Restricted options necessarily undermine autonomy

The first type of argument against allowing MAiD for persons who are chronically but not terminally ill states that if oppression restricts one’s options, autonomy has necessarily been reduced, although it may not be entirely undermined. While each proponent of this view has a slightly different story to tell, they all make having full-fledged autonomy dependent on a socially ideal state of affairs, where the agent has meaningful access to a wide enough range of options.19–23 For example, Mackenzie argues that self-determination is one of three dimensions of autonomy, and self-determination requires freedom from oppression and substantive social equality.24 On this sort of view, any choice to have MAiD in unjust circumstances would not be substantially autonomous because the agent does not have substantive social equality. The prima facie consequence of this view is that if a person is not autonomous, then at most, respect for their decision is no longer warranted, and at the very least, the status of the person’s choice is called into question.

These types of views are criticised by Khader, who argues that feminist theories of relational autonomy that claim that autonomy is necessarily reduced by restricted opportunities for self-determination can be too easily taken to warrant reduced respect for the actions and choices of oppressed agents.3 We agree with Khader that restricted opportunities will often be insufficient to reduce autonomy. Of course, oppression can sometimes reduce autonomy, for instance, by subjecting one to coercion, drastically reducing hope, substantially constricting what one can imagine as a possibility for action, or limiting one’s access to the information required to make an informed decision. When an agent’s options are restricted due to oppression, this should certainly be condemned. However, the fact that someone’s options have been restricted is insufficient to reduce autonomy, because oppressed persons can and often do retain capacities and self-regarding attitudes necessary for autonomy, such as the ability to understand, appreciate the consequences of their choice, the capacity to value, the ability to reflect on the values guiding their decision, as well as the attitudes of engaged hope, self-worth and self-trust.

We agree with Mackenzie that access to substantive equality is important but disagree that such conditions are or should be understood to be requirements for autonomy. One reason is that such views provide the conceptual framework to justify ‘promiscuous paternalism’.3 This is true despite the fact that some individual proponents of these views hesitate to conclude that paternalistic intervention is the correct course of action, even though they maintain that autonomy has been compromised significantly.

Another problem with this view is that it would disqualify many ordinary choosers from the realm of autonomy. To this end, it is important that an approach to autonomy be practically applicable in bioethical contexts. It is not feasible to expect medical professionals to assess the extent to which the person in their care has had their options restricted because of oppression and neither is it directly relevant. Medical professionals are properly positioned to detect a reduction in the capacities and self-regarding attitudes that their patients need to make autonomous decisions. If they were to attempt to assess the impact of oppression on patients who maintain the relevant capacities and self-regarding attitudes, we think they are likely to overreach and behave paternalistically to people solely because they have experienced oppression. Mackenzie herself argues that, although she thinks that having one’s options for self-determination restricted due to oppression directly and necessarily reduces autonomy, in medical contexts patients’ decisions should not be overruled on this basis alone and the ‘most salient dimensions’ of autonomy should be competency and authenticity.24

Decisions to request MAiD by people with chronic conditions will certainly be impacted by the opportunities available to them. In an ableist society which values and is clearly built for people who are physically and mentally healthy, independent and able-bodied, and who live generally pain-free, people with disabilities and those living in chronic pain and suffering are not sufficiently valued. Further, they often experience significant barriers in achieving effective access to long term care—from general support to safe pain management—that would improve the quality of their lives. We agree that these constitute serious injustices. However, to claim that access to ideal social conditions is a requirement for autonomy is to mistake conditions for substantive justice for the conditions required for autonomy.25 Further argument is needed to relevantly connect injustice (in the form of restricted options) to one’s actual autonomy capacities. We consider two such arguments below.

Oppression undermines internal autonomy capacities

While we think it is theoretically unsound to make autonomy contingent on access to ideal social conditions, we recognise that oppressive circumstances can be severe enough that agents despair, and do not take any action in pursuit of their goals. In another piece, arguing that certain self-reflective attitudes like self-worth and self-trust are required for autonomy, Mackenzie presents an example of one such kind of case. Mrs. H has cancer, has had part of a leg amputated, and her husband has left her due to her disability and regards her as an embarrassment.26 Mrs. H, as described by Mackenzie, has lost all sense of self-worth, does not trust herself, and does not want any of the treatment options her medical team proposes. She wants to die. We agree with Mackenzie’s suggestion that Mrs. H as so described is not fully autonomous. However, the judgement that she is not autonomous is grounded in the fact that in this particular case, we have an agent who lacks self-trust, has no self-worth and has evidently lost engaged hope. She is not declared non-autonomous simply because her values have been shaped by sexist norms of traditional femininity, and she has arguably existed with a restricted set of options for some time.

We would, therefore, recommend interventions aimed to increase her sense of hope, such as putting her in contact with people successfully coping with a diagnosis like her own rather than interventions aimed at transforming her values. While Mrs. H’s case is, we think, an excellent example of how oppressive socialisation can contribute to undermining capacities needed for autonomy, particularly insofar as her self-worth is significantly damaged by the combination of her illness and her husband’s attitudes and actions, her case is relevantly unlike patients who are the target of this paper: those who are choosing MAiD in part due to unjust social conditions. Whereas Mrs. H’s refusal of care comes from a radically undermined capacity for self-trust, low self-worth and lack of engaged hope due to neglectful close interpersonal relationships, these persons are choosing MAiD due to a lack of alternatives. In these cases, there is no evidence to suggest that they do not trust themselves, or that they despair completely. Unlike Mrs. H, they demonstrate a considerable level of engaged hope: they are taking an active role in responding to their situation and are motivated to realise a goal. The fact that this goal is arguably grim and reflects a severely unjust social landscape does not detract from the legitimacy of their agency. It is an indictment of the options available, not of the status of their autonomy, given the options they have to work with.

To explore this argument further, we turn to Ho’s argument that the presence of oppressive circumstances renders decision-making in the context of oppression non-autonomous, because of the impact of oppressive norms.27 28 Like Mackenzie, who focuses on the impact of sexist norms on Mrs. H’s autonomy, Ho’s argument is that unjust social circumstances—in this case, ableism—undermine autonomy in end-of-life choices. However, unlike Mackenzie, the cases Ho considers are arguably analogous to those that come under consideration in light of the changes to MAiD brought on by Bill C-7.

Ho analyses several examples of people who chose to die via passive euthanasia, discussing Larry McAfee (1985), Kenneth Berghsted (1969), Dan Crews (2010) and Elizabeth Bouvia (1983–1986). McAfee and Berghsted became quadriplegic after accidents (1985 and 1969, respectively), and requested that they be removed from a respirator and allowed to die. Crews, who was disabled at the age of three, asked to be allowed to die in his early twenties. Bouvia had cerebral palsy, and in her mid-20s (1983), made a request to a hospital in Irvine, California, that she be allowed to die by starving to death.27 We grant that a relevant dissimilarity between these cases and the ones that are the target of this paper is that the former are instances of passive euthanasia, while the latter involve active euthanasia. However, the topic of this analysis is not to debate the ethicality of euthanasia per se, and we do not argue that passive and active forms are equivalent. Instead our focus is on making a choice to die (passively or actively) in the context of injustice, due to a lack of alternatives; in both types of cases, we have just this.

There are several common threads running through the cases analysed by Ho. First, each of the people in these cases were eventually granted the legal right to die. The second similarity, as Ho points out, is that no one was in a state of physical deterioration because of their disability. In each case, what prompted the request to die was arguably non-medical: the inability to afford or access care that sustained the quality of life they had before. And here, we have a case very similar to the 2022 MAiD cases we started with: the request for death is due to one’s social circumstances—importantly, a lack of options—rather than ‘strictly’ medical, such as a change in one’s health state, or a deterioration in one’s executive functioning. A third commonality is that each of the people involved went to considerable lengths to advocate for a solution that they wanted to be able to access. Bouvia’s persistence is particularly notable: she petitioned the hospital in Irvine, California, to be allowed to die. She was admitted but treated against her will, she petitioned the hospital’s decision to a lower court judge, appealed the judge’s decision, and was placed in another care facility where she was force fed again. Her physicians stated that she was ‘clearly intending suicide,’ and that she could live an estimated 15–20 years with current treatment. She sued the hospital and went before the Superior Court of Los Angeles, only to be turned down again. She appealed again, and finally in 1986, the California Court of Appeal granted her the right to decide for herself to refuse treatment.29

The ability of the people discussed by Ho to advocate for themselves and navigate the courts and medical bureaucracies is arguably evidence of engaged hope, self-worth and a stronger commitment to pursue what they want than many ‘ordinary’ choosers do, choosers that we have no problem evaluating as autonomous. There is no evidence of diminished decision-making capacity or autonomy in these cases.

What about the possibility that people with disabilities who are choosing death because of limited options may be doing so due to internalised ableism? That is, perhaps these cases do resemble cases like that of Mrs. H. Perhaps Crews, McAfee, Bergstedt, Bouvia, and those currently electing for MAiD in Canada, are in part doing so because they have internalised an incorrect belief that life with a disability is less worthy than life without one. First, operating on the default assumption that people with disabilities have internalised such a downgraded sense of self-worth such that their testimony regarding their own experience is doubted constitutes another kind of ableism in the form of epistemic injustice. Their testimony is discounted because of their group membership—they are identified as people with disabilities and the fact that they have likely encountered ableism is thought to make them unreliable when it comes to giving the reasons for their decisions. When people living with disabilities state that their quality of life is intolerable to them due to a lack of access to care they ought to have, the ethically sound response is to criticise the social conditions that have downgraded their quality of life, not question their self-reflexive attitudes or how authentic or legitimate their evaluation of their quality of life is. Second, it is just not clear without question begging that it is accurate that persons who grow up with disabilities have, in fact, internalised ableism to such a degree that their autonomy should be questioned or assessed as deficient.

What should we think about the fact, then, that in each case outlined above, and with the cases emerging in the wake of Bill C-7, decisions towards death are being prompted not by development of a clinical condition, but by problems in access to needed healthcare, and more fulsome social and financial support for dependency needs? Is it that they have internalised an ‘ableist ideology’, whereby certain impairments due to disability are ‘more legitimate than other reasons for one to desire death’ as Ho has interpreted these types of decisions?27 This is an understandable conclusion to draw—after all, each of these agents is making decisions in an ableist environment, the lack of long-term, robust and holistic care for disabled people being just further evidence of this.

We propose a different understanding of these kinds of choices. Each of these persons had a disability and decided to pursue assisted death due to a social reason. Does this mean they think their own lives are not worth living because of their disability, while more abled lives are? We answer no. It is reasonable to think that each of these agents viewed their suffering as insurmountable, and unlikely to change. Rather than betraying an ‘ableist bias,’ their decisions can be more charitably and respectfully interpreted as an accurate assessment of their situation. It is one thing to identify an ableist bias in a person or a policy where there is no lived experience with a disability, and quite another to attribute ableism to a person who has intimate experience living with their disability, and to on this basis question the legitimacy of their decisions regarding their own care. Trusting a person with a disability’s subjective reports of their own quality of life is not mutually exclusive with also arguing in support of and advocating for a world where the structural barriers that undermine their quality of life do not exist.

Third, even if investigating the authenticity of their evaluation of their quality of life to detect whether an agent had internalised nefarious bias were an appropriate response, it would be practically impossible to assess this reliably. It is difficult enough for individuals to accurately assess the degree to which internalised oppression might operate in their motivational landscape, and medical teams are not well-positioned to detect the presence of internalised oppression within the values sets of those requesting MAiD. We agree with Khader that this seems like a recipe for unwarranted paternalistic interference.3 While neither Mackenzie nor Ho argue that decisions made by oppressed people should be disregarded for that reason alone, their arguments provide support to those who would argue for that conclusion.

If there are reasonable doubts regarding the autonomy of persons seeking MAiD, assessors should, as with all patients requesting an end to their lives, look for signs of coercion, confusion, a lack of self-trust so severe that it undermines self-governance in the service of what they care about, or a state of despair that entirely erodes hope, such that they do not pursue a course of action at all. Given that MAiD is being requested, the latter seems unlikely.

Oppression leads to hopelessness, which undermines autonomy

Finally, what about the argument that persons choosing within restricted options are hopeless? Unjustly restricted options do not necessarily undermine hope to the point of despair, but of course, we certainly recognise the possibility that such oppressive circumstances could lead some to despair. The case of Mrs. H as articulated by Mackenzie is one such example. However, persons requesting MAiD are arguably acting with engaged hope in seeking what they consider to be the best option of the options available to them.

To act with engaged hope one must have goals and take an active role in responding to one’s situation. There are several things such hope does not require. It does not require the person to act on their own or to ignore the views of others. Most projects, including seeking medical aid in dying, involve support from others and this is no threat to autonomy so long as those others do not withhold crucial information or act coercively. What matters is that the autonomous person’s goals and what they hope for are uncoerced and reflect a basic sense of self-trust. For example, someone who requests MAiD in part to spare their family the experience of their suffering or to ensure they can leave behind resources to support social causes they care about is not lacking autonomy if they do not feel pressured to do so, and if they do not feel undeserving of having what they care about taken seriously.

Finally, and crucially for the purposes of our paper, having engaged hope does not require the person to think they can achieve an ideal outcome in their course of action. Acting with engaged hope only requires that some outcomes are better, and others worse, according to the person’s own lights. For instance, Sathya Darya Kovac would have preferred to continue to live for several more years with her ALS if she were provided with more hours of paid home care to enable her to live at home with a companion animal. However, she chose MAiD as an alternative to either living in a shared residence with round the clock care or living in her own home with insufficient home care.30 Her choice was tragically shaped by social conditions, but there is no evidence that a lack of hope undermined her autonomy and her determined pursuit of access to MAiD is instead evidence that she had a goal that mattered deeply to her and persisted in pursuing it.

Each individual considered in this paper lacked access, not only to better options but also options that they should have had as a matter of justice. Individuals in unjust circumstances can have the capacities and attitudes they need for autonomy undermined by leading them to lose self-worth and hope, or by denying them access to information about their options. However, we think that any assessment of their autonomy should focus on whether there has been a loss of those capacities and attitudes, and not simply on evidence that their options have been restricted due to oppression, or that they live in a society that devalues them and so they might have internalised oppressive attitudes. The presumption that autonomy is necessarily undermined by oppression is problematic and produces a conception of autonomy that is impossible to assess in medical contexts. Moreover, the persons who are the subject of news and legal proceedings, those battling to be able to access an option they desire, are rarely the persons whose autonomy has been compromised in the first place.