1

Kant begins the Religion by describing a dispute between those who accept the view, characteristic of historical religions, that ‘the world lieth in evil’ and the recent pedagogues and philosophers who have begun to argue, on the other hand, that ‘the world steadfastly … forges ahead in the very opposite direction, namely from bad to better’ (R, 6: 19–20). His strategy for intervening in this debate is to investigate how one can be judged as either good or evil in the first place. He argues that our moral status is determined by our free choice of maxims. Specifically, Kant endorses a view called rigorism which claims that agents must be either good or evil, but not a mixture of both, based on the maxims that they adopt.

Kant argues for this view by appealing to what has become known as the incorporation thesis. He claims that the answer to the question of whether human beings are good or evil ‘is based on the morally important observation that freedom of the power of choice has the characteristic, entirely peculiar to it, that it cannot be determined to action except so far as the human being has incorporated it into his maxim’ (R, 6: 24).Footnote ⁵ Briefly put, the incorporation thesis states that incentives cannot determine an agent to act except insofar as she incorporates these incentives into a maxim. By ‘incentives’, Kant has in mind both the desires that come from the lower motivational faculties and the moral law which has its source in the faculty of reason (Schapiro Reference Schapiro2011: 149). He employs the incorporation thesis in the following argument:

1. 1. The moral law can be a sufficient incentive for action.

2. 2. If the moral law is a sufficient incentive for a subject S’s action, then their action is good.

3. 3. Self-love can also be a sufficient incentive for action.

4. 4. The distinction between the incentives of self-love and morality is exhaustive.

5. 5. For any particular action A, only one of the incentives of self-love or the moral law can be sufficient for S to A.

6. 6. If S’s sufficient incentive for A is something other than the moral law, then it must be self-love. (4, 5)

7. 7. When self-love suffices for S to A in morally significant circumstances, the action A is evil.Footnote ⁶

8. 8. A person’s actions must be either good or evil. (2, 6, 7)

This argument moves rather quickly – it takes up just about half of a page of the Cambridge edition of the Religion Footnote ⁷ – largely because it relies on several assumptions that Kant has defended in his earlier writings, and I will not attempt to give a further defence of it here.Footnote ⁸ For our purposes, the main difficulty with this argument is that it does not seem to establish rigorism of one’s character. Proposition (8) is a conclusion that a person’s actions must be either morally good or evil. However, one may object that a person can do both good and evil deeds and resultingly have a character that is either both good and evil in parts or neither good nor evil but something in between. As presented, this argument does not rule out this possibility, but Kant intends for rigorism to apply to both actions and agents. How can he secure this conclusion? The argument for rigorism of character is also quite brief:

Kant argues by assuming for reductio that there is a person who is both good and evil. This person is not necessarily motivated by the moral law, because for at least some actions he takes self-love and not the moral law to be a sufficient incentive to act. However, Kant holds that the moral law obligates us with unconditional necessity (G, 4: 416). Accepting the authority of morality means that one must always act out of duty in the relevant circumstances. The person who is both good and evil would simultaneously have both the maxim characteristic of a good will – stated roughly, ‘Always perform one’s moral duty when this is required’ – and the evil maxim ‘Self-love can be a sufficient ground to act, even if this contravenes one’s duty.’ This is contradictory; one cannot rationally hold both these maxims at the same time and so one cannot be good or evil in parts.Footnote ⁹

Kant does not think that a person could abjure self-love entirely, nor could one fail to be receptive to the moral law.Footnote ¹⁰ Because we are constantly confronted by the two incentives of morality and self-love in our practical engagement with the world, we must choose how to prioritize them. Furthermore, this rule or principle of action has a propositional structure and hence can be expressed as a maxim that describes which incentive a person ‘makes the condition of the other’ (R, 6: 36). Because this maxim has a higher level of generality in relation to other maxims, some commentators have called it a fundamental maxim.Footnote ¹¹ Kant writes that a good fundamental maxim should be understood as making the moral law ‘the supreme condition of the satisfaction of [self-love]’ (R, 6: 36). Thus, the good fundamental maxim can be modelled as:

Kant thinks that we have imperfect duties as well as perfect duties. We have a wide obligation to fulfil our imperfect duties, which means that there are various possible ways that we can discharge these obligations. An outsized commitment to self-love that would impede a person from fulfilling their imperfect duties to the extent required would belie an evil character even though the obligation to fulfil imperfect duties is indeterminate.

While the good fundamental maxim makes acting on self-love conditional on such actions being morally permissible, the person with an evil fundamental maxim denies this requirement, holding:

Intentionally acting against what morality commands, even one time, would entail that one has an evil character at that time.

Kant notes that these fundamental maxims describe a general attitude that an agent has towards the rational authority of the moral law and calls this an agent’s Gesinnung or disposition. Because Kant thinks that a person is either good or evil based on their Gesinnung, it itself must be freely adopted, as Kant holds that a person can only be morally judged by that which can be imputed to them (R, 6: 20). He defines the Gesinnung as ‘the first subjective ground of the adoption of the maxims’ which ‘applies to freedom universally’ (R, 6: 25). The Gesinnung is a ground (Grund) of particular maxims.Footnote ¹² Kant writes, for example, regarding the evil fundamental maxim: ‘In order, then, to call a human being evil, it must be possible to infer a priori … an underlying evil maxim, and, from this, the presence in the subject of a common ground, itself a maxim, of all particular morally evil maxims’ (R, 6: 20). As Paul Guyer notes, the fundamental maxim serves as the ‘basis of which’ an agent adopts more particular maxims (Reference Guyer2016b: 181). In this sense, the good or evil Gesinnung serves as a necessary condition for good or evil actions respectively and is logically prior to these actions.Footnote ¹³

If Kant conceives of a Gesinnung as a ground that applies to our free actions universally, then it seems that a person with a good Gesinnung could not act evilly and a person with an evil Gesinnung could not act from duty. Most commentators want to attribute a more moderate view to Kant where a person can have a good Gesinnung but nevertheless act evilly or have an evil Gesinnung but act from duty.Footnote ¹⁴ One passage suggests such a moderate interpretation:

This passage appears to say that a person who is not a ‘morally good human being’ may have the moral law as her sole incentive in action sometimes but not always, while the morally good person always does. However, Kant continues directly: ‘We can say of the first that he complies with the law according to the letter (i.e. as regards the action commanded by the law); but of the second, that he observes it according to the spirit (the spirit of the moral law consists in the law being of itself a sufficient incentive)’ (R, 6: 30). Kant is saying here that an agent with an evil fundamental maxim would act, at best, in conformity with the moral law whereas only those with a good fundamental maxim can have the moral law as a sufficient incentive, which is what dutiful action requires.Footnote ¹⁵

Kant also argues that the Gesinnung can be chosen. His commitment to transcendental idealism means that there can be no temporal origin of our moral character, but rather our character must be grounded in a free act that occurs outside of space and time, which he at one point calls an ‘intelligible deed’ (R, 6: 31). Kant accepts this because he holds that no matter how wicked a person has been or what the current natural circumstances are, it must be possible for an evil person to adopt a good Gesinnung. This commitment raises a difficulty for Kant because the choice of transforming one’s moral disposition, conceptually, cannot have its basis in a person’s current Gesinnung. In this sense, the choice to transform one’s Gesinnung must be groundless. The possibility that one can transform their Gesinnung is in tension with viewing the Gesinnung as a universal ground of one’s actions – or so it would seem. Kant is aware of this challenge and responds directly to it:

Kant appeals to the principle ‘ought implies can’ to explain this, noting that we have a duty to adopt a good fundamental maxim so this must be possible: ‘the command that we ought to become better human beings still resounds unabated in our souls; consequently, we must also be capable of it’ (R, 6: 45).Footnote ¹⁶ Moral reform must be really possible even for an agent who has an evil character because otherwise someone who is evil would have no duty to become good, and this contravenes one of Kant’s fundamental philosophical commitments.

Kant concedes that we cannot conceive of how the Gesinnung can be transformed. But he notes that what is conceivable does not necessarily track what is really possible. Freedom is, famously for Kant, an ‘inscrutable faculty’ (CPrR, 5: 47), and we can not cognize the grounds by which the moral law can serve as a determining ground of our will (CPrR, 5: 72). Nor can we conceive of how an agent who has been endowed with a predisposition to the good has chosen an evil character, any more than we can understand how someone who is evil would choose the good. But we know that both of these choices are really possible, the former because experience has testified to this fact and the latter because it is demanded by pure practical reason. Kant’s theory of freedom entails that the choice of either an evil or good disposition is always really possible for human beings.Footnote ¹⁷

2

The freedom to choose one’s Gesinnung is of crucial importance because Kant holds that, because of the universal propensity to evil, all human beings – even those who seem most morally upright – must be thought to initially possess an evil fundamental maxim. Hence no person is exempted from the task of replacing an evil fundamental maxim with a good one.Footnote ¹⁸ Although the intelligible choice of one’s disposition is inscrutable, Kant does not shy away from discussing the process of becoming a good human being in detail. The main passage where he discusses this is quite complex, and because it will serve as the foundation for much of the rest of my interpretation, I will present it in full:

At the start of this passage, Kant reiterates his claim that the transformation of one’s character is really possible. He then claims that both a ‘revolution … in the mode of thought’ and ‘a gradual reformation in the mode of sense’ are required to become a good human being (R, 6: 47). The former makes ‘one receptive to the good’ but does not yet suffice for possessing a good character; it must be supplemented by ‘incessant laboring and becoming’ (R, 6: 48). While human beings can only hope to continually progress ‘from bad to better’ they can hope that God, who can determine the nature of one’s fundamental maxim, can judge this progress to be ‘a unity’ (R, 6: 48). I will use the term ‘change of heart’ to describe an agent’s adopting a good fundamental maxim. I will call an agent’s attempt to possess a good character through incessant labouring ‘moral labour’.

Initially, this account may seem puzzling. In the quoted passage, Kant contends that ‘reversing the supreme ground of one’s maxims’ is necessary but not sufficient for being good. One becomes a good person through ‘incessant laboring and becoming’ (R, 6: 48). However, as we have seen earlier, Kant has argued that only ‘when a human being has incorporated into his maxim the incentive implanted to him for the moral law, is he called a good human being’ (R, 6: 45n.; see also R, 6: 44). So, Kant appears to give two answers to the question of how to become a good human being. It looks like Kant wants to have things both ways, claiming that the transformation of one’s character occurs both through adopting a good fundamental maxim and gradually through a continuous process of moral labour. I will call this the ‘inconsistency problem’.

Commentators have tended to emphasize one of these two aspects of Kant’s thought regarding moral transformation. Daniel O’Connor claims that a ‘change of disposition is by sudden revolution (conversion) not by gradual reformation’ (Reference O’Connor1985: 300). Mavis Biss contrasts the ‘revolutionary approach’ of the Religion with the ‘gradualist view’ of the Doctrine of Virtue (Reference Biss2015: 3). On the other hand, some commentators have argued that Kant is not committed to the intelligible choice of a disposition. Henry Allison argues that Kant does not think that the choice of a Gesinnung is ‘like a choice of a disposition or character in a full-blown psychological sense’ (Allison Reference Allison1990: 142). On Allison’s view, the Gesinnung should be understood as more like a Denkungsart, or a set of principles. Julia Peters on the other hand denies that we can choose our Gesinnung at any one moment. On Peters’ view, the Gesinnung is not a disposition that precedes and causally determines us to act either out of duty or self-interest depending on whether it is good or evil. The good Gesinnung, for Peters, is not ‘fully present at the time at which a particular moral choice is made’ (Reference Peters2018: 507). Peters claims, rather, that a good Gesinnung must be constructed through a series of dutiful moral deeds over time (Reference Peters2018: 516). Both Peters and Allison reject the idea that one’s character and subsequent empirical history is fixed by a choice of disposition.Footnote ¹⁹

Kant can hold that moral reform requires both a ‘change of heart’ and gradual reform by appealing to his transcendental idealism. His rigorism suggests that we must understand the change of Gesinnung as a revolution from evil to good; there are only two Gesinnungen and, though Kant thinks there are several grades of evil (R, 6: 29–30), the Gesinnungen do not admit of degrees. But Kant never does say that the change of Gesinnung occurs instantaneously; this itself is a temporal term that is inapplicable at the noumenal level. The coherence of Kant’s programme would indeed be threatened if he were describing two distinct undertakings that both independently suffice for securing a good disposition: one ‘noumenal’ choice that occurs instantaneously and a second, phenomenal struggle that occurs over time. But Kant need not endorse a ‘revolutionary’ view that reform happens instantaneously by contrast with a ‘gradualist’ account. Rather, because the choice of disposition is not spatio-temporal, he is not committed to an understanding of this that would conflict with a theory of phenomenal moral reform. On the contrary, the change of disposition needs to be compatible with how we are to understand the efforts to become good that we undertake as empirical agents.

In part two of the Religion Kant further clarifies that becoming a good human being requires us not only to achieve but to preserve a good disposition. He notes that although we must always judge our continual moral conduct in this life to be ‘defective’, because even an uninterrupted series of good deeds is not enough to show that one accepts the unconditional obligation of morality, God can judge the ‘disposition from which [our conduct] derives … to be a perfected whole’ (R, 6: 67). Being a good human being, which Kant sometimes describes as being ‘well pleasing to God’, requires our ‘endless progress’ in the phenomenal world to be ‘a unity’ (R, 6: 48). That is, to be a good human being an agent’s conduct throughout her life (following a moral conversion) must be grounded in a good Gesinnung. As I have argued in the last section, Kant’s theory of freedom implies that a person with a good disposition can nevertheless relapse into evil, and such a person would not be pleasing to God unless he or she restores and maintains her good disposition. In my following analysis, I will take care to distinguish between having a good Gesinnung or disposition and being a good human being (or, equivalently, being ‘pleasing to God’), as the latter term describes a person whose post-conversion conduct is grounded in a good Gesinnung.

Commentators have questioned whether Kant can say anything plausible about how agents can possess a stable moral character. As previously noted, Michalson argues that because Kant thinks that one's character is outside of space and time, he is unable ‘to give sense to any notion of continuity, over time, in the life of the moral agent … he cannot show how a “previous” act or moral condition would be relevant to a “present” act’ (Reference Michalson1990: 85). And O’Connor also raises an objection that the commitment to the noumenal will in his critical philosophy leaves Kant unable to account for how temporal factors such as moral education or one’s efforts to control their emotions can lead to a stable disposition, because ‘nothing outside or inside’ one’s will can determine that it act in one way or another (Reference O’Connor1985: 294). O’Connor concludes that Kant’s account of the change of heart involves an implausible conflation between Sartre’s conception of a fundamental project, which implies the possibility of radical change regardless of one’s current status, and Aristotle’s account of a hexis, which is a continuous and stable moral disposition cultivated through moral education and the proper moral practices (Reference O’Connor1985: 293–6).Footnote ²⁰ Because Kant insists on the freedom to choose one’s disposition no matter one’s previous history, he seems to have no way to explain how an agent’s past moral efforts would lead her to preserve her good disposition when she possesses it. I will call this the ‘stability problem’.

Kant does have resources to respond to this problem. Although he denies that we should look for a temporal origin for a human being’s character, he does argue that there are specific empirical practices that can give coherence to an agent’s moral life. Furthermore, as we will see, these practices are morally significant – they can play a role in an agent either adopting or maintaining a good moral disposition. To see how Kant can think this, we will have to look in detail at his theory of phenomenal moral reform.

3

For Kant, phenomenal moral reform has two stages: in the first stage an agent undergoes a revolution of thought, which I will also refer to as a ‘transformation of attitude’. This is accomplished by making a commitment to the moral law. In the second stage of moral reform an agent continuously labours to live up to her moral commitment; this involves the cultivation of moral emotions.Footnote ²¹ This section will discuss the first stage of Kant’s account and will focus in particular on the moral significance of the transformation of attitude.

Much of Kant’s discussion of the moral transformation at the end of part one of the Religion describes various empirical practices. Kant claims that the moral revolution which leads a person to become good is part of ‘moral discipline’ (R, 6: 51). In the central passage 6: 47–8, he claims that this ‘revolution’ that must accompany the ‘gradual reformation in the mode of sense’ is a revolution ‘in the mode of thought (Denkungsart)’ (R, 6: 47). So, although Kant contrasts the gradual reformation of sense with the revolution of thought, these both refer to actions that we undertake in the empirical world and are part of a process of phenomenal reform. Kant re-emphasizes this when he notes that ‘a human being’s moral education must begin, not with an improvement of mores, but with the transformation of his attitude of mind and the establishment of a character’ (R, 6: 48). A transformation of one’s ‘attitude of mind’ suggests not an intelligible deed but an agent’s attempt in the empirical world to radically restructure the way that she relates to her moral duties. Because Kant’s discussion is confined to the phenomenal realm here, this should be understood as the establishment of empirical character.

Kant discusses the formation of empirical character in most detail in the Anthropology where he describes character as the ‘property of the will by which the subject binds himself to definite practical principles that he has prescribed to himself irrevocably by his own reason’ (7: 292). Because one can bind oneself to both good or evil principles, one can have either a good or evil empirical character. But, of course, what is relevant for Kant’s discussion in the Religion is the establishment of a good character, whereby a subject forms a commitment to the moral law in its purity. Kant reiterates that this establishment of empirical character can be considered a type of transformation: ‘one may also assume that the grounding of character is like a kind of rebirth, a certain solemnity of making a vow to oneself; which makes the resolution and the moment when this transformation took place unforgettable to him’ (Anth, 7: 294). Education and teaching cannot ‘bring about this firmness and persistence in principles gradually, but only, as it were, by an explosion which happens one time’ (Anth, 7: 294). The reason, presumably, why this cannot happen gradually is because it requires an agent to endorse a general and unitary commitment to morality, and this is something that must occur at one time. As Kant argues in the Religion, this decision must be ‘single and unalterable’ (R, 6: 48); the agent who makes a genuine commitment to morality must intend to live up to this commitment throughout his entire life as this is what one’s moral obligation demands.Footnote ²² Such a decision recontextualizes the practical conduct that follows it. The agent who attempts to establish a good character could live with an unwavering adherence to morality and as such would become a good human being. The person who fails to do this is evil, just like he was before this transformation, but this failure is contextualized as a (culpable) inability to follow through with the commitment that he made in utmost seriousness.

We should understand this decision, which Kant sometimes describes using religious terminology as the decision to become a ‘new man’ (R, 6: 48), as expressing a commitment to morality. This commitment, as something done qua phenomenal agent, is not identical with the intelligible choice of a good disposition. In this sense, there is an ambiguity in Kant’s discussion of what it means to endorse a good maxim in the Religion. Considered at the noumenal level, it describes the intelligible act to make the moral law one’s incentive in action. But considered empirically endorsing a good maxim is aspirational; this is why Kant claims in Religion, 6: 47–8 that changing one’s maxims makes one ‘receptive to the good’ but does not imply that a person is good as such (R, 6: 48).

Nevertheless, Kant clearly thinks that such a decision is morally significant – it is a necessary part of becoming a good human being. The reason why he thinks this, however, is initially not entirely clear. At one point, Kant claims that absent making a commitment to the moral law an agent can be at best ‘legally good’ (R, 6: 47), which he understands in the Religion as acting merely in conformity with morality. The person who seeks to become good without a transformation of attitude would not be genuinely acting from duty. But this answer is question-begging; it is not clear why continuous dutiful action requires an agent to make such a moral commitment. Kant comes closer to a satisfying response when he notes that, without transformation of attitude, an agent will ‘fight vices individually, while leaving their universal root undisturbed’ (R, 6: 48). He makes a similar point in the Anthropology, noting that ‘wanting to become a better human being in a fragmentary way is a futile endeavor, since one impression dies out while one works on another; the grounding of character, however, is absolute unity of the inner principle of conduct as such’ (7: 294–5). Kant’s suggestion is that by making a moral commitment an agent accepts a principle that serves to unify his or her conduct.

This point can become more plausible still if we understand the transformation of attitude as involving more than an abstract commitment to morality; the revolution of thought grounds specific moral duties which an agent undertakes in an effort to combat certain vices. As Kant makes clear, one of the main sources of evil arises due to agents repressing or otherwise not acknowledging the extent of their moral vocation. Certain commentators, most notably Laura Papish, have argued that self-deception and evil are closely connected for Kant (Reference Papish2018: 87–115). This is plausible because Kant denies that human beings are diabolical, which he takes to mean that they do not incorporate ‘evil qua evil’ into their maxim (R, 6: 37).Footnote ²³ In the Groundwork, Kant argues that we ‘like to flatter ourselves by falsely attributing to ourselves a nobler motive [than the one that actually guides our actions]’ (4: 407). In the Religion, he discusses a related form of self-deception which he calls ‘deliberate guilt’ (R, 6: 38). This is guilt in which an evil heart deceives ‘itself as regards its own good or evil disposition’ by ‘not troubling itself on account of its disposition’ but instead taking itself to be ‘justified before the law’ (R, 6: 38). Deliberate guilt may characterize someone who knows that she has done an evil deed but quiets the demands of her conscience by considering herself to be good despite these failings. A genuine commitment to possessing a good character requires an agent to reject the incomplete and exculpatory picture of her conduct that she has constructed for herself and take responsibility for her past evil actions.Footnote ²⁴ Accordingly, this moral commitment requires an agent to combat the specific vice of downplaying or ignoring one’s moral transgressions. The transformation of attitude, then, grounds particular duties of moral self-examination. It is unsurprising that in the Anthropology Kant claims that character ultimately requires possessing a maxim of truthfulness: ‘the only proof within a human being’s consciousness that he has character is that he has made truthfulness his supreme maxim’ (7: 295). The call to establish one’s character requires an agent to examine his motives in order to discern that he is genuinely living up to his moral commitment. Insofar as deliberate guilt is an expression of a vicious state of mind, establishing a good character in an attempt to resist this is itself a necessary part of moral reform.

However, here this proposal faces an objection. Kant claims that we cannot know our fundamental maxim. As such, we can never be certain that we are indeed acting out of duty. This scepticism is best understood as scepticism about the ultimate grounds of our actions, and it threatens to undermine the very coherence of a duty of self-examination.Footnote ²⁵ Fully addressing this issue would require a more extensive discussion than what I can give here, so I will confine myself to a few brief points. To start, we should not think that this scepticism by itself undermines the possibility and practical significance of moral self-examination. While Kant thinks that we can never know that we are acting purely from duty, as no matter how dutiful our actions may seem it is always possible that we are being covertly guided by self-love, self-examination can reveal that we are evil. This is because Kant presupposes that we have some basic knowledge of our intentions and our reasons for action. A person can know that they have, for example, made a cutting remark to offend a colleague – knowledge of this sort often does not require much if any sophisticated reflection at all. When reflection reveals violations of the moral law, moreover, it is very rare for wrongdoers to be able to plausibly claim that they genuinely had a good maxim and have unfortunately made an exculpatory error. For Kant, then, there is an asymmetry between good and evil conduct: whether we are genuinely acting from duty is opaque, but evil actions can be transparent to us.Footnote ²⁶ Self-examination can reveal that an agent must redouble his efforts to become good.

Here is, briefly, what self-examination involves for Kant.Footnote ²⁷ The general commitment to morality engendered by the revolution of thought does not specify what one’s particular duties must be; self-examination involves determining what one’s commitment to morality requires. Moral self-examination conceived in this way is not primarily about a person peering into the ultimate motives that guide her. But it can help the agent who undertakes it understand the nature and extent of her moral obligations. So, the self-examination that is required by a genuine commitment to morality is not undermined by the scepticism of grounds of action, which for Kant is an inextirpable feature of moral life. Kant’s discussion of conscience supports this proposal. Kant claims that ‘an erring conscience is an absurdity’ (MM, 6: 401).Footnote ²⁸ His point here is not that all of our motives for action are transparent to us but rather: ‘While I can indeed be mistaken at times in my objective judgment as to whether something is a duty or not, I cannot be mistaken in my subjective judgment as to whether I have submitted it to my practical reason’ (MM, 6: 401). While we can be mistaken about whether we are acting for duty or promoting a good end, we cannot be mistaken about whether we are assessing whether our actions are indeed morally permissible.Footnote ²⁹ Kant is not surreptitiously overstepping his own epistemic restrictions in his discussion of self-examination.

4

Kant’s discussion of a transformation of attitude reveals one part of his response to the stability problem. It suggests that, insofar as we have a coherent moral identity, it must be constructed by forming a commitment to morality, which is a transformative and significant episode in an agent’s life, and continuously striving to live up to this commitment. As we will see now, Kant also argues that the agent who engages in this continuous moral struggle aims to instantiate certain ethical ideals, and that such moral labour has an influence on the moral psychology of the agent that undertakes it. Specifically, in the Metaphysics of Morals he claims that agents cultivate virtue through continuous moral struggle.

It takes Kant some time to arrive at this position, however, because his views about the nature and efficacy of moral labour shift over time. In the Critique of Practical Reason, Kant writes that we must hope to attain ‘the complete conformity of dispositions with the moral law [as] the supreme condition of the highest good’ (CPrR, 5: 122). He defines ‘complete conformity’ to morality as ‘holiness’ and argues that such holiness may be found through ‘endless progress toward’ this ideal (CPrR, 5: 122). Kant then argues that because we require endless progress to instantiate the ideal of holiness, we must postulate the immortality of the soul. This proposal however has several serious problems. As Guyer points out, it is unclear why Kant presents holiness an ethical ideal (Guyer Reference Guyer and Höwing2016a: 167). If Kant denies, as he does, that human beings can be holy then it is hard to see how gradual progress can lead a person to become holy even if such progress were extended indefinitely into another life.

It is tempting to think that Kant has rejected this account by the time that he wrote the Religion. In the Religion Kant makes no explicit reference to the immortality of the soul; his view that God can take one’s progress to be a unity can be interpreted as replacing the doctrine of immortality, because it entails that we can be pleasing to God if our continuous moral conduct in this life is grounded in a good fundamental maxim. However, there are signs that Kant has not completely abandoned his views from the second Critique. In part one of the Religion he writes: ‘The original good is holiness of maxims in the compliance to one’s duty … whereby a human being, who incorporates this purity into his maxims, though on this account still not holy as such (for between maxim and deed there still is a wide gap), is nonetheless upon the road of endless progress to holiness’ (R, 6: 46–7). This passage draws a subtle distinction between holiness of maxims and being holy in general.Footnote ³⁰ The person who has holy maxims is motivated to act solely out of duty by pure practical reason. Holiness of one’s maxims does not make one holy as such, and indeed this is not possible for human beings. However, at the end of this quote Kant returns to the claim that adopting such a maxim would lead one to be on the road to holiness. Thus, in the Religion, Kant maintains that holiness is an ethical ideal and the relationship here between endless progress and holiness remains extraordinarily opaque, just as it was in the second Critique.

In the Doctrine of Virtue of the Metaphysics of Morals Kant once again revises his views about the ethical ideal towards which our conduct is directed.Footnote ³¹ He replaces the ideal of holiness with the ideal of the autocratic will that has the strength to execute its moral duty. Anne Margaret Baxley notes that autocracy is ‘the ideal form of moral self-governance for merely finite rational beings’ (Reference Baxley2010: 49). The autocratic agent possesses virtue, which Kant understands as ‘the moral strength of a human being’s will in fulfilling his duty … insofar as this constitutes itself an authority executing the law’ (MM, 6: 405). Unlike holiness, which characterizes an agent who does not experience any impulses that would impel her to act contrary to the moral law, the virtuous person still represents moral laws as commands that she must dutifully follow. This shift implies that we reconceive the continual moral struggle as a struggle to develop the strength required to put one’s moral intentions into practice.

Kant describes the person who lacks virtue as follows: ‘weakness in the use of one’s understanding coupled with the strength of one’s emotions is only a lack of virtue … which can indeed coexist with even the best will’ (MM, 6: 408). While Kant does not employ the language of the Gesinnung here, his observation can bear on his discussion of the transformation of disposition in the Religion. His thought is that even a person who possesses a good will or, in the parlance of the Religion, a good Gesinnung, is not virtuous. This is because virtue must be developed over time.

Here is why Kant thinks this. Kant holds that virtue involves the cultivation of moral emotions. The primary moral emotions that Kant discusses in relation to virtue are moral feeling and conscience. In describing moral feeling, Kant notes:

For Kant, we do not have a duty to possess moral feeling, since this is a capacity that humans are endowed with in general; we are susceptible ‘to feel pleasure or displeasure merely from being aware that our actions are consistent or contrary to the law of duty’ (MM, 6: 399). Kant’s theory of the moral emotions is teleological in the sense that he takes these emotions to develop from our particular dutiful actions. His discussion of beneficence illustrates this:

It is not the case that we ought to start with the cultivation of emotions and hope that these emotions would lead to dutiful action. Rather, we should employ our faculty of the understanding to determine how to fulfil particular duties. Dutiful action, Kant claims, would lead to the development of the moral feeling of love. Similarly, we have an indirect duty to not ‘avoid the places where the poor who lack the most basic necessities are to be found but rather to seek them out, and not to shun sickrooms or debtor’s prisons and so forth in order to avoid sharing painful feelings’ (MM, 6: 457). Though we may experience painful feelings by seeking out those who are less fortunate, this experience is a vivid reminder of our imperfect duties of charity.

There is much more that can be said about Kant’s account of the moral emotions.Footnote ³² But this brief discussion shows how it informs Kant’s theory of moral reform and reveals how he can answer the stability problem. As we have seen, the transformation of attitude requires an agent to combat the vices of self-deception such as deliberate guilt. I have argued that duties of self-examination can be discharged by determining what this moral commitment requires and performing particular good deeds based on this commitment. In Kant’s account of virtue, he argues that an agent cultivates moral emotions through particular dutiful actions. By performing good actions an agent will strengthen her faculties of feeling and her responsiveness towards her moral duty, which will allow her to more easily demonstrate virtue when her will is tested in the future.

5

Kant’s account of moral reform seems to involve an unstable combination of two core commitments. The first is that one is always free to adopt a good or evil disposition by an act of the noumenal will that occurs outside of space and time and cannot be determined by natural or temporal factors, and the second is that one becomes a good human being through an incessant temporal struggle. I have argued that a person’s Gesinnung is the result of a free intelligible choice, and that one is always free to adopt either a good or evil disposition. But I have also argued that this thesis is not in tension with Kant’s claim that moral reform requires incessant labouring and becoming. If we understand the Gesinnung as a disposition to prioritize the moral law over self-love, there is no reason to think that this cannot be secured over time – though not in time – through continuous moral struggle. Following a conversion, one is judged by God to be a good human being only if her subsequent action is grounded in a good Gesinnung.

In the Groundwork, Kant notes that even the good will that achieves no good ends would ‘like a jewel … shine by itself, as something that has full worth in itself’ (4: 394). In his later practical writings he argues that stability of character comes from setting and achieving morally good ends; Kant thinks that we must construct and preserve our moral identity through the proper moral practices that structure our ethical life. The agent who undergoes a transformation of attitude and strives to put her moral principles into practice will over time acknowledge the ‘perceived … efficacy of these [good] principles’ on her conduct, which would give ‘cause to infer … a fundamental improvement in [her] disposition’ though – in accordance with Kant’s epistemic restrictions – such an inference remains merely a ‘conjecture’ (R, 6: 68).

The account of moral reform that Kant develops does not imply that he abandons the view that we can always choose our moral disposition at any point in our natural lives. Kant’s theory of freedom implies that there is always a precarity to human beings’ efforts to become good. Because he thinks that cognition is limited to the sensible conditions of space and time, one cannot ever know that he or she genuinely possesses a good disposition, and the human capacity for self-deception suggests that moral failures may become unnoticed or entrenched for agents who do not continuously affirm their moral commitment. In this way, Kant’s transcendental idealism informs, rather than threatens, a rich conception of the moral life.