Responsibility and collaboration

Philosophy and Public Affairs

Volumn 28, Issue 3, 1999, Pages 169-204

  Kamm, F M ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33750011288     PISSN: 00483915     EISSN: 10884963     Source Type: Journal    
DOI: 10.1111/j.1088-4963.1999.00169.x     Document Type: Article

References (50)

1. From "A Critique of Utilitarianism," in Utilitarianism: For and Against, by J.J.C. Smart and Bernard Williams (Cambridge: Cambridge University Press, 1973), p. 98.
2. A rule-consequentialist may answer differently. Even an act-consequentialist may answer differently if he believes the consequences of killing in this case would be to encourage other evil people to make innocent bystanders into killers by threatening to kill a greater number (when they would not otherwise threaten). I shall assume that the case can be imagined so that it will not have any further negative consequences, and is an isolated and soon-forgotten episode.
3. There are really two other agents in this case: the Captain and Pedro. I will treat them as one, for my purposes, imagining that the person who makes the offer to Jim and the person who would actually kill the Indians are the same, namely Captain. Alternatively, we could imagine that Pedro willingly follows the Captain - Williams does not say Pedro is urging Jim to shoot so he need not - and Captain would himself shoot if Pedro did not. Under the Nuremberg Doctrine, a soldier like Pedro has a duty not to carry out the orders of someone like Captain. Does this raise a problem for Captain's responsibility for murder if Pedro shoots? Can Captain say that he cannot be held to account for a deed that would not have occurred had the subordinate been doing his duty? It seems clear that the wrong act of ordering a killing can indeed make someone responsible for a killing, though another prohibited act is necessary for the killing to occur. However, since Pedro's killing of the twenty could also have been prevented if he had not shot them, Pedro also has positive responsibility for their deaths. Who has the heavier burden of responsibility? Is it the last person whose omission could have prevented the killing or the person who failed to interfere with that last person who, it was known, would not omit to kill? Presumably, a consequentialist would decide this question by considering the consequences of each assignation of responsibility. For example, if we hold most responsible the last person who could have altered the outcome (by omitting to act), this may encourage people to leave to the last moment the prevention of bad consequences. If we hold most responsible someone further up the line, we encourage nipping disasters in the bud.
4. Notice that (ii)(a) requires that some of the Indians will be better off. Without this condition, (i)(a) plus Pareto optimality would permit killing in the following case: Captain asks Jim to kill all twenty Indians, or else Captain will do it. Only if Jim kills all twenty will this save an innocent bystander from dying of natural causes. This is a case in which Jim's killing improves a situation and makes no one worse off (it is Pareto optimal). (I owe this case to Michael Otsuka.) But we may doubt that Jim's shooting is permissible in this case. This is because it does not rescue any of those who face one threat, but instead uses them to save someone from a different threat.
5. For my analysis of the case in which the Bystander must be killed, see Morality, Mortality, Vol. 2 (New York: Oxford University Press, 1993), chapters 9 and 10, and "Toward the Essence of Nonconsequentialism" (forthcoming). For my analysis of the Trolley Problem, see chapters 6 and 7, and "Toward the Essence of Nonconsequentialism." I am aware that if participants imagine themselves sufficiently ex ante to the situation they are in, it may have been in the interest of everyone, even the person who turns out to be the by-stander in the Bystander Case, to agree to the bystander's being shot. This is because there was a time when no one knew who the bystander would be and each then had a greater chance to live if the bystander is shot. I will not discuss this issue here, but do so in chapter 11 of
6. Morality, Mortality, Vol. 2. The necessity doctrine in the law, which allows someone to produce a lesser evil to avoid a greater evil, unfortunately, does not carefully distinguish among these various cases.
7. See Glanville Williams, Criminal Law: The General Part (London: Stevens Sons, Limited, 1961).
8. We could change the case so that Captain says that if Jim kills one particular Indian, Gonzalez, Captain will not kill the others. Now it will not be ex ante in the interest of each that someone be killed. Does this change the permissibility of Jim's acting when Captain makes his offer? I do not believe so, even though it is morally significant that Gonzalez's death is now merely a means of saving the others and was not the upshot of any scheme that was ex ante in Gonzalez's own interest. The Gonzalez Case is based on one that Julia Driver presented in her lecture "Negative Responsibility," given at City University Graduate Center, May 3, 1996; it was reflection on that case that led me to write this paper. In Driver's case, Mary is the mother of five children. If Joe kills her, he will also kill her five children. We can stop Joe by killing Mary in particular ourself. There is, however, an important difference between Mary and anyone Jim would kill. We might assume that a mother would want to sacrifice her life to save her children, especially when she would die anyway. This could rightly play a role in our deciding whether to kill her. But the desire to sacrifice one's life to save others is not the same as the desire to take a chance of being the one who will be shot to save others because this reduces one's own chance of dying. This difference in motivation might reasonably play a role in determining whether we are willing to kill Gonzalez. This is because, in Mary's case, at the time we kill, we kill someone who wants to die to save others, but in the Gonzalez Case, at the time we kill, we do not kill someone who wants to die to save others.
9. I owe this example to Leo Katz.
10. I refer to them as the Captain Cases, not Jim and the Indians Cases, since I wish to focus on the role of the Captain's offer.
11. Those who think that collaborating with evil makes acting in W worse than acting in C1 will think that Williams picks an unfairly hard case, which is not typical of consequentialism, as C1 is.
12. To put in relief the cases I have presented, now consider cases in which life and death are not at issue, but the distinctions between C3 and C1 to which I have pointed is also in play. Suppose Albert is about to deliberately tell a story that will embarrass twenty people. However, he makes Jim an offer: if Jim will tell a story that embarrasses only one of the people, Albert will desist (A1). In an alternative scenario (A2), Jim knows with certainty that he can stop Albert only if he tells a story that will embarrass one of the people, though he must act alone without an offer from Albert. My sense is that in A1, Albert will have positive responsibility for Jim's act and that acting in A1 is morally favored over acting in A2. Ned Block suggested that the use of non-life-and-death cases might better highlight the factors I believe are at work in the life-and-death cases, since some might think life and death would swamp them. Richard Arneson, Liam Murphy, and Sigrun Svavarsdottir helped me construct the cases.
13. By "completely," I mean he has all the moral responsibility and accountability; he alone has it. I also mean that he has full moral responsibility and accountability, i.e., he has it to the highest degree. Two people could share full responsibility, but then neither would have complete moral responsibility. The assignment of all positive moral responsibility and accountability for negative consequences to Captain assumes that Jim chooses a way to kill the one Indian that has no unnecessary negative effects. What if Captain's specific offer is to desist if Jim shoots three of the twenty, but Jim is aware (through his brain scanner) that the captain will also desist if he kills either (a) two of the twenty, or (b) just one bystander who would die soon anyway. Given that it would be permissible for Jim to kill the three if none of the other alternatives were available, Jim may now carry out the best possible of the acts whose other conditions favor permissibility, knowing that positive responsibility for them will also be Captain's. This is so, even if responsibility for these acts would not have been the Captain's without his offer, and his offer did not specify (a) or (b).
14. In distinguishing between causal responsibility, being to blame, and liable for further responses to the act, I am relying on distinctions drawn by Joel Feinberg in "Action and Responsibility," and "Sua Culpa," in his Doing and Deserving (Princeton: Princeton University Press, 1970). He also says of the person who is to blame that the act can be charged to him or placed as an entry to his record (reputation) (p. 128). I say it can be placed at his doorstep.
15. An analysis of collaboration might distinguish between accomplices and perpetrators, distinguishing them by the fact that the perpetrator is the one causally responsible. I am disagreeing with this analysis: I am suggesting that the perpetrator is not the one causally responsible.
16. And so understandably crucial in giving rise to positive responsibility, as Jerome Schneewind emphasized to me.
17. Cannot Captain also become responsible for the negative consequences of Jim's act by deliberately putting much information in Jim's way about the connection between his act and Captain's possible behavior? What if Jim is even told about the contingency point blank by the Captain without the latter making an offer? Perhaps positive responsibility for Jim's act is located in part in Captain in these cases, relative to ones where Captain does not do anything to bring about Jim's act. But it does not shift completely, as when Jim accepts an offer. When Jim acts without Captain's claim to be committed - a commitment that cannot be made by these inducements to act - he is not like Captain's Agent and, as we shall see, this is important.
18. As Christopher Grau pointed out to me, someone who does harm that he could have avoided entirely, but does less rather than more simply in order to do less, may deserve credit for the good consequences of fewer harmed. If we take Williams's description of Captain non-ironically, Captain's motivation is best described, I think, as follows. He seems to be the sort of person who thinks peasants can be treated as toys. It is really a way of honoring a guest, to let him shoot one of them. The guest's involvement in the sport is such a wonderful occasion that some of the toys can be given a gift (of life).
19. For example, Thomas E. Hill, Jr., in "Moral Purity and the Lesser Evil," reprinted in his Autonomy and Self-Respect (Cambridge: Cambridge University Press, 1991).
20. If Jim did the right act in C1, there is no fault in the act, so how could it be his fault that the death occurred? (By contrast, the Captain's behavior is full of fault in virtue of which a death occurs.) But the fact that Jim's act is right does not exclude its having bad consequences for which he might be morally accountable. It also does not exclude its being faulty in one aspect, i.e., if it wrongs someone in the course of doing the right thing.
21. It is important to emphasize that this apportioning of responsibility for the consequences does not depend on the view that there is a fixed amount of responsibility for any act and that fixed amount can either be had by one person or had in diminished amounts by different parties. Many people may each be maximally responsible for negative consequences of an act (or for the occurrence of an event), for example, if many villains participate in killing someone.
22. Possibly, whether Captain has sole positive responsibility for negative consequences may also depend on Jim's reasons for shooting; if they have nothing to do with saving the greater number from Captain's threat, positive responsibility may not lie totally with Captain. For example, suppose Jim would not have killed because of Captain's offer if he had not also always had a desire to kill an Indian. Since Captain's offer provides the opportunity for Jim's acting, responsibility will still be Captain's in part. I owe this point to Claire Finkelstein.
23. I owe this point to Michael Otsuka.
24. We already discussed one case where Jim kills an innocent bystander, above. Another one arising from natural disaster, with no offer from potential victims, will be discussed in Section VII.
25. Samuel Scheffler raised the issue of whether Jim has a positive reason for action in Captain's responsibility. Similarly, we should not intend that a freak of nature not interfere with Captain's shooting the twenty, in order that he then deserve punishment for murder. This is so even though he deserves to deserve it.
26. For further discussion of the distinction between doing something in order that x come about and doing something because x will come about, see Morality, Mortality, Vol. 2, pp. 176-78,
27. my "Toward the Essence of Nonconsequentialism," and "The Doctrine of Triple Effect and Why a Rational Agent Need Not Intend the Means to His End" (forth-coming in Proceedings of the Aristotelian Society, June 2000).
28. Recall that in C5, when we interfere with the consequences of an act Captain has already done, there is no offer necessary for his having moral complete responsibility. It is only when we reasonably believe that Captain will do something, and act to forestall it, that the offer is necessary.
29. Of course, a reformed criminal would also prefer that the twenty not have been killed. This means he prefers that Jim have acted in C1 and have moral responsibility for the negative consequences of the act. For even if Jim acted wrongly in killing in C1, his wrong would not be as great as the one Captain would have at his doorstep if he were morally responsible for killing one Indian or twenty Indians when it was in his power that no one be killed. It is understandable for even a reformed Captain to think it this way. But it is not appropriate for Jim to prefer his acting in C1 to acting on an offer just because his wrong in C1 would be less than the wrong Captain becomes responsible for if he makes an offer. After all, only one Indian would get killed in either case and the Captain deserves to have moral responsibility if anyone does.
30. At this point, it is useful to consider a possible relation between American law and the analysis I have provided. In their discussion of the Model Penal Code on Duress (2.09), Kadish and Schulhofer (in The Criminal Law and Its Processes [Little Brown & Co., 1995], pp. 903-7) note (but do not approve of the fact) that the Code permits the necessity defense to both natural and manmade threats, but permits duress as an excuse only when there is a "do-it-or-else" threat, not a nonperson-caused threat. They quote (in
31. Kadish and Schulhofer, Criminal Law, p. 906) the Model Penal Code's final commentary (Model Penal Code and Commentaries, Comment to 2.09 [1988], pp. 378-79): ". . . there is a significant difference between the situations in which an actor makes the choice of an equal or greater evil under the threat of unlawful human force and when he does so because of a natural event. In the former situation, the basic interests of the law may be satisfied by prosecution of the agent of unlawful force; in the latter circumstance, if the actor is excused, no one is subject to the law's application." I have suggested that it is the desirability of a villain having responsibility for murder which lies behind giving weight to an offer. There is a bias against permitting action when this will not happen. The Model Penal Code commentary may seem to be expressing a similar view. However, as we have seen, I also wish to claim that the interest in having a villain be responsible for murder need not imply that when there is no villain, there is also a bias against acting.
32. He says "a Gresham's Law operates by which the bad acts of bad men elicit from better men acts which, in better circumstances, would also be bad." Morality: An Introduction to Ethics (New York: Harper and Row, 1972), p. 104.
33. Samuel Scheffler suggested this.
34. I believe Williams means another agent's subsequent act. If so, would this imply that he thinks Jim does have negative responsibility for the deaths of twenty if he does not kill one in C5? But presumably he should not think this. On some of these issues, see Ann Davis, in "The Priority of Avoiding Harm," in B. Steinbock, ed., Killing and Letting Die (Englewood Cliffs, NJ: Prentice-Hall, 1980), pp. 173-214.
35. From "A Critique of Utilitarianism," pp. 97-98. I will put to one side two interesting issues raised by this case: (i) why there is a strain on the children if George's wife goes to work when George could take care of them; (2) whether and when it is morally wrong to go after a job because one knows one will do it less well than someone else, i.e., with the intention of, at least in part, sabotaging the goals of one's firm.
36. I discuss a hypothetical case here because I am not sure there really was anyone like this, although I believe Albert Speer (Hitler's architect) may have conceived of himself in this way.
37. I am grateful to Susan Wolf for this information.
38. I owe this point to Richard Arneson.
39. I have discussed these issues in Morality, Mortality, Vol. 1 (New York: Oxford University Press, 1993).
40. For more on this, see Morality, Mortality, Vol. 2, and "Toward the Essence of Non-consequentialism."
41. I have dealt with this issue in detail in Morality, Mortality, Vol. 2, and more recently in "Toward the Essence of Nonconsequentialism."
42. A mistake to avoid in considering this issue is treating the paperwork representation of what a Selector would do as governed by the same moral principles as apply to actual treatment of people. Consider three possible types of paperwork, where "A," "B," and "C," are letters representing three different people, (1) "A" and "C" are on a written list and then someone writes in "B" ahead of them. (2) "A" and "C" are at the top of a list, with "B" next. Someone crosses the first two out, and so "B" is at the top of the list. (3) "A" and "C" are alone on a list; someone crosses them out and puts in "B." There need not be any moral difference in what one may do based on which list one starts with and what one does to it.
43. This is hypothetical. I have been told that the Nazis did not use Jews to kill other Jews. I thank Doris Bergan for information about the Judenrat.
44. Here I focus only on problems raised by collaboration. I have discussed other aspects of this issue elsewhere. See my "A Right to Choose Death?" Boston Review 22 (Summer 1997): 20-23;
45. "Physician-Assisted Suicide, the Doctrine of Double Effect, and the Ground of Value," Ethics (April 1999);
46. and "Physician-Assisted Suicide, Euthanasia, and Intending Death," in M. Battin, R. Rhodes, and A. Silvers, eds., Physician Assisted Suicide: Expanding the Debate (New York: Routledge, 1998), pp. 28-62.
47. On this, see Margaret Battin's "Euthanasia: The Fundamental Issues," in her The Least Worst Death (New York: Oxford University Press, 1993).
48. Brock, "Voluntary Active Euthanasia," reprinted in his Life and Death (New York: Cambridge University Press, 1993), p. 211.
49. Brock, "Voluntary Active Euthanasia," p. 230.
50. Though it may show it is not a form of conduct that is completely beyond the pale, since then one would not even be permitted to have an Agent do it.