Criticism One

It is a central concern of much of the recent philosophical literature concerning moral evil to distinguish evil from mundane wrongdoing.Footnote ²⁰ This distinction is crucial to any proposed definition of evil because it demonstrates that evil is its own moral category identifying a feature of moral interaction that is irreducible to other moral categories such as bad, wrong, unvirtuous or unjust.Footnote ²¹

The motivation behind drawing this conceptual distinction rests on the insight that some acts or persons ought to be described by a term of moral opprobrium not satisfied by terms like ‘wrong’ or ‘bad’. Certain acts such as genocide, torture and forced starvation, and perhaps individuals capable of such acts, are not properly described as ‘wrong’ or ‘bad’ no matter how many times we preface these terms with ‘very’ or ‘really’. The moral category of evil performs an explanatory role insofar as it properly identifies the morally worst kind of acts.Footnote ²² Hence, the term ‘evil’ purports to have ontological and normative priority because it describes an extant category of agential action and affords apt normative judgements of acts falling under that category.

If it is indeed appropriate to distinguish between evil and mundane wrongdoing, then Kant’s theory of evil, it is charged, is apparently ill equipped to do so. Under the categories of evil described by the three propensities, Kant includes acts that are, in fact, not evil outside of a Kantian context.Footnote ²³ For example, recall Kant’s description of the propensity for impurity. The impure individual conforms her acts to the moral law, but does so also out of sensuous incentives rather than solely out of duty to the moral law. Consequently, according to Kant, an individual is evil if she tells the truth in order to gain a good reputation rather than solely out of duty to the moral law. Yet, outside of this Kantian context, telling the truth for non-moral reasons is not morally worse than other acts of mundane wrongdoing, and therefore cannot be an example of evil. Telling the truth to establish a good reputation is altogether ordinary and not of ‘nightmarish horror’. Accordingly, many of the acts that Kant refers to as evil can easily be captured by the category of mundane or ordinary wrongdoing. In fact, it is difficult to comprehend how any of the acts categorized as ‘impure’ by Kant could be acts that are evil in contrast to merely wrong. Since acts of impurity will always conform to the moral law (but also be done for wrong reasons), they seem to be excluded by definition from the horrendous or extremely harmful acts that characterize evil.Footnote ²⁴ As we have already noted, for the concept of evil to have explanatory value it ought to identify acts, motives, intentions or traits that are morally worse than merely wrong ones. Otherwise, the term ‘evil’ is simply emotional hyperbole.Footnote ²⁵

Kant’s conflation of evil with mundane wrongdoing is confirmed not only by the relevant passages in the Religion but also in the Critique of Practical Reason.Footnote ²⁶ Here Kant distinguishes between two senses of ‘good’ and their respective negative analogues. He contrasts (a) the good (das Gute) and evil (das Böse), and (b) the good (das Wohl) and the bad (das Übel or das Weh) (CPrR, 5: 59–60). Under this description, to refer to something as bad in the sense of (b) is to call something harmful to one’s well-being.Footnote ²⁷ It is meant to refer to disagreeable, yet entirely accidental untoward and unfortunate circumstances. In contrast, to refer to something as evil in the sense of (a) is to call an action morally wrong, ‘resulting not from natural contingency but from the direct action of an agent’s will – when she wrongly subordinates the principle of morality to the principle of self-love’ (Garcia Reference Garcia2002: 195).

With this distinction juxtaposed with his discussion of evil in the Religion, we see that what Kant refers to as ‘evil’ should really be categorized under the broader class of ‘all immoral acts’. Thus it seems that, under a strict Kantian framework, there is no unique conception of ‘evil’ in contrast to mere or mundane immoral acts.Footnote ²⁸

Reply One

Is Kant guilty of conflating mundane wrongdoing and evil? Briefly put: yes. As just discussed, Kant’s theory of evil is in fact a discussion of all of the possible ways of committing immoral acts generally speaking. One commits immoral acts through the propensities to frailty, impurity or depravity. Although it is possible to identify acts of evil as one kind of immoral act falling under these broad categories, Kant does not carve out a special category of evil in distinction to mundane wrongdoing.

Nevertheless, it would be uncharitable to Kant to describe this conflation as culpable. We need to keep in mind Kant’s main goal in the Religion. His principal objective is to explain religion in terms of rational morality. As indicated by the title, the book purports to explore what reason can provide us in the way of religion without resorting to divine revelation. In the Preface to the second edition Kant writes:

Kant’s purpose is precisely indicated by the book’s title. He intends to investigate religion within the bounds of reason. Since religious belief supported by revelation is empirical in its source, it is the appropriate content of theological, but not philosophical inquiry. As a philosopher, Kant’s goal is to remove ‘everything empirical’ (Rel, 6: 12–13) from his inquiry into religion and religious doctrine. (Nevertheless, he believes that between his methodology and biblical scripture ‘there is not only compatibility but also unity, so that whoever follows the one (under the guidance of moral concepts) will not fail to come across the other as well’, Rel, 6: 13.)

What is more, his goal is to pursue this investigation for a specific audience of late eighteenth-century German Lutherans. Kant expects his audience to hold certain religious beliefs that merge with their life as rational moral agents. His aim is to convince them that there is ‘no fundamental conflict between these two aspects of their moral life’ (Wood Reference Wood2014: 31). To this end, he adopts certain language that was more common in a religious rather than a secular philosophical setting. It is understandable that he chose to focus on evil (das Böse), which plays a central role in Christianity generally and Lutheranism in particular. Finally, Kant’s discussion is less focused on what the nature of evil is, but on transcendental evil – how evil occurs, how its occurrence is compatible with freedom and how we can understand it through reason rather than revelation.Footnote ²⁹

With these facts in mind, we see that it was simply not Kant’s goal to explicate the difference between evil and mundane wrongdoing. It is perhaps anachronistic in a way to expect him to have pursued a conceptual distinction that has proven to be philosophically significant through contemporary discussions of moral evil. Rather than asking if Kant distinguished between evil and mere wrongdoing, it is more philosophically generous to ask whether Kant could have distinguished between these two concepts given what he says about evil specifically and immorality more generally.

It is possible to answer this question affirmatively. Although Kant left the conceptually possible distinction between evil and mundane wrongdoing unexplored, Ernesto Garcia argues that one can reasonably modify the Kantian ethical theory in order to locate ‘a fundamental qualitative difference between evil and more ordinary immoral actions … in which … a “material” (as opposed to purely “formal”) difference exists between the respective maxims of the immoral and the evil agent’ (Garcia Reference Garcia2002: 194). It is a hallmark of Kantian ethics that the form of the maxim and not its matter determines the moral worth of an action (G, 4: 399–400). The moral law obligates us formally to recognize humanity as an objective end comprised of all rational beings and regarded as an ‘end in itself’ (G, 4: 429). Conversely, immoral actions have the character of positing some ‘material end’ that is not determined by pure reason, but rather a posteriori by natural incentives (G, 4: 427–8). The principle that ought to govern our action, namely, the principle that recognizes the objective end of humanity as an end in itself (i.e. the categorical imperative), is subordinated to the principle of self-love. As a result, the agent uses the humanity of others as a mere means to the realization of some concrete state of affairs.

According to Garcia, evil actions distinguish themselves from merely immoral ones by emanating from a specific kind of material end. Immoral actions are characterized by using the humanity of others merely as a means to some contingent end, but ‘in evil actions we seek out to directly violate the humanity of another person itself’ (Garcia Reference Garcia2002: 201). Whereas the moral law commands us to regard humanity as the necessary formal end of moral action, in evil action ‘the object of our willing just consists in the mistreatment of the humanity of another person qua human’; evil action has as its material end ‘the violation of humanity itself’ (ibid.). Understood in this way, Kant’s practical philosophy can be interpreted to make room for a distinction between evil and mere wrongdoing.

When we consider acts like genocide, torture and forced starvation, the direct violation of humanity itself is an apt characterization and pinpoints why such horrendous acts deserve to be categorized as evil rather than simply wrong. If the direct violation of humanity is central to what makes an act evil, then Kant’s claim that recognizing humanity as an objective end in itself is central to morality (and immorality) indicates that his theory is indeed an appropriate one for understanding the nature of evil. A unique kind of direct violation of that which is central to morality distinguishes evil from mundane wrongdoing. Thus it seems as if Kant can provide us with the conceptual tools to locate the identifying characteristic of evil as distinct from mere wrongdoing. If so, then we can conclude that Criticism One is not substantial enough to persuade us to abandon Kant’s theory of evil. Be that as it may, Reply One is admittedly not fully satisfying. Although it provides us with the best way to handle the criticism, we are nevertheless left with the impression that we have merely evaded the issue rather than solved it directly. As a result, Reply One is not sufficient on its own to convince Kant’s critics of the continued relevance of his account. For this, we need to move on to the next two criticisms and my responses.

Criticism Two

Many critics argue that Kant’s theory of evil is overly simplistic and narrow in its reduction of evil to the subordination of the moral law to self-love. With Kant’s account in mind John Kekes writes:

Even if we grant to Kant that the prioritization of self-love is a probable root of evil, it seems implausibly reductionist to claim that it is the only one. The seemingly unending list of cruelties pervading human history are too numerous and varied to attribute to a single cause. A quick glance at this history indicates that evil can be done out of malice and hate, but also greed, boredom, ambition, revenge, fear and obedience.Footnote ³⁰ Therefore, to claim that evil is attributable only to the subordination of the moral law to self-love is to ignore the manifold ways that evil occurs in moral interaction.

Reply Two

There are two responses to this criticism that may help save Kant’s theory. First, we must precisely identify Kant’s understanding of the root of evil – that is, the subordination of the moral law to the principle of self-love.Footnote ³¹ Selbstliebe is most commonly translated into English as self-love, and on occasion as self-interest.Footnote ³² Rather than investigate the propriety of the translation, I find it more revealing to examine its use in order to discern its meaning. ‘Self-love’ understood in a limited, non-Kantian, and perhaps more common, way refers mainly to a self-indulgent hedonist or egoist. For Kant, in contrast, it refers to the whole array of non-moral, natural incentives that navigate towards happiness.Footnote ³³ To see this, recall that Kant’s definition of evil unfolds within the framework of his analysis of the conflict or struggle between two very general categories of incentive: the moral and the non-moral. ‘Inclination’ is simply the Kantian term for non-moral incentives, and ‘self-love’ is a general term that is meant to encompass the pursuit of objects of inclination. As Allen Wood writes, these terms are ‘merely placeholders for whatever non-moral incentives might be chosen in preference to those of morality. Kant is not imposing any limits on what one can have an inclination to will’ (Wood Reference Wood2014: 36; see also Louden Reference Louden2000: 138–9). Since Kant uses ‘self-love’ as a broad term covering all non-moral incentives, and argues that evil comes about due to prioritizing non-moral incentives over the moral law, his theory can apparently accommodate the notion that there are many roots of evil.

Although this first response to Criticism Two is convincing on several levels, it remains vulnerable to an objection. One might resist the interpretation offered in the preceding paragraph and insist that all sensuous inclinations can indeed be reduced to the single root of self-love. It is not psychologically far-fetched to claim that the common motivations behind acts of evil – e.g. greed, boredom, ambition, revenge, fear and obedience – are fundamentally manifestations of self-love. Addressing this possibility, I offer an analysis that reinterprets Kant’s theory using the terminology of contemporary moral philosophy and suggest in what way self-love alone is a plausible origin of evil.

The following argument proceeds in two main steps. First, I suggest that evil may be most adequately understood as resulting from an agent’s rendering irrelevant overriding normative reasons for action. Second, I argue that insofar as such overriding reasons are second-personal reasons, self-love explains why an agent may render them irrelevant.

We begin this discussion by assenting to the quite credible suggestion that an agent’s reasons for acting are germane to actions of evil. Eve Garrard agrees and argues that evil ought to be understood in terms of the phenomenon of ‘silencing’ as it relates in two distinct ways to reasons.Footnote ³⁴ ‘Metaphysical silencing’ occurs when in the presence of reasons that make an action right (or good) other considerations cease to be reasons at all. Alternatively, ‘psychological silencing’ is ‘a total failure [on the part of an agent] to see that certain considerations are reasons at all’ (Garrard Reference Garrard2002: 329–30).

As Garrard describes it, an evil action occurs when an agent psychologically silences significant reasons against her acting. Reasons are significant when they serve to metaphysically silence other considerations that in differing circumstances could have reason-giving force. In other words, an agent performs an evil act when she psychologically silences reasons which are themselves metaphysical silencers.

The term ‘metaphysical silencer’ seems conceptually (and metaphysically) loaded, but the rationale behind it is simple: certain considerations lose their reason-giving force in the presence of strong reasons to do the morally right thing. For example, the possibility of saving money is an acceptable reason for buying one box of cereal over another.Footnote ³⁵ In contrast, it is not a reason at all to refrain from buying a much-needed medicine for my child who might die a painful and prolonged death without the medicine. Some reasons are so powerful that they have the effect of metaphysically silencing alternative considerations, which in other contexts might have reason-giving force. In the context of saving my child’s life, saving a modest amount of money is simply no reason at all.Footnote ³⁶

I find Garrard’s theory to be compelling, but her talk of ‘silencing’ is unnecessarily confusing, psychologically and metaphysically onerous, and difficult to prove. Rather than utilize the notion of ‘psychological silencing’, or refer to the capacity of reasons to ‘metaphysically silence’ other considerations, I believe we can capture the same insights while avoiding the excessive implications by using the phrases ‘rendering irrelevant’ and ‘overriding’.Footnote ³⁷ On this account evil occurs when an agent renders irrelevant, that is, consciously or perhaps even unconsciously rejects, the presence of overriding normative reasons against her acting. Reasons are overriding when they turn alternative potential reasons into mere considerations, depriving them of any reason-giving force. To put the proposed idea more simply, an agent performs an evil act when she renders irrelevant overriding normative reasons for action.

Two questions arise. What kinds of reasons are normatively overriding? And how could an agent render such reasons irrelevant? To answer the first question we first observe that moral agents find themselves in relationships of moral address and demand with one another.Footnote ³⁸ Moral agents address each other normatively by making claims, requests or demands on others and responding to others’ claims, requests or demands. Normative claims, requests or demands from others provide a moral agent with agent-relative reasons for action. Stephen Darwall calls such reasons second-personal reasons:

Agents in the moral community are moral agents precisely insofar as they are addressees and addressors of moral claims, requests and demands. In other words, moral agents engage in reciprocal relations of moral address, and the claims, requests and demands that make up this address provide agents with reasons for action.

In contexts of great moral import such reasons are overriding. For example, under non-moral circumstances there are reasons for me to perform actions that satisfy my individual preferences or bring about some degree of pleasure for me. However, if the means to achieve these ends require the torture of another person, then clearly the second person’s demands not to be tortured override my reasons for performing an action that brings about some degree of pleasure for me. Indeed, because second-personal reasons are by definition normatively significant they will often override non-moral reasons, and they always override in circumstances of great moral import.Footnote ³⁹

Despite the fact that second-personal reasons are often overriding, some individuals render these reasons irrelevant. I propose that we can explain this process through an appeal to self-love. Individuals who have an immutable preference for the satisfaction of their own inclinations are individuals who are prone to reject second-personal reasons to act. The demands of others play absolutely no role in their practical reasoning. These individuals are moral solipsists who prioritize first-person considerations even when such considerations have lost all reason-giving force.Footnote ⁴⁰ They allow considerations of self-love to render second-personal reasons irrelevant. Hence, evil acts are performed when an agent, out of self-love, renders irrelevant overriding second-personal reasons for action.

Understood in this way, self-love is indeed the root of moral evil because it allows the normative claims, requests and demands of others to go unaddressed. Since the addressing of these claims, requests and demands is the essence of moral agency, and the meeting of these claims, requests and demands when appropriate is the core of morally praiseworthy action, we can conclude that Kant is correct to claim that the prioritization of self-love over the moral law is the single cause of immoral and evil action.

Criticism Three

It has been widely objected that by focusing solely on the intelligible adoption of a supreme principle for action, Kant’s theory fails to give heed to the most salient feature of an evil act: the actual empirical harm and suffering of the victim. By defining evil solely as weakness of will, impurity of will and depravity of will – the subordination of the moral law to the principle of self-love – an agent’s action would not need to cause any suffering or produce any victims at all for the act to be evil. Can evil be defined without the infliction of suffering? Daniel Haybron suggests that someone might qualify as evil despite never bringing about any harm:

Certainly this individual is despicable. Yet, contrary to his intention, Haybron’s terminology betrays a significant characteristic of evil. As he describes her, the individual is not evil, but vile. Not only her misanthropy, but the joy she feels in witnessing others’ suffering makes her contemptible. However, the inability to cause and inflict suffering on others frees her from the moniker of evil. As Paul Formosa writes: ‘there is an important moral difference between merely wishing and fantasizing about evil … and actually intending and inflicting evil. While the quadriplegic remains a vile human being, she is not an evil person until she turns her evil fantasies into evil intentions and evil acts’ (Formosa Reference Formosa2007: 235).

Indeed, numerous philosophers argue that harm, specifically a certain kind or degree of excessive or life-wrecking harm, is a necessary condition of an evil act.Footnote ⁴¹ For example, Peter French notes that ‘victimization is the identifying characteristic of evil’ (French Reference French2011: 62–3). Arne Johan Vetlesen argues that evil is to ‘intentionally inflict pain and suffering on another human being, against her will, and causing serious and foreseeable harm to her’ (Vetlesen Reference Vetlesen2005: 2). Each of these theories identifies a necessary harm component in evil acts that characterizes the act as evil rather than merely wrong. Even if the motives, intentions and maxims of an agent are relevant to discussions of the nature of an evil act, the harm inflicted on the victim is as well. To see this, just imagine the silly example of someone stealing a candy bar from a store out of sheer malice. It is clear that the malicious intention is not sufficient to characterize the act as evil. The element of extreme, prolonged or life-wrecking harm is missing.Footnote ⁴² If empirical harm or suffering is a necessary element of evil, and Kant excludes this element from his theory, then his approach is incomplete at best, and false at worst. Criticism Three makes prima facie the strongest case in favour of rejecting Kant’s account of evil. A theory of evil without victims is no theory of evil; it is a theory concerning the potential for evil.

Reply Three

There are two responses that defenders of Kant’s theory can develop. The first response requires a brief review of Kant’s notion of a maxim to indicate that he can or does accommodate harm and the suffering of evil’s victims into his account of evil. The second response explains why a focus on the quality of will of the evildoer is essential to an accurate theory of evil.

The suggestion that Kant’s emphasis on the intelligible adoption of maxims reflects a complete disregard for the significance of empirical actions, and ipso facto empirical harm, is a caricature of his view and is simply false. Maxims, according to Kant, are subjective volitional principles of action (G, 4: 399–400; cf. Wood Reference Wood1999: 40). In other words, they are principles that motivate the will and are practical by their very nature.

The confusion arises by conflating Kant’s insistence that the moral worth of an action is to be determined solely by the formal maxim that gives rise to it, and not in the end to be attained by it, with the idea that a noumenal maxim stands somehow in isolation from its phenomenal counterpart – the action (see G, 4: 399–401). However, the mere fact that the moral worth of an action is determined by its formal maxim does not divorce a maxim from its subsequent action. Insofar as a maxim ‘is a general rule by which [an individual] will conduct himself’ (Rel, 6: 24), it is inherently connected to an individual’s conduct or action. Even if the moral worth of a maxim is determined purely by virtue of its form, ‘every maxim involves an end which serves as its matter’ (Wood Reference Wood1999: 52; see G, 4: 436, CPrR, 5: 21).

Consequently, even if it is true that in his theory of evil Kant’s main concern is with the quality and corruption of the will of the agent, he is not blind to the simple fact that empirical actions follow directly from an agent’s maxims.Footnote ⁴³ Because empirical actions follow from the adoption of maxims, it is almost tautological that the adoption of maxims is significantly phenomenal. As a result, it is unfairly dismissive of Kant’s view to argue that his theory holds no regard for the actual harm suffered by victims of evil.

One might reply that even if Kant’s theory does pay heed in this way to empirical suffering, there is no doubt that his analysis remains principally transcendental rather than empirical – that is, concerning those characteristics that make the doing of evil possible. By identifying the transcendental features of evil, he directs our attention to evil’s possibility rather than its actual occurrence. Nevertheless, I reply that a focus on the former does not entail believing that the latter is inconsequential. In fact, Kant attends to empirical evil directly in his discussion of the ‘vices of culture’. He observes that we have constant anxiety that other people will attain superiority over ourselves. Through a continuous comparison with others there arises a desire for superiority over them. Out of this rivalry emerge ‘envy, ingratitude, joy in others’ misfortunes, etc.’ (Rel, 6: 27), all of which occur in the phenomenal world.Footnote ⁴⁴

Even so, there is a significant reason why Kant does not focus his account of evil on empirical harm. We must keep in mind not simply that Kant’s account is transcendental rather than empirical, but also on the reasons why. What is so morally troublesome about the vices of culture, and all of evil, is that they hide behind reason and are perfectly compatible with good empirical conduct.Footnote ⁴⁵ Recall Kant’s emphasis on the ways in which immorality is cloaked in self-deception. The agent may perform morally legal acts while being motivated by immoral incentives. Far from trivializing the actual suffering of victims of evil, Kant seeks to end it by identifying its source and possibility. The prevention of empirical suffering is only the beginning of the struggle against evil, not its conclusion. For this reason, his approach to evil focuses on individuals’ quality of will and evil’s transcendental conditions.

This judgement renders Kant’s theory highly relevant for contemporary conceptions of evil and morality, and identifies a crucial point relevant to discussions of evil and the moral psychology of perpetrators of evil. Morally legal acts – that is, acts that outwardly adhere to the moral law – are not constitutive of morality because they do not reflect the moral quality of an agent’s character. The absence of empirical suffering is not an indication of moral goodness because it is merely contingent whether an individual acts according to the moral law when she has not prioritized the moral law in her will. This leads to the possibility, if not the likelihood, that the individual will act contrary to morality in different circumstances. Consequently, appearances of morality are no guarantee against future occurrences of evil. Especially insofar as an individual can be deceived about the moral goodness of her own character, evildoing gains a tenacious foothold when an agent fails to act out of duty to the moral law and acts only in accordance with it.

This insight acquires both gravity and credibility when we consider it alongside the large-scale mass atrocities of the past century. Empirical studies have demonstrated that the people who perpetrated the Holocaust or the genocide in Rwanda were ordinary individuals rather than diabolical monsters.Footnote ⁴⁶ Nevertheless, it remains puzzling how ordinary people can so easily commit extraordinary acts of evil. Although Kant and his contemporaries could not have fathomed the massacres that in large part defined the twentieth century, his theory of evil helps us understand the ease with which ordinary individuals became perpetrators of the morally worst kinds of acts. Without a morally good moral character, the fact that an agent has successfully acted in accordance with the moral law in the past is no reason to think she will do so in the future. Rather, it is good reason to think that she will not.Footnote ⁴⁷ Precisely because people are not diabolical, but rather tend to be mistaken about the moral quality of their wills, are easily motivated by sensuous incentives and commonly misunderstand the nature of morality and evil, they can become perpetrators of evil.

Contemporary criticisms of the Kant’s theory often miss this point and its relevance for the prevention of future evil, thereby neglecting the importance his theory still holds for understanding the phenomenon of evil. The constant struggle in the human will regarding the prioritization of moral, non-moral or immoral incentives is definitive of moral character. Whichever incentives win out in this struggle constitute one’s moral character. Without a moral character dedicated to morality, humans will tend to act in evil ways.