Entretien avec Jeff McMahan sur Jonathan Glover et l’éthique du faire-mourir

Dans le cadre de ce numéro spécial consacré à Jonathan Glover et son éthique appliquée, nous avons demandé à Jeff McMahan de revenir sur l’influence qu’exerça sur lui l’ouvrage Questions de vie ou de mort (Labor et fides, 2017), publié il y a quarante ans dans sa version originale. Jeff McMahan, qui fut l’étudiant de Glover, a depuis lors développé sa propre éthique du faire-mourir. Nous avons voulu savoir ce qu’il retenait de la philosophie de Glover, dont il reconnaît le caractère pionnier en éthique appliquée. In this special issue on Jonathan Glover and his applied ethics, Jeff McMahan was asked to review the influence of Questions of Life and Death (Labor et fides, 2017), published forty years ago in its original version. Jeff McMahan, Glover’s former student, has since developed his own ethics of killing. We wanted to know what he had learned from Glover’s philosophy, which he recognized as a pioneer in applied ethics.

Les éditeurs suivront les recommandations et les procédures décrites dans le Code of Conduct and Best Practice Guidelines for Journal Editors de COPE. Plus précisément, ils travaillent pour s'assurer des plus hautes normes éthiques de la publication, y compris l'identification et la gestion des conflits d'intérêts (pour les éditeurs et pour les auteurs), la juste évaluation des manuscrits et la publication de manuscrits qui répondent aux normes d'excellence de la revue.
The editors follow the recommendations and procedures outlined in the COPE Code of Conduct and Best Practice Guidelines for Journal Editors. Specifically, the editors will work to ensure the highest ethical standards of publication, including: the identification and management of conflicts of interest (for editors and for authors), the fair evaluation of manuscripts, and the publication of manuscripts that meet the journal's standards of excellence.

Références Annexe A An Interview with Jeff McMahan on Jonathan Glover and the Ethics of Killing
In this special issue devoted to Jonathan Glover, it was important to highlight his influence on some of the later philosophers. By giving an "applied" character to moral philosophy, in other words by taking a position on concrete subjects (such as abortion, euthanasia or the death penalty) where previous philosophers had tended to limit their analyses to semantic questions, Jonathan Glover undoubtedly helped to pave the way for a path that others have not failed to follow. Among them is a very prominent philosopher today who does not hide his immense debt to Glover, namely Jeff McMahan, the White's Professor of Moral Philosophy at Oxford University. The aim here was to review with Jeff McMahan the decisive role he himself gives to Questions of Life and Death in his decision to devote himself to questions of applied ethics. Moreover, the two philosophers have in common that they confront the fundamental question "how is it wrong to kill?". So it was interesting to compare their answers and to reflect with Jeff McMahan on the reasons why Glover's theses do not fully satisfy him.

Jeff McMahan, thank you very much for accepting this interview on the occasion of the French release of Jonathan Glover's seminal book Causing Death and Saving Lives. You already had the opportunity to say that your decision to become a philosopher yourself was largely due to Jonathan Glover. You even wrote that you thought of your book The Ethics of Killing as "Son of Causing Death and Saving Lives" [1, p.71]. Could you tell us why? What was so appealing to you in the way Glover practiced philosophy?
Between 1976 and 1978 I studied philosophy and politics at Oxford at the undergraduate level. My main interests were moral and political philosophy but I found much of what I was reading even in those areas rather arid. But within a year of its being published, I read Jonathan's Causing Death and Saving Lives and found in it exactly what I had been looking forthough I did not entirely understand what that was until I read the book. In it, Jonathan reasons about moral and political issues with the scrupulous care and objectivity demanded by matters of life and death, which is literally what the issues he addresses are. Nothing, it seemed to me, could be more important than that. And as it has turned out, those same issueswar, abortion, euthanasia, causing people to exist, saving those at risk from natural or human threats, the distinctions between doing harm and allowing harm to occur and between harming as a means and harming as a side effect, and so onhave dominated the work that I have done throughout my career. To Jonathan it was wholly alien to see these moral and political problems as challenging puzzles on which one could exercise one's cleverness and ingenuity; yet they are nevertheless highly demanding intellectually. The combination of passionate engagement with moral issues and dispassionate analysis of arguments that characterizes Jonathan's work has always been the model that I have sought, with varying degrees of success, to emulate.
My next question is about the method to be followed in moral philosophy. In your books, you seem to seek principles capable of capturing most of our intuitional responses, at least those "that are deeply and pervasively held to be presumptively reliable" [2, p.104]. But sometimes you also proceed through adjustment and modification of both principles and intuitions until consistency and harmony are achieved. 7

Thus, one may have the impression that your method is quite similar to what Glover called an "interplay between responses and beliefs" [3, p.26-28] and to what
John Rawls had described before as the method of "reflective equilibrium" [4]. Is that correct? Do you think it is the best alternative to the strict "intuitive approach" and the pure "theoretical approach"?
You describe the method I follow very well. Our moral intuitions are just the ways in which matters seem to us morally. I see no reasonable alternative to beginning our moral inquiries by consulting our intuitions about general problems, particular cases, and moral principles and then seeking to achieve consistency, coherence, and maximum plausibility among these intuitions. Because everyone's moral beliefs are to some extent inconsistent and confused, serious moral thinking inevitably involves revising some intuitive beliefs, abandoning others, and progressively refining our moral principles. In the ideal condition that Rawls calls "reflective equilibrium," moral principles unify, illuminate, and provide deeper rationales for our intuitive judgments about particular problems and cases. Consistency and coherence are not, however, merely aesthetic features of our beliefs in reflective equilibrium; rather, they are necessary if our beliefs are to be true.

Do you think there might be cases in which people's intuitions conflict for real? For example, one of the main theses of your book Killing in War, namely the claim that we should not consider all combatants as equal in a war (because some of them are "liable to be attacked", whereas others fighting for a just cause are not) is far from being an obvious intuition for many people. You even had to admit that the opposite thesis (the equality of combatants) "seems to be intuitively compelling to most people" [5, p.35]. So how do you solve this kind of conflict? How do you know when an intuition should be rejected?
Intuitions that are both common and strong can nevertheless be mistaken. In the American south during much of the 18 th and 19 th centuries, for example, most white people believed that human slavery was morally justified. Both in the past and at present, most people have believed that it is morally permissible to cause animals to suffer and then to kill them as means of gratifying people's taste for meat. That these intuitive beliefs are mistaken is suggested by various considerations. One is that they are inconsistent both with other even more compelling moral beliefs and with plausible general principles. To try to reconcile these beliefs with other moral beliefs and with general principles, one is forced to appeal to implausible claims about the moral significance of race or species. Another relevant consideration is that these intuitions can be explained in part by the interests of those who have them. The belief that it is permissible to own slaves enabled slave owners and others who shared in their culture to persist in a practice that was highly beneficial to them. Each was supported in his beliefs by the agreement of others whose interests coincided with his own. The intuitions of many slave owners and meat eaters have also been shaped by theological doctrines of ancient origin that are very unlikely to be true. All of these factors, along with others, tend to undermine the credibility of the intuitions.
The explanation is, however, quite different in the case of the common intuition about war that you citenamely, that combatants on both sides in a war are morally permitted to fight. As you note, I think this intuition is mistaken as well. But I think the reason why people have this intuition is that, at least at present, the general consequences of people's having this belief are likely to be better than the consequences of people's having and acting on the correct moral belief, which is that it is morally impermissible to fight and kill in pursuit of aims that are unjust. People generally, and soldiers in particular, tend to believe that their own country's wars are just. Because of this, if soldiers believed that it was morally wrong to fight in unjust wars, soldiers on both sides would tend to believe that their enemies were not just soldiers like themselves but evildoers who might deserve to be killed in war and to be punished in the aftermath of war. This could make wars even more ferocious than they already are. Also drawing a sharp, simple line between combatants as legitimate targets and non-combatants as morally immune to attack helps to minimize the destructiveness of war. These are among the reasons why the law holds that combatants on both sides are legally permitted to fight. And at least at present they remain good reasons. Yet, people are inclined to think that what the law permits in an activity as serious as war must also be morally permissible. That is a natural inference but I think it is mistaken.

Just like Glover, you faced the fundamental question "What is wrong with killing?". Glover gave both direct and indirect reasons. The first kind of reasons appealed to the notions of worth-while life ("It is wrong to reduce the length of a worth-while life" [3, p.113]) and autonomy ("it is wrong to kill someone who wants to go on living, even if there is reason to think this desire not in his own interests"). Then he mentioned other reasons related to the wrongful side effects of killing. What do you think of this account for the wrongness of killing?
Although acts of killing are usually bad not only for the person killed but also for others, the bad side effects are not the main reason why killing people is normally wrong. Yet, the second of Jonathan's claimsthat killing is wrong because it deprives the victim of further life that would be worth livingconflicts with the common intuition that intentional killings of innocent people do not differ significantly in the degree to which they are wrong. Most people, for example, do not think that the intentional killing of an innocent 50-year-old is substantially less seriously wrong than the killing of an innocent 30-year-old. Yet, assuming that each victim would otherwise have lived to roughly the same age, the killing of the 30-year-old inflicts a far greater loss of life worth living and should therefore be substantially more seriously wrong. Jonathan's further claim that killing involves a violation of the victim's autonomy may help with this problem, but raises other problems of its own. Suppose, for example, that a person whose subsequent life would or could be worth living is depressed and wishes to die. Would her murder by a stranger be less seriously wrong because it would not violate her autonomy, since her will is not opposed to her being killed? It does not help to respond by saying that her depression renders her non-autonomous, for that still leaves it true that the murder involves no violation of autonomy, so that one of the reasons why Jonathan thinks killing is normally wrong does not apply in this case. If the harm-based reason and the autonomy-based reason are additivethat is, if these reasons combine to form an overall stronger reasonthen the killing of the depressed person must be less wrong than the killing of another person with comparable prospects for the future who does wish to continue to live. And that seems implausible. Moreover, if the reasons are additive, the problem of unequal wrongness remains even in cases in which autonomy is violated, for the autonomy-based reason simply strengthens to an equal degree the harm-based reasons that are of unequal strength. But if the harm-based and autonomy-based reasons are not additive, it is unclear what Jonathan's view implies about cases in which one reason applies but the other does not. Suppose, for example, that one wrongdoer is about to kill a newborn infant while another is about to kill a 95-year-old who, if not killed, would inevitably die tomorrow. The first killing would not violate autonomy but would cause a great loss of life worth living. The second would violate a person's autonomy but would deprive the victim of only a very brief period of good life. Which killing would be more seriously wrong? Which killing ought one to prevent if one could prevent only one? That it seems to have no determinate implications for these questions is not a decisive objection to Jonathan's view. But it does suggest that Jonathan's account of the wrongness of killing is incomplete.
You defended what you called the "embodied mind account" of personal identity, namely the idea that the criterion of personal identity is "the continued existence and functioning, in non branching form, of enough of the same brain to be capable of generating consciousness or mental activity". Contrary to Glover, you find it crucial to start by an inquiry on personal identity and the basis for egoistic concern before being able to give a correct account of the wrongness of killing. Could you explain why?
Suppose we apply Jonathan's account of the wrongness of killing, which we have both just described, to the practice of abortion. Usually the side effects of abortion are good, as the only people who are affected are the biological parents of the fetus, and normally they both desire the abortion. And the fetus itself is not autonomous. So the only reason for thinking that abortion is wrong, according to Jonathan's account, is that it deprives the fetus of future life that would be worth living. But even that reason does not apply unless the fetus is one and the same individual as the later person whose life would be worth living. And to know whether that is the case, we must understand personal identity, for one cannot understand when we begin to exist without understanding what the conditions of our existing and continuing to exist are. There is also the question of how much good life is lost when a fetus is killed. The obvious answer is that an entire human life, and not just a part of such a life, is lost. This suggests that on Jonathan's view, strictly interpreted, the harm-based reason for not killing a fetus is much stronger than the harm-based reason not to kill an older child or an adult. That seems implausible, however, because the death of a fetus does not seem as serious a misfortune for the fetus as, for example, the death of a young adult is for that person. I think the solution to this problem is to reject the idea that the extent of the misfortune of death is determined by how much good life is lost. Rather, the misfortune of death is determined by both the amount of good life lost and the degree to which an individual at the time of death would have been related in relevant ways to herself in the future. The relevant relations are, in my view, those that are constitutive of personal identitythough unlike personal identity itself, they are matters of degree. The more strongly they would have held within the life, the worse the death is, other things being equal. (Although you correctly quote me as saying that personal identity is a matter of the functional continuity of the brain, that is correlated with certain forms of psychological continuity. It is because a fetus is only very weakly psychologically related to itself in the future that its loss of future life in being killed is a lesser misfortune for it. It is only weakly related in the ways that matter to the life that it loses through death.) So personal identity is important in both these ways to understanding the morality of abortion, which is one of the more puzzling problems of killing.
Glover is known for having challenged the acts and omissions doctrine. He finally considered that there was no decisive reasons to consider killing as morally more objectionable than letting die, when both have the same consequences. You made a major contribution to determine whether an act of withdrawing aid or protection counts as killing or letting die [6]. Does it mean that the distinction between acts and omissions, killing and letting die, is morally important to you?
I do accept that the distinction between doing harm and allowing harm to occur has moral significance in many instances. In accepting this view, I am mainly guided by intuition rather than theory. As far as I can tell, people's intuitions in all cultures in all times have been responsive to this distinction. Suppose that each of two drivers is taking two injured people who are strangers to him to the hospital and must get them there immediately if their lives are to be saved. One driver sees an unconscious woman beside the road who will die unless he stops to save her. But saving the woman will prevent him from delivering the two injured people to the hospital in time for them to be saved. He drives on and allows the woman to die. This seems permissible to almost everyone. It is permissible for him to save the two strangers rather than allow them to die as a side effect of saving one other stranger. The second driver sees an unconscious woman lying in the road ahead and cannot maneuver around her. To avoid running over her, he must stop the car and drag her out of the road. But that would prevent him from delivering the two injured people to the hospital in time. He therefore runs over and kills the unconscious woman. Relatively few people think that this is permissible. People's intuitions about these cases do not seem to derive from a distorting or discrediting source, such as self-interest or some implausible theological doctrine. Support for the general significance of the distinction between doing and allowing does not, moreover, come solely from intuitions about cases. It can also be found in considerations of responsibility. In my examples, the woman beside the road would have died even if the driver had not been present; but the woman in the road would (we may suppose) have been safe had the driver not run her over. But even though I think the distinction between doing and allowing has moral significance, I also think there can be reasonable disagreement about how much it matters. It may well not matter as much as common sense intuition supposes.
It seems like you never really expressed at length your ideas on the death penalty. In the Preface of The Ethics of Killing, you informed your readers that capital punishment would be examined later in a forthcoming volume. However you suggested that there were "cases in which an individual has done something that has lowered the moral barriers to harming him, or compromised his status as inviolable, or made him liable to action that might result in his death" [2, p.VII]. Were you referring only to self-defence and fighting in war? Or did you mean that, at least on a theoretical level, a man guilty of murder is liable to be punished by death?
I was referring mainly to defensive action both at the individual level and in war but the comment does also apply to capital punishment. This is because I think the only plausible justifications for punishment are ones grounded in principles of defense or compensation (or reparation). It is possible in principle that executing someone who is morally responsible for having committed a crime could be a necessary and proportionate means of preventing his criminal action from causing serious harm to others in the future. This might be the case if, for example, the failure of the law to fulfill its threat to execute this person would diminish the credibility of its deterrent threats in other cases, thereby decreasing the security of innocent people generally. If this were true, the criminal might be morally liable to be executed, not in retribution for the harm he has done in the past, but as a means of preventing his past action from leading to the deaths of other innocent people in the future. The justification would, in essence, be defensive. But I think, as I will explain below, that this form of justification seldom applies in practice.
You also gave us a clue on the way you consider the death penalty in an article about torture. You claimed right from the start that "the morality of torture is similar to the morality of capital punishment. There are in principle, and probably in practice, certain rare instances in which either would be morally justified. At the same time, morality itself demands that both be categorically banned in law" [7, p.241]. Does it mean that you agree with Glover that the death penalty should be abolished mainly for consequentialist reasons, and not from an absolutist point of view?
Yes, as both my last remark and the passage you quote suggest, I agree with Jonathan that the death penalty should be abolished, though primarily for reasons concerned with the consequences of having a practice of execution and not because I think that execution can never be justified in principle. Cases in which executing a person might be necessary to prevent his past criminal action from threatening innocent people in the future are rare. But in jurisdictions that practice capital punishment, executions of innocent people are not rare. And most executions inflict terrible harms as a side effect on innocent people who are specially related to the executed person, such as his children, parents, and partner. Of course, lengthy prison sentences also cause harm as a side effect to those who love or depend on the person imprisoned. But these harms are normally less severe. So in practice the wrongful harms that allowing capital punishment is likely to cause greatly outweigh the harms that executions might prevent. This means that the death penalty is overall disproportionate as a mode of punishment. It therefore ought to be constitutionally prohibited. This is compatible with its being morally justified on rare occasions. But because the law cannot reliably distinguish those cases from the much more common cases in which it is not justified, the law must prohibit it in all cases.
Since this special issue is to be regarded as a tribute to Jonathan Glover, I wish we could end this interview on a more personal note. Obviously Jonathan was not only your professor. He has been your friend for a long time now...
Not only was Jonathan my inspiration in philosophy, but he and Vivette were Sally's and my inspiration as parents as well. The Glover children -Daniel, David, and Ruthalways sat at the table with us and were encouraged to participate in the conversation, though no encouragement was ever needed by that point. Jonathan is known as a utilitarian, though not as one who is altogether doctrinally orthodox. Although most of us who have studied philosophy were taught early on to beware of utilitarians, who would lie to us, break their promises, betray us, and even extract our vital organs for transplantation if by doing so they could promote the greatest good, it has been striking to me that the professed utilitarians of my acquaintance in general do far better by the standards of ordinary common sense morality than their opponents in ethics theory who go on about human dignity, integrity, humanity, fidelity, honor, and so on. Here I can cite Jonathan Glover as Exhibit A: no one could hope for a kinder, more sympathetic, reliable, or generous friend and mentor than he.