Professors Mendz and Kissane have written an informative article on what they call “the conditions that can deeply affect agency.”Reference Mendz and Kissane1 These are conditions such as depression, demoralization, distress, and family dysfunction. Near the end of the article, these authors summarize what they believe they have reviewed and accomplished: “In this study, three mental conditions and a life situation were reviewed which may compromise personal autonomy through loss of agency owing to factors that could escape the understanding of affected individuals, their carers and treating health professionals.” This statement is an accurate presentation of the content of the article.

This precis may seem odd given the title of the article: ‘Autonomous Agency and Euthanasia or Assisted Suicide.’ The article contains little on agency, autonomy, assisted suicide, or euthanasia. There are brief synopses of the laws on euthanasia or assisted suicide in countries including the Netherlands, Belgium, and Canada, each of which has seen important changes in attitudes and sometimes in law. But there is no attempt to analyse or evaluate these laws or the practices in these countries. The connections they make between autonomous agency and these laws and practices is straightforward: These laws and practices generally assume that an autonomous decision can be made in either first-party decisions by patients or second-party decisions in the case of surrogate decision makers. But Mendz and Kissane argue that this assumption is questionable when underlying mental conditions and life situations compromise autonomy, including very subtle influences of this soret. Here are the final two sentences in their article:

This statement needs textual interpretation. The authors seem to mean that in biomedical ethics and related areas of law, we have erected various principles and rules, such as those of informed consent, that are designed to respect and protect autonomous choice. These laws assume a specific capacity to process information and make autonomous decisions. If a physician assists a patient in dying without establishing his or her decisional capacity beyond a reasonable doubt, one would have failed to respect the person’s autonomy, which entails true agency and not merely the appearance of it.

It is unclear to me that this line of argument and conclusion move beyond what is already well understood and established in contemporary biomedical ethics and in the law of countries that allow physician assistance in dying. Mendz and Kissane would be making a new contribution to current literature only if they provide a relevant account of “true agency,” but their article lacks such an account. They do not have a close analysis of this capacity or of a related capacity such as autonomy. For such an account these authors seem to rely heavily on the literature in their references. In particular, for their accounts of agency, autonomy, and capacity assessment, Mendz and Kissane seem to depend on the writings of authors such as Alfred Mele, Sarah Buss, Jason Karlawish, and Paul Appelbaum — all excellent writers with excellent accounts of these basic notions. To their credit, Mendz and Kissane do appropriately cite these sources in their references.

I find questionable some of the ways these authors try to connect psychological concepts such as the desire to be self-governing (the desire to perform autonomous actions) with euthanasia and assisted dying. They say that this desire is “ordinarily absent from persons who request euthanasia. They find themselves in a contingency in which they regard death as their only escape; this is autonomy-undermining reasoning.” Whether this desire is “ordinarily absent” is an empirical hypothesis that I suspect has never been adequately tested, but the real problem is that there is no reason to think that persons are engaged in “autonomy- undermining reasoning” merely because they request help in dying when they judge it to be their only escape. To me this is like saying that people committed autonomy-undermining reasoning when they jumped from the burning trade center towers in New York City on 9-11. They had very good reasons to jump, as do many patients who request some form of help in dying.

Mendz and Kissane seem to make a mistake I have frequently seen in discussions of the justification for complying with requests for assistance in dying. It is assumed that since some patients are psychologically affected in some way by underlying conditions, any patient similarly situated must be psychologically motivated or at least causally influenced. There are many good reasons to request assistance in dying, and there is no reason to presume that anyone who wishes to die does not have good reasons or does not have the necessary capacities to make an autonomous choice. Good reasons are found in many last-resort situations. These can often be avoided by better social or medical policies and practices, including improved palliative care, but not all cases can be resolved in this way. Control of pain and suffering is a moral imperative, but even great progress in control of pain and suffering will not prevent last-resort situations in which individuals understandably seek to control their dying as they see fit. The good doctor may be the one who benevolently helps the patient in these situations, not the doctor who refuses to help while lacking adequate evidence of incapacity. The fact that the autonomous requests of patients for aid-in-dying should be respected in some circumstances does not entail that all cases of physician- assisted death at the patient’s request are justifiable. That’s a bigger problem for another occasion.