1.1.

Consider some representative passages discussing Kant’s prohibition on suicide and its relation to assisted dying:

1. i. “Although Kant did not address the question of assisted suicide, it follows from his arguments regarding suicide that assisted suicide is impermissible” (Gunderson, Reference Gunderson2004, 277).

2. ii. “I contend that Kant’s argument does not establish an obligation to forego suicide (and hence, an obligation on the part of others not to assist in suicide)” (Cholbi, Reference Cholbi2018, 144, my emphasis).Footnote ⁶

3. iii. “An agent who takes his own life acts in violation of the moral law, according to Kant; suicide, and, by extension, assisted suicide are therefore wrong” (Brassington, Reference Brassington2006, 571).

4. iv. “from the fact that it is permissible for an agent to do something it does not follow that it would be permissible for someone else to assist him […]. However, the entailment does seem to hold for impermissibility: if the act would be wrong then it would be wrong to assist it” (Sumner, Reference Sumner2011, 85n).

The last passage is not discussing Kantian moral commitments, but the reasoning is the same. If suicide is impermissible, MAID is ruled out too.

It is natural to read these passages as claiming something like the following: “on the assumption that committing suicide is wrong, killing someone on their request is aiding or enabling the performance of their impermissible action and is therefore also wrong.” This suggestion can neatly explain the presumption that MAID is wrong whenever suicide is wrong. However, I will argue that this way of thinking about MAID is not available to Kant given the way that he thinks about violations of duties to self in general and the duty prohibiting suicide in particular. The remainder of Section 1.1 clarifies Kant’s prohibition on suicide. In Section 1.2, I turn to the question of how this relates to providing MAID.

Kant’s prohibition on suicide is best understood in the context of his claim that humanity has dignity and is an end in itself. An entity is an end in itself when it has the capacity for morality (4:435, 5:87, 6:434–35). For Kant, ends in themselves are one of the two radically different kinds of ends that structure our actions.Footnote ⁷ In general, an end is “that which serves the will as its objective ground of self-determination” (4:427); that is, an end determines how an agent should act. Relative ends are given to us by inclination and are states of affairs to be brought about by our action. The satisfaction of the sum total of our inclinations is our happiness (4:399, 8:282–83). In contrast to this, ends in themselves are existing entities—that is, persons—that restrict the ways in which we may pursue our relative ends (4:427–30). For Kant, the value of our relative ends is conditional upon our having a good will (4:393, 5:93, 6:481).Footnote ⁸ When we act in a way that fails to respect the dignity of ends in themselves, we thus fail to meet a necessary condition for our relative ends having value for us. This means that we act irrationally when the means we deploy in pursuit of relative ends fail to respect ends in themselves. While the achievement of the ends contributes to our happiness, our happiness lacks value when it is brought about via immoral means.

In the Groundwork, Kant says of a would-be suicide that, “If he destroys himself in order to escape from a trying condition he makes use of a person merely as a means to maintain a tolerable condition up to the end of life” (4:429). The suicidal person destroys themselves, and thus the humanity in them, for the sake of, for example, “a tolerable condition,” or to avoid servitude (27:603), or for the sake of honor (27:371), or for another end of inclination. In doing so, they treat themselves and their rational nature as something that can be disposed of in pursuit of their relative ends, and thus they undercut a necessary condition for the value of those ends. A consequence of this is that, as Cholbi puts it, “prospects of future suffering or despair are irrelevant to the moral permissibility of suicide” (Cholbi, Reference Cholbi2010, 493). As these prospects concern our relative ends, our pursuit of them is limited by our moral obligations. Suicide is unconditionally prohibited because there are no ends for which the wilful killing of oneself can serve as a permissible means.Footnote ⁹

It is important that we understand Kant’s prohibition in the light of his non-consequentialism, according to which the “objects of moral assessment are not objects or states of affairs, but willings” (Herman, Reference Herman1993, 94, my emphasis). For Kant, all free actions (i.e., all actions imputable to the person who performed them) have a maxim. Maxims specify the end to be pursued by our action and the means by which we realize that end (6:389). Actions are assessed based on their maxims. Moreover, different maxims might result in the same externally manifesting behavior. Two actions that are alike in their external behavior, and which bring about identical consequences, might nevertheless differ in terms of their permissibility; one might be the result of a permissible maxim, the other an impermissible maxim (4:407). Since the maxim determines the act-type tokened by the agent’s action, the same external behavior will not always token the same act type. Kant’s focus on maxims has interesting implications for the duty prohibiting suicide. I focus on two here.

First, it will sometimes be compatible with the prohibition on suicide to perform an action even though you know it will lead to your death—that is, when death is a foreseen but unintended consequence of the action. For example, in the second Critique, Kant describes a person who must give false testimony or else be immediately executed (5:30). Kant’s point is about the relation between freedom and the moral law. The person in this situation judges “that he can do something because he is aware that he ought to do it” (ibid.). Since he knows that he ought not to lie, he also knows that not lying is an option for him. Important for us, however, is the fact that a person who refuses to lie in this situation because lying is wrong would not violate their duty to self. Their maxim does not specify their death as the means for bringing about one of their ends. They do not aim for their death by performing the action and so are not the author of it (6:81n, 27:629, 27:376).Footnote ¹⁰

Second, it is possible to violate the prohibition even when you do not bring about your death. For example, consider a person who tries to commit suicide as a way of avoiding future suffering. To do this, they decide to stab themselves. Unbeknownst to them, however, the knife they use is a harmless stage prop. While they cause themselves no harm at all by their attempt, they still violate the duty prohibiting suicide. Their maxim, “I kill myself to avoid future suffering,” is not made permissible by the ineffectiveness of the means they choose for the achievement of their end. As Kant is reported to have said in his lectures: “It is the intention to destroy oneself that constitutes suicide” (27:371).Footnote ¹¹ Recall that a person commits suicide when they pursue a course of action with the intention of bringing about their death, and that course of action results in their death (in the way they intend). For Kant, a person violates the prohibition on suicide merely by pursuing a course of action with the intention of bringing about their death. They need not be successful for it to constitute a duty violation.

1.2.

With the nature and scope of Kant’s prohibition on suicide in mind, I turn now to the question of how duties to self are violated in our interactions with others. I will argue that some requests and instances of consent constitute violations of duties to self.

To begin, consider consent that constitutes the violation of a duty to another. For example, say that I promise you that I will not lend anyone my copy of the second Critique. After making the promise, a friend asks if they can borrow it for an upcoming seminar, and I agree. While I have a duty to you not to lend my book out, my consent is normatively transformative. To think otherwise would require asserting that my friend would wrong me by taking the book, even though I have given them permission. This is implausible. With that said, I wronged you by giving them permission. This wrong is explained by the fact that I made it permissible for my friend to take the book when I said that I would not. That the wrong is explained in terms of giving permission also indicates that I break my promise even if they do not take the book.

Duties to self can also be violated by acts of consent. Consider self-promising. I will assume that it makes sense to say that we can make promises to ourselves, and that we thereby generate a duty to self to do what we promised.Footnote ¹² For example, I promise myself that I will work through the weekend to get ahead on the term’s teaching preparation. Having made myself the promise, a friend asks if they can come around to watch the Big Game on Sunday, and I agree. Here, I violate the promissory duty to self, even if the friend does not show up. This is because I make it permissible for another to interfere with the completion of the activity I promised myself I would complete.

Finally, Kant claims that we have a duty to self not to forfeit the respect we are owed from others (6:435).Footnote ¹³ But imagine that I am very eager to be noticed by a certain group of people. To get that attention, I convince a friend to demean me in front of them so that I can gain their sympathy. In enlisting the help of my friend, I consent to being demeaned. We consent to another’s action both by explicitly giving permission and by making requests or issuing directives.Footnote ¹⁴ Now imagine that the plan falls through. The people whose attention I desire do not show up at the time I expect, and my friend never acts on my consent. I have still violated my duty to self. Just as above, the duty is violated in giving consent to an action or form of treatment that I ought not make permissible, and not at the point at which the consented-to action takes place. In this way, making a request that is not acted upon can be explained in terms similar to those of a failed attempt.

So, one way in which we violate a Kantian duty to self is by consenting to or requesting a certain kind of treatment from another. However, this means that acting on the consent of a patient who asks for MAID is not aiding the patient’s violation of their duty to self.Footnote ¹⁵ For Kant, the patient has already violated their duty just by seeking out and requesting MAID. That this is so can be seen from the fact that the patient would violate their duty to self even if the healthcare practitioner ends up not acting on their request. It is enough that the patient attempts to engage their help in bringing about their death. Recall that a natural understanding of the wrongfulness of MAID, assuming the impermissibility of suicide, is that it aids or enables the duty violation of another. Given the familiar features of Kantian non-consequentialism outlined above and his account of duties to self, this cannot be his understanding of the morality of the interaction between the patient and their healthcare practitioner.

2.1.

The conclusion of my argument so far is only that healthcare practitioners do not enable or aid their patient’s duty violations. This leaves open the possibility that they act wrongly in some other way. We act wrongly not only when we enable or causally contribute to wrongdoing but also when we otherwise support it or are complicit with it or turn a blind eye to it, and so forth. In most jurisdictions in which it is legal, MAID involves an ongoing relationship between the healthcare practitioner and the patient. This relationship will include discussions of the patient’s prognosis and treatment options, evaluations of the patient’s decisional capacity, and perhaps also discussions with and support for the patient’s family. Healthcare practitioners might also be the first to alert patients to the possibility of MAID, for example, as part of informing them of the full range of treatment options open to them. If suicide is impermissible, healthcare practitioners plausibly act wrongly by engaging in this broader relationship and supporting the patient’s course of action. In doing this, we might think that they endorse the patient’s ends and, on the Kantian view, their wrongful maxim. In the remainder of this section, I outline one promising way to characterize the wrongfulness of providing MAID given the impermissibility of suicide.

The form of wrongdoing I focus on is expressive wrongdoing. A person commits an expressive wrong when the meaning expressed or intended by their action (or omission) is morally impermissible. Expressive wrongs are not a function of the states of affairs brought about by the action but rather of the action qua expression of a set of beliefs or attitudes.Footnote ¹⁶ Expressive wronging as a category thus captures a large range of wrongs that are not explained by appeal to the states of affairs brought about by one’s action or to one’s causal role in another’s wrongdoing. This makes it a good fit for considering the healthcare practitioner’s action in this context. A person commits an expressive wrong, to use an example amenable to Kantian ethics, by acting in a way that expresses a lack of equal respect for a group or individual. This is plausibly what happens in hate speech, which, in most cases, depicts its target as lacking the basic moral standing of the speaker.Footnote ¹⁷ According to an expressive analysis, hate speech is wrongful (at least in part) in a way that is not reducible to or explained by the harmful consequences that arise from that expression. We might also think this is what happens when a person ignores the obvious plight of another who is suffering. Failures of attention or action can also express a lack of equal respect.

Kantian ethics is attentive to the expressive nature of action. For example, Allen Wood has suggested that what the formula of humanity version of the categorical imperative “fundamentally demands of our actions is […] that they express proper respect or reverence for the worth of humanity” (Wood, Reference Wood1999, 141). An action that expresses contempt for humanity, in oneself or another, will be wrongful even when it fails to bring about any negative consequences. We have already seen this in the case of failed violations of duties to oneself. The person who attempts to kill themselves for the sake of inclination thereby expresses that they do not ascribe the dignity to rational agency that Kant thought they must, or else that they do not respect that dignity.

This is not to say that Kantian concerns align completely with the contemporary discussion of expressive wrongs. A central feature of the contemporary discussion is that an action’s meaning is set by the social, historical, and linguistic context of the speaker, such that, for many instances of action, the expressive meaning of that action will be objectively determinable.Footnote ¹⁸ This is necessary for an action’s meaning to be amenable to legislative control. For Kant, however, the expressive meaning of an action is most plausibly given by the agent’s maxim. The socially determined meanings of the action will only be relevant insofar as they are incorporated into the agent’s practical reasoning. In most cases, these will align, but they will not always do so. For that reason, the contemporary and Kantian accounts differ in how they attribute expressive meaning to an action.

By acting together with the patient and helping to realize the patient’s intended outcome, healthcare practitioners might be taken to commit an expressive wrong on a Kantian account because their actions show support for the patient’s wrongful maxim. While not causally contributing to the patient’s wrongdoing, healthcare practitioners might nevertheless be complicit in that wrongdoing. This will not characterize every interaction between a patient and their healthcare practitioner in these contexts. For example, a healthcare practitioner might provide MAID to a patient they believe does not satisfy the conditions because that patient hounds them unrelentingly with requests for aid in dying. Assuming that the healthcare practitioner has communicated their disagreement with the patient, the provision of MAID in this circumstance would not constitute support for the patient’s decision in the relevant sense. The healthcare practitioner would be acting wrongly, but not in the way characterized by expressive wrongdoing.Footnote ¹⁹ With that said, many instances of a healthcare practitioner providing MAID will also be instances of support for the patient, insofar as the healthcare practitioner agrees with and adopts the patient’s end as their own. In cases in which this occurs, the healthcare practitioner violates the Kantian requirement to respect the dignity of humanity and so acts wrongly. In short, by acting jointly with the patient to bring about the outcome the patient hopes will result from the duty violation, healthcare practitioners might thereby be taken to express support for it. Such action is impermissible in Kantian ethics. This indicates that even if the healthcare practitioner does not wrong the patient (because the patient requests the treatment, see Section 3.2), and does not aid the duty violation itself, they nevertheless act impermissibly.

2.2.

We ought to take the expressive concerns raised in the previous section seriously. They constitute a significant barrier to the permissibility of MAID in Kantian ethics. However, Kant’s political philosophy shows that this barrier is not insurmountable. In fact, the legalization of MAID can, in some circumstances, insulate healthcare practitioners from the charge of ethical wrongdoing. This is because certain forms of legalization involve authorizing healthcare practitioners to act on their patients’ requests. In authorizing them, the law does not make expressive wrongs morally permissible. Rather, it changes the nature of the healthcare practitioner’s action such that they no longer commit those wrongs. That is, legalization can change what is expressed by the healthcare practitioner’s action, and thus its moral status.

In this context, it is helpful to recall Kant’s discussion in “An Answer to the Question: What is Enlightenment?.” In that essay, he outlines the barriers to enlightenment that must not be imposed by the sovereign. Kant argues that individuals must be free to express their disagreements with the state without fear of punishment and coercion. The public use of one’s reason, that use one makes “as a scholar before the entire public of the world of readers” (8:37), must not be the subject of state censure. To the contrary, what Kant calls the “private use of reason” can permissibly be constrained. One’s private use of reason “is that which one may make of it in a certain civil post or office with which he is entrusted” (ibid.).Footnote ²⁰ As examples, Kant lists a military officer who must obey orders despite disagreeing, a citizen who must pay taxes despite finding them unjust, and a clergyman who must preach the teachings of the church even if he finds them erroneous. The officer, citizen, and clergyman are all contractually bound to obey a higher authority. They are “carrying out another’s commission” (8:38).Footnote ²¹ This does not mean that all of their actions are strictly prescribed by that authority. A clergyman might be permitted to choose which lessons to teach his catechism class and the subjects of his sermons. He might have discretion in terms of how he acts in his position. What he is not permitted to do, however, is use his role within the church to criticize his mandate.

As part of this discussion, Kant claims that those who occupy civil positions ought not to be concerned that the actions they take as part of their role do not reflect their own beliefs, or might even stand in tension with them. Of the clergyman, Kant says: “there is nothing in this that could be laid as a burden on his conscience. For what he teaches in consequence of his office as carrying out the business of the church, he represents as something with respect to which he does not have free power to teach as he thinks best, but which he is appointed to deliver and prescribe in the name of another” (ibid.). Clergymen are not to use their position in order to preach the creed that they believe to be true. They are employed to share the teachings of the church to which they belong, and do not act wrongly when doing so even if they do not believe those teachings. There are limits of course. Kant claims of a clergyman who disagrees with the creed of the church, that if it was “altogether impossible that the truth may lie concealed” in the teachings or if they are “contradictory to inner religion,” “he would have to resign” (ibid.). Short of this, however, he must not violate his contractual obligations. The same goes for the officer who believes his orders are imprudent and the citizen who believes the taxes they owe to be unjust.

The roles Kant discusses are all civil positions. In distinguishing between the private and public uses of reason, he is drawing on the distinction between persons and offices. While individuals as members of the public may criticize what they take to be imprudent, unjust, or untrue, the same is not the case for those same individuals as holders of an office. As Ripstein puts it, “An official is permitted only to act for the purposes defined by [their] mandate. The concept of an official role thus introduces a distinction between the mandate created by the office and the private purposes of the officeholder” (Ripstein Reference Ripstein2009, 192).Footnote ²² Insofar as the official acts on their mandate, the actions they take are to be understood as representing the higher authority with whom they have contracted.

Let us return to MAID. Healthcare practitioners do not occupy civil positions, but they do (in some jurisdictions) occupy public positions that are governed by a mandate. This is why the legalization of MAID can make a difference to the expressive nature of the healthcare practitioner’s action. In those jurisdictions, they act in the name of the state and not for private purposes. Following Cordelli, I take two necessary conditions for an agent A to act in the name of the principal P whom they represent to be as follows: (i) P has validly granted A the authority to do so, and (ii) A does not perform the action for reasons that are excluded in virtue of acting under P’s authorization (Cordelli, Reference Cordelli2020, ch. 5).Footnote ²³ To better understand this account, it is instructive to consider a concrete example of one way in which MAID has been legalized.

It is at least possible that healthcare practitioners in Ontario meet the conditions for representative agency outlined above and thus act in the name of the state rather than as private individuals when they provide MAID. Healthcare practitioners are validly authorized by the province to act on the patient’s request by virtue of the legislation decriminalizing MAID and the patient’s right to access healthcare services. The authorization is valid because, as I defend further in Section 3, the legalization of MAID is consistent with Kantian principles of justice. This does not mean that the state must do so. For example, those states unable to implement the necessary safeguards would not act wrongly by failing to legalize MAID.

Moreover, it is possible that healthcare practitioners who act on their patients’ requests do so without acting on reasons that are ruled out by virtue of their acting on the state’s authorization.Footnote ²⁶ For example, they might act on their patient’s request because the patient is entitled under law to access the service, or simply because it is a way for them to earn money in their profession. Insofar as they act for reasons that are not ruled out by the state’s authorization, the expressive nature of their action is altered. Absent legal authorization, their action is plausibly an instance of expressive wrongdoing as described in Section 2.1. In jurisdictions like Ontario, however, healthcare practitioners who provide MAID do so as agents of the province and not as private individuals. Just as with the clergyman described above, this does not mean that they lack discretion in the way in which they fulfil that mandate. However, just so long as they do not violate their professional obligations, their actions should be taken as representative of the province and not their private will.

There are many other examples of a person’s institutional role altering the moral permissibility of their actions. For example, a defence lawyer ought to work toward the acquittal of even those clients whom they have good reason to believe are guilty of the crimes they have been charged with. This is because, within a just court system, the role of the defence lawyer is not to make a judgment about the guilt or innocence of their client, but to defend them and try to secure their release. Assuming that the institutions are just, and that the relevant just procedures have been followed, a person would normally act wrongly by trying to secure the release of a person they know to be guilty. The fact that the lawyer acts in an institutional capacity makes the otherwise wrongful act permissible (McMahan, Reference McMahan2006, 384–385). Healthcare practitioners’ actions can be seen in a similar light.

The above example also helps to address a possible objection. I have mentioned that healthcare practitioners are not duty-bound to provide MAID. They have rights of refusal on grounds of conscience or religion. While some jurisdictions require dissenting healthcare practitioners to help the patient access the relevant services, this is not always the case.Footnote ²⁷ We might think that this means that each instance of a healthcare practitioner providing MAID merely represents their private will, thus exposing them to the charge of ethical wrongdoing.Footnote ²⁸ However, I think we need to distinguish between the act of dissenting, which is private, and action in accordance with a mandate. Here is where it helps to return to defence lawyers. In many jurisdictions, even publicly appointed defence lawyers may refuse to represent a client on grounds of fundamental ethical disagreement. Such refusals are indeed private acts, expressing the beliefs and commitments of the individual. However, that does not mean that each instance of defending a client is equally a private act. By refusing, a lawyer is choosing not to act on the authorization available to them in that particular case. However, if they accept, the expressive meaning of their action will be set by their institutional role. Kant’s own example of the clergyman works similarly. If they quit due to a fundamental disagreement with the tenets of the church to which they belong, this is a private act. However, if they remain in their position, they speak in the name of those whom they represent.Footnote ²⁹ Healthcare practitioners likewise have a choice of whether to offer MAID, and they might decline to act on the request of a patient on grounds of fundamental disagreement with the practice. However, if they do offer MAID in jurisdictions in which it has been legalized as it has been in Ontario, the meaning of their action is not determined by their private will.

3.2.

A second objection is due to Ripstein. He claims that it is impossible for one person to successfully consent to be killed by another on Kant’s view due to the formal features of consent. If correct, healthcare practitioners would wrong their patients by administering MAID, and the practice would be prohibited in a Kantian state. While Ripstein’s argument concerns consenting to be killed in general, I will restrict my response to the context of MAID.Footnote ³⁶

Ripstein’s argument proceeds as follows:

The idea is the following. When I give you a right to do something to me, I also undertake an obligation to you (ibid. 118). For example, if I’ve given you permission to touch my hair, I would wrong you by pushing you away as if in self-defence when you did that very thing. This is because part of what is involved in giving you the permission is undertaking an obligation not to treat you as an aggressor. When the action in question is being killed by the other, this means that consent amounts to undertaking an obligation to allow oneself to be killed (i.e., not to resist). Moreover, Ripstein claims that “murder […] is a form of biological slavery, since the murderer decides whether the victim will continue to exist” (ibid. 140). By undertaking an obligation to permit another to kill us, we are allowing them to decide whether we will continue to exist, and this is a power no person can have over another.

Ripstein’s argument is significantly weakened when we appreciate that a waiver can operate in different ways. Not all waiver entails undertaking an obligation that the other gets to decide for us whether the consented-to action will occur. In many cases, at least, we retain control over the rights we waive and may revoke the permission given to another should we choose to do so.Footnote ³⁷ Consent makes it the case that consent-receiver no longer has a directed duty to consent-giver to omit the consented-to action. But the consent-receiver is not entitled, simply by virtue of having their directed duty waived, to the successful completion of the action. This is because the waiver itself does not always create a claim right in the consent-receiver that the consented-to action takes place.

A waiver that occurs in the context of a contract is different. At least in standard cases of contract, consent-giver does create a claim right in the consent-receiver that corresponds to a duty in the consent-giver to perform or omit some action. Failing to act in the way agreed to would wrong consent-receiver. Consent-receiver also has control over whether the action occurs due to the fact that they can choose to enforce or waive their claim against consent-giver.Footnote ³⁸ If MAID were modeled along the lines of this kind of contractual transfer, then it would be true that the healthcare practitioners would be able to choose for the patient. However, we should not characterize the practice this way. Rather, patients are best understood as waiving their rights while also retaining the possibility of revoking their consent without wronging their healthcare practitioners. Indeed, that patients may revoke consent is a central feature of the practice of MAID, and of consent in medical settings more generally. So Ripstein is wrong to think that the consent-giver is a thing for the consent-receiver’s use.

Ripstein might object that this does not address his central concern, which is that the form of the interaction requires the consent-giver to have the status of both “thing” and “person.” We might think that this, and not the obligation one incurs, is why consenting to being killed by another is ruled out on the same grounds as slavery contracts. An important part of Ripstein’s argument is his claim that “two people can only act together in a way that is consistent with their freedom provided that they unite their purposes while preserving their separate purposiveness” (Ripstein, Reference Ripstein2009, 135). Consensual killing falls foul of this requirement on Ripstein’s view, and this explains why it is ruled out.

I do not think this helps Ripstein’s case. While Ripstein speaks of the preservation of purposiveness, he does not believe that our mere capacity to set and pursue ends provides a normative foundation for relations of Right in Kant’s political philosophy.Footnote ³⁹ Instead, this capacity is a necessary condition for agents to stand in relations of Right to each other: “In talking about purposiveness, my concern [in Force and Freedom] was solely to characterize the manner in which the human beings who stand in relations of right are able to do so because of features that they have apart from those relations” (ibid., 317). When Ripstein speaks of purposiveness he means “to be filling out the idea of choice, in the sense in which innate right provides for its coexistence” (ibid., 317–318). However, the sense of choice provided by the innate right is a moralized one, according to which our capacity for choice is infringed only when we are hindered from performing an action that we have a right to perform. It is a sense of choice concerned with protecting against instances of being subordinated to the will of another. This means that the mere reduction in our nonrelational capacity to set and pursue ends as a result of an interaction with another is insufficient to rule that interaction out on Kantian grounds. Instead, it needs to be the case that our capacity for choice is wrongfully reduced by that interaction, and this only happens when our rights are violated. Viewed in this way, while MAID is a form of interaction in which one person’s ability to stand in rightful relations to others is reduced completely, it is not thereby a form of interaction in which they are wronged. This is because the patient is not subordinated to the will of the healthcare practitioner.

For these reasons, Ripstein’s formal argument against MAID fails. It is not true that in consenting to MAID, a patient lets their healthcare practitioner decide whether they will continue to exist. Nor is it true that in consenting, they make themselves a thing for the healthcare practitioner’s use. It is important to note that the failure of this argument does not imply that one person can consent to be killed by another under any circumstances.Footnote ⁴⁰ This is because a formal ability to consent to be killed in no way entails a lack of other constraints on valid consent. For example, we might think that the consent giver must have a sufficient reason for waiving their right, or that they must actually (or be able in principle to) reflectively endorse the waiver, or that the waiver must be “autonomous” in some other sense.Footnote ⁴¹ None of these are formal constraints of the kind that Ripstein endorses, and yet they would significantly reduce the circumstances in which a person could successfully consent to being killed by another. That is, even if consensual killing is in principle possible, the bar to successful consent might still be high. While it is beyond the scope of this paper to specify the conditions more precisely, it strikes me as plausible that even a high bar for valid consent can be met by those whose medical conditions qualify them for MAID.