, Kant is asserting that the common grounding of morality needs to precede any philosophical or ‘scien[tific]’ reflections on that fact of reason which grounds morality. Kant then goes on to describe this common human judgement as a species of judgement that operates not so much by moral feeling from the Doctrine of Virtue as Baxley articulates it, but via the more conflicted second Critique experience of the moral feeling of respect:

Kant is asserting here, contra Allison, that from the common point of view we can rely upon the moral feeling of respect even for the very justification of moral principles. Allison, of course, who avoids the common story Kant tells, is also going to avoid this story. But those more committed to a central role for feeling in Kantian virtue need to welcome this alternative, common story of the grounding of morality as one that permits more freedom in integrating Kant's many discussions of moral feelings. We must, of course, also admit that the full story of Kant's common grounding of philosophy has not yet been told, and that there are various obstacles to overcome in welcoming it. Yet, were Baxley to take this common route, she would not need to do so much twisting and turning in her understanding of moral feeling to make it clean enough to welcome into an a priori morality. The future of Kantian ethics rests, I think, on the willingness of more commentators to welcome the fuller integration of feeling into the grounding of a priori morality which Kant's common grounding of morality encourages.