Kant’s Casuistical Questions

1. Introduction

One of the most interesting features of Kant’s late practical philosophy is the casuistical questions in the Metaphysics of Morals (1797). These are eight small sections Kant added to each of his discussions of specific ethical duties in the first half of the Doctrine of Virtue, the ‘Doctrine of the elements of ethics’. In one of them, for instance, Kant asks us to consider the following situation: ‘An author asks one of his readers “How do you like my work?” […] The author will take the slightest hesitation in answering as an insult. May one, then, say what is expected of one?’ (MS 6:4319).¹ In another section, shortly thereafter, Kant asks his readers: ‘How far should one expend one’s resources in practicing beneficence?’ (MS 6:454). Whether or not we believe that we have good answers, Kant’s casuistical questions point to ethical difficulties that all of us have faced and will continue to face in our lives. Strikingly, these difficulties are familiar to everyone with even a modest amount of ethical sensitivity despite their being embedded in a seemingly long-gone cultural and historical context. The most perspicuous difficulty demonstrated by Kant’s casuistical questions is the temptation to rationalize away the requirements of our immediate moral judgment in favor of our apparently reasonable inclinations. Thus, one of Kant’s many insights in the Metaphysics of Morals is that moral challenges cannot be resolved by downplaying their difficulty; only by taking their difficulty for us seriously can we make any attempt at better understanding—and dealing with—the challenges of our ethical lives.

Kant’s casuistical questions have received comparatively little scholarly attention so far.² Moreover, they have traditionally been treated in ways that make it difficult to appreciate the different moral challenges they raise. Kant’s ethics is commonly assumed to aim primarily at figuring out what is truly the right thing to do. On this assumption, it is plausible to expect that the casuistical questions are practice examples for the application of a test or decision procedure for right action. Consequently, the casuistical questions have often been interpreted either as difficult cases of moral conflict (i.e., ‘conflicting grounds of obligation’, in Kant’s terms), which do not lend themselves to easy resolution by maxim-testing (Esser 2008; O’Neill 2002: 343; Koch 2003: 121–88),³ or as questions that test the reader’s comprehension of the decision procedure (Schüssler 2012, 2021). However, Kant’s casuistical questions do not easily fit this picture: few of them actually concern conflicting grounds of obligation, and many seem too difficult to be mere tests of the reader’s comprehension of a decision procedure.

In this article, I argue that we should read the casuistical questions against the background of the therapeutic purpose of Kant’s moral philosophy. Locating Kant’s conception of moral philosophy within a longstanding tradition of ethical reflection—one aimed at shaping human character and freeing us from our own forms of self-entrapment—helps us understand the point of the casuistical questions as they emerge in the Doctrine of Virtue. They articulate instances of the characteristic sort of moral difficulties that human beings face—difficulties which can be traced to certain sources. Like in the practice of a therapist, the first part of dealing with this sort of problem lies in first achieving an understanding of its origins, which in turn moves us closer to overcoming it. On my proposed reading, Kant seeks to provide an overview of the different species of moral difficulties that human beings face. Thus, the significance of Kant’s casuistical questions lies not so much in their potential answers but in the difficulties they highlight.

In the first section of this article, I provide a concise overview of the initial puzzle about Kant’s casuistical questions and their contemporary reception. In section 2, I outline the historical context of Kant’s treatment of casuistry, which will provide a helpful background for my proposed reading of the casuistical questions and their role in the ‘Doctrine of the elements of ethics’. For ease of discussion, section 3 reproduces Kant’s casuistical questions. There, I also argue that Kant’s questions are remarkably diverse and defy the categorizations available in previous literature. Finally, section 4 argues that Kant’s casuistical questions should be read—against the background of the therapeutic purpose of Kant’s ethics—as highlighting the diverse difficulties human beings face in their ethical lives.

2. Kant’s Casuistical Questions: An Inconsistency or a Puzzle?

At first glance, Kant’s casuistical questions either give rise to an inconsistency or a puzzle. In his discussion of the divisions of the Doctrine of Virtue, Kant defines casuistry as the practice of making ‘determining judgments’ in the context of imperfect duties. Since imperfect duties such as ‘be benevolent’ or ‘help your neighbor’ allow a certain latitude in how they are to be complied with, they further require determining judgment, i.e., ‘casuistry’. Such further determining judgment is a context-dependent practice, rather than a theory, and thus cannot be explained ex ante. Consequently, this practice of casuistry cannot, strictly speaking, be part of a metaphysics of morals (which, for Kant, must have the status of a science [Wissenschaft]), but can only be added to the metaphysics of morals as explanatory illustrations (‘scholia’):

[E]thics, because of the latitude it allows in its imperfect duties, unavoidably leads to questions that call upon judgment to decide how a maxim is to be applied in particular cases, and indeed in such a way that judgment provides another (subordinate) maxim (and one can always ask for yet another principle for applying this maxim to cases that may arise). So ethics falls into a casuistry which has no place in the doctrine of right. Casuistry is, accordingly, neither a science nor a part of a science; […] casuistry is not so much a doctrine about how to find something as rather a practice in how to seek truth. So it is woven into ethics in a fragmentary way […] and is added to ethics only by way of scholia to the system (MS 6:411).

Importantly, this brief explanation of casuistry makes clear that such scholia have ‘no place in the doctrine of right’ because there are no imperfect duties in the sphere of right. By explicitly restricting casuistry to imperfect duties, Kant here evidently rules out casuistry in the case of perfect duties. Yet, six out of eight sets of casuistical questions are devoted to perfect duties.

One possible way of avoiding this seeming inconsistency is by differentiating casuistry as explanatory illustrations pertaining to imperfect duties—and ruled out in the case of perfect duties—from casuistical questions. Not only does such a distinction avoid an inconsistency, but it also fits with Kant’s use of the terms—which consistently refers to the sets of questions added to his discussions of duties as ‘casuistical questions’, not as ‘casuistry’.⁴ However, this differentiation opens up a puzzle: if the casuistical questions are not part of what Kant described as casuistry, then what are they? And what philosophical purpose are they supposed to serve?

This initial puzzle regarding the possible purpose of casuistical questions about perfect duties is particularly salient in light of Kant’s background assumptions about moral judgment. There is a long tradition of casuistry in the history of western philosophy. This perceived need for casuistry usually arose for ethical theories that restricted detailed moral knowledge to an elite of trained experts, who would work out all the minute rules through difficult casuistry. By contrast, Kant notoriously believed that no one needs moral philosophy or instruction in order to know what is right and wrong (KrV A43/B61–A44; KrV A830/B858–A831/B859; GMS 4:403–4; TP 8:288; HN 20:44); in his examples of moral judgment, Kant emphasized that common moral agents can make moral judgments ‘on the spot’, without the need for any elaborate casuistical method (TP 8:284–87; MS 6:480–81; KpV 5:36). If Kant’s casuistical questions are not about developing an expert taxonomy of derivative rules and exceptions, then what purpose could they serve within his ‘doctrine of the elements of ethics’? In other words, how should we understand their philosophical significance?

How, then, does the existing literature navigate between the seeming inconsistency and the general puzzle about Kant’s casuistical questions? So far, the most popular solution has been to attribute confusion to Kant. Thus, David James (1992) has speculated that Kant really meant to say ‘broad’ duties in his explanation of casuistry, rather than what he actually said (namely, ‘imperfect’ duties).⁵ Similarly, Yvonne Unna (2003) and Elke Schmidt (2023) have suggested that perfect duties may be included in Kant’s casuistical questions because they prohibit particular actions but do not prescribe particular actions,⁶ and may thereby still allow for some latitude in terms of which actions the agent should perform—an aspect shared with imperfect duties. Consequently, on Unna’s (2003) reading, too, Kant was simply confused in excluding imperfect duties from casuistry in his explanation in the introduction to the Doctrine of Virtue.⁷ Following James’s suggestion just mentioned, Tatiana Patrone (2013: 490) even dismisses the seeming inconsistency by saying that ‘Kant’s take on his own position with respect to casuistry is misleading’.

To my knowledge, Rudolf Schüssler (2012, 2021) has so far been the only interpreter who has made a serious effort at taking Kant by his word(ing). On Schüssler’s reading, Kant’s casuistical questions are not instances of Kantian casuistry. In large part, this interpretation fits well with the text of the Metaphysics of Morals: throughout the Doctrine of Virtue, Kant is consistent in never calling his casuistical questions instances of casuistry. Distinguishing between casuistry (i.e., the search for ‘subsidiary maxims’ in the case of imperfect duties, which allow significant latitude) and casuistical questions helps explain why there can be casuistical questions regarding perfect duties. Kant’s explanation of casuistry, which explicitly rejects the possibility of casuistry regarding perfect duties, simply does not apply to casuistical questions.⁸

However, taking seriously this distinction between casuistry and casuistical questions leaves us with the more general puzzle about Kant’s casuistical questions: if the questions are not part of what Kant described as ‘casuistry’, then what philosophical purpose are the casuistical questions supposed to serve? On Schüssler’s proposed reading, Kant’s casuistical questions are testing the reader on whether they have really understood Kant’s ethics: they are ‘didactic classroom exercises designed to test how much basic ethical knowledge pupils have acquired’ (Schüssler 2021: 1014). More specifically, the purpose of the casuistical questions would be to test the reader’s comprehension of the respective duties’ scope.

Although this answer has the considerable virtue of being theoretically consistent, it still retains a shortcoming: numerous Kant scholars have found the casuistical questions difficult (or even impossible) to answer. For instance, some authors have taken Kant’s casuistical questions to be difficult but closed questions about conflicting grounds of obligation (Esser 2008; Koch 2003: 121–88);⁹ some have argued that only some questions are closed ones, while others are open (James 1992);¹⁰ some have thought that the casuistical questions are about conflicting grounds of obligation and are genuinely open (O’Neill 2002: 343; Timmermann 2000: 49; Kim 2009; Gregor 1963);¹¹ and others have argued that Kant’s casuistical questions are open and difficult questions about duties permitting a certain latitude (Schmidt and Schönecker 2019; Schmidt 2023). Unless we want to summarily dismiss these scholars as seriously confused, the very fact that so many senior Kant scholars have found the casuistical questions difficult—and have found such widely different interpretations of them—strongly suggests, at least to me, that their philosophical significance unlikely consists in testing the reader’s comprehension of Kant’s ethics. Whatever didactic purpose the casuistical questions serve, this purpose cannot be exhausted by testing the reader’s comprehension of the scope of the respective duties like a final take-home exam.¹² After all, it would be awkward to suppose that Kant’s casuistical questions test the reader on whether they have really understood the scope of the duties discussed, only for Kant scholars themselves to consistently fail this test.

Fortunately, neither of these interpretive options—that Kant would have been seriously confused about his own description of casuistry, or that Kant scholars would be seriously confused about Kantian duties—is strictly necessary. Instead, I suggest that we see the philosophical significance of Kant’s casuistical questions not in their potential answers, but in the specific way they highlight the complexity of ethical life. On the alternative interpretation I offer below, we can differentiate between casuistry and casuistical questions as well as provide a more sympathetic answer to what purpose ties together the casuistical questions. Before moving on to this option, it will be helpful to briefly synthesize some of the historical context surrounding Kant’s discussion of casuistry.

3. Casuistry and the Purpose of Moral Philosophy

Kant wrote his casuistical questions during a time of popular disdain for the practice of casuistry. The analysis of ‘cases of conscience’ had been a common practice in scholastic philosophy already, but the rise of the early modern practice of casuistry accelerated especially after the Council of Trent (1545–1563), which had decreed that sinners needed to confess all the circumstances that might influence the nature of their sinful actions. Consequently, the second half of the sixteenth and the seventeenth centuries saw a surge of commentaries dealing with collections of various cases of conscience, taking into consideration all the possible, minute background circumstances of an individual’s action.¹³ Intimately connected with this practice of casuistry was the doctrine of probabilism, developed by Bartolomé de Medina in his 1577 commentary on Aquinas’s Summa Theologiae. De Medina argued that in difficult ethical situations one could permissibly follow an opinion deemed less likely to be correct—without committing a sin—if that opinion had been stated or defended at some point by a serious authority with good arguments.¹⁴ This theoretical background of probabilism gave rise to an (allegedly) widespread practice of justifying people’s unconscientious actions by retrospectively finding some professional opinion that would justify it. Thus, the practice of casuistry soon acquired the reputation of attempting to justify lax morals, or even immoral conduct, by recourse to hair-splitting and factitious distinctions.¹⁵ By the middle of the seventeenth century, casuistry had both become firmly identified with the doctrine of probabilism and been subjected to harsh criticism from rigorist thinkers. Partly due to prominent criticisms of authors such as Pascal and Rousseau, among many others, the term ‘casuistry’ became almost synonymous with spurious moral reasoning by the mid- to late eighteenth century.¹⁶

Rousseau gives a particularly memorable expression to this popular antipathy toward casuistry in his Emile, first published in 1762. In a section titled “The Creed of a Savoyard Priest”, he writes:

[…] I do not derive these rules [of conduct] from the principles of the higher philosophy, I find them in the depths of my heart, traced by nature in characters which nothing can efface. I need only consult myself with regard to which I wish to do; what I feel to be right is right, what I feel to be wrong is wrong; conscience is the best casuist; and it is only when we haggle with conscience that we have recourse to the subtleties of argument. […] Conscience is the voice of the soul, the passions are the voice of the body. Is it strange that these voices often contradict each other? And then to which should we give heed? Too often does reason deceive us; we have only too good a right to doubt her; but conscience never deceives us; she is the true guide of man; it is the soul what instinct is to the body; he who obeys his conscience is following nature and he need not fear that he will go astray (1961: 249–50).

As this passage reminds us, Rousseau dismissed casuistry not merely because he was an epistemic egalitarian about moral knowledge; Rousseau also had a highly pessimistic opinion of philosophy, which he believed could do more evil than good for people’s moral compass. He even suggested that we start relying on philosophical arguments about ethics ‘only when we haggle with conscience’. In this way, Rousseau followed the popular antipathy toward casuistry and turned it against popular moral philosophy in general.

Kant famously followed Rousseau in his epistemic egalitarianism about moral knowledge;¹⁷ but, as John Callanan (2019) has shown, Kant did not follow Rousseau in his more pessimistic belief that philosophy would corrupt the common moral knowledge that is already available to people in their uncorrupted, natural state. While Rousseau believed that natural human feeling would be enough as a moral guide in human conduct, Kant believed that reason—and importantly, philosophy—still has a significant role to play.

According to Kant, the project of unearthing the supreme principle of practical reason is also a therapeutic exercise, bringing us closer to who we truly are as rational beings, and helping us overcome the subconscious, corrosive parts of our finite, rational nature. Kant most clearly expressed this therapeutic conception of the practical purpose of moral philosophy in the Groundwork of the Metaphysics of Morals (1785). There, Kant claimed that beings like us are easily led astray by the ‘natural dialectic’, which arises from the conflicting demands of our inclinations and the demands of morality, and which can lead people to rationalizing away the requirements of morality:

The human being feels within himself a powerful counterweight to all the commands of duty, which reason represents to him as so deserving of the highest respect—the counterweight of his needs and inclinations, the entire satisfaction of which he sums up under the name happiness. Now reason issues its precepts unremittingly, without thereby promising anything to the inclinations, and so, as it were, with disregard and contempt for those claims, which are so impetuous and besides so apparently equitable (and refuse to be neutralized by any command). But from this there arises a natural dialectic, that is, a propensity to rationalize against those strict laws of duty and to cast doubt upon their validity, or at least upon their purity and strictness, and, where possible, to make them better suited to our wishes and inclinations, that is, to corrupt them at their basis and to destroy all their dignity—something that even common practical reason cannot, in the end, call good. In this way common human reason is impelled […] to go out of its sphere and to take a step into the field of practical philosophy […] (GMS 4:405).¹⁸

Kant’s account of this natural dialectic is not simply a conflict between inclination and reason; rather, this natural dialectic is itself a rational process. As finite beings, we necessarily take an interest in our happiness. And our happiness is a genuinely valuable end—which makes pursuing our happiness appear so reasonably ‘equitable’. And since it is hardly an option for finite, dependent beings like us to shape our inclinations at will, we take a pathological interest in justifying them. In other words, we are prone to rationalizing in favor of our inclinations and against our moral judgment not simply because we are suffering from overly strong inclinations, but precisely because we are rational beings who take an interest in everything that is reasonable.¹⁹

Moral philosophy can help us overcome this corrosive aspect of our rational, finite nature because it unearths the supreme principle of practical reason, thus bringing this principle into reflected self-consciousness and even ‘closer to feeling’.²⁰ In this way, moral philosophy can bring us closer to who we truly are as rational beings. We need the resolve to trust our capacity for pure practical reason and pursue what is intrinsically good even when it conflicts with our happiness. Kant believed that philosophy could help people find this resolve to rely on their rational capacities and to trust their immediate moral judgments, instead of rationalizing away the requirements of morality and relegating their decisions to inclinations. Moreover, while bad moral theory risks reinforcing our rationalizations, moral philosophy done well deprives us of such tools and may even help us recognize our own errors.²¹

Thus, while Kant agrees with Rousseau about the accuracy of common moral knowledge, he did not believe that we should therefore either give up or turn against moral philosophy. On Kant’s view, moral philosophy can be practical because it is therapeutic. Both the highest principle of practical reason and its main obstacles are already within us, simply because we are finite, rational beings. Thus, the project of unearthing the supreme principle of practical reason is not a merely metaethical or speculative exercise; for Kant, it is also a therapeutic exercise, helping us overcome the corrosive parts of our finite, rational nature.

On a high level of abstraction, all moral philosophy aims at virtue. To call the purpose of Kant’s moral philosophy ‘therapeutic’ is to emphasize the specific way in which moral philosophy helps us strive for virtue. Rather than offering a new kind of knowledge by means of a ‘method of ethics’, Kant’s moral philosophy helps us strive for virtue by bringing us closer to something that is within us all along (our capacity for immediate moral judgment as rational beings), and helps us overcome something that is also within us (our tendency as finite beings to rationalize against the requirements of morality).

With this approach, Kant is in venerable company: ‘Empty is that philosopher’s argument’, Epicurus tells us in a fragment, ‘by which no human suffering is therapeutically treated. For just as there is no use in a medical art that does not cast out the sickness of bodies, so too there is no use in philosophy, unless it casts out the suffering of the soul’ (Us. 221 Porphyrius ad Marcellam 31 p. 209, 23 Nauck).²² On this conception of the practical purpose of moral philosophy, we are less in need of a ‘method of ethics’ and instead require resources for a therapeutic way of reflection: a kind of reflection that can help us overcome our self-incurred forms of unenlightenment or self-entrapment.²³

Because of their view of moral cognition as egalitarian (i.e., accessible to every human being of a healthy mind and heart), Kant and Rousseau both rejected the idea of a practice of casuistry that would allow people to quibble with their conscience by means of hair-splitting and factitious distinctions. Thus, the Vigilantius’s lecture notes record Kant as saying that probabilist casuistry is the ‘practice of cheating or chicaning conscience through sophistry, insofar as one makes it [conscience] appear mistaken’ (V-MS/Vigil 27:620, translation mine).²⁴ Indeed, for Kant, probabilist casuistry is the perfect example of the human tendency to rationalize against the moral law.²⁵ But unlike Rousseau, Kant also leaves room for casuistical questions. What, then, could provide a more promising starting point for explaining their different respective attitudes toward casuistry and casuistical questions than their different views about the practical purpose of moral philosophy? As I suggest in the remainder of this essay, this therapeutic purpose of Kant’s moral philosophy—through which he distinguished himself from Rousseau in the Groundwork—allows us to trace an alternative reading of Kant’s casuistical questions and their philosophical significance.

4. The questions

Kant adds eight small sections of casuistical questions to some (albeit not all) discussions of specific ethical duties and corresponding vices. Since, to my knowledge, no discussion of Kant’s casuistical questions has yet reproduced them collectively, and for ease of reference in the following discussion, it is worth restating Kant’s casuistical questions under the specific section titles in which they appear. The following quotations are given in the order they appear in Kant’s text, omitting the preceding discussions of the respective duties and vices.

1. On killing oneself

   • - Is it murdering oneself to hurl oneself to certain death (like Curtius) in order to save one’s country?—or is deliberate martyrdom, sacrificing oneself for the good of all humanity, also to be considered an act of heroism?

   • - Is it permitted to anticipate by killing oneself the unjust death sentence of one’s ruler—even if the ruler permits this (as did Nero with Seneca)?

   • - Can a great king who died recently be charged with a criminal intention for carrying a fast-acting poison with him, presumably so that if he were captured when he led his troops into battle he could not be forced to agree to conditions of ransom harmful to his state?—for one can ascribe this purpose to him without having to presume that mere pride lay behind it.

   • - A man who had been bitten by a mad dog already felt hydrophobia coming on. He explained, in a letter he left, that, since as far as he knew the disease was incurable, he was taking his life lest he harm others as well in his madness (the onset of which he already felt). Did he do wrong?

   • - Anyone who decides to be vaccinated against smallpox puts his life in danger, even though he does it in order to preserve his life; and, insofar as he himself brings on the disease that endangers his life, he is in a far more doubtful situation, as far as the law of duty is concerned, than is the sailor, who at least does not arouse the storm to which he entrusts himself. Is smallpox inoculation, then, permitted? (MS 6:423–24, emphasis in original)

2. On defiling oneself by lust

   • - Nature’s end in the cohabitation of the sexes is procreation, that is, the preservation of the species. Hence one may not, at least, act contrary to that end. But is it permitted to engage in this practice (even within marriage) without taking this end into consideration?

   • - If, for example, the wife is pregnant or sterile (because of age or sickness), or if she feels no desire for intercourse, is it not contrary to nature’s end, and so also contrary to one’s duty to oneself, for one or the other of them, to make use of their sexual attributes—just as in unnatural lust?

   • - Or is there, in this case, a permissive law of morally practical reason, which in the collision of its determining grounds makes permitted something that is in itself not permitted (indulgently, as it were), in order to prevent a still greater violation?

   • - At what point can the limitation of a wide obligation be ascribed to purism (a pedantry regarding the fulfillment of duty, as far as the wideness of the obligation is concerned), and the animal inclinations be allowed a latitude, at the risk of forsaking the law of reason? (MS 6:426, emphases in original)

3. On stupefying oneself by the excessive use of food or drink

   • - Can one at least justify, if not eulogize, a use of wine bordering on intoxication, since it enlivens the company’s conversation and in so doing makes them speak more freely?

   • - Or can it even be granted the merit of promoting what Horace praises in Cato: virtus eius incaluit mero?

   • - The use of opium and spirits for enjoyment is closer to being a base act than the use of wine, since they make the user silent, reticent and withdrawn by the dreamy euphoria they induce. They are therefore permitted only as medicines.—But who can determine the measure for someone who is quite ready to pass into a state in which he no longer has clear eyes for measuring?

   • - Although a banquet is a formal invitation to excess in both food and drink, there is still something in it that aims at a moral end, beyond mere physical well-being: it brings a number of people together for a long time to converse with one another. And yet the very number of guests […] allows for only a little conversation (with those sitting next to one); and so the arrangement is at variance with that end […] How far does one’s moral authorization to accept these invitations to intemperance extend? (MS 6:428, emphases in original)

4. On lying

   • - Can an untruth from mere politeness (e.g., the ‘your obedient servant’ at the end of a letter) be considered a lie? No one is deceived by it.

   • - An author asks one of his readers ‘How do you like my work?’ […] The author will take the slightest hesitation in answering as an insult. May one, then, say what is expected of one?

   • - If I say something untrue in more serious matters, having to do with what is mine or yours, must I answer for all the consequences it might have?

   • - For example, a householder has ordered his servant to say ‘not at home’ if a certain human being asks for him. The servant does this and, as a result, the master slips away and commits a serious crime, which would otherwise have been prevented by the guard sent to arrest him. Who (in accordance with ethical principles) is guilty in this case? Surely the servant, too, who violated a duty to himself by his lie, the results of which his own conscience imputes to him. (MS 6:431)

5. On avarice

   • - [M]iserliness is not just mistaken thrift, but rather slavish subjection of oneself to the goods that contribute to happiness which is a violation of duty to oneself since one ought to be their master. It is opposed to liberality of mind (liberalitas moralis) generally (not to generosity, liberalitas sumptuosa, which is only an application of this to a special case), that is, opposed to the principle of independence from everything except the law, and is a way in which the subject defrauds himself. But what kind of a law is it that the internal lawgiver itself does not know how to apply?

   • - Ought I to economize on food or only in my expenditures on external things? in old age, or already in youth? Or is thrift as such a virtue? (MS 6:433–34, emphases in original)

6. On servility

   • - Is not the human being’s feeling for his sublime vocation, that is, his elation of spirit (elatio animi) or esteem for himself, so closely akin to self-conceit (arrogantia), the very opposite of true humility (humilitas moralis), that it would be advisable to cultivate humility even in comparing ourselves with other human beings, and not only with the law?

   • - Or would not this kind of self-abnegation instead strengthen others’ verdict on us to the point of despising our person, so that it would be contrary to our duty (of respect) to ourselves? Bowing and scraping before a human being seems in any case to be unworthy of a human being.

   • - Preferential tributes of respect in words and manners even to those who have no civil authority—[…] does not all this prove that there is a widespread propensity to servility in human beings? (MS 6:437, emphases in original)

7. The duty of beneficence

   • - How far should one expend one’s resources in practicing beneficence? Surely not to the extent that he himself would finally come to need the beneficence of others.

   • - How much worth has beneficence extended with a cold hand (by a will to be put into effect at one’s death)?

   • - If someone who exercises over another (a serf of his estate) the greater power permitted by the law of the land robs the other of his freedom to make himself happy in accordance with his own choices, can he, I say, consider himself the other’s benefactor because he looks after him paternalistically in accordance with his own concepts of happiness? Or is not the injustice of depriving someone of his freedom something so contrary to duty of right as such that one who willingly consents to submit to this condition, counting on his master’s beneficence, commits the greatest rejection of his own humanity, and that the master’s utmost concern for him would not really be beneficence at all? Or could the merit of such beneficence be so great as to outweigh the right of human beings?—I cannot do good to anyone in accordance with my concepts of happiness (except to young children and the insane), thinking to benefit him by forcing a gift upon him; rather, I can benefit him only in accordance with his concepts of happiness.

   • - Having the resources to practice such beneficence as depends on the goods of fortune is, for the most part, a result of certain human beings being favored through the injustice of the government, which introduces an inequality of wealth that makes others need their beneficence. Under such circumstances, does a rich man’s help to the needy, on which he so readily prides himself as something meritorious, really deserve to be called beneficence at all? (MS 6:454, emphases in original)

8. The duty of sympathy

   • - Would it not be better for the well-being of the world generally if human morality were limited to duties of right, fulfilled with the utmost conscientiousness, and benevolence were considered morally indifferent? It is not so easy to see what effect this would have on human happiness. But at least a great moral adornment, benevolence, would then be missing from the world. This is, accordingly, required by itself, in order to present the world as a beautiful moral whole in its full perfection, even if no account is taken of advantages (of happiness).

   • - [U]niversal love of one’s neighbor can and must be based on equality of duties, whereas in gratitude the one put under obligation stands a step lower than his benefactor. Is it not this, namely pride, that causes so much ingratitude? seeing someone above oneself and feeling resentment at not being able to make oneself fully his equal (as far as relations of duty are concerned)? (MS 6:458)

Kant’s casuistical questions are remarkably diverse. The first six duties to which he adds casuistical questions are perfect duties to oneself, while the last two duties are imperfect duties to others. At first sight, some of Kant’s casuistical questions seem to pose genuinely difficult, if not unanswerable, questions about conflicting grounds of obligation, like some of the questions concerning suicide (1). By contrast, some casuistical questions seem to concern conflicting grounds of obligation, but also appear to be answerable, like the questions concerning lying (4). The questions regarding lust (2) can seem difficult to answer from a Kantian perspective, but they do not obviously concern any conflicting grounds of obligation. Still other questions appear to be both answerable and also do not obviously concern conflicting grounds of obligations, like the questions regarding the use of food and drink (3) and the questions about servility (6). In addition to these differences, some of Kant’s casuistical questions regarding the duty of beneficence (7) seem clearly answerable, while other questions regarding the same duty seem both difficult and even concern conceptual, rather than practical, difficulties. Some entire sets of casuistical questions do not appear to raise any moral questions at all, but only indirectly practical or even theoretical ones, like the questions regarding sympathy (8). Lastly, the questions on avarice (5) may sound as if they were about an imperfect duty, although they explicitly concern a perfect duty toward oneself.

Although Kant’s questions regarding suicide present genuinely difficult situations of conflicting grounds of obligation, Kant himself answers at least three out of the five questions on suicide in his own writings (Anth 7:259; Refl 15:972–976; V-MS/Vigil 27:603, 629).²⁶ Similarly, Kant also provides an immediate answer to his third casuistical question on beneficence—whether one may deprive others of their freedom of choice in order to paternalistically impose one’s own conception of happiness on them: ‘I can benefit him only in accordance with his concepts of happiness’ (MS 6:454, original emphasis). And Kant also provides at least part of an answer to his first casuistical question on beneficence—how much of one’s resources one should spend for beneficence: ‘Surely not to the extent that he himself would finally come to need the beneficence of others’ (MS 6:454).

Finally, Kant’s casuistical questions do not seem to be examples of casuistry, i.e., cases in which we need to find subsidiary maxims for the application of other, more general, maxims. In his introduction to the Doctrine of Virtue, Kant had stated that ethics,

because of the latitude it allows in its imperfect duties, unavoidably leads to questions that call upon judgment to decide how a maxim is to be applied in particular cases, and indeed in such a way that judgment provides another (subordinate) maxim (and one can always ask for yet another principle for applying this maxim to cases that may arise). So ethics falls into a casuistry […] (MS 6:411–12).

While some of Kant’s casuistical questions might be read in a way that makes them fit this picture—like, for instance, the questions concerning how much of one’s resources should be spent in beneficence, or how much alcohol one can consume without stupefying oneself through excess—most of Kant’s casuistical questions do not fit this description of casuistry. Interestingly, the casuistical questions which we would have expected to best fit this paradigm—namely, the questions concerning imperfect duties—do not fit this explanation at all. Thus, Kant’s casuistical questions are most likely not casuistry—indeed, Kant is very consistent in his use of these two terms for apparently different practices.²⁷

After this brief initial look at Kant’s casuistical questions, we can already make the following preliminary observations. Kant’s casuistical questions have little, if anything, in common: neither the type of difficulty involved, nor the answerability of the question, nor even the nature of the question (whether directly or indirectly practical) remains consistent throughout. Consequently, it would be misleading to describe them as instances of conflicting grounds of obligation (as not all concern such conflicts), or as uncertain cases that require subsidiary maxims for their solution (since only few might require subsidiary maxims), or as examples of simply unanswerable ethical questions (as some of them can be answered); it would also be insufficient to describe them as questions that test the reader’s understanding of Kant’s ethics (as some of them remain genuinely difficult even to Kant scholars).

One thing, however, that Kant’s casuistical questions share is their sensitivity to ethical difficulties.

5. Highlighting the Complexity of Ethical Life

One likely reason why Kant’s casuistical questions have appeared so strange to commentators is that most recent readers have also assumed a specific view about the purpose of Kant’s moral philosophy. On this commonly assumed view, Kant’s ethics is a theory of right action—that is, a ‘method of ethics’ (to use Sidgwick’s term) which will provide a test or decision procedure and thus help us resolve difficult questions about what we should do.²⁸ On this view, it is reasonable to expect that the casuistical questions are exercise examples which the theory’s test will allow us to answer. However, as we saw above, this way of reading the casuistical questions forces a uniformity on them that these questions simply do not have. And importantly, this way of reading the casuistical questions also eclipses the questions themselves: on such a view, their purpose would merely lie in their answers because they are supposed to be mere exercise examples for a decision procedure.

By contrast, I suggest that we see the philosophical significance of the casuistical questions not in their answer, but in the difficulties they highlight. If moral philosophy is to cast out ‘the suffering of the soul’, it must first be sensitive to what kinds of suffering beings like us face. In more Kantian jargon: if moral philosophy is supposed to help us overcome the self-deceptive aspect of our nature, then it should give us examples of what self-deceptive reasoning itself can look like²⁹ and be sensitive to the complexities of ethical life that tend to underlie such self-deceptive reasoning in the first place.³⁰

Thus, on the view I propose, the kind of reflection that brings us closer to virtue cannot be the algorithmic application of the Categorical Imperative; rather, reflection that brings us closer to virtue starts with the very rejection of the idea that such an algorithmic application of a test or decision procedure could do justice to the complexity of ethical life. Only by acknowledging the full range of difficulties beings like us face can we also work against the temptations of rationalization and self-deception and move closer to virtue. And, arguably, Kant’s casuistical questions demonstrate just that.

Kant already indicated one of these difficulties in his discussion of the ‘natural dialectic’ in the Groundwork: the human propensity to place their inclinations over their knowledge of what is moral because the demands of our inclinations ‘are so impetuous and besides so apparently equitable’ (GMS 4:405). In the Doctrine of Virtue, Kant picks up this difficulty again in the casuistical questions. For instance, Kant’s first three questions concerning lust are paradigmatic examples of this difficulty. The first and last questions concerning the excessive use of drink also deal with this propensity of human nature to succumb to the natural dialectic. Similarly, the second question on lying involves a case in which one is tempted to tell a falsehood in order to avoid inadvertently insulting an author, and thus to avoid an uncomfortable situation. One might also add the second question regarding suicide to this category. Moreover, these questions highlight not merely the difficulty of following through with a moral demand in the light of contravening inclinations; they also highlight that the demands of our inclinations are not merely ‘impetuous’, but are represented to us as ‘so apparently equitable’. After all, is the pleasure of the kind of lively conversation that only happens after a couple of drinks really not a perfectly reasonable enjoyment? And aren’t many forms of sexual pleasure genuinely good even though they do not take ‘nature’s end’ into account? On Kant’s view, the inclinations that tempt us to these options are not merely strong, they also appear reasonable because there is nothing in itself wrong with pleasure. This makes it all the more tempting to rationalize away the arduous requirements of morality in favor of our inclinations.

Another version of this difficulty comes up in Kant’s last question regarding suicide. There, he asks if smallpox inoculation is morally permissible.³¹ Although this question is sometimes read as an example of conflicting grounds of obligation (insofar as the obligation in question seems to pull in two different directions), it cannot be an example of such a conflict, for the only obligation under consideration here is to preserve one’s life. The casuistical question asks whether one may pursue this moral end by taking the comparatively small risk of inoculation with a mild variant of smallpox, or whether one should instead avoid such risks and hope not to contract smallpox at all. Thus, at first sight, the difficulty in question concerns the appropriateness of specific actions in the pursuit of one and the same moral end. Kant answers this question in his handwritten remains, claiming that the maxim of a person who risks their life either willingly or by indifference can be accused of suicide (Refl 15:972–76).³² However, in these notes, Kant also points out that the conflict underlying this casuistical question is actually one between the ‘prudence in consideration of one’s advantage’ and the requirements of morality not to endanger one’s life. Kant’s answer is that considerations of advantage (increasing one’s chances of surviving smallpox) cannot outweigh requirements of morality (not to risk one’s life). As one would expect from the rigorist philosopher, Kant thought that no matter how certain the gained advantage might be, means contrary to the moral law must not be used in pursuing one’s advantage. Thus, although this casuistical question initially seems to concern either conflicting grounds of obligation or the appropriateness of a particular action to a moral end, the underlying difficulty turns out to be the human propensity to place a foreseen advantage over morality. Ironically, the very fact that the conflict appears, at first sight, to be moral underscores Kant’s claim that the natural dialectic makes us rationalize away the requirements of morality in light of the demands of our inclinations.

Another difficulty thematized in Kant’s casuistical questions is our frailty. In his third question on suicide, Kant refers to the belief that Frederick II of Prussia had carried with him a fast-acting poison, as Kant remarks, ‘presumably so that if he were captured when he led his troops into battle he could not be forced to agree to conditions of ransom harmful to his state’ (MS 6:423). As so many others, this casuistical question is not about conflicting grounds of obligation. From a Kantian perspective, the duties at stake appear to be (i) not to kill oneself, and (ii) not to agree to anything harmful to the state and its citizens. But these grounds of obligations are not themselves contradictory. Rather, the problem arises because of human vulnerability and frailty. The two obligations in question only give rise to a conflict when we assume that human frailty would make it impossible to comply with the latter duty without ending one’s life. Of course, only the most pathologically megalomaniacal among us could seriously believe that they could withstand the pressure of psychological and/or physical torture. What actions, then, are morally permissible when we expect that harm might potentially befall others because of our frailty? We might not share Kant’s seeming optimism about the capacity of rational beings to overcome their limitations by the force of pure reason, but even if we do not, Kant’s question remains perfectly reasonable. We might excuse, understand, even admire the suicide of someone who fears that they would not withhold torture and thinks suicide the only realistic option for protecting others. But Kant’s question perspicuously highlights the moral difficulty: is it merely to accept one’s vulnerability and frailty if one does not even try to overcome them? Or is an intention that presumes we would not be able to overcome these factors the same as to renounce our ability to do the right thing altogether?

A further difficulty concerns self-conceit. In his first two questions on servility, Kant points out that self-esteem is dangerously similar to arrogance. In an almost proto-Freudian fashion, Kant here suggests that one easily deceives oneself about the actual nature of one’s feeling. How, then, should we deal with such a propensity for self-conceit? Kant first asks if it would not be advisable for humans to cultivate humility by comparing ourselves with others in order to counteract our propensity to portray our arrogance as self-esteem. But he then asks if such a habit of cultivating humility by comparing ourselves with others might not make others view us as inferior, and thus appear in conflict with our duty of self-respect. The options adumbrated by Kant in these two questions—comparing ourselves with others and making others thereby despise us—are far less significant than the underlying problem. Indeed, once we take the problem seriously, we can see that the appropriate response is, first of all, to take the difficulty seriously, and to cultivate a habit of reflecting especially on those feelings that are so prone to self-deception.

Some of Kant’s casuistical questions also deal with the (seemingly more banal) difficulty of drawing a line. In his last question regarding lust, Kant asks at what point one may allow a latitude to our inclinations, and at what point compliance with a wide duty becomes pedantry. Similarly, in his second question on avarice and miserliness, Kant asks on which expenditures, and at what point in life, one ought to economize. And regarding beneficence, Kant asks how much of one’s resources one should give to others. All of these questions concern the necessity of a determination in situations that do not seem to have a clear limit or border. As such, the difficulty in these questions might seem quite trite at first sight. But what is more interesting are the competing considerations. For what makes these cases difficult is not simply the indeterminacy of the situation but the conflicting goals in question. We want to give lust a latitude because of our sexual drive; we want to minimize beneficence because we desire our own comfort; we economize on expenditures for ourselves because we want to keep or accumulate wealth. Unlike some other casuistical questions, Kant neither answers them in the Metaphysics of Morals nor anywhere else, nor does he seem to provide the resources for answering them. Evidently, what matters for Kant in these questions is not a solution, but an accurate awareness of the difficulty. In other words, what appears to matter in these casuistical questions is not simply the difficulty of drawing a line, but a more complete awareness of what makes the difficulty of drawing a line an ethical difficulty for human beings: the tension between rational principles and conflicting inclinations that reason rightly represents as equitable.

Another difficulty concerns conflicting grounds of obligation. In two of his most famous casuistical questions, Kant asks if one may sacrifice oneself for the good of all humanity (as in the legend of the Roman soldier Marcus Curtius) or kill oneself at the onset of hydrophobia in order to protect others. Contrary to the questions discussed above, these two casuistical questions do not concern the natural dialectic, human frailty, or our propensity for self-deceit. Instead, these two questions concern two equally applicable grounds of obligation: self-preservation and the well-being of others. Interestingly, Kant also answered these questions negatively in his lectures (V-MS/Vigil 27:629).³³ Although conflicting grounds of obligation only make for two casuistical questions, and although Kant eventually answers them in his lectures, they demonstrate that Kant was aware of the fact that conflicting grounds of obligation are a genuine moral difficulty—even where they can allegedly be resolved.

Finally, several casuistical questions are not concerned with the ethical difficulties from the agent’s perspective but concern difficult philosophical questions about moral obligation and political institutions. In this sense, these questions are better understood as only indirectly practical. For instance, in the final two questions regarding lying, Kant asks if one can be held legally responsible for the consequences of a lie. Although Kant’s example in the last casuistical question on lying concerns a lie by a servant who may feel contractually obliged to obey their employer’s order, there can be little doubt that Kant did not think of this situation as one of conflicting grounds of obligation, or as giving rise to any significant moral difficulty at all. Instead, Kant’s focus is on the relation between the immorality of lying and the permissibility of holding people legally accountable for their lie. Thus, the question asks not how one should act, but what rightful legal institutions would look like. While such legal liability may have practical consequences by deterring vicious or morally frail law-subjects, how one should act from an ethical perspective is not at issue. Moving even further away from directly ethical concerns, several casuistical questions concern the correct understanding of abstract concepts, as well as the correct classification of specific types of actions and attitudes. For instance, in his brief discussion of avarice, Kant asks in what sense we should understand the term ‘law’ in our talk about the moral law, given that the ‘internal lawgiver’ themselves can have difficulties in applying the law. In his discussion of servility, Kant asks if empirical evidence does suggest that it might be a widespread, and presumably unconscious, propensity in human beings. Kant also asks if testamentary bequests can be classified as beneficence, and whether the philanthropic endeavors of rich people in a deeply unjust society can be considered beneficence at all. Finally, Kant asks if pride might be the unconscious cause of ingratitude.

So far, commentators have often tended to neglect Kant’s theoretically oriented casuistical questions. However, I believe these questions are just as insightful as Kant’s practical questions. After all, Kant’s moral philosophy seeks to be therapeutic by unearthing the rational principles that underlie our moral cognition and their systematic interconnection. Consequently, we might expect that Kant’s casuistical questions highlight not merely the difficulties that his moral philosophy should help us to resolve, but also some of the difficulties in producing such a philosophy.

Kant’s casuistical questions might be remarkably diverse, but they all share a sensitivity to ethical difficulties. And, as we should expect from good moral philosophy, the ethical difficulties Kant’s casuistical questions highlight are themselves quite diverse, mirroring the complex nature of human life. Without such complexity, there would be little space for rationalizing and self-deception.³⁴ If moral philosophy is to help us overcome this corrosive aspect of our nature, it should not merely tell us what self-deceptive reasoning looks like but also be sensitive to the difficulties that underlie it.

6. Conclusion

Kant’s casuistical questions vividly demonstrate the complexity of ethical life. This complexity is not exhausted by puzzled agents asking themselves what they should do. It involves the self-deceived agent, the frail agent, those struggling with the natural dialectic, the morally unlucky who cannot act without going against some ground of obligation, and those struggling to find the right theoretical understanding of their practical situation.

Since many of Kant’s casuistical questions emphasize the natural dialectic as a cause for moral difficulty, they also forge a direct bridge between the first and final pages of Kant’s moral philosophy. In section I of the Groundwork, Kant had remarked that ‘[t]he human being feels within himself a powerful counterweight to all the commands of duty, which reason represents to him as so deserving of the highest respect […]’. And, therefore, practical wisdom ‘needs science, not in order to learn from it but in order to provide access and durability for its precepts’ (GMS 4:405). With this view about the practical purpose of moral philosophy, Kant distanced himself significantly from thinkers like Rousseau, who, though they shared Kant’s epistemic egalitarianism about moral cognition and his rejection of baroque casuistry, had a much more pejorative view about the role of scientific and philosophical reasoning for ethical life.³⁵ Kant’s project of a metaphysics of morals is an attempt to analyze the a priori principles that underlie our common moral cognition and their systematic interconnection. Such an analysis should also provide us with the resources for a therapeutic way of reflecting, a reflection that brings us closer to our true nature as rational beings. But in order to do this, Kantian moral philosophy must be sensitive to the specifically human difficulties we face in our ethical lives. Only if we take these diverse difficulties seriously can moral philosophy get closer to an improvement of our situation.³⁶

Competing Interests

The author has no competing interests to declare.