Kant, casuistry and casuistical questions

Interest in the role of casuistry and casuistical questions in Kant’s Doctrine of Virtue ( DV ), i.e. the second part of the Meta-physics of Morals , has grown in recent years. My own position is formulated in Schuessler (2012, in German), the main thesis of which will be retained here in an updated form and with some shifts of emphasis. I hold that the casuistical questions concerning perfect duties in the DV are not intended to represent casuistry in Kant’s sense. Casuistry and casuistical questions areneitherequivalentinthe DV nordotheyservethesamepur-poses.

section concludes the paper.Throughout I will comment on the practices of early modern casuistry, which are often not well understood by Kant scholars, and emphasise that Kant's use of casuistical questions reflects the Pietist practices with which he grew up.

CASUISTRY, CASUISTICAL QUESTIONS AND THEIR PLACE IN THE DOCTRINE OF VIRTUE
Kant refers to casuistry (Kasuistik) in just one extended passage of the DV.Besides this, the DV contains 20 casuistical questions (kasuistische Fragen) whose purpose is not explained in the passage on casuistry but in the Doctrine of Method (DM) of the DV. 1 Let us first look at what Kant has to say about casuistry. 2 the introduction to the DV, among other issues, Kant discusses foundational concepts for a division of the DV into several branches. 3In a long remark, set apart in the text, he explains why he distinguishes between a Doctrine of Elements (DE) and a Doctrine of Method (DM), whereas no such distinction was drawn in the first part of the Metaphysics of Morals, the Doctrine of Right.The distinction is rendered necessary according to Kant because ethical duties are wider than duties of right.That is, they allow for greater latitude in application.The application of all duties, no matter whether they are ethical or juridical, requires subsumption.Subsumption is an activity of judgment (Urteilskraft) which determines whether a case falls under a rule.The relevant rules in a Kantian ethical context are laws or maxims.Subsumption is therefore necessary for deciding whether we ought to follow a law or maxim in a given particular case.
Kant asserts that subsumption in the DV is bound to differ from subsumption in the Doctrine of Right, and introduces casuistry in this context: Hence, the doctrine of right, which by its nature must determine duties strictly (precisely), has no more need of general directions (a method) as to how to proceed in judging than does pure mathematics; instead, it certifies its method by what it does -But 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, which has no place in the doctrine of right.(Kant, 2017, p. 180   [6:411]) It is important to note with respect to this passage that Kant is not claiming that ethics 'falls into' casuistry because its duties generally allow for more latitude in application than duties of right.Kant assumes a difference of latitude sufficient to give rise to casuistry only for imperfect duties.Perfect ethical duties, it seems, can with respect to subsumption be dealt with in the same way as duties of right.There is, of course, some discussion among Kant scholars on how to understand the difference between Kant's distinction of narrow versus wide and perfect versus imperfect duties (see Denis, 2001, p. 30).However, we may presently restrict our attention to what is required for understanding Kant's remarks about casuistry.He seems to think that for duties of right it is easy to determine whether a case falls under a rule or not.His reference to mathematics suggests that subsumption in these fields is as easy to judge as whether a particular manifestation of a geometric figure falls under the concept of triangle.For imperfect duties, by contrast, a much more elaborate process of judgment is often required because we do not see immediately whether, for instance, refraining from helping a particular beggar violates an imperfect duty of love (i.e. a Liebespflicht, a duty of charity or beneficence).
Perfect ethical duties sit uncomfortably between the two kinds of duties which Kant explicitly addresses.However, taking him at his word, perfect ethical duties are to be grouped with duties of right, and for the latter the subsumption of cases under a duty can be judged without casuistry.This claim shows how the term 'latitude' in the quoted passage should be understood.It involves the fact that imperfect duties are always valid but do not always call for action (i.e. they are valid semper, sed non ad semper in traditional terminology).This temporal and circumstantial latitude in complying with imperfect duties is the main reason for the use of subordinate maxims.Yet, another possible latitude exists with respect to the interpretation of concepts.There can be more or less room for interpretation whether a case falls under a concept.This is a latitude to which Kant here is apparently not referring, and with good reason.While the distinction between always and not always being obliged to act is clear, it is not at all clear that all legally binding concepts are less fuzzy in application than all perfect ethical duties, and these in turn less fuzzy than all imperfect ethical duties.Cases of an imperfect duty of respect might be more easily delimited than cases of a particular property right, e.g.lend-lease.In any case, Kant's restriction of casuistry to imperfect duties strongly suggests that he only has the first kind of latitude in his sights when justifying the use of casuistry.
For compliance with imperfect duties, further rules are helpful and sometimes even required.Agents need to identify these rules and then allow themselves to be guided by them.Kant specifically tasks judgment with finding subordinate maxims. 4One might think, for instance, of a maxim of giving 10% of one's income to the poor.Not giving anything to a beggar at one's door (who is not about to starve or is otherwise in 'extreme necessity' , a state of immediately life-threatening neediness in the Christian ethics of charity) would then clearly not constitute a violation of duty for a person who already has spent more than 10% of her income on charity.Kant identifies this more complex procedure of subsumption which includes the use of submaxims with casuistry.Other forms of subsumption, by contrast, might rely on a direct perception that an act is of type x and that it conflicts with a law that prohibits x.Here, no conscious use of a rule for guiding the subsumption is necessary, and consequently casuistry has no place.At this point, I would like to add that Kant here retains aspects of the traditional practice of casuistry, which existed for centuries before the DV and had boomed in the 17th century (see more below).Casuists were specialised moral theologians who typically used subordinate rules to decide how morally problematic cases of conduct were to be judged.Based on subordinate rules and area-specific principles they could claim that an action x was morally permissible, whereas a deceptively similar action y was not.
After having justified the existence of casuistry, Kant continues: Casuistry is, accordingly, neither a science nor a part of a science; for in that case it would be dogmatics, and 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, not [presented] systematically (as dogmatics would have to be), and is added to ethics only by way of scholia to the system.(Kant, 2017, p. 180 [6:411]) I take this to be a quite puzzling statement.Kant asserts that casuistry is not a scientific endeavour.It is more a practice of seeking moral truths than a doctrine for finding them.The emphasis is here on the distinction between a doctrine and a practice.Kant should therefore be interpreted to claim that casuistry cannot be done in an 'algorithmic' way.
There is no precise method for finding subordinate maxims, which one might learn by reading a handbook.Casuistry is more of an art, a practice depending on experience and training with cases.This is exactly how it was learned in the early modern era, beyond the acquisition of some general rules that functioned as a shared toolset for casuists.For this reason, it might be considered 'fragmentary' , in the sense of something that can be shown with examples but does not allow for a general theoretical explication.The reference to scholia, the explicative comments on text passages in traditional commentaries, seems to stand in a certain tension with this interpretation.Yet, Kant probably had exemplifying rather than theoretically explicative scholia in mind.
Next, Kant turns away from considerations concerning casuistry and speaks about the DM of the DV.He emphasises that reason rather than the faculty of judgment informs the theory of ethics and grounds its practical exercise.Kant then elaborates on methods for teaching ethics, and especially for testing whether pupils have understood an ethical subject matter correctly.
All these considerations help us, or so at least Kant seems to think, better to understand the architectonic of the DV offered a page later.According to Kant, ethics has the following branches: 5 Ethics (ethische Prinzipien eines Systems der reinen praktischen Vernunft) • Doctrine of elements (Elementarlehre) • Dogmatics

• Ascetics
Casuistry is here a major branch of the DE, whereas casuistical questions have their systematic place in the DM, specifically as a method of catechising.Kant later confirms this separation when speaking about the teaching of ethics (ethische Didaktik) as the first part of the DM.He claims: In this catechistic moral instruction it would be most helpful to the pupil's moral development to raise some casuistical questions in the analysis of every duty and to let the assembled children test their understanding by having each say how he would solve the tricky problem put to him.(Kant, 2017,   p. 244 [6:483]) Casuistical questions are thus characterised as a teaching instrument.Teachers should use them to test pupils and promote their moral development, which in this context mainly means their understanding of duties.
Taken together, Kant's assertions concerning the role and systematic place of casuistry and casuistical questions in the DV indicate that casuistry and casuistical questions are not two equivalent ways of referring to the same subject matter.Much of the literature on Kant's practical ethics has tended to overlook this important point.In fact, answering casuistical questions need not involve casuistry as defined by Kant.Casuistry is preoccupied with the practice of applying imperfect duties in the real world.As such, it belongs to the DE.Casuistical questions are classroom exercises.
They can be raised for imperfect as well as for perfect duties.The systematic role of casuistical questions in the DM as an instrument of catechetics does not prevent their application to subject matter from the DE, and there is no restriction in this respect as to the kind of duties for which they can be raised.Kant in fact discusses casuistical questions for perfect as well as imperfect duties in the DV, which would be inconsistent if casuistical questions were invariably a way of doing casuistry in the sense stated by Kant.
A look back at the early modern tradition of casuistry helps to resolve some of the present entanglements.Casuistry was not just a didactic exercise in the early modern period but another term for applied moral theology. 6In modern diction, it was a very elaborate practical ethics.As such, it told princes whether they might wage a just war in given cases (Protestant and Catholic casuistry alike), it allowed or prohibited vestments with multicoloured sleeves (Calvinist casuistry) or permitted Christians to compromise with the Emperor of China concerning the vestments of priests (Catholic casuistry).Hence, casuistry strove to offer practicable moral guidance beginning with quotidian moral problems and ranging to the most impactful questions of international politics or economic governance.Kant alludes to this traditional role of casuistry by listing it as the second main branch of the DE.
At the same time, casuistical questions had a longstanding use in classroom exercises and for the training of professional casuists.(Casuistry could be studied as a subject with its own professors and curriculum at some early modern universities and colleges.)Of course, casuistical questions could address the problems with which professional casuists dealt as confessors or as moral counsellors of the mighty.In other words, casuistical questions could be treated in a classroom with the same problem-solving or problem-analysing mindset as in real moral practice.However, it was also possible to use casuistical questions merely to test what pupils or students had learned about the basics of practical ethics.In this respect, the use of casuistical questions did not differ much between Catholic and Protestant casuistry, both of which reached their all-time apogee roughly between 1580 and 1720.Professional casuistry remained a main conduit of practical ethics in Catholicism throughout the 18th century, or at least until the 1760s.Protestant casuistry experienced a far steeper decline after 1700, with a notable exception, which is of particular interest with respect to Kant.
Pietism, a religious movement that was highly successful in Prussia in the 18th century, used casuistical questions in a peculiar way in classroom exercises even for quite young pupils.Kant grew up in a Pietist environment, and it has been argued-justly I think-that his philosophy displays Pietist influences, even though Kant did on the whole view Pietism critically. 7Pietists emphasised the importance of true faith and a pure heart rather than appreciating 'Jesuitical' refinements in treating moral problems.Hence, for them, inculcation of the right attitude of a sincere person mattered far more than ferreting out what might still be permissible in a moral borderline case. 8Pietist moral education included catechetical instruction for quite young children, but Philipp Spener (1635-1705), the founding father of Pietism, also recommended moral discussion groups for university students where cases of conscience could be discussed. 9It is not documented that Kant ever formed such a group among his students.Nevertheless, Kant's treatment of casuistical questions hints at a usage similar to that envisaged by Pietists.Primarily, casuistical questions show whether a pupil is ready to follow the moral law with the right attitude.Yet, of course, a pupil's ability to correctly subsume cases to duties also mattered.Moreover, perfect duties are more important in this respect than the imperfect duties which allow for more latitude.Latitude implies that there may be alternative subordinate maxims which agents might follow.Given large enough latitude, the maxims might propose different actions (as in cases of giving to the poor).The DE only requires that an agent finds a subordinate maxim in cases where there is large enough latitude.
Not finding and following a permissible maxim would imply a moral mistake.However, Kant does not demand in this context that all agents follow the same subordinate maxims.Note that otherwise finding these maxims could, contrary to what Kant says, be turned into a science.This pluralism renders concrete subordinate maxims quite unsuitable for testing whether pupils have the right moral understanding, since pupils are required to understand only that they need to find a suitable submaxim.Correctness of understanding is better tested by asking what morality tells persons not to do.We should bear this in mind when we discuss Kant's actual casuistical questions in more detail.For the moment, let us sum up what we have found so far.
For Kant, casuistry and casuistical questions are situated in different branches of the DV.Casuistry and casuistical questions do not necessarily follow the same modus operandi or serve the same purposes.Many didactical casuistical questions can be answered without practising casuistry, at least as Kant characterises it.Moreover, the similarities between Kant's exposition of casuistical questions and Pietist educational practices suggest that Kant, in fact, makes scant use of true casuistry in his examples for teaching in the classroom.(The latter kind of didactical use would be more characteristic of Catholic higher education and of Protestant theological education in the 17th century.)

A VIEW OF THE LITERATURE ON KANT AND CASUISTRY
Studies of Kant's practical ethics and even the literature on the DV usually fail to distinguish between casuistry and casuistical questions. 10 Most of this literature interprets Kant's remarks concerning casuistry and his casuistical questions without situating them in the historical context of early modern practices of casuistry, except for passing references to this tradition.Of the few writings whose focus squarely rests on Kant's casuistry, Matson (1968), Kittsteiner   (1988) and Patrone (2013) offer general assessments, without a detailed discussion of the casuistical questions.For Matson, casuistry comprises all sorts of action-guiding judgments intended to comply with theoretically grounded duties.He mainly discusses examples from the Groundwork and On a Supposed Rght to Lie.The DV is not Matson's concern.
Kittsteiner, an expert on Protestant ethics of conscience, claims that Kant positions himself as an anti-casuist with his theory of strict perfect duties.That is, Kant's conception of perfect duties is (among other things) a conscious attempt to prevent the confusing jungle of traditional casuistical judgments from arising.This is, in my view, a correct assessment and a warning to Kant scholars who think that a Kantianised theory of practical wisdom might be used to soften Kant's perfect duties in application.Such tendencies neglect the fact that early modern casuistry amply documented the absence of agreement in practical ethics.Disagreement abounds with respect to the solution of all but a few of the 20,000 cases assembled in Antonino Diana's gigantic Moral Solutions (Resolutiones Morales, 1633). 11nt realised that only a very 'narrow' interpretation of perfect duties, which leaves no room for different choices of action, prevents intersubjectively conflicting application in practice.Too much leeway will breed disagreement and conflict.This is exactly what an ethics of pure reason is meant to avoid.Kittsteiner regards the remnants of casuistry in the Metaphysics of Morals as hardly relevant, without engaging in earnest with the casuistical questions of the DV.
Patrone's discussion of Kant's casuistry mainly strives to show that the categorical imperative is not a decision procedure.She recognises that casuistry, as the term is introduced in the DV, is unsuited to 'resolving conflicts between inconsistent grounds of obligation or in determining the specific requirements generated by imperfect duties' (Patrone, 2013, p. 491).However, claiming that the educational use of casuistical questions can replace the categorical imperative as an instrument for the guidance of actions is hardly what Kant had in mind.
The first notable in-depth discussion of the casuistical questions in the DV is James (1992), who charts the field for subsequent investigations.James is aware that the number of 20 questions highlighted in the title of his paper is a matter of interpretation.He explicitly admits that alternatives to his reckoning are possible because Kant sometimes seems to split one question into several formulations, and also seems to throw in some merely rhetorical questions.
It is, of course, difficult to determine what a merely rhetorical question is in the context of classroom interrogation.
What might be a mere rhetorical question for an expert can give children pause to think.If we concentrate on the eight bundles of questions which appear under the header 'casuistical questions' in the DV, a narrow reckoning will in my view lead to 27 and a wide reckoning to 36 questions.The number 20 thus refers to thematic topics of questions rather than to questions in a literal sense.Nevertheless, it seems reasonable to focus on the thematic topics of Kant's casuistical questions and the expression they find in their best formulations rather than to bother with each and every formulation in the DV.
In contrast to the present investigation, James does not distinguish systematically between casuistry and casuistical questions.Accordingly, he considers Kant's presentation of both as deeply muddled.James notes the contradiction between Kant's claim that casuistry is only concerned with imperfect duties and the ample discussion of casuistical questions concerning perfect duties.James also finds a contradiction in Kant's claim that casuistry is no science and his remark that casuistical questions can help a person to attain a state of science with respect to the assessment of duties.Yet, if casuistical questions need not imply the use of casuistry, the apparent contradiction vanishes.
It is then possible to ascribe the practice of casuistry to the faculty of judgment, buttressing the claim that casuistry is no science.Science builds on the faculty of reason, and guidance by reason is the highest aspiration in theoretical moral education and the classroom exercises of the casuistical questions.It is therefore not necessary to accuse Kant of a blatant self-contradiction so long as we take him at his word and abandon James's identification of casuistry and casuistical questions.We will address this point in more detail below when discussing specific casuistical questions.Unna (2003) deals only with a subset of Kant's casuistical questions, those pertaining to 'self-disembodiment' , or, for short, suicide. 12However, her claim that answers to the casuistical questions concerning suicide can be found in Kant's published and unpublished writings is important for understanding casuistical questions and their general purpose.Scholars (like Schuessler (2012)) who agree with Unna can use Kant's own answers to his casuistical questions concerning perfect duties to show that they are not meant as casuistry in the sense of opening a space for differentiated and varying answers depending on circumstances.They are rather raised to inculcate a strict and rigorous understanding of perfect duties.Kim (2009) focuses on the DV but without distinguishing between casuistry and casuistical questions.Except for a comment on the questions on suicide, there is no deeper discussion of the casuistical questions.The aim is mainly to understand Kantian casuistry against the background of the literature on traditional and modern casuistry.Kim emphasises, correctly in my view, that Kant's casuistry is not meant to offer exceptions to rules but to guide agents in the application of wide duties.Oggionni (2017) takes issue with the view that Kant elsewhere gave firm answers to the casuistical questions of DV.Her purpose is to make room for the 'softening' interpretation of Kant's perfect duties which has spread in contemporary Kantianism.To this end, she emphasises the difficulties inherent in using Kant's unpublished writings as Unna (2003) and Schuessler (2012) do.There can be no doubt, of course, that all interpretations of Kant's casuistical questions remain to some extent uncertain.The relationship between the casuistical questions of the DV and remarks in Kant's other writings, which can be interpreted as answers, cannot be determined with certainty.These are, however, usual concomitants of philosophical interpretation.Oggionni's merely general reservations offer, in my view, no sufficient grounds to claim that the casuistical questions concerning perfect duties in the DV are meant to be solved through casuistry, as Kant understands it.None of these questions require the introduction of subordinate maxims which modify the stringency of perfect duties or balance conflicting grounds of obligation.This view is not only supported by disparate remarks of Kant's which may be taken as answers to casuistical questions, but also by Kant's explicit claim that casuistry deals with imperfect duties, and by the role that casuistical questions played in Pietist educational practice.Hence, the interpretation which Oggionni opposes both coheres with the context in which Kant wrote and allows us to take Kant at his word.
Höffe (2020) discusses conscience, probabilism and three casuistical questions on lies, avarice and servility.With respect to the three questions, he offers an explanation why Kant, contrary to what one might expect, does not invoke the categorical imperative to solve the cases.Staying close to Kant's text in his analysis, Höffe claims that Kant does not use the methods of casuistry in his treatment of the cases.Hence, Höffe, like Schuessler (2012), distinguishes between casuistry and casuistical questions.The added reference to conscience is interesting because Kant mainly required subjective moral certitude to attain a good conscience.On this basis, Kant was able to declare that an erroneous conscience is impossible because we can know whether we feel subjectively certain in believing a moral proposition.Casuistical questions can help to reveal whether an agent or a pupil has the required subjective certainty. 13suistry and casuistical questions are also often briefly addressed in general commentaries on the DV and in works on Kant's virtue ethics or pedagogy (e.g.Morris, 2020, included in this issue).In this literature, casuistry and casuistical questions are usually not distinguished, and it is assumed that casuistry plays the same role in Kant as in the tradition of early modern casuistry.The aim of the latter is supposedly to chart a moral course in conflicts of obligation, or in a more Kantian formulation, in cases where grounds of moral obligation conflict.Esser (2008), Louden (2000), O'Neill (2002)   and many others emphasise the importance of this concern for Kant's ethics in general and for the DV in particular.
By resorting to casuistical questions, according to this interpretation, Kant acknowledges that reasons for obligation (rationes obligandi) can conflict and that his ethics is able to determine the balance of the conflicting reasons (Esser,   2008; O'Neill, 2002, p. 343).It is necessary to speak of reasons for obligation here, because for Kant, moral obligations cannot conflict.True moral dilemmas therefore cannot exist (Kant, 2017, p. 19 [6:224]).In apparent conflicts between obligations, one side always prevails all things considered, and at least in a given case, the weaker side cannot generate an action-binding obligation.This was, indeed, a common assumption in traditional natural law theories and in the casuistical tradition.
In principle, it would be possible on this basis to resolve conflicts between obligating reasons in any given case straightforwardly by weighing the reasons and ascribing obligating force to the weightier side, just as Esser, Louden and O'Neill assume.This would open a door to a softening of Kant's apparent moral rigorism through a case-by-case weighing of reasons.On this view, the casuistical questions of the DV suggest that Kant approved of case-by-case weighing but, unfortunately, he personally too often came to rigoristic results.We might therefore discard Kant's actual weighing as the idiosyncrasy of an old Prussian professor and accept weighings that appear more plausible to us without abandoning the modus operandi of Kant's ethics.To modern minds, this storyline holds considerable appeal, and it has gained some prominence among Kant scholars.
Since I do not consider the storyline correct and would like to convince you of its wrongness, it seems best to oppose it right away before we turn to the casuistical questions of the DV in more detail.As indicated, Kant could have proceeded in the way described.Yet, he did not.And he had reasons for his reluctance arising from the early modern casuistical tradition.Early modern casuistry, whether Catholic or Protestant, strove to chart and to delimit the space of permissible actions. 14In this endeavour, it did not rely on a freewheeling application of Aristotelian practical wisdom to cases of conscience.Early modern casuists did not start with a comparison of paradigmatic cases.Before early modern casuists approached a case, they already inhabited a whole edifice of moral theology in which cases of conscience were embedded.Moral theology contained theological, metaphysical, psychological, ethical and epistemological principles, rules and law-like generalisations.It also had a specific architectonic.All this informed a casuist's judgment of cases no less than the immediate case-specific moral reasons (rationes or motiva rationis in the Latin of the sources) encountered in controversial cases.Moreover, casuists were aware that they were not alone in judging cases.
Their social epistemologies attributed weight to the decisions and arguments of other competent reasoners, which comprised not only other moral theologians but also 'experts in an art' .At this point, the problem of disagreement became acute in early modern casuistry.Casuists realised that they disagreed with respect to the solution of nearly all problems of practical ethics.(This may be considered a truism because cases with an easy consensual solution do not give rise to problems of practical ethics.)In the 17th century, the main concern of theoretical discussions concerning casuistry was therefore how the rampant disagreement between casuists should be handled.Hundreds of specialised treatises contributed to this discussion.As already indicated, the abundance of disagreement was also underlined by huge case collections like Diana's Moral Solutions from 1633.
Looking back to the casuistical tradition, Kant could see that different moral agents would weigh conflicting reasons differently in problematic cases.The result would not only be moral disagreement but most likely conflict.Kant aspired, at least concomitantly, to avoid disagreement and conflict by striving for aprioristic certainty in ethics. 15reement in practice was particularly important with respect to the prohibitive duties that protected the core of human dignity (i.e. the perfect duties).A permission to weigh conflicting reasons with respect to this core would again invoke the quagmire of disagreements whose sordid effects the tradition of casuistry had amply documented, or so Kant believed.In his view, ethics had to offer a shared orientation in action at least with respect to the perfect duties, and this is precisely what the modern 'Neoaristotelian' approach to Kant jeopardises.Of course, a modernised Kantian approach which steers closer to Aristotle and puts more trust in the consensus fostering powers of practical wisdom (or fears disagreement less) might be more attractive today. 16However, there will always exist an alternative Kantian approach which tries to minimise the occasion for moral disagreements, and this approach will steer closer to Kant.
Defenders of the Neoaristotelian approach to Kant may object that his casuistical questions at least show that some problematic forms of conduct which are closely related to prohibited actions are nevertheless licit.Permissibility here hinges on the action or maxim not falling under the moral law's prohibition.It is certainly true that we need to attend closely to Kant's distinction between action types if we want to determine how he conceives the scope of perfect duties.For Kant, in agreement with large swathes of the Christian moral tradition, the withholding of information, simulation and dissimulation should not generally be classified as lying (Mahon, 2003).Modern ethicists may judge this issue differently.In any case, as the above quotation from Kant (2017, p. 180 [6:411]) shows, the faculty of subsumptive judgment which determines under which category of action type an action or a maxim of action falls does not require the application of casuistry in the case of perfect duties.Moreover, in Kant's view, practical wisdom has no strong role in subsumption, which does not depend on a weighing of reasons of obligation but just asks whether a case falls under a concept.The concepts which Kant uses (like simulation, dissimulation and lying) are given by the history of practical moral thought.They are not invented by the agent while comparing cases.The casuistical questions in the DV suggest that Kant wants to use moral education to forestall freewheeling interpretation, which in his view is wont to be undermined by rationalisations and the influence of self-love.Moral agents need to learn not to infringe the scope of strict prohibitions, e.g. by interpreting lies as mere simulation or some cases of suicide as self-sacrifice for the common good.The casuistical questions concerning perfect duties in the DV are therefore not meant to teach how hard cases can be classified so as to avoid subsumption under perfect duties but rather to fortify subsumption under perfect duties against forms of rationalising that nibble away at its scope.

THE CASUISTICAL QUESTIONS CONCERNING PERFECT DUTIES TO ONESELF
Six of the eight clusters of casuistical questions in the DV deal with perfect duties to oneself.This must appear utterly strange if casuistical questions invariably involve casuistry.After all, Kant explicitly restricts casuistry to the field of imperfect duties.The simplest and best solution from an exegetical perspective is to acknowledge that casuistical questions need not involve casuistry in the sense described by Kant.In fact, in the passage quoted in the first section above, Kant recommends using a catechetical and Socratic method in moral education.These methods involve a question-and-answer exchange between teacher and pupils documenting whether the pupils grasp what they are supposed to learn.When discussing the task of a moral catechism in the DV's Doctrine of Method, Kant again emphasises the utility of casuistical questions: In this catechistic moral instruction it would be most helpful to the pupil's moral development to raise some casuistical questions in the analysis of every duty and to let the assembled children test their understanding by having each say how he would solve the tricky problem put to him.(Kant, 2017,   p. 244 [6:483]) Since Kant is talking about tricky problems here, he might appear to allude to the use of casuistry.However, this is by no means a safe conclusion.A catechism is usually designed to imbibe the basics of faith and not meant to delve into sophisticated theological interpretations.Moreover, Kant offers an example of a catechetic Q&A exchange right before formulating the quoted passage.The teacher asks: Suppose, for example, that a situation arises in which you could get a great benefit for yourself or your friend by making up a subtle lie that would harm no one: what does your reason say about it?Pupil: That I ought not to lie, no matter how great the benefits to myself and my friend might be.(Kant,   2017, p. 242 [6:481]) The example shows that never to violate a perfect duty such as the duty not to lie was a possible right answer to a tricky casuistical question in the classroom context.To answer correctly, the pupil need not identify cases of simulation which might be morally permissible because they do not involve lies (in the way casuists were wont to do).In short, to answer the teacher's casuistical question, the pupil need not resort to casuistry.Kant is explicit about this also in his lectures on pedagogy.The teacher is to ask moral questions from everyday life in which it needs to be decided whether an action is morally right or not.Should we delay repayment of a debt if the money would help to alleviate great suffering?Is it permissible to use lies in cases of emergency?Kant both times answers with a resounding 'No!' (Kant, 2007, p. 477 [9:490]).He does not resort to subtle maxims in order to make distinctions.Kant just wants to hear: 'Message understood!I am never allowed to do this!' Against this background, we should not assume in advance that Kant wants to hear anything else with respect to the casuistical questions concerning perfect duties to oneself.Kant never uses the term 'casuistry' in this context, speaking only of casuistical questions.The burden of proof is on commentators who argue that the presence of casuistical questions concerning such duties shows that Kant wants to make compliance with perfect duties flexible.His six clusters of questions pertain to suicide (self-disembodiment), lust (defiling oneself), intemperance (stupefying oneself), lying or honesty, avarice or thrift, and servility.I will not address all six clusters here but focus on suicide to show that the existence of the questions should not be taken as proof of an intention to relinquish exceptionless obligations or to apply casuistry in their application.With respect to almost all questions concerning suicide in the DV, Yvonne Unna has shown that Kant offers answers elsewhere in his published works or lectures.In all these cases he claims that suicide can never be morally right. 17Admittedly, he sometimes seems to defend the moral integrity of agents who commit suicide.In a lecture on the metaphysics of morals, Kant adverts to the noble attitude of the Stoics with respect to suicide and the case of the Roman hero Marcus Curtius, who sacrificed his life to save his fatherland.As Livy records, Curtius jumped into an abyss to placate the Gods who appeared to threaten Rome with destruction through a landslide (Kant,   1997, p. 370 [AA27.2.1:629]; Unna, 2003, p. 459).Kant regards Curtius's sacrifice as equivalent to suicide and therefore as morally wrong.Although Kant sympathises with the hero, respect for the moral law, as Kant understands it, retains the upper hand, and leads to a negative answer.However, would Kant also have disallowed the self-sacrifice of a soldier who throws himself on a grenade to save his comrades?In such cases, the line between forbidden suicide and permitted sacrifice for the common good becomes razor thin.In the soldier's case, the grenade is thrown by others, and therefore the deadly mechanism is not triggered by the agent himself.This may suffice to regard shielding others with one's own body as permissible, even if it leads to certain death, while Curtius's superstitious deed is classed as impermissible suicide.Kant's casuistical questions concerning suicide are, among other things, about drawing the line in tricky cases.Yet, such questions of subsumption of cases under a rule are not what Kant thinks casuistry is generally concerned with.
In the case of Seneca, who committed suicide to forestall murder by Nero, Kant expresses his sympathy for the agent quite openly.He recognises that suicides like Seneca's aim at upholding the person's honour and dignity, and such considerations are dear to Kant.Yet, in the end, the dignity of humankind and the moral law prevail.Kant professes not to demand a defence of the respective suicidal actions. 18Suicide is always morally wrong for Kant.The single question on suicide in the DV which Kant does not answer negatively elsewhere concerns the planned but not executed suicide of Frederick II (called the Great) of Prussia (Kant, 2017, p. 191 [6:423]).Frederick carried a box with poison in case he was captured by enemies in the Seven Years War.Kant does not tell us whether he considered this plan as morally condemnable.It would certainly have been very risky to publicly take a stance on this matter.However, Kant's documented answers to all the other questions concerning suicide indicate with considerable likelihood what he really thought of the allegedly great Frederick. 19ubtlessly, Kant's intransigence with respect to perfect duties makes it more difficult to defend him today.A bit more casuistry, a greater willingness to brook exceptions, would draw Kant's ethics closer to present-day moral attitudes.Modern Kantians may be justified in venturing beyond Kant in this respect, but they should not claim to have the original Kant on their side.It is important to understand that from the perspective of his age, Kant was not incomprehensibly narrow-minded when taking an exceptionless stance with respect to perfect duties.The casuistical questions concerning perfect duties to oneself deal with what traditionally was prohibited as mortal sin.In Christian ethics, it is always wrong to commit a mortal sin.Kant, although a philosopher of the Enlightenment, was not a revolutionary with respect to moral content.The DV ostensibly transfers Christian sins into the framework of a secular ethics (and even the subbranches of the DV [i.e.dogmatics, casuistry, catechetics, ascetics] mirror those of early modern textbooks of moral theology and thus betray theological roots).In the process, mortal sins remain strictly prohibited.This is not idiosyncratic for Kant in his place and time.In Kant's view, attempts to circumvent strict duties by way of casuistry had led to shocking results in the centuries before him.The case-wise weighing of reasons had given rise to subterfuges and the immoral practices of Jesuitry, and the only way to prevent sliding down this very slippery slope was to be strict with respect to perfect duties even in application.
The only casuistical question concerning perfect duties to oneself that apparently spoils this reading deals with self-intoxication.Here Kant faces the problem that a zero-tolerance strategy was not backed by received Christian morality.Moreover, insisting on strict abstinence from alcohol would have meant the end of convivial intellectual exchange as Europe knew it.And while you cannot commit suicide to a still healthy degree, consuming a moderate amount of alcohol was certainly not considered as intoxication in Kant's time.In this context, Kant contemplated what size of dinner parties was conductive to convivial intellectual exchange without motivating drunkenness (Kant,   2017, p. 195 [6:428]).Large parties, with much singing, dancing and drinking, were certainly not helpful for a deeper exchange of views.Kant quotes the British author Chesterfield's suggestion of taking the number of muses, that is, nine, as the limit.Here, indeed, Kant seems to use a maxim ('Never attend dinner parties with more than nine guests') in the way his conception of casuistry in the DV suggests.However, the maxim is not concerned with determining the application of a perfect duty.The perfect duty demands that one not intoxicate oneself, and the size of a dinner party does not tell us anything about intoxication.The maxim comes only into play because Kant considers convivial exchange as a legitimate reason to consume alcohol, and the maxim regulates convivial exchange, not intoxication.
Convivial exchange serves the purpose of self-cultivation and politeness in dealings with others.It is thus linked to an imperfect duty, for which casuistical determination of subordinate maxims is in order.The maxim of the number of muses is therefore not a counterexample to the claim that Kant's treatment of perfect duties to oneself does not involve casuistry.

IMPERFECT DUTIES AND THE CASUISTICAL QUESTIONS CONCERNING DUTIES OF LOVE
If we take Kant at his word, the place of casuistry in the DV would be among the imperfect duties to oneself or to others.Yet, Kant does not raise casuistical questions in the sections on imperfect duties to oneself or duties of respect to others.However, he asks casuistical questions with respect to two of three categories of the (imperfect) duty of love.How are we to interpret this incompleteness?Moreover, is Kant at least in these cases engaging in casuistry, or do the casuistical questions again only serve didactic purposes?
Kant formulates casuistical questions only for beneficence and sympathy.It is difficult to see why gratitude should be absent.In fact, Kant touches upon the duty of gratitude in the questions concerning the other duties of love.In my view, he just uses material in this context which he had at hand from his lectures and does not care much for honouring architectural symmetry.This is a sign that he probably did not consider the casuistical questions concerning duties of love, or imperfect duties in general, as very important.We will come back to this point.It is easier to find a reason why Kant would add a casuistry specifically to the duties of love.Kant distinguishes not only between perfect and imperfect duties but also between narrow and wide duties.Narrow duties bind actions directly, while wide duties bind actions only indirectly via maxims (Kant, 2017, p. 163 [6:390]; Denis, 2001, p. 30).Moreover, some wide duties are wider than others.The widest duties quite obviously stand most in need of structuring through subordinate maxims because they allow for most latitude in actions.That is, the wider a duty, the better it is suited for casuistry, and Kant assumes that duties of love are the widest duties of all.Hence, we should expect casuistry if at all then for duties of love, and there, indeed, we find casuistical questions.
However, at closer inspection, the questions which Kant offers seem again to be mere classroom exercises with hardly a connection to moral casuistry.Questions like '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?' do not ask whether an action is morally right or not (Kant, 2017, p. 221 [6:458]).They merely reveal whether pupils have understood some not directly action-relevant moral subject matter.Another question, pertaining to a conflict between merit and human rights (original: Menschenrecht; Cambridge translation: right of human beings), is straightforwardly answered by Kant emphasising the priority of human rights (Kant, 2017, p. 218   [6:454]).There is no need for differentiation and submaxims in this case, and Kant presumably simply wants to hear that his addressees understand what has greater moral weight.The only question which seems to involve a submaxim of sorts pertains to beneficence.How far may we go using our financial fortune to benefit others?The answer is that we should not venture so far as to risk becoming needy ourselves (Kant, 2017, p. 218 [6:454]).This looks like a submaxim of beneficence and might therefore count as an instance of casuistry.
We may therefore conclude that Kant offers some casuistry precisely in the place which his own words suggest.It should also be said, however, that Kant's treatment of financial assistance to the needy differs considerably from the view that casuistry is mainly concerned with the solution to hard moral cases.A general memento concerning the limits of one's financial abilities is not a solution to a hard moral case.Kim makes a related observation: Kant's casuistry is not meant to find exceptions to a rule in hard cases but to guide agents in the exercise of imperfect duties of love and (possibly) respect (Kim, 2009, p. 338).It should be noted, however, that Kant's limitation of personal financial risk is not telling agents how to allocate help.Kant is therefore not so much offering an action-guiding submaxim as a restriction on the choice of submaxims.He may nevertheless have regarded this as part of casuistry's task.

CONCLUSION
In the DV, casuistry and casuistical questions do not have the same functions.Casuistical questions can but need not embody casuistry.Most of Kant's casuistical questions in DV, and in particular the questions concerning perfect duties, do not employ casuistry.They are didactic classroom exercises designed to test how much basic ethical knowledge pupils have acquired.For perfect duties, pupils are mainly meant to document that they understood the exceptionless bindingness of the duties.Moreover, the pupils are to cultivate their inner moral will more than acquire competence in solving tricky moral cases in the manner of 17th-century casuists.In this, Kant follows Pietist catechetical practices instead of reverting to the mainstream practices of early modern casuistry.
These points remind us not to fall for interpretations which allot practical wisdom too important a role in Kant's understanding of perfect duties.It may seem attractive to combine Kant and Aristotle, and to some extent a convergence is borne out by the role of virtues in Kant's ethics.However, Kant does not use casuistry or the casuistical questions of the DV to dilute the exceptionless character of perfect duties.This may render his ethics less attractive to modern readers, but we should at least appreciate the weight of Kant's motives.In the early modern era, the enormous flourishing of casuistry documented that agreement between moral theologians even of the same creed and training was hard to come by in practical ethics.Kant's aprioristic ethics and its strict perfect duties are, among other things, attempts to overcome this chaotic situation by insisting on the powers of pure reason and without resorting to force and external compulsion.Kant thinks he has offered us a way out of the rampant moral disagreement that traditional casuistry had uncovered.Moreover, he intends to erect a bulwark against the nearly unlimited possibilities of selfserving rationalisations which the early modern boom of casuistry had fostered.He would have thwarted his intention by applying traditional casuistry in the field of perfect duties.At the same time, he had good reason not to trust practical wisdom as a remedy against moral disagreement.The historical record had shown that in applied ethics moralists hardly agree on what practical wisdom demands in practice.
The case is different for imperfect duties, especially of love.Here, it seems innocuous if different persons use different submaxims for implementing a duty of love.Kant restricts this pluralism by highlighting binding grounds of obligation, such as avoiding the risk of becoming needy oneself, but does not altogether eliminate the latitude of maxim choice.Hence, casuistry as an art of finding suitable submaxims has a place in the exercise of imperfect duties.For this reason, some of the casuistical questions concerning imperfect duties may indeed represent casuistry, although others seem clearly again to be classroom exercises with the purpose of testing elementary moral knowledge.

ACKNOWLEDGEMENTS
I am grateful to two anonymous reviewers and the participants in the 'work in progress' meetings organised by the editors for insightful criticism and helpful suggestions.

ENDNOTES
1 The assumed number of casuistical questions in the DV depends on the method of counting.I follow the approach of James (1992). 2 Kant occasionally referred to casuistry in his published works and lectures before the DV (for a short discussion, see Schuessler, 2012).He mainly depicted (true) casuistry as concerned with moral bagatelles (micrologia), while condemning (false) 'Jesuitical' systems of casuistry, and above all moral probabilism, as dangerous subterfuges (Höffe, 2020; Schuessler  2020).Compared to his earlier comments, Kant ascribes greater importance to casuistry in the DV.I will not analyse the respective changes in Kant's views here.