Duties to Oneself, Duties of Respect to Others Allen Wood
1. Kant’s division of duties
One of the principal aims of Kant’s Metaphysics of Morals, especially of the Doctrine of Virtue, is to present a taxonomy of our duties as human beings. The basic division of duties is between juridical duties and ethical duties, which determines the division of the Metaphysics of Morals into the Doctrine of Right and the Doctrine of Virtue. Juridical duties are duties that may be coercively enforced from outside the agent, as by the civil or criminal laws, or other social pressures. Ethical duties must not be externally enforced (to do so violates the right of the person coerced). Instead, the subject herself, through her own reason and the feelings and motives arising a priori from her rational capacities -- the feelings of respect, conscience, moral feeling and love of other human beings, must constrain herself to follow them (MS 6:399-404).1 Among ethical duties, the fundamental division is between duties to oneself and duties to others.
Within each of these two main divisions of ethical duty, there is a further division between duties that are strictly owed, requiring specific actions or omissions, and whose violation incurs moral blame, and duties that are wide or meritorious, the specific actions not strictly owed, but deserving of moral credit or merit. Kant treats these latter as ‘duties’ (eschewing any category such as ‘supererogation’) because the actions in question are conceived as fit objects of self-constraint – things we can make ourselves do through the exercise of reason and the moral feelings arising from the application of practical reason to our faculty of desire. Regarding duties to oneself, this division is between ‘perfect’ and ‘imperfect’ duty; regarding duties to others, the strict or narrow duties are called ‘duties of respect’ while the wide or meritorious ones are called ‘duties of love.’
We may represent the major divisions of Kant’s system of duties in the following diagram:
The aim of this essay is to discuss the three classifications of duties that appear leftmost on the bottom line of this diagram: Ethical duties to oneself (both perfect and imperfect) and duties of respect to others. Juridical duties and duties of love to others are not part of our topic here.
It is important to recognize, however, that Kant’s Metaphysics of Morals does not attempt to cover all the ethical duties that we have. This is because Kant confines the ‘metaphysics’ of morals only to those duties that are generated by applying the principle of morality to human nature in general. But many of our duties, as Kant recognizes, arise from the special circumstances of others, or our relations to them, and especially from the contingent social institutions defining these relations.
In Kant’s German idealist followers, Fichte and especially Hegel, the system of ethical duties came to be defined, or even superseded, by an account of the social structure (Fichte spoke here of ‘particular’ duties, Hegel of the ‘rational system of ethical life’ that is supposed to replace a ‘doctrine of duties’ in his system of objective spirit). Perhaps some people, who might call themselves as ‘cultural relativists,’ could even think that all ethical duties arise solely out of such social institutions relations. More cosmopolitan and universalistic, Kant holds that there are universal duties that we have, both to ourselves and to others, simply as human beings, and he regards these as in some sense the foundations of all our duties, within which we also acquire duties in consequence of social customs, institutions and relationships. Some of these duties might be to ourselves, though most will no doubt be to others; some will no doubt be narrow and others wide; and some may in effect convert wide duties into narrow duties, as when responsibilities to others convert our wide duty of beneficence into a narrow duty to contribute in determinate ways to the welfare of our family or friends or clients in some professional relationship. Kant holds that we have duties based on social institutions and relations, and that they are important; but they fall outside the scope of what he intends to cover in the Metaphysics of Morals.
2. Kant’s theory of duties is his theory of moral reasoning
For some reason, reading Kant’s Groundwork for the Metaphysics of Morals tends to create in you a kind of reflex-reaction. To the stimulus ‘Kantian ethics’, you tend to emit the response ‘universalize your maxims’. Some Kantians have even thought that the very essence of Kantian ethics is the use of a ‘CI-procedure’ (involving the testing of maxims for universalizability) to decide what to do, or even to ‘construct’ all ethical truth.2 All who would understand Kant’s actual theory of moral reasoning, however, ought to begin by performing a bit of minor surgery on themselves, severing the nerve that connects the stimulus with this reflexive response. In place of this reaction, we should think of Kant’s theory of moral reasoning as a theory about the way our different duties bear not only on our individual actions, but on our maxims and on the choices through which we, as self-governing rational beings, shape our lives and give meaning to them.
It may be true that for Kant every action conforming to duty involves a maxim that can be willed to be a universal law; it is certainly true for Kant that the fundamental principle of duty is a law given universally for all rational beings by the idea of the will of every rational being. But because they are propositions about the philosophical foundations of morality, Kant does not think that either of these propositions tells us very much about the structure of everyday moral reasoning. Instead, as Kant presents things in his final work on moral philosophy, the Metaphysics of Morals itself, the normal procedure of moral reasoning depends on the constraints of duty, on the wide variety of duties we have, and on the different kinds of duties which ought to determine our choices in various ways.
This point about the need to supplement the formula of universal law with a theory of duties, based on a different formulation of the principle of morality, is clearly stated early in the Doctrine of Virtue:
“[In] the formal principle of duty, in the categorical imperative ‘So act that the maxim of your action could become a universal law,’…maxims are regarded as subjective principles which merely qualify for a giving of universal law, and the requirement that they so qualify is only a negative principle (not to come into conflict with law as such). – How can there be beyond, this principle, a law for the maxims of actions?...
“For maxims of actions can be arbitrary [willkürlich], and are subject only to the limiting condition of being fit for a giving of universal law, which is the formal principle of actions. A law, however, takes away the arbitrariness of actions” (MS 6: 389).
The law that goes beyond the merely formal principle of duty has to do with the ‘matter of choice’, namely with its ends.
“Only the concept of an end that is also a duty, a concept that belongs exclusively to ethics, establishes a law for maxims of actions by subordinating the subjective end that everyone has to the objective end” (MS 6:389).
“The supreme principle of the doctrine of virtue is: act in accordance with a maxim of ends that it can be a universal law for everyone to have. – In accordance with a this principle a human being is an end for himself as well as for others, and it is not enough that he is not authorized to use either himself or others merely as means;… it is in itself his duty to make the human being as such his end” (MS 6:395).
Here it is clear that the “supreme principle of the doctrine of virtue” is more closely allied to the formula of humanity as end in itself than to the formula of universal law. It is also clear that this principle will establish duties via establishing that there are certain ends which it is our duty to have – to which ends Kant gives the name ‘duties of virtue’ (MS 6:394-395).
The ends that are duties to have in accordance with this principle are of two kinds: Our own perfection, and the happiness of others. Regarding the former, Kant says:
“The capacity to set oneself an end – any end whatsoever – is what characterizes humanity (as distinguished from animality). Hence there is also bound up with the end of humanity in our own person the rational will, and so the duty, to make ourselves worthy of humanity by culture in general, by procuring or promoting the capacity to realize all sorts of possible ends” (MS 6:392).
This argument rests our duty to make our own perfection into an end firmly on the formula of humanity as end in itself. Regarding our duty to make the happiness of others our end, the argument is different:
“The reason it is a duty to be beneficent is this: since our self-love cannot be separated from our need to be loved (helped in case of need) by others as well, we therefore make ourselves an end for others; and the only way this maxim can be made binding is through its qualification as a universal law, hence through our will to make others our ends as well” (MS: 6:393).
This argument, while clearly alluding to the idea that humanity is an end in itself, also has evident parallels with the argument used in the fourth illustration of the formula of universal law in the Groundwork, where appeal is also made to the fact of human interdependence, that our self-love cannot be rationally separated from our need to be helped by others (G 4:423).
A closer look, however, reveals that the two arguments are decisively different, and that the formula of universal law could not serve as the basis in this case. For there the question is only whether the maxim of refusing (on principle) to make the welfare of others our end can be willed without contradiction to be a universal law (or a law of nature). Since it can’t, it is impermissible to adopt it. But as Kant has noted, the formula of universal law (as a merely formal principle of duty) is only a negative test for maxims, and cannot give rise to any moral laws. Even if the maxim of principled non-assistance is impermissible, it might still be the case that helping or not helping others are equally permissible policies in general, that it should be possible to adopt no maxim making the happiness of others an end. But that is precisely what the present argument is supposed to rule out.
“It does so by asking not whether the maxim of principled non-assistance can be thought without volitional conflict to be a universal law, but instead what we necessarily will to be an actual universal law consequent on our rationally necessary volition that we be an end for others. This question is the one posed not by the formula of universal law, but by the formula of autonomy, in those formulations which require us to act only on maxims that include at the same time the volition that they actually be universal laws (G 4:437-438, 440, 447). For it is only through such maxims that we can regard ourselves as legislating universally for all rational beings, in accordance with the idea of the will of every rational being as universally legislative” (G 4:431).
None of this need come as a surprise to an attentive reader of the Groundwork. For Kant tells us that his search for the supreme principle of morality, which proceeds from the formula of universal law (and of the law of nature) through the formula of humanity as end in itself, to the formula of autonomy (and of the realm of ends) constitutes a progression – a progression in which one of the formulas combines the other two in itself (MS 6:436). This evidently refers to the fact that the formula of autonomy was derived by combining the formal principle of duty (the formula of universal law) with the principle specifying the matter of duty (the end in itself) (MS 6:431, 436). So we might have known that the formula of universal law, as the earliest stage of the progression, would also be the most provisional, least adequate and (as a merely formal principle) the poorest in content of the three formulas, while the formula of humanity would be the one from which insight into the matter of duty (the ends that are duties) could be had most easily, and the formula of autonomy would be the purest and hence the ‘universal formula’ which is the final touchstone of moral judgment – as Kant says it is, and as he also makes it in the Critique of Practical Reason and the Metaphysics ofMorals (KpV 5:30, MS 6:225). But this is not the place to try to explain why readers of the Groundwork have so often and so mistakenly given priority to the formula of universal law in interpreting Kant’s ethics. 3
3. Obligatory ends (duties of virtue) as the ground of ethical duties.
In general, the easiest way to make out the distinctions needed for Kant’s taxonomy of ethical duties is through the use of the formula of humanity as end in itself. A duty d is a duty toward (gegen) S if and only if S is a finite rational being and the requirement to comply with d is grounded on the requirement to respect humanity in the person of S. A duty is wide or imperfect (or, if toward others, a duty of love) if the action promotes a duty of virtue (that is, an end it is a duty to set); an act is required by a strict, narrow or perfect duty (or a duty of respect to others) if the failure to perform it would amount to a failure to set this obligatory end at all, or a failure to respect humanity as an end in someone’s person. An act violates a perfect duty (or duty of respect) if it sets an end contrary to one of the ends it is our duty to set, or if it shows disrespect toward humanity in someone’s person (as by using the person as a mere means). No corresponding account of these matters seems derivable from the formula of universal law; that is a further reason for regarding it as a merely provisional formula, poorer in consequences than the formulas of the same moral law that Kant derives later in the progression.
This also shows how there might be narrow or perfect ethical duties, even though all ethical duties, as duties of virtue, are fundamentally wide duties. For the duty to promote an end involves not only a duty to refrain from adopting the maxim of refusing in principle to promote it, but also a duty to refrain from setting all ends that opposed to the obligatory end – specifically any end of decreasing one’s own perfection (or doing anything that makes you less worthy of humanity), or making the unhappiness of any person your end (as happens in the “vices of hatred”: envy, ingratitude and malice) (MS 6: 458-461). We thus have a perfect duty to avoid any action that involves these forbidden ends, and also a narrow or perfect duty to perform any action whose nonperformance would amount to the principled renunciation of the obligatory end.
In grounding duties of virtue on the ends of our own perfection and the happiness of others, Kant does not mean to say that we have a duty to maximize our own perfection or the happiness of others. Rather, these duties, he argues, are wide duties, duties that determine us to make something our end, but leave us with latitude (or ‘play-room’) regarding how far we promote the obligatory ends, and which actions we take toward them (MS 6:390-394). All such actions are meritorious, but their omission is not blameworthy, unless it proceeds from a principled refusal to adopt the end in question (MS 6: 390).
Kant’s theory regards the active pursuit of any end of these descriptions (the development of any talent or gift or capacity in ourselves, the contribution to anyone’s happiness, or any component of their happiness) as in general meritorious (unless, of course, it proceeds by way of the violation of a strict, narrow or perfect duty). It is up to us to decide which such ends to include in our plan of life. Our relation to others in determinate social institutions (the aspect of morality Kant deliberately leaves out of a ‘metaphysics’ of morals) may render some of these imperfect duties perfect (caring for our children, or parents, or friends in determinate ways may be perfect duties, for the neglect of which we might be blamed).
Moreover, it is apparently also up to us, to some extent, to decide how wide or narrow to make a duty of this kind: “The wider the duty, therefore, the more imperfect is a human being’s obligation to action; as he nevertheless brings closer to narrow duty (duties of right) the maxim of complying with wide duty (in his disposition), so much the more perfect is his virtuous action” (MS 6:390). Thus if I commit myself to perfect myself in certain determinate ways, this can create something approaching a perfect duty to actions that promote this perfection (thus a devoted musician or athlete might be blamable for failing to practice or keep in condition, in ways that an a casual amateur at these pursuits would not be.) But one might have expected that an ethics of autonomy would leave a lot to individuals in determining their lives, including determining the content of their duties.
Regarding most narrow duties, including perfect duties to ourselves and duties of respect to others, however, Kant seldom appeals to the ends of our own perfection and the happiness of others. He more often goes behind the back of these obligatory ends, so to speak, appealing to something even more fundamental – to the worth of humanity as end in itself, and the requirement that we show respect for it in our actions.
Humanity as end in itself is not an “end” in the sense of a future state of affairs to be brought about, but rather a value whose worth we are required to acknowledge expressively in our actions – an “end” only in a somewhat broader sense: that for the sake of which we act. This kind of end creates obligations to avoid acting in any way that shows a lack of respect for the worth or dignity of humanity. We could describe this by speaking about the value of ends, as by saying that conduct contrary to duty treats humanity as if it had value “only as a means” to some end to be produced, or treats the dignity of humanity as having lesser value than something whose value is mere price. But in order to be guilty of this conduct, it is not sufficient that you treat humanity as a means or use humanity in seeking something whose value is mere price. For doing all that might still be compatible with also treating humanity as an end in itself and as having dignity, if your conduct also expressed that valuation of humanity. However you use humanity in relation to other ends, the crucial question always comes down to whether your conduct expresses respect for the dignity of humanity or, on the contrary, betrays a lesser valuation of humanity than its dignity demands.4
A few of the perfect duties to oneself that we will be discussing appear to be based on the thought that their violation would be incompatible with making one’s own perfection an end, but even there, the deeper reason why failing to set that end is contrary to duty is that this failure shows disrespect for the dignity of humanity as an end in itself.
4. Duties to Oneself
In the Anglophone tradition of moral philosophy, the concept of a duty to oneself is commonly applied to alleged duties to promote one’s own welfare. But this is not what the concept means in Kant’s ethical theory.5 Kant does not regard one’s own happiness as a fundamental duty of virtue at all, because in general our own happiness is something we inevitably pursue without the constraint of duty (MS 6:386). But he does think that I have a duty to promote my happiness whenever failing to do so might tempt me to violate other duties (G 4:399), and also insofar as my happiness falls under the heading of my perfection (MS 6:387). Further, though duty may sometimes require us to sacrifice our happiness, Kant thinks it cannot be lawful to adopt the general maxim of sacrificing one’s own happiness for the sake of others, since this maxim would destroy itself (frustrate the happiness of all) if made a universal law (MS 6:393). But none of these points has anything at all to do with “duties to oneself” as Kant understands that notion. Duties to oneself are not about self-interest but about self-perfection and being worthy of one’s humanity.6
Kant begins by facing squarely the question whether the concept of a duty to oneself is contradictory, since it seems to make a constraining person (or auctorobligationis) the same as the person constrained (the subjectum obligationis), permitting the latter (in his person as the former) always to release himself from the obligation, thus making it fundamentally null and void (MS 6:417). The response is to deny that the author of the obligation is identical to its subject. Rather, what is distinctive about the concept of an imperfectly rational and self-governing being (a being with ‘personality’ in the Kantian sense) is that this concept involves that of a relation between two persons who are combined in one and the same being. I contain in myself both the person of the rational legislator, whose law is necessary, objective and binding on all rational beings, and the person of the finite, imperfect being who has the capacity to obey this law, but also the possibility of failing to obey it.
Kant employs here his distinction between the sensible and the intelligible (MS 6:418), but it is doubtful whether that distinction, with its metaphysical baggage, makes any sense at all in this context. The point is rather that the Kantian conception of self-legislation is misunderstood when it is interpreted as the subjectivizing of moral value, as by those who present it as the metaethical thesis that it is we humans who “construct” the moral law or “confer value” on things through our choices. The moral legislator for Kant is rather (as he puts it in the Groundwork) the “idea” (or pure rational concept, to which no experience can be adequate) of every rational being as giving universal law (G 4:431), while the subject of these laws is a finite and fallible being whose volitions are subject to this law, and have objective value only through conformity to it. The welfare and the choices of such a being do have value, but only because the moral law makes this being an end in itself.
Perfect duties: arising from our animal nature.The basic division within duties to oneself, as noted already in our diagram in § 1, is between perfect and imperfect duties. Kant describes the former as “limiting (negative) duties” that “forbid a human being to act contrary to the end of his nature and so have to do merely with his moral self-preservation” (MS 6:419). These duties are in turn divided into those arising from our animality and those arising from our moral nature (MS 6:420). Under the duties to ourselves regarding our animality, Kant includes the duty to preserve our lives (and forbidding killing oneself), the duty forbidding “defiling oneself by lust” and the duty forbidding “self-stupefaction through food and drink” (MS 6: 422-427). Some of Kant’s views about these matters seem quaint (or worse) to us today, but this may pose an obstacle to our recognition that his categories, and even his general approach, may still make a lot of sense, even if we come to very different conclusions from his on the moral issues themselves.
Suicide. Kant approaches issues regarding our animality through the idea that our animal predispositions have natural purposes, and respecting ourselves requires treating this purposiveness with respect, rather than simply making everything serve our inclinations. His views proceed from the proposition that the fundamental natural purpose of self-love is the preservation of the life of the individual (G 4:422). We are unlikely to accept this judgment today, and may even be skeptical about the whole idea of natural teleology. But we should, and probably do, have views about the value of our lives and of the role of sexuality in human life, views that might support the thought that we show disrespect for ourselves when we act in disregard of that value, and we should therefore also be susceptible to Kant’s argument that suicide is at least sometimes an act that shows blamable self-contempt (MS 6:422-423).
In some places, Kant seems to be aware of (though never wholly to accept) the idea that suicide might be compatible with, or even a necessary expression of, the preservation of our own dignity – when we face the prospect of a life deprived (by disease or by the mistreatment by others) of the conditions under which our human dignity can be maintained (MS 6:423, VA 7:258, VE 27:374). In light of this idea, it seems one-sided for Kant to suggest that any act of suicide constitutes a denigration of one’s person and a case of treating it as a mere means (G 4:422, MS 6:422-423).
Sexuality. Kant’s abhorrence of sexual activity for pleasure, which he regards as inherently degrading to human beings, is likely to excite only amusement or indignation among us (MS 6:.277-280, 424-426). Yet we ought to think a little more deeply about the matter. The thought that sex is only “for pleasure” is not directly false, but it is shallow. Hedonism, the general doctrine that pleasure is good, is likewise not false but shallow in a way that makes it profoundly misleading about the nature of value and the relation of pleasure to value. Hedonism at least neglects the truth (insisted on even by a hedonist such as Mill) that pleasures differ in quality, and thereby also directly denies the deeper truth behind this (which Aristotle recognized) that the value of pleasure is not fundamental, but derivative, based on the value we place on the functions and activities with which a given pleasure is associated, and on the value of their proper ends. To say that we engage in sex “for pleasure” is misleading because simply elides the crucial distinctions here, and thereby fails to acknowledge the importance of the manifold, sometimes ambiguous connections of sexual pleasure with human life, and the qualitatively different values, even among pleasures, that these involve.
Kant’s discussion of the duty against sexual “self-defilement” in the Doctrine of Virtue is premised on the claim, which few nowadays are likely to accept, that the sole natural purpose of sexual desire is reproduction of the species (MS 6:424-426). In another context, however, Kant observed that sexuality has role in human life different from, and more important than, the role it has in any other animal species, because it is precisely not confined to its periodic reproductive function, and because it involves both the operation of the imagination, the attempt to excite and control desire of another, and also the attempt to make oneself the object of another’s desire while at the same time retaining the other’s respect. In the desire excited by sexual refusal Kant finds the origins of all morality (MA 8:113; cf. VA 7:152).7
Thus it is sad that Kant did not apply these acute and adventurous observations to his thinking about the morality of sex. Perhaps he did not know how to do so without entering on paths of thought he could not comfortably follow where he could glimpse that they naturally lead. But we know that he was on the right track. The nature of sexual pleasure often has little or nothing to do with the “natural purpose” of reproduction and everything to do with the expression of distinctively human meanings, such as self-concealment, self-revelation, self-withholding and self-bestowal, both in the flesh and in the imagination, and the possibilities of intimacy that arise out of the ambiguities and transitions between offering and refusing oneself, giving and taking, possessing in reality and in the imagination.
Yet it is also precisely for these reasons – and not at all because of any animal reproductive function – that some forms of sexual activity can indeed involve the degradation of one person by another, and also to self-degradation. Kant had to be aware of this, since so many of his pronouncements about sex are admirably (if also often misguidedly) sensitive to just these terrible possibilities (MS 6:277-278, 425).8 Issues about the ways that sexual activity might involve a violation of duties to ourselves as well as others are obviously more subtle than Kant ever acknowledged, but he was not mistaken in viewing these issues as turning on duties of self-respect based on the purposiveness (albeit individual and social as well as natural) of sexual desire and sexual self-expression. The noise in our heads provoked by Kant’s wrongheadedness on the surface too easily prevents us from listening for the ways he gets things right at a deeper level.
Gluttony and drunkenness. It should be less controversial that we show disrespect for our humanity when we show contempt for our capacities, as by harming or depriving ourselves of them. This is the way Kant understands the violation of duty to ourselves involved in drunkenness or gluttony (MS 6:427). Kant is aware, of course, that what counts as a loss of capacity in one respect or in one context may not in another, or it may even count as an enhancement. In his “casuistical questions” he suggests that the slight intoxication afforded by the consumption of wine may even promote our healthy sociability at a dinner party, for example, by enlivening conversation, making it less reserved and more candid (MS 6:428).