1. Introduction
In the Lectures on Ethics, Kant is reported to have described the historical practice of senicide (i.e., killing the elderly) among certain Inuit communities as a ‘loving service’ (Liebesdienst) carried out from ‘true filial love’ by children to their parents.Footnote 1 Moreover, he reportedly claims that the practice was ‘to some degree justified’, as its aim was to spare the elderly from starvation – a fate he calls a ‘more ignominious death’ (L-Eth-H, 27: 43; L-Eth-Mr2, 29: 622).Footnote 2 These remarks are surprising in light of central tenets of his moral philosophy, such as the claim that humanity exists as an end in itself (G, 4: 429) and the view that persons have an obligation to preserve themselves (MM, 6: 421–8), from which it also follows that they may not consent to being treated in ways that violate their dignity as rational beings (MM, 6: 236, 420–37, 236).Footnote 3
This paper considers whether Kant’s reported evaluation of Inuit senicide is consistent with the core commitments of his mature moral philosophy. His verdict, I argue, sheds new light on a central claim of his mature ethics – that persons are obligated to preserve themselves for as long as they remain a subject of duty (MM, 6: 422). On my reading, Kant regarded Inuit senicide as ‘to some degree’ justified insofar as circumstances beyond their control rendered it impossible for the elderly in question to continue living an honourable and virtuous life. I argue that, on Kant’s interpretation of the circumstances, it is consistent with his ethics to hold that such a life fell beneath the dignity of a rational being and that consenting to having one’s life ended on that basis may have been morally permissible.
Drawing on cultural anthropological sources, I begin in section 2 by reconstructing the context in which certain Inuit communities are said to have practiced senicide.Footnote 4 In section 3, I turn to Kant’s brief reported remarks on Inuit senicide in the Lectures on Ethics. In section 4, I argue that Kant’s ethics can consistently prescribe obligatory acts of self-killing and allow for merely permissible self-killing acts, provided they do not constitute self-murder (Selbstmord), as Kant defines it, and thus do not violate the duty against it. In section 5, I examine what Kant might have meant by ‘ignominious’ in the context of his conclusions regarding Inuit senicide and consider whether this notion might serve as a justification for the permissibility of ending one’s life that is consistent with his mature ethical system. I argue that, to the extent that the predicament of the Inuit elderly can be characterized as anticipating an ‘ignominious death’ under Kant’s interpretation of their circumstances, the practice is best understood as a merely permissible instance of assisted self-killing: while those subject to senicide were no longer bound by a duty of self-preservation, they were likewise under no obligation to surrender their lives.
2. Inuit senicide contextualized
The cultural anthropological data seems to paint the following picture of the general, shared rationale for the historical practice of senicide among particular Inuit communities. In times of hunger, scarcity, and forced migration to new hunting grounds, family members who could no longer contribute to the group’s survival – particularly the infirm elderly – were commonly perceived as burdens, both by themselves and by their family:Footnote 5
The Eskimos of Baffin Island have great respect for the aged and treat them well. But when a woman becomes so old that she is a burden, she may calmly resign herself to death, allowing herself to be walled into a snow hut and left to die. She thinks this is better; the tribe agrees (Weyer Reference Weyer1932: 138).
[T]he usual reasons for giving up one’s life were sickness, suffering, and the feeling of uselessness (which may have been in large part a reflection of attitudes from the individual’s social environment) (Leighton and Hughes Reference Leighton and Hughes1955: 329).
Like the Stoic who argues, metaphorically, that if the chimney smokes, one should get out of the house, the Eskimo justifies suicide, especially if age or infirmity renders one useless and a burden (Weyer Reference Weyer1932: 248).
These judgements arose in part from the perception that it was impossible for able-bodied family members to distribute energy and resources in such a way as to both sustain themselves and care for the elderly – without jeopardizing the entire family’s chances of survival (Rasmussen Reference Rasmussen1931: 143; Pousset Reference Pousset2023: 22; Leighton and Hughes Reference Leighton and Hughes1955: 329). To relieve their families of this perceived ‘burden’, it was customary (and culturally encouraged) for the infirm elderly to request that their lives be ended by their kin. Fulfilling such a request was understood as ‘proof of devotion’:
Often, the old people themselves asked for their death; the Igloolik even knew a whole begging ritual for their own death (Pousset Reference Pousset2023: 22)Footnote 6 .
As elsewhere … when the Labrador Eskimos put an old person out of the way, it is generally in accordance with the wishes of the one concerned, and is thought to be proof of devotion (Weyer Reference Weyer1932: 139).
Participants in the practice reportedly took comfort in the belief that the souls of those who ‘met forms of violent death’ would travel to the most desirable afterlife (Leighton and Hughes Reference Leighton and Hughes1955: 328; Weyer Reference Weyer1932: 247–54; Hawkes Reference Hawkes1916: 132; Pousset Reference Pousset2023: 22) – a place thought to be free of the suffering of earthly existence.
From what I can tell, the data leaves room for interpretation regarding the sense in which the infirm elderly were considered a burden to their families, and thus the sense in which senicide functioned to ‘ameliorat[e] the stress which accompanies scarcity of food’ (Weyer Reference Weyer1932: 137). Two interpretations seem plausible in light of the anthropological record.
The first – which we might call the dead weight interpretation – is suggested by several contemporary ethicists, including James Rachels (Rachels Reference Rachels2003: 24).Footnote 7 On this view, the infirm elderly were seen as a burden (by themselves and by others) because they consumed precious resources while being unable to contribute to the acquisition of new ones, thereby reducing the chances of survival for others. On this interpretation, senicide had the function of ‘ameliorating stress’ by improving the survival odds of able-bodied family members. However, this raises a question: if the goal was merely to conserve resources, why not keep the infirm elderly alive – provided they could keep up during migrations – but simply refrain from feeding them until food became more abundant? After all, this would seem less taxing than the emotional and moral weight of actively killing one’s parents.
While I do not pretend to have a definitive answer to this question, it is worth noting that several anthropological sources report a widespread fear among historical Inuit communities of vengeful spirits of deceased kin, believed to interfere with the living by affecting both their health and their success (Weyer Reference Weyer1932: 245–7; Hawkes Reference Hawkes1916: 119; Jennes Reference Jenness1922: 171, 178). This may have provided a strong incentive for families not to ‘anger’ the spirits of the infirm elderly – for instance, by taking care of them for as long as they chose to remain alive. The act of requesting to be killed may thus have functioned as a way for the elderly to prevent the emergence of such resentment.
Alternatively, on what we might call the emotional burden interpretation, the sense in which the infirm elderly were considered a ‘burden’ to their families was primarily emotional rather than physical. On this view, the Inuit aimed to increase the community’s chances of survival by rationing food strictly among able-bodied family members, depriving the elderly of nourishment while scarcity persisted. In such a context, the emotional toll of watching loved ones starve – particularly when those loved ones requested death, and when honouring such a request was culturally understood as ‘proof of devotion’ – may have been heavier than the grief of actively hastening their death. This interpretation also finds support in an account by anthropologist Knud Rasmussen, who describes the case of an Inuit woman who survived being ‘left out in the snow’ for several days without food. Although she survived, she was still regarded as a burden by her kin group, even though her relatives had ceased to care for her (Rasmussen Reference Rasmussen1931: 143–4).Footnote 8 On the emotional burden interpretation, then, senicide functioned to reduce the kin group’s emotional stress by allowing the elderly to die in a way that was considered more merciful than the likely alternative of starving to death.
3. Kant’s remarks on Inuit senicide in the Lectures on Ethics
3.1. Kant’s ‘less ignominious death’ interpretation of Inuit senicide
From what is documented of Kant’s (reported) position on Inuit senicide, it is clear that he had a somewhat different understanding of the practice’s underlying purpose and motivation than is suggested by the cultural anthropological data discussed above. In the Lectures on Ethics delivered between 1763 and 1764, Kant’s student Johann Gottfried Herder reports him as saying that ‘the Eskimoes, who murderFootnote 9 [ermorden] their parents as a loving service [Liebesdienst] to them, are to some degree justified [haben in gewißermassen Grund], since they foresee a more ignominious [schmälichern] death for them in the hunting that is necessary for survival’ (L-Eth-H, 27: 43). A similar statement – albeit without explicit reference to moral justification – appears in the Mrongovius lectures, where Kant is reported to have said the following:
The Esquimaux, when their parents are decrepit and no longer capable of working, strangle them; and the old also make preparation for this. But the children do it from true filial love, because in winter they are absent for many weeks out hunting, and during that time the old might starve (L-Eth-Mr2, 29: 622, emphasis mine).
Much like the cultural anthropologists cited earlier, Kant understands the practice as taking place within a context of hunger and scarcity. However, the idea that the elderly were perceived as a ‘burden’ to their families is entirely absent from his remarks. Nor is there any indication that he understood senicide as a means of increasing the community’s chances of survival, controlling the population, or sparing children the agony of watching their parents starve to death. As such, his account does not neatly align with either the dead weight interpretation or the emotional burden interpretation.Footnote 10 Instead, Kant characterizes Inuit senicide as an ‘act of love’ from children to their parents and as ‘to some degree justified’ – with the sole intention of granting them a death that is ‘less ignominious’ than the alternative of dying from starvation. I will refer to this interpretation as the less ignominious death interpretation.
These passages from the Lectures on Ethics may strike us as a milder take on Inuit senicide than one might expect from Kant during this period. Elsewhere, Kant is reported to have followed ‘the French’ in describing ‘Eskimoes’ as ‘very repulsive in appearance [sehr abscheulich von Gesicht]’ and savage and spiteful in their customs [wild und boshaft an Sitten]’ (PhyG, 9: 432). He does not, however, appear to consider senicide among the Inuit to be a practice rooted merely in savagery or spite.
Nevertheless, these remarks raise more questions than answers with regard to where and how senicide fits within Kant’s moral philosophy. For one thing, the very idea that an action could be to some degree justified is puzzling, since Kant is not known for having a gradual conception of justifiability. Moreover, in light of his strong objections to the idea that rational nature may dispose of itself as a mere means to a discretionary end (MM, 6: 423; G, 4: 421–2), one might well question whether, on Kant’s account, the prospect of facing a ‘more ignominious’ death could ever constitute a morally legitimate reason to end one’s life.
3.2. Should one be ashamed of starving to death?
Nowhere does Kant elaborate on what he means by an ‘ignominious death’, nor does he explain why voluntarily dying by senicide would be – or be considered – ‘less ignominious’ than dying from starvation.Footnote 11 ‘Ignominious’ (schmählich) generally denotes something along the lines of ‘shameful’ or ‘disgraceful’. Typically, when Kant discusses shame and self-contempt, he frames them as emotional responses appropriate to the recognition of one’s own moral failure (cf. G, 4: 426; MM, 6: 420, 429, 482–3). Making oneself ‘contemptible in one’s own eyes’ is usually associated with violating the dignity of humanity in oneself as a person (MM, 6: 429). This raises the question whether Kant meant to suggest that shame or self-contempt would have been appropriate for the Inuit elderly if they were to die of starvation – and that the less shameful and more honourable alternative would be to choose to be killed. If so, it would follow that dying of starvation in this context would amount to a moral failure. On that reading, the Inuit elderly would have had a duty to request their own death.
This seems implausible. Since human beings are subject to the duty of self-preservation (MM, 6: 420, 422), it seems to follow that a person who chooses to die of starvation despite access to food and water would indeed be in violation of this duty and ought to be ashamed. But this is clearly not analogous to the case of someone who dies of starvation due to resource scarcity – which Kant recognizes as the predicament faced by the Inuit elderly. In the Critique of Pure Reason, Kant affirms that an action can be a duty only if it is physically possible (A547-8/B575-6). In other words: ought implies can. Thus, if Kant’s characterization of the Inuit elderly’s death by starvation as ‘ignominious’ is to be interpreted as suggesting that it is shameful, this does not align with his mature view of when shame or self-contempt is appropriate. Given that the Inuit elderly were physically unable to preserve themselves, they cannot be said to have violated a duty to the self.
This – along with the sheer implausibility of suggesting that those dying from starvation due to deprivation and infirmity should be ashamed rather than seen as victims of a grave misfortune – casts serious doubt on the idea that this is what Kant intended. Moreover, this interpretation finds little support in the cultural context. As noted in section 2, several anthropological sources report that, at the time under discussion, the Inuit believed that the souls of those who died in conditions of great suffering would enter the most desirable afterlife. In other words, dying from starvation or hardship was not culturally understood as ‘taboo’, nor is there any plausible indication that Kant believed otherwise. It is thus worth considering whether Kant may have meant to express something more in line with his established ethical framework.
In the next section, I discuss considerations that point towards a more plausible interpretation of what Kant may have meant by ‘ignominious’ in the context of his verdict on Inuit senicide and how this concept might support the justification of a self-killing act. In section 5, I develop this interpretation further and argue that it might serve as a possible justification of Inuit senicide within the terms of the less ignominious death interpretation.
4. Kant’s position on suicide
Since, according to the anthropological data, it was customary for senicide to be requested by those Inuit elders to whom the practice applied, I understand these cases as a form of assisted self-killing. Kant reportedly remarks that ‘the old make preparations’ for being strangled to death (L-Eth-Mr2, 29: 622), which suggests that he believed that the elderly took part in carrying out their own deaths. It therefore seems likely that Kant also conceived of Inuit senicide as a form of assisted self-killing.
My method for analysing whether Kant’s ethics allows for the moral permissibility of senicide is to examine whether it allows for the justifiability of requesting that another end one’s life on the basis of reasons connected to old age. I take for granted the following assumption: if it is morally permissible for an elderly member of the community to make such a request, then it must also be permissible for another member of the community to carry it out – provided the act is, as a matter of moral principle, conditional upon the actual consent of the elderly community member in question (cf. Kleingeld Reference Kleingeld2020: 400–1).
As is well known, Kant deems suicide a violation of a perfect duty to oneself (MM, 6: 422–4). Moreover, while he holds that we have a duty to assist others in realizing their ends when they are unable to do so themselves, his ethics forbids us from helping others to pursue ends that are themselves impermissible (6: 450). How, then, could the prospect of living an inevitably ‘ignominious’ life until death ever serve as a justification for ending one’s own life – or for consenting to have it ended by another?
It is important to note that, in arguing for the impermissibility of suicide, Kant uses the German term Selbstmord (or Selbstmorde), which literally translates as ‘self-murder’ (cf. G, 4: 429; MM, 6: 421–4). ‘Murder’ is clearly a morally loaded concept, whereas ‘killing’ is not (cf. Sticker Reference Sticker, Lyssy and Yeaomans2018 : 181). As will become clear in the next section, there is sufficient textual evidence to suggest that the circumstances and motivations underlying an act of self-killing play a significant role in determining whether that act should be classified as an impermissible instance of self-murder on Kant’s account.Footnote 12 Hence, in order to assess whether Inuit senicide qualifies as self-murder, we must first clarify what, in Kant’s view, distinguishes a permissible act of self-killing from one that is morally impermissible.
4.1. Suicide as a duty
Martin Sticker has proposed that Kant’s moral theory allows for the possibility that Inuit senicide was permissible for the Inuit, though not for others. While Kant’s account categorically and universally prohibits acts of (self-)murder, Sticker suggests that whether particular cases of killing should qualify as instances of (self-)murder may depend, to some extent, on ‘linguistic customs and societal attitudes’ (Sticker Reference Sticker, Lyssy and Yeaomans2018: 181–2). If the Inuit collectively and sincerely believed that senicide was at times justified, then – according to Sticker – it may have been a permissible, non-murderous practice for them (ibid., 181–182).
However, if we assume that one and the same practice can be morally impermissible in one community yet permissible in another on Kant’s account, we undermine the principled, universalist character of his ethics. If a case of self-killing is not to be classified as an instance of self-murder, then – on Kant’s account – this must be due to the underlying principle on which the act is willed, not to prevailing cultural norms or linguistic conventions. If, for example, medically induced euthanasia for terminally ill patients is not in principle a form of (assisted) self-murder, then this must hold regardless of whether a given society approves or disapproves of it. The same applies to senicide: if there are conditions under which senicide does not qualify as an instance of (assisted) self-murder, then these must be grounded in principle, not social consensus. To determine whether Inuit senicide can be justified on Kantian grounds, we must identify the principle on which a self-killing act could justifiably be willed – provided such a principle exists – and examine whether the Inuit practice aligns with it.
After arguing in the Metaphysics of Morals that suicide always violates a duty to the self, Kant follows with a series of ‘casuistical questions’. He asks, for example: ‘Is it murdering oneself to hurl oneself to certain death (like Curtius) in order to save one’s country? Or is deliberate martyrdom, sacrificing oneself for the good of all mankind, also to be considered an act of heroism?’ (MM, 6: 423). He goes on to pose related questions: Is it self-murder for a king to poison himself fatally in order to avoid being coerced into agreeing ‘to conditions of ransom harmful to his state’? For a person to kill himself to escape an unjust death sentence? Or for someone to kill himself to avoid harming others while suffering from the onset of hydrophobia (i.e., rabies)?
As is typical with Kant’s casuistical sections, he does not provide definitive answers to these questions. Nevertheless, there are moments where he appears to express admiration for those who choose death under certain morally charged circumstances. For instance, he writes that when a convict is offered the choice between death and forced labour, ‘the man of honor would choose death, and the scoundrel convict labor’ (ibid.). The ‘man of honor’, Kant continues, values honour more highly than life, ‘while the scoundrel considers it better to live in shame than not to live at all’ (MM, 6: 334). This closely resembles one of his reported statements during the Collins lectures on moral philosophy, likely held around 1775 (Kuehn Reference Kuehn, Denis and Sensen2015: 51):
So if certain persons, in all innocence, were to be accused of treason, though among them there were really a few men of honour, along with others of the baser sort, having no inner worth, and if all these people were together condemned to die, or to undergo a life-sentence of penal servitude, and each had to choose which of these punishments he preferred, it is perfectly certain that the honourable ones would choose death, and the worthless ones the penal service (L-Eth-C, 27: 376).
In the Critique of Practical Reason, Kant approvingly cites a passage from Juvenal’s Satires:
If summoned to bear witness in some dubious and uncertain cause, though Phalaris himself should dictate that you perjure yourself and bring his bull to move you, count it the greatest of all iniquities to prefer life to honour and to lose, for the sake of living, all that makes life worth living (CPrR, 5: 159, translated from Latin).
The bull of Phalaris is said to have been an ancient bull-shaped torture and execution device in which victims were roasted alive. According to Kant, this classical example ‘lets the reader vividly feel the force of the incentive hidden in the pure law of duty as duty’ (CPrR, 5: 159). The duty in question appears to be the duty not to deceive, even under the threat of gruesome death. A similar message is conveyed in Kant’s famous so-called ‘gallows example’, where a person is threatened with hanging unless they give false testimony against an honourable man. Kant insists that it is possible for the man at the gallows to ‘overcome his love of life’ and refuse to lie, because he knows he ought not to give false testimony – even if doing so would cost him his life (CPrR, 5: 30).
What these examples seemingly have in common is the idea that one is morally required to give up one’s own life if preserving it would require violating one’s moral integrity. If the price of staying alive is the corruption of one’s moral character, then that price, according to Kant, must not be paid. As Michael Cholbi rightly notes, ‘[m]orality and living honorably are necessary’ for Kant, ‘while life itself is not’ (Cholbi Reference Cholbi2000: 169; cf. L-Eth-C, 27: 373, 376-7).
It should be noted that these examples do not, strictly speaking, describe instances of self-killing – at least not in the conventional sense. None of the men in the examples wills his own death, either as a means or as an end. Rather, death is a foreseeable consequence that they are morally obliged to accept on pain of violating their moral integrity. It may therefore seem that these cases are not directly relevant to Kant’s position on suicide and self-killing in general.
However, there is textual evidence to suggest that, for Kant, the same principle applies to certain instances of self-killing. In the Collins lectures on moral philosophy, Kant is reported to have acknowledged the possibility of cases wherein a man ‘must’ put an end to his life, namely those in which he ‘can no longer live in accordance with virtue and prudence’:
Suicide [Selbstmord] can also come to have a plausible aspect, whenever, that is, the continuance of life rests upon such circumstances as may deprive that life of its value; when a man can no longer live in accordance with virtue and prudence, and must therefore put an end to his life from honourable motives (L-Eth-C, 27: 371).
As in his mature works, Kant here recognizes that the prospective loss of one’s moral integrity may oblige a person to forfeit their life. In this instance, however, he explicitly speaks in terms of ending one’s own life and choosing suicide (Selbstmord) from ‘honourable motives’, rather than treating death as an acceptable consequence of a moral decision.
It is admittedly puzzling that Kant also reportedly insists in the Collins lectures that suicide is impermissible under any conditions:
There are many conditions under which life has to be sacrificed; if I cannot preserve it other than by violating the duties to myself, then I am bound to sacrifice it, rather than violate those duties; yet, on the other hand, suicide [Selbstmord] is not permitted under any condition (L-Eth-C, 27: 372, italics mine).
On closer inspection, however, Kant appears to be using the word ‘suicide’ (Selbstmord) in two distinct senses. When he refers to ‘suicide from honourable motives’, he has in mind the deliberate sacrificing of one’s own life to preserve one’s moral integrity. When he speaks of suicide as never being permitted, he seems to mean acts of self-killing motivated by inclination – to escape suffering, despair, or unhappiness, for example (L-Eth-C, 27: 373). The way in which Kant reportedly speaks of suicide ‘from inclination’ in the Lectures is consistent with his mature works. Happiness, Kant reportedly says, is unfit as a ground for suicide, as ‘there is no necessity that, so long as I live, I should live happily’ (27: 373). Moreover, by choosing ‘suicide’ I dispose of myself as a thing, which is impermissible as I am not a thing but a being endowed with humanity (27: 372-3).Footnote 13 In the case of ‘honourable suicide’, then, it makes sense to say that one is not disposing of oneself as a thing; rather, one is affirming the supreme value of morality by sacrificing one’s life to uphold morality. By contrast, suicide ‘from inclination’ treats life and personhood as disposable in the service of happiness.
One might object that Kant’s early position, as expressed in the Collins lectures, should not be taken as representative of his mature view. It is possible that he later changed his mind, or that the lectures reflect not his own views but rather those of his predecessors (e.g., Baumgarten). However, a passage from the Critique of Practical Reason strongly suggests that Kant retained this position in his mature philosophy:
The action by which someone tries with extreme danger to his life to rescue people from a shipwreck, finally losing his own life in the attempt, will indeed be reckoned, on one side, as duty but on the other and even for the most part as a meritorious action; but our esteem for it will be greatly weakened by the concept of duty to himself, which seems in this case to suffer some infringement. More decisive is someone’s magnanimous sacrifice of his life for the preservation of his country; and yet there still remains some scruple as to whether it is so perfect a duty to devote oneself to this purpose of one’s own accord and unbidden, and the action has not in itself the full force of a model and impulse to imitation. (CPrR, 5: 158)
Kant offers this passage against viewing morality in terms of heroic or ‘magnanimous’ gestures. Even when acting beneficently, we must not lose sight of our duties to ourselves (CPrR, 5: 158). He admits to having reservations about the idea that meritorious actions that can be expected to lead to one’s own death could contain ‘the full force of a model and impulse for imitation’, as they may involve an ‘infringement’ of a duty to the self. This makes sense: if we are obliged to take our duties to preserve and perfect ourselves seriously, we cannot at the same time be obliged to endanger our lives whenever others are in need.
What is important to note here is not Kant’s reluctance to endorse these acts as moral exemplars – as ‘containing the full force of a model for imitation’ – but rather his comparative judgement that the actions of the man who ‘magnanimously’ sacrifices himself (i.e., directly wills his own death) to preserve his country are ‘more decisively’ creditable to duty than the actions of the man who attempts to rescue people from a shipwreck ‘at the greatest danger to his life’. This suggests that what matters morally is whether my choice not to sacrifice my life, be it by my own hand or that of others, somehow comes at the cost of violating a duty.Footnote 14 If not, then self-killing, risking one’s life, or allowing oneself to be killed (provided it is within one’s power to resist) are generally forbidden.Footnote 15 This logic applies to most cases, on Kant’s account (cf. Cholbi Reference Cholbi2015: 610), including the examples described in CPrR, 5: 158. In those rare cases where life can be preserved only by doing wrong, it must be relinquished.
Hence, some of the earlier-mentioned casuistical questions concerning self-killing – such as the self-poisoning king and the man suffering from hydrophobia – should be assessed in the same way Kant assesses the examples of the man at the gallows and the bull of Phalaris. If continued life is conditional on the commission of an immoral act, then one is morally obliged to forfeit that life, even if doing so requires killing oneself. On Kant’s account, such cases do not qualify as instances of self-murder.
In sum, for Kant, killing oneself is obligatory only if it is a necessary means of preserving one’s moral integrity. When performed on this basis, and as a matter of moral principle, the act does not qualify as an instance of self-murder. In section 3.2, I argued that it is implausible to claim that Inuit elders had a duty to opt for senicide on Kant’s account, given that involuntary starvation does not amount to a violation of duty. This conclusion now rests on firmer ground. The question that remains, however, is whether Inuit senicide might nevertheless have been merely permissible on Kant’s account.
4.2. Suicide as merely permissible
According to Kant, it is a contradiction to suppose that self-murder is permissible for rational beings. He maintains that ‘[m]an cannot renounce his personality as long as he is a subject of duty’ and that ‘it is a contradiction that he should be authorized to withdraw from all obligation, that is, freely to act as if no authorization were needed for this action’ (MM, 6: 422).
While Kant’s arguments against suicide (Selbstmord) are often regarded as puzzling or unconvincing (cf. Brassington Reference Brassington2006), the argument under consideration appears plausible. There does indeed seem to be a contradiction in the idea that I could be under an obligation to perfect myself, preserve myself, and promote the happiness of others while simultaneously being permitted to end my life at will (thereby ‘withdrawing from all obligation’). This would seem to imply that I was never genuinely bound by those obligations in the first place.Footnote 16
What is particularly important for our purposes, however, is not the argument’s plausibility but the fact that it reveals a sufficient condition that Kant deems necessary for the applicability of the duty not to ‘dispose’ of oneself at one’s discretion: namely, that one remain a subject of duty in general.
A self-killing act that is merely permissible, on Kant’s account, would thus have to be an act that is neither morally forbidden nor a duty. At first glance, the criterion just outlined may appear to rule out any form of merely permissible (assisted) self-killing. However, this is not necessarily the case.Footnote 17 Michael Cholbi, for instance, has argued compellingly that Kant’s ethics permits those anticipating severe dementia to request euthanasia once the condition has fully developed. On Cholbi’s reading, ‘[t]he death of rational agency coincides with the death of the person and the termination of moral status’ (Cholbi Reference Cholbi2015: 609). Since such individuals are no longer persons in the morally relevant sense, they are no longer subjects – or direct objects – of moral obligation. Accordingly, once dementia has taken its course, they are no longer under a duty to preserve their lives (ibid).Footnote 18
Since, from the absence of an obligation to preserve one’s life, it does not follow that one is under an obligation to end one’s life (Cholbi Reference Cholbi2015: 609), Cholbi takes himself to have identified a case in which (assisted) self-killing is morally permissible (or ‘optional’, as he calls it) (p. 610). In the next section, I argue that a similar consideration may be at play in Kant’s verdict on Inuit senicide.
5. What should Kant’s ethics make of Inuit senicide?
5.1. An interpretation of ‘ignominious’
We are now in a position to draw preliminary conclusions about how we should understand Kant’s verdict on Inuit senicide. In light of Kant’s views on honour and virtue in relation to duties concerning giving up one’s life, I propose that his use of the term ‘ignominious’ in the present context should be understood as expressing the idea that a life deprived of honour and virtue is a life not worth living – one the preservation of which cannot be a moral duty. On this interpretation, Kant used the word ‘ignominious’ to signify the ‘disgracefulness’ of living such a life, as a life deprived of (opportunities for) honour and virtue does not befit the dignity of a rational being.
If this is correct, then anticipating one’s own ‘ignominious death’ should be understood as approximating the end of one’s own life under circumstances in which honour and virtue are unattainable. As Kant sees it, virtue signifies moral strength of will: the disposition to do what duty requires from the motive of duty (MM, 6: 405). An honourable person is one who acts virtuously. Thus, to say that the prospect of dying ‘ignominiously’ provides legitimate grounds for ending one’s life is to say that one may take one’s own life – or permit others to do so – if, within the confines of one’s continued life, it would be impossible to live as a virtuous agent.
This interpretation has the merit of not tying Kant’s use of ‘ignominious’ to the ludicrous notion that one ought to be ashamed of dying from starvation. Instead, ‘ignominious’ signifies the disgracefulness of a dignified, rational person’s being condemned to a life the preservation of which is not morally necessary. In most of the textual examples discussed above, Kant writes of people who are presented with a tempting opportunity to disgrace themselves by violating their moral integrity. I suggest, however, that his remarks on Inuit senicide indicate that he regarded the practice as justified insofar as the elderly in question were condemned to live (and die) under ‘disgraceful’ conditions (i.e., conditions unbefitting of a dignified being) through no fault of their own.
5.2. On the ‘dead weight’ and ‘emotional burden’ interpretations
Contrary to Kant’s less ignominious death interpretation, on both the dead weight and the emotional burden interpretation, the Inuit practiced senicide in order to relieve the community of a burden during periods of hunger and scarcity. But would Kant’s ethics permit this?
In the Metaphysics of Morals, Kant clearly states that because humanity exists as an end in itself, a person’s worth cannot be measured by his usefulness as a means to his own ends or the ends of others (MM, 6: 434-5). It is important to emphasize this point, as it rules out the possibility that, on Kant’s view, those who are no longer able to care for themselves or others thereby lose their moral standing as ends in themselves. Hence, the moral status of infirm Inuit elders is not compromised by their inability to ‘make themselves useful’. Accordingly, on Kant’s account, no person may consent to being killed on the basis of being perceived – either by themselves or by others – as a (physical) burden on others, as the dead weight interpretation has it. This conclusion is reinforced by Kant’s critique of the Stoic who chose to die ‘because he could be of no more use in this life’:
The Stoic thought it a prerogative of his (the sage’s) personality to depart from life at his discretion (as from a smoke-filled room) with peace of soul, free from the pressure of present or anticipated ills, because he could be of no more use in life. But there should have been in this very courage, this strength of soul not to fear death and to know of something that man can value even more highly than his life, a still stronger motive for him not to destroy himself, a being with such powerful authority over the strongest sensible incentives, and so not to deprive himself of life. (MM, 6: 422)
At the same time, there is a recognizable parallel between the Inuit elder who turns to senicide to increase her family’s chances of survival and Kant’s account of the mythological Marcus Curtius, who ‘hurled himself to certain death’ to save his country (MM, 6: 423). Curtius, like the Inuit elder, sacrifices his own life in order to save others. Is such a sacrifice morally permitted?
I suspect that Kant’s position on Curtius’s deed resembles his ambivalent stance in the Critique of Practical Reason, where he describes acts of self-sacrifice – such as risking one’s life to save others – as ‘creditable to duty’, and even for the most part ‘meritorious’, while also raising concerns about possible infringements of a duty to the self (CPrR, 5: 158). As previously argued, self-sacrifice is only a duty, on Kant’s account, if it is a necessary means of fulfilling an end that is also a duty. Since only the concept of duty can justify sacrificing one’s own life for others, Kant is, I suspect, ultimately – though perhaps reluctantly – committed to the view that Curtius’s self-sacrifice was impermissible. The same verdict should apply to the Inuit elder who sacrifices herself ‘for the greater good’.
In any case, on Kant’s account it does not befit the dignity of a rational being to be degraded (by oneself or others) to the status of a mere ‘burden’ to others, no matter the circumstances – let alone to consent to being killed on such a basis. The same conclusion applies to the emotional burden interpretation. Kant insists that rational nature, which possesses a dignity beyond all price, must never be sacrificed as a means to one’s own happiness (or the reduction of discomfort), which merely has a ‘market price’ (G, 4: 421; cf. G, 4: 435, MM, 6: 422-3). Indeed, Kant does not appear to recognize suffering as such as a sufficient reason for ending one’s life (let alone the indirect suffering of children who pity their parents’ predicament). This is underscored by the earlier example of the Stoic: one may not take one’s own life merely to free oneself ‘from the pressure of present or anticipated ills’ (MM, 6: 422; cf. G, 4: 398). This idea can also be found in the Collins lectures on moral philosophy, where Kant is reported to have said: ‘I must not let myself be deterred from living by any fate or misfortune, but should go on living so long as I am a man and can live honourably’ (L-Eth-C, 27: 375). On my interpretation of what Kant means by ‘ignominious’ and how this concept may be used to justify (assisted) self-killing, it is only when one’s suffering impedes one’s capacity to participate in moral life (i.e., to live honourably) that suffering indirectly becomes a morally relevant consideration in this regard.
5.3. On Kant’s ‘less ignominious death’ interpretation of Inuit senicide
In section 4.2, I argued that Cholbi’s claim – that self-killing may be morally permissible (though not obligatory) for individuals anticipating severe dementia – bears a significant resemblance to how Kant’s ethics might evaluate Inuit senicide in light of the less ignominious death interpretation. At first glance, this comparison may seem puzzling. In Cholbi’s case, the individuals in question suffer from advanced dementia and, as a result, are no longer persons in the Kantian sense. By contrast, there is no indication in the anthropological literature – or in Kant’s reported commentary – that those Inuit elders singled out for senicide were no longer in possession of the rational capacities that, for Kant, constitute personality. Requesting senicide thus cannot have been morally ‘optional’ for the Inuit elderly on the grounds that they were no longer persons. On the surface, then, the two cases may appear fundamentally disanalogous.
This dissimilarity is less significant than it first appears, however. On Kant’s mature account, the dignity of human beings lies in their status as ‘subject[s] of a morally practical reason’ (MM, 6: 434; cf. CPrR, 5: 87). What distinguishes human beings as ends in themselves is that which sets them apart from other animals, namely their rational nature: their capacity to set ends independently of inclination and in accordance with rational principles (Kleingeld Reference Kleingeld2020: 391; cf. G, 4: 428–31, 434–5; CPrR, 5: 87). In other words, persons possess dignity, a worth beyond all price, by virtue of their autonomy and their capacity to participate in moral life – in short, by virtue of their humanity.
When we unpack Kant’s thesis that ‘[m]an cannot renounce his personality as long as he is a subject of duty’ (MM, 6: 422) against the background of this context, we can see that what is actually doing the work of grounding the duty against ‘optional’ self-killing is not the human being’s status as a person as such but his standing as a subject of duty. Cholbi’s argument establishes that those suffering from severe dementia cease to be subjects of duty because they are no longer persons. But this does not imply that the absence of personhood is a necessary condition for no longer being a subject of duty (although it is certainly a sufficient condition).
As I see it, Kant’s verdict on Inuit senicide implies that his ethical system is open to the possibility that a person may retain full rational capacities and yet no longer be a subject of duty in general. The notion of being confined to an ‘ignominious’ life until death is a prime example of such a case. As I have argued, Kant uses the term ‘ignominious’ to describe a life in which honour and virtue are unattainable. Those condemned to such a life are effectively barred from acting from duty – not because they lack reason or personhood, but because external, bodily, or environmental constraints render their good will practically impotent.
If external circumstances make it impossible for me to govern myself by self-authorized ends, and thus for me to use my good will, then I am no longer able to participate in moral life. If participation in moral life becomes a permanent impossibility,Footnote 19 then there is no plausible sense in which I can still be said to be under moral obligation, let alone subject to the duty of self-preservation.
To be clear, it is only when agents are evidently (i.e., with a sufficient degree of certainty) condemned to live under undignified conditions for the rest of their lives that they are no longer subject to the duty of self-preservation. In other words, they must be ‘beyond help’ with regard to their inability to escape their undignified predicament. Since one’s dignity as a human being grounds a claim to be treated with respect (i.e., as an end in itself) by others and a duty to respect (and thus preserve) oneself, one is obligated, if possible, to seek help from others if necessary to escape one’s undignified predicament. Others are morally obligated to help if they can do so without excessively sacrificing their own well-being (MM, 6: 393). Importantly, it should not matter, on Kant’s understanding of the good will, whether one succeeds in acquiring the necessary help:
If with its greatest efforts [the will] should yet achieve nothing and only the good will were left (not, of course, as a mere wish but as the summoning of all means insofar as they are in our control) – then, like a jewel, it would still shine by itself, as something that has its full worth in itself. (G, 4: 394)
What matters is that the capacity to use one’s good will (i.e., the ‘summoning of all means insofar as they are in [one’s] control’) – in this case, by seeking help – remains intact.
If my argument is successful, then the anticipation of an ‘ignominious’ life until death may, on Kantian grounds, constitute a legitimate basis for ending one’s life in a manner consistent with Kant’s mature ethics. The remaining question, then, is whether, on Kant’s interpretation of the circumstances, it is accurate to characterize the lives of the elderly under consideration as ‘ignominious’.
According to Kant’s reported interpretation, if senicide were not practiced by the Inuit, the infirm elderly would be forced to remain behind, alone and without food, while the rest of the group went out hunting for weeks at a time.Footnote 20 They would, in effect, be condemned to starve (and presumably freeze) to death. The question, then, is whether it was still possible for the elderly to live honourably and virtuously under these circumstances – that is, to will (not merely wish for) moral ends.
The parallel between Cholbi’s example of those suffering from dementia and Kant’s reported interpretation of the Inuit case is that both examples describe a situation in which human beings physically endure but are, to a significant extent, condemned to do so as mere ‘playthings’ of nature rather than persons. In the Groundwork, Kant writes that ‘[a]utonomy’, the will’s property of being a law to itself, ‘is the ground of the dignity of human nature and of every rational nature’ (G, 4: 436). Autonomy requires independence of the will from natural causes (4: 446; cf. MM, 6: 213-4). In the case of dementia, this independence is (arguably) lost because the human being no longer exists as a rational agent, whereas the Inuit elderly retain the rational capacities that constitute personality but, for all morally relevant intents and purposes, nevertheless remain ‘at the mercy’ of external, natural forces beyond their control. To the extent that their good will is permanently deprived of control over ‘summonable means’ (G, 4: 394), they are thereby rendered unable to live as – and thus, in a morally relevant sense be – free, autonomous persons. Hence, it can be argued that, just as Kant’s ethics can permit (assisted) self-killing in the dementia case Cholbi describes, so too can it permit it in the Inuit starvation case. Crucially, however, this act would be permissible at most, not obligatory. Unlike the man at the gallows, who must choose between committing a grave moral wrong to preserve his life and dying to preserve his moral integrity, the Inuit elder must choose between expediting the end of a life that is unbefitting of a dignified, rational being – the preservation of which is no longer morally necessary – and enduring that life to its natural conclusion.
While Kant is reported to have said that a person ‘must put an end to his life from honourable motives’ when it becomes impossible to live in accordance with virtue and prudence (L-Eth-C, 27: 371, italics mine), this conclusion appears excessively strong in the present context – as Kant himself seems to acknowledge in his reported statement that Inuit senicide is to some degree justified, a stance far removed from regarding it as obligatory. After all, there is no dishonour in continuing to live a life the prolongation of which is no longer morally obligatory. As Cholbi observes, it is a mistake to infer that the absence of a duty to preserve life entails a duty to disrespect or destroy it (Cholbi Reference Cholbi2015: 609). In all other cases, Kant deems it a duty to sacrifice one’s life only when doing so is necessary to avoid moral corruption (cf. p. 610). But the Inuit elder who chooses not to expedite death does not thereby become a bad person, even when there is no prospect of his life’s suddenly becoming worth living, as far as morality is concerned. Moreover, Kant’s reported idea that some Inuit communities practiced senicide to spare their elders a more ignominious death suggests that he regarded the elders’ actual deaths – though comparatively less ignominious – as still marked by indignity.Footnote 21 This stands in stark contrast to his moral praise for those who choose death to escape moral corruption, whose acceptance of their own death ‘from honourable motives’ appears not to be ignominious at all.
For the Inuit elders in question, then, the choice was not between living as a bad person and not living at all – as it was for the man at the gallows – but between not living as a person (through no fault of one’s own) and not living at all. While the man at the gallows preserves his moral integrity by dutifully choosing death over committing injustice, the same cannot be said of those subject to senicide, as in their case virtue and honour have supposedly been rendered unattainable in both life and death. Kant’s verdict on Inuit senicide, I argue, points to the following possible Kantian line of reasoning: while such a life is deprived of the moral agency necessary to ground a duty of self-preservation, the choice to end it may be permissible, though not morally required.
This line of reasoning is not decisive, however, as Kant’s reported claim that Inuit senicide was justified to some degree suggests that he harboured reservations about the practice’s permissibility.Footnote 22 There may be several reasons for Kant’s hesitation, or at least his apparent refusal to endorse the practice unequivocally. For instance, he may have judged it insufficiently certain that the group would not encounter food sources shortly after abandoning the elderly to their fate. This suggests that Kant did not consider it sufficiently certain that the elderly would have been condemned to an ‘ignominious death’ had they not opted for senicide. Indeed, this reservation seems to have some basis in fact: the anthropological record includes instances in which Inuit groups returned to elders they had previously left behind, taking them in after unexpectedly securing food (cf. Pousset Reference Pousset2023: 22). Additionally, Kant may have questioned whether, even given the extreme nature of the circumstances, the Inuit elderly were fully incapacitated as moral agents (cf. R, 6: 62). Given their alleged abandonment, bodily infirmity, and lack of a source of nutrition, they were evidently no longer capable of fulfilling duties to others or discharging positive duties to themselves – such as the duty to cultivate their natural talents (MM, 6: 444–5) or actively preserve their own lives (6: 421). Their negative duties to themselves, however, appear to have remained perfectly intact. If we accept that virtue, as ‘the moral strength of a human being’s will in fulfilling his duty’ (6: 405), can also be exercised by not willing immoral ends from the motive of duty – not just by positively willing moral ends – then it would appear that the Inuit elderly were not fully incapacitated as moral agents, even given their predicament.Footnote 23
Nonetheless, Kant’s assessment of Inuit senicide suggests that he saw in it something redeeming – something absent from cases of (assisted) suicide motivated by self-love. Throughout his works, Kant consistently condemns the latter as instances in which one ‘throws away’ one’s humanity for the sake of happiness. Notably, such language is entirely absent from his reported remarks on Inuit senicide, prompting the question of why this might be. The most plausible explanation, I suggest, is that Kant regarded the Inuit case as involving a more fundamental moral consideration than that which underlies (assisted) self-killing motivated by considerations pertaining to happiness. Whereas the Stoic who believes he is ‘of no more use’ and the ‘unhappy person’ contemplating suicide lack the desire – but not the means – to treat themselves with the respect due to a rational being, Kant seems far less certain that the same can be said of the Inuit elderly, whom he describes as nearly immobile, abandoned without food, and awaiting imminent death beyond the reach of help.
6. Concluding remarks
Kant’s reported view on Inuit senicide represents a rare – perhaps even unique – instance in which he considers the permissibility of a non-obligatory (assisted) self-killing act in terms other than his characteristic portrayal of an agent coldly calculating whether continued life will yield the happiness he desires.
Kant’s verdict, I propose, suggests the outlines of a possible Kantian argument: namely, that a person may consent to ending her life when it has become – and is expected to remain – impossible to realize virtuous ends, such that continued existence no longer embodies the dignity that grounds the duty of self-preservation. If such consent is morally possible, it should be possible for others to permissibly act upon it.Footnote 24
Even with the relevant moral consideration identified, however, Kant’s reported assessment of Inuit senicide remains ambiguous. Hence, it falls to contemporary Kantian ethicists to assess whether his framework ultimately permits ending one’s life in anticipation of an ‘ignominious’ death and, if so, to delineate the conditions under which a person might be said to confront such a prospect. If my argument holds, it invites further exploration of the relationship between Kant’s notion of dignity and the duty of self-preservation, particularly in the broader context of end-of-life decision-making.
Acknowledgements
I am grateful to the Netherlands Organization for Scientific Research NWO for financial support, and I thank Pauline Kleingeld, Hanno Sauer, Vinicius Carvalho, Luke Davies, Fiorella Tomassini, Tijn Smits, and two anonymous referees for invaluable comments on earlier drafts of this paper.
Open access
