Kant’s Universal Law Argument Against Suicide

As is well known, Kant’s universal law test for maxims is meant to serve as a ‘canon’, or authoritative standard, for moral judgement (G, 4: 424). In both the Groundwork and the second Critique, he uses suicide to illustrate the test, and is satisfied that it does its work, ruling suicide out. The test condemns any maxim that, universalized, generates a contradiction. Condemnations of suicide in the Metaphysics of Morals and in the Collins lectures do not formally apply the test, but they invoke suicide’s contradictoriness, suggesting that Kant has something like it in mind in those texts as well (MS, 6: 422 and VE, 27: 369).

Here is Kant in the Groundwork, illustrating the test as offered in the first formulation of the categorical imperative, namely, ‘act only in accordance with that maxim through which you can at the same time will that it become a universal law’ (G, 4: 421):

A suicide maxim, universalized, ‘could not possibly be a law of nature’ since a nature with such a law ‘would contradict itself and would therefore not subsist’. Suicide is ruled out. In the Critique of Practical Reason, Kant again uses suicide (along with lying testimony) to illustrate the universal law test:

Here again a maxim of ‘disposing freely of my life’ (KpV, 5: 44) cannot be universalized; an ‘arrangement’ with such a law ‘would not be an enduring natural order’. But why not?

Interpretations of the Argument on which it Fails

Kant’s argument that a suicide maxim cannot be universalized finds itself in trouble right away because, on standard interpretations of the universalization test, it can be. Recall the Groundwork maxim: ‘from self-love, I make it my principle to shorten my life when its longer duration threatens more troubles than it promises agreeableness’ (G, 4: 422).Footnote ⁵ How does this maxim fare on ‘contradiction in conception’ and ‘contradiction in will’ interpretations of the test suggested by Kant at G, 4: 424 and elaborated by Onora O’Neill, among others?Footnote ⁶ A universal resort to suicide when life threatens more trouble than agreeableness would not, on the one hand, generate a ‘contradiction in conception’. It would not in any way thwart my own suicide maxim; my success as a suicide does not depend on anyone else’s hewing to life as, for instance, my success as a false promisor depends on others hewing to truth (G, 4: 422). Nor does a universalized suicide maxim, on the other hand, somehow thwart my interests overall, producing a ‘contradiction in will’. My suicidal will is intent on renouncing further interests, or has only an interest in eliminating pain, and suicide will accomplish both; a universalized suicide maxim thus would not contradict my general interests the way universalized maxims of neglecting talents or withholding aid, for instance, would (G, 4: 422–3 and 4: 423, respectively). The result? While the universalizability test, interpreted along standard lines, produces desired results for Kant’s other Groundwork examples, it fails for suicide. It seems easy to universalize a suicide maxim without contradiction.

What if, as many have, we attend instead to nature in Kant’s discussions of suicide and try to interpret the universalizability test accordingly?Footnote ⁷ Perhaps suicide is a case where we test neither for contradiction in conception nor for contradiction in willing, Kant’s G, 4: 424 remarks notwithstanding, but for contradiction with nature itself. Though a version of this strategy will ultimately be mine, we must proceed with caution. This is so for many reasons.

First, if ‘nature’ is taken as the Newtonian system of spatiotemporal objects and events governed by mechanistic causal law, a universalized suicide maxim generates no contradiction: such a system would not fail to subsist or otherwise be contradicted at all if everyone who wanted to committed suicide. Indeed, it is impossible to imagine any human maxim that, universalized, would contradict this ‘nature’, if only because anything that happens in this nature must, perforce, follow some mechanistic causal law, and that law would necessarily fit comfortably within Newtonian nature.Footnote ⁸

Second, Kant cannot be objecting that the maxim, universalized, contradicts our nature, understood as our living animal existence. Morality sometimes requires the sacrifice of just this, as in the case where we should prefer execution to bearing false witness (KpV, 5: 30).Footnote ⁹ Kant’s moral project as a whole strenuously resists the idea that agreement with our living animal existence is ever by itself an adequate ground for moral claims.Footnote ¹⁰ If a suicide maxim is ruled out, it cannot be because, universalized, it ‘contradicts’ life or our own natural existence.

Third and finally, the idea that ‘nature’ should be understood teleologically here also will not work. It is nonetheless the idea that commentators have found most promising.Footnote ¹¹ Granted, the idea is suggested by Kant’s own language: ‘a nature whose law it would be to destroy life itself by means of the same feeling whose destination [or function or purposeFootnote ¹²] is to impel toward the furtherance of life would contradict itself and would therefore not subsist as nature’ (G, 4: 422). Suicide here seems to go against nature’s purposes. Further support for a teleological interpretation can be found in the Metaphysics of Morals:

It looks like murdering oneself (among other things) is condemned here for violating nature’s purposes in giving us the impulses it has given us. Is this why a suicide maxim cannot become a universal law of nature? Alas, no.

A desire to commit suicide may confound us, given the widespread impulse to self-preservation. But Kant cannot be claiming that suicide, or anything else mentioned above, is actually inconsistent with nature, teleologically conceived. Teleology is a regulative idea for Kant: it helps organize our investigations of nature, but nature as we find it, not as we wish or imagine it to be. We may, and should, ask what nature’s purposes are, but where things (e.g. tapeworms, tumultuous dreams) do not fit our story, we must revise that story, not reject the things (KU, 5: 379–80); we err profoundly, for Kant, if we start declaring things actually found in nature teleologically unfit or ‘unnatural’.Footnote ¹³ Kant is elsewhere eloquent in speculating on self-destructive human drives that, seemingly purposively, propel civilization forward (KU, 5: 429–34). A law according to which a quest for agreeableness sometimes tends towards self-preservation and sometimes towards self-destruction presents a teleological puzzle, but does not contradict nature, teleologically conceived, any more than it contradicts the laws of physics. Nature teleologically conceived can withstand a universalized suicide maxim.

A Practical Nature

Just after introducing the Groundwork’s universal law formulation of the categorical imperative, Kant writes:

Notice Kant’s gloss of ‘nature in the most general sense’ as ‘the existence of things insofar as it is determined in accordance with universal laws’ (G, 4: 421). Kant describes ‘nature in the most general sense’ in these terms in the second Critique as well: ‘nature in the most general sense is the existence of things under laws’ (KpV, 5: 43). This is crucial. It points to the possibility that the ‘nature’– the system of universal law governed existence – that is being nominated to play a role in the universalizability test need not be the spatiotemporal world of material objects described by Newtonian physics nor the system of purposes that makes up nature teleologically conceived. It must be an existence of things under laws, yes. But that existence, those things, that law governed whole, need not be any of Kant’s theoretical ‘natures’ but can be rather, as I will argue, a specifically ‘practical nature’.

The idea may be unfamiliar, but is certainly present in Kant’s texts. In the second Critique, Kant writes about ‘a nature under the autonomy of pure practical reason’ (KpV, 5: 43) and ‘the idea of a nature not given empirically and yet possible through freedom, hence a supersensible nature to which we give objective reality’ (KpV, 5: 44).Footnote ¹⁶ He notes that where ‘teleology considers nature as a kingdom of ends, morals considers a possible kingdom of ends as a kingdom of nature’ (G, 4: 436n.; Kant’s emphases). Indeed, the Groundwork’s idea of a realm or kingdom of endsFootnote ¹⁷ picks out precisely what I am calling ‘a practical nature’ via the idea of a ‘systematic union of rational beings through common objective laws’ (G, 4: 433). We see in all of this a non-theoretical idea of an existence of things under laws, a community of free, self-legislating rational wills that stands as a lawful whole. Further references to such a nature, under various names, can be found throughout Kant’s practical texts.

Such a nature is ‘practical’ because it is the idea of the world, or a world, as arranged by and understood in terms of practice, that is, as arranged by and understood in terms of the free rational activity of our wills. It is thus not conceived theoretically, whether along Newtonian or teleological lines; it is not, that is, a nature we study as a given object.Footnote ¹⁸ It is, rather, a nature constituted by our free actions and the practical moral laws governing them, which Kant also frequently calls the ‘laws of freedom’ (e.g. at G, 4: 387, KpV, 5: 65, MS, 6: 214). Through these laws our wills ‘confer on the sensible world the form of a whole of rational beings’ (KpV, 5: 43). Practical nature is a lawful whole or system of nature not given, but made.

Kant goes quite a way toward articulating the possible shape of such a practical nature in the Metaphysics of Morals, which derives various morally demanded forms of ownership, public authority, self-care and attention to the well-being of others. Along the way, he address particulars as specific as the nature of books (MS, 6: 289–90), the problem with concubinage (MS, 6: 278–9), the role of prayer (MS, 6: 441), and legitimate and illegitimate forms of ridicule (MS, 6: 467). More work articulating the shape and particulars of a possible practical nature is found, for instance, in the Religion’s development of the idea of the church as a kind of ethical government (Rel, 6: 93–202) and in essays that imagine global realizations of moral ideals, such as ‘Idea for a Universal History with a Cosmopolitan Purpose’ (I, 8: 15–31) and ‘Perpetual Peace: A Philosophical Sketch’ (ZeF, 8: 341–86). Depending on the degree to which we comprehensively develop, refine and adhere to an articulated system of moral laws, or laws of freedom, a practical nature will be realized in our institutions, practices, self-understandings and relationships.

In this nature we, qua creatures with free rational wills, together with the actions, events, mutual understandings and other objects that issue from the activity of our free rational wills, are the things the existence of which is determined in accordance with universal (practical) laws.Footnote ¹⁹ In the Metaphysics of Morals, for instance, morally demanded systems of property rights and contract law give rise to practical entities like deeds (MS, 6: 273–6), money (MS, 6: 286–9) and marriages (MS, 6: 277–80). Sales, paydays, weddings and the people who have them are thus also all among the things whose existence is determined by the laws of a practical nature.Footnote ²⁰

As must now be clear, the practical nature I sketch here traffics intimately with empirical, theoretical facts, incorporating current empirical givens into the determination of its objects. This is because laws of freedom determine moral relations between wills that are not only free and rational but also embodied in particular ways with particular needs in particular environments. This can be confusing, but is essential: a practical nature, a ‘kingdom of ends on earth’, relies on empirical facts without being a fundamentally theoretical entity. The ultimate justification for laws in a practical nature, and so its deep shape and structure, will never rest on empirical facts, but always on considerations of pure practical reason. Nonetheless, any effort at articulating a practical nature cannot ignore empirical facts, on pain of being a merely formal exercise. Kant is not ultimately interested in a merely formal exercise, but rather in the possibilities for a systematic whole of free rational wills who are also embodied human beings.

Thus empirical, anthropological facts play an important role. Despite individual differences, there is a fairly predictable common array of sensuous pleasures and pains to which human beings are subject and, accordingly, a fairly predictable common array of corresponding desires and fears. This will be crucial for a system of reliable moral incentives. In a practical nature populated by, say, angels, the free rational wills whose activity is to be coordinated would all always do the right thing simply because it is right. But we – free rational embodied human wills – are not like that, and the moral law’s sway is too unreliable to neglect the other resources at our disposal.

We can indeed read the Metaphysics of Morals as elaborating a system of imperatives, discovered by practical reason and expressing laws of freedom, that are suitable to be enforced by juridical law, social norms, conscience and whatever other forces can be brought to bear without violating fundamental respect for free rational self-legislation (see MS, 6: 214–18). Kant explicitly describes good juridical law as a system of mutual coercion (MS, 6: 232–3), and insists we cultivate emotional and psychological skills that can assist rational responsiveness to moral considerations (MS, 6: 387, 419). Widespread success in coercing, cajoling and encouraging ourselves and others to do the right thing thus requires systems of incentives that reckon on anthropological facts. Failure to establish and institute such systems would render human moral community only fleetingly, locally, realizable – a temporary practical nature of particularly good people. But for Kant, full respect for the moral law and for ourselves demands more.Footnote ²¹

Thus the viability of a practical nature depends not only on our commitment to moral law but also on the reliability of a set of back-up incentives based in broad anthropological facts about common sensuous pleasures and pains, desires and fears. This will become very important below. We must grant the necessity of such incentives even as we resist collapsing the moral idea of a practical nature into a theoretical idea, or collapsing specifically moral judgement into theoretically informed prudential judgement. But we are used to resisting this collapse when we think with Kant.

We have seen, then, that the nature in Kant’s requirement that a maxim be suitable to be a universal law of nature need not be theoretical but can be practical. We have also seen that any conception of a realizable practical nature must take empirical facts into account, and in particular must use anthropology to devise incentives. Now, how does any of this help save Kant’s universal law argument against suicide?

Can a Suicide Maxim Become a Universal Law in a Practical Nature?

My argument will be that a practical nature could not sustain universalization of a suicide maxim because a suicide maxim represents the rejection of two basic commitments that are necessary conditions of the possibility of such a nature. Or, in short: a suicide maxim rejects all bases for practical law.

The first problem is that a suicide maxim subordinates a person to happiness; it represents a subject’s preference for the experience of ‘agreeableness’ over existence as a creature with a free rational will. The maxim thus eschews commitment to the only thing – the value of humanity, or personality, or the capacity for free rational willing itself – that must be committed to above all else, as a non-negotiable end in itself. The second problem is that, in eschewing self-preservation and the array of ordinary sensuous commitments it entails, it undermines the reliability of practical incentives designed to back up moral motivation. A suicide maxim thus cannot be universalized without wreaking havoc on conditions of the possibility of a system of practical law.

This, at any rate, is what I aim to show below. In showing this, and as my comments already suggest, I do in a sense look beyond universalizability as a stand-alone test. On my view, to ask seriously about universalizability is always to ask about it in conjunction with the idea of a system that (a) actively respects all free rational wills as ends in themselves and that (b) sees them all as potential rational legislators in a shared practical world (G, 4: 436–8).Footnote ²² Universalizability, in other words, is always tested against a backdrop of considerations that are brought to the fore in other formulations of the categorical imperative (the formulas of humanity and the kingdom of ends). As Kant himself remarks, the three Groundwork formulations of the categorical imperative are ‘at bottom the same’ and express the same law (G, 4: 437). I take him at his word, and understand each formula to have its own forte: the universal law test is particularly useful in displaying the self-undermining or otherwise morally incoherent features of problematic maxims.Footnote ²³ My aim here thus cannot be to show that the universal law test will do all the work of barring suicide ‘by itself’, but rather to show, contra the literature, that it nonetheless displays the morally incoherent, and so immoral, features of suicide, and indeed does so in a way that is interesting and illuminating.

The following passage from the Collins lectures provides a starting point.Kant’s topic is suicide, and he makes an aside. ‘Freedom’, he says, is ‘the inner value of the world’ (VE, 27: 344). ‘But’:

What is Kant saying here? Unless he gives himself rules, ‘it is uncertain whether man will not use his powers to destroy himself, and others, and the whole of nature’. The viability of any large-scale order for us depends on our following freely adopted laws. If, like ‘the whole of non-free nature’, we had no choice about which laws governed us, our individual choices would be ordered by theoretical principles of pleasure and survival. But we are free, and if we refuse to adopt and restrict ourselves according to ‘objective’ rules, that is, rules suitable to be held in common by rational wills,Footnote ²⁴ all kinds of lawlessness will result.Footnote ²⁵

It is crucial that Kant cannot here be worried about lawlessness in nature theoretically conceived, since nature theoretically conceived will always be lawful. He can only be worried about practical lawlessness, lawlessness in a practical sense. Back to the context: suicide, Kant is suggesting here, is paradigmatic of the refusal of rules that can keep human order. Only in this sense can the use of freedom threaten ‘savage disorder’. This is how I read the passage; what follows shows how it illuminates Kant’s universal law argument against suicide.

A Suicide Maxim Rejects a Non-Negotiable, Come-What-May Commitment to Free Rational Willing

Recall the maxim Kant proposes: ‘I make it my principle to shorten my life when its longer duration threatens more troubles than it promises agreeableness’ (G, 4: 422). Here is the first problem. The suicide’s maxim proposes that the life of an embodied free rational will is only worthwhile as long as it promises agreeableness. But this subordinates the value of free rational willing per se to agreeableness, and any maxim that does this will, universalized, violate the conceptual foundation of a possible practical nature. The maxim thus cannot be thought as a universal law of (practical) nature without contradiction.

The reason for this is that a practical nature, to be a systematic whole, needs an unconditioned end. An unconditioned end is one with intrinsic, ultimate and therefore non-negotiable, value; it is an end that cannot be swapped for another, or subordinated, or regarded as a mere means to further ends. This end needs to be something determinate, not a generic grab bag open (like happiness or agreeableness) to idiosyncratic and conflicting interpretations, and it needs to be an end everyone both can and must adopt (G, 4: 399–400). Only thus can it inform the objective rules needed to keep savage disorder at bay. For Kant, free rational willing itself alone fits this bill (G, 4: 401; see also G, 4: 393, 396, 436–8). Its status as unconditional value organizes the entire system that honours it, the practical nature that is the system of morality, governed by laws of freedom, lending the whole conceptual coherence and stability.

This demand is not just architectonic. The aim of Kant’s moral philosophy is to bring into being what we can now see as a ‘nature’ that respects each as an end in herself, refusing to regard free rational willing as something that can be traded off against other goods. Besides ruling out suicide from despair, this consideration explains Kant’s controversial rejection of heroic and Stoic suicide. Both heroic and Stoic suicides represent failure, for Kant, to recognize a non-negotiable end and defend it to the death, come what may; both engage in unacceptable cost/benefit analyses, putting pay-off above a struggle that would convey a non-negotiable commitment to the activity of free rational willing itself. Kant accordingly rejects the suicides of Cato, Lucretia and Curtius, as well as Stoic ‘time to depart’ suicide generally (VE, 27: 374, 371, 629; MS, 6: 422). Better, he thinks, to have gone down fighting. This makes sense: without a strict ban on suicide, Kant would open his system to competing, different-in-kind ultimate aims. Because a suicide maxim subordinates free rational willing to agreeableness, its universalization violates a conceptual condition of the possibility of a coherent, stable moral law-governed human community; a system containing a universalized suicide maxim could therefore not ‘subsist as nature’ (G, 4: 422).

As some will already have noted, this account significantly recasts the universalization test. The maxim fails to universalize not in terms of the suicide’s own actual interests, whether in suicide or in general, but in terms of an interest in a realizable practical nature that she necessarily ought to have as a free rational will. The maxim fails as a law she can regard herself as making as a legislator already committed to such a nature. In this regard, the account pulls away from some constructivist readings, relying on the thought that there is a good – free rational willing itself – to which we must be committed not as a result of good practical reasoning, but as its condition.Footnote ²⁶ Commitment to this good entails commitment to whatever practical nature is needed to respect and realize its value; we all have ‘objectively necessitating’ reasons to commit to a practical nature, even if we are subjectively unimpressed by the prospect. This account is therefore not moved by the suicide’s own insistence that she really wants to die, or that she would be more than happy to quit the practical demands of moral community altogether. Such insistence, far from troubling the Kantian case against suicide, would serve as further evidence of suicide’s immorality – to want to die, and to want out of moral community altogether, is to want to quit morality itself.Footnote ²⁷

There may seem to be a tension between my focus on practical nature and the fact that Kant always addresses suicide as a duty to self (G, 4: 421–2; MS, 6: 421; VE, 27: 340). My account may seem to cast suicide as an affront to a kind of community, making suicide sound like a violation of a duty to others. But for Kant, all moral duties have the same root and same fundamental justification. Distinguishing duties to self from duties to others is, in my view, just a way of classifying and getting a grip on particular obligations to particular objects, not a way of distinguishing fundamentally distinct kinds of obligation. If we are moral, we command ourselves to respect, preserve and promote the complex of freedom, rationality and will that is distinctive of human beings and that lends us our unconditional worth. A failure to command ourselves thus with respect to our own lives – a violation of a duty to self – is also always, for Kant, a failure to commit to the possibility of morality, and so of moral community (see MS, 6: 417–18). Put another way: my focus on a practical nature shows how universalization entails commitment to a kingdom of ends, and how commitment to a kingdom of ends entails duties to self. This, I hope, makes clear why I think the tension here is only apparent.

A Suicide Maxim Rejects an Anthropologically Typical Commitment to Self-Preservation

We now have an interpretation on which a universalized suicide maxim contradicts a necessary conceptual condition of a practical nature. But there is both more to want and more to be had. We might, for example, want more explanation of Kant’s alarmed and harsh condemnation. It seems unlikely to issue entirely from the conceptual point. After all, we subordinate the value of the unconditioned end to agreeableness in all sorts of violations and, though morally blameworthy, these violations do not occasion the same apoplectic response: a suicide maxim may place agreeableness over free rational willing, but is this really tantamount to threatening savage disorder? Kant himself even grants that agreeableness by itself, when subjected to careful prudential calculations, can produce quite a bit of order, perhaps enough for us to live reasonably good lives; he concedes arguments advanced by both Hobbes and Hume, acknowledging in his famous remark that even a nation of devils could solve the problem of setting up a state (ZeF, 8: 366).Footnote ²⁸ Kant of course wants more, and endorses acting towards the agreeable only when this does not involve subordinating free rational will, but, as we have also seen, prudentially enforced incentives play an important supporting role in establishing moral community (see also e.g. ZeF, 8: 366–7). Prioritizing agreeableness does not have to lead to out-and-out catastrophe, even if it is always morally suboptimal. The conceptual claim does not seem to fully explain Kant’s alarm.

We might also want to understand why suicide is among the very few cases where Kant reliably invokes not only ‘universal laws of nature’ (read now as practical nature), but also violation of something like our sensuous, animal nature. The other cases are self-mutilation (MS, 6: 421, 423), overindulgence in food and drink (MS, 6: 420, 427), and bestiality and non-heterosexual sex (MS, 6: 277, 420, 425) (see also VE, 27: 345–7 and 385–92 on all these cases). Kant cannot, as we have already seen, consistently claim that any of these contradicts nature in any theoretical sense. They all occur in that nature, after all. But he is clearly tempted to make something of traditional claims about their ‘unnaturalness’. This may in part explain why so many suppose he turns to teleology as a moral touchstone – something the same readers quickly note as a problem. I think we can, in fact, explain why Kant so often invokes our sensuous, animal nature in his condemnations of suicide, and I think we can do this in a way that helps explain his alarm.

Over and above his concern with the conceptual demand for an unconditional end, Kant’s invocations of ‘unnaturalness’ and his alarm both stem, I think, from the suicide maxim’s rejection of the ordinary array of sensuous desires that flow from an anthropologically typical commitment to self-preservation. My claim here will be that a failure to have this commitment is tantamount to rejecting a condition of the possibility of a realized practical nature.

The following passage from the Collins lectures is helpful:

What is Kant’s thought here? The passage opens in a way that feels teleological: suicide is wrong because it goes against the natural drive to self-preserve. But the topic, again, is really the free choice of rules to live by. Damaged trees, animals and other living bodies are not free; they follow natural laws, pursuing their own growth, sustenance, reproduction and preservation relentlessly and without choice. We, in contrast, live freely; we are not forced by nature to pursue self-preservation, though we typically have an impulse to. The thing that evokes revulsion with horror here is the use of freedom to pursue a self-destructive deviation from this typical impulse. Such a use of freedom is horrifying because it coincides with being utterly ungovernable, master of everyone but subject to none, free from any restraint whatsoever: ‘for anyone who has already got so far as to be master, at any time, over his own life, is also master over the life of anyone else’. A subject who opts to use freedom for self-destruction is immune to coercive incentives: ‘for him, the door stands open to every crime, and before he can be seized he is ready to spirit himself away’. As I read it, this passage offers the beginning of an argument that suicide is inconsistent with the possibility of a moral law-governed human community.

The argument is this. A maxim generally committed only to agreeableness is bad enough, but a maxim that specifically misses or overrides the ordinary sensuous commitment to self-preservation presents, for Kant, a special kind of danger. A person acting on such a maxim cannot be motivated to comply with any law: such a person turns away not only from moral considerations, but also from ordinary prudential ones. The best and most reliable order-producing principle in human affairs is the moral law, but prudential self-preservation is a powerful second. A maxim, a will, that turns away not only from a commitment to free rational willing but also from a commitment to self-preservation turns away from all possible sources of practical order, exploiting freedom to, as it were, introduce lawlessness.

A strong desire for self-preservation can either be missing or be overridden. A subject missing the inclination is profoundly and problematically idiosyncratic; it will be nearly impossible for others to know what, if anything, could reliably serve as a sensuous incentive for such a subject, making her a kind of wildcard, ‘master over the life of anyone else’ (VE, 27: 372). A subject who uses freedom to override an inclination for self-preservation is just as bad. This subject uses freedom against sensuous nature (in the inclination for life) only to allow sensuous nature (in the desire to avoid the disagreeable) to best (the moral demands of) freedom. While empirically and theoretically possible, this last strategy is practically incoherent, oscillating as it does between commitments to sensuous nature and freedom in the same act. Either way, in missing or in actively overriding the ordinary background inclination for self-preservation, a suicide maxim rejects an attachment to life that is a necessary condition of a system that depends on our collective ability to push and pull each other in reliable ways. A maxim that rejects the attachment to life in the name of agreeableness cannot be universalized if a practical nature is to be realized.

How does attention to this second problem help explain Kant’s willingness to invoke suicide’s ‘unnaturalness’? I think we should take the term loosely and colloquially, as picking out anthropological facts, related to embodiment, that we neither expect nor can readily explain. Here a failure to have, or a willingness to override, a fierce animal attachment to life is ‘unnatural’. Its deviance from the norm thwarts our efforts to use the anthropological facts that are generally true about people in formulating effective, widely available, collectively enforced aids to moral self-governance. Suicide, like overindulgence or non-heterosexual sex, is thus not in any philosophically serious sense ‘unnatural’. But the deviation suicide anyway represents makes reckoning on incentives for suicidal subjects impossible. Suicide’s ‘unnaturalness’, in this sense, is thus itself part of the problem.Footnote ²⁹

A suicide maxim represents both a failure to properly value what is most valuable and, adding insult to injury, rejects the sensuous basis for a system of incentives that could at least constrain action. Kant remarks in the lectures that ‘those who defend and teach the legitimacy of suicide inevitably do great harm in a republic’ (VE, 27: 374). This is so, my interpretation would suggest, because they advocate free rejection of the crucial sensuous bases for a system of incentives in a realizable practical nature. Such a free rejection is alarming, striking a blow at the very possibility of order in human affairs. A maxim that threatens this in principle is indeed, for Kant, horrifying.