1 For a canonical exposition of the theory, see the Pastoral Letter of the US Catholic bishops, The Challenge of Peace (London: CTS/SPCK 1983), 24-30.

2 The category of preventive war includes preemptive war, which is war fought to prevent an imminent threat.

3 There may be cases in which disarming a defeated aggressor would be wrong. The paradigm of such a case would be characterized by three factors: the aggression of which the adversary was guilty is fully or partially excused by mitigating circumstances; the likelihood of further aggression is low; and there is reason to believe that the adversary will have need of its military forces for the purpose of justified self-defense.

4 Can the presence of a sufficient JA ‘activate’ contributing JAs, making it permissible to pursue them, even if one chooses not to pursue the sufficient JA? For example, if state S engages in aggression against one’s enemy, so that there is a sufficient JA for war against S, may one go to war against S, not to defeat its aggression against one’s enemy, but to prevent it from going on to attack one’s ally or one’s own country? We believe so. The occurrence of the offense is alone sufficient to lower certain moral barriers.

5 It may be that, if Iraq’s aggression had not been reversed, this would have had a harmful effect on the world’s oil markets, causing damage to many economies and hence much human suffering. If so, then the prevention of this effect should be included among the goods that the war might have sought to achieve. It is not clear, however, whether this was a separate possible JA or whether it was subsumable within the JA of ending the Iraqi occupation of Kuwait. How serious the threat was is also a matter of speculation. Would Iraq have boosted production or boosted prices and how would other oil-producing nations have responded?

6 Lackey, Douglas ‘Bush’s Abuse of Just War Theory,’ in Cady, Duane L. and Werner, Richard eds., Just War, Nonviolence, and Nuclear Deterrence (Wakefield, NH: Longwood 1991), 278Google Scholar

7 Kavka, Gregory ‘Was the Gulf War a Just War?’ Journal of Social Philosophy 22 (1991), 23-4CrossRefGoogle Scholar

8 Ibid., 24. In another context, Lackey too appears to suggest that P requires a comparison between war and an alternative when he claims that P ‘states that a war cannot be just unless the evil that can reasonably be expected to ensue from the war is less than the evil that can reasonably be expected to ensue if the war is not fought.’ See Lackey, Douglas P. The Ethics of War and Peace (Englewood Cliffs, NJ: Prentice Hall 1989), 40Google Scholar. Further down the same page, however, he offers a revised formulation: ‘[A] war for a just cause passes the test of proportionality unless it produces a great deal more harm than good.’ He italicizes the intended revision, apparently unaware of having shifted from a comparative to a noncomparative version of P and thereby illustrating our earlier contention that writers often presuppose different interpretations of P at different points in their work.

9 Kavka, 24

10 While Kavka seems to abandon this interpretation in principle in response to one of the objections we cite below, he nevertheless follows it in trying to work out the implications of P for the Gulf War (Kavka, 24).

11 Complications arise when the threatening sequence consists in action that the agent would or might abort. If, for example, the driver of the truck would have swerved to avoid the child, then it may seem less clear that the passerby’s action saves the child. We believe, however, that the analysis of this case is the same: the passerby saves the child. The fact that the driver would have swerved does not show that there was no threat that the passerby averted. (What if the driver swerves at the same moment that the passerby snatches the child back? In that case they both save the child.)

12 For reasons stated below in Section III.2.2, causally remote good effects that are not part of the JC are excluded from consideration by a further, different restriction on P.

13 The Challenge of Peace, 29

14 Ramsey, Paul The Just War: Force and Political Responsibility (Lanham, MD: University Press of America 1983), 193Google Scholar

15 Johnson, James Turner Can Modern War Be Just? (New Haven: Yale University Press 1984), 25Google Scholar

16 We are indebted to Thomas Hurka for the example of economic benefits, which forced us to see the need for this restriction. Despite its absence in the quotations cited above, this restriction is assumed by various writers in the tradition. Anthony Kenny, for example, says that Just Cause holds that ‘war … must be waged to right a specific wrong,’ while P holds that ‘the good to be obtained by the righting of the wrong must outweigh the harm which will be done by the choice of war as a means.’ See Kenny, Anthony The Logic of Deterrence (London: Firethom Press 1985), 9Google Scholar.

17 Ibid., 10

18 For a fuller defense, see McMahan, Jeff ‘Innocence, Self-Defense, and Killing in War,’ Journal of Political Philosophy 2 (1994)CrossRefGoogle Scholar, and McMahan, The Ethics of War (New York: Oxford University Press, forthcoming)Google Scholar. Relevant arguments may also be found in McMahan, ‘Self-Defense and the Problem of the Innocent Attacker,’ Ethics 104 (1994).

19 See McMahan, ‘Self-Defense and the Problem of the Innocent Attacker’ and McMahan, ‘Innocence, Self-Defense, and Killing in War.’

20 While pure partiality was clearly one reason for the policy, the minimization of US casualties was also sought for its instrumental value in maintaining public support for the war.

21 It might be argued that Bush’s claims prior to the war about the magnitude of the Iraqi nuclear threat show that the administration was aware of the extensiveness of Iraq’s nuclear program. There are, however, two reasons for believing what the experts said at the time—namely, that Bush was exaggerating the perceived threat for propaganda purposes. One is that the US announced during the war that it had eliminated the Iraqi nuclear weapons program. But in fact much remained for UN inspectors to discover in the aftermath of the war. The announcement therefore reveals that the US was unaware of those elements of the program that were subsequently discovered by the UN. This is confirmed by the response of the US intelligence community to the UN revelations. Rather than claiming credit for their prescience, members of the intelligence agencies expressed surprise and conceded that the international intelligence community had been guilty of an extraordinary failure.

22 It may also be that there are goods that are in fact JAs that are not recognized as such prior to a war. These too, if actually realized by the war, count in favor of the war on the retrospective calculation even though they were not aimed at in the war.

23 See Clark, Ramsey The Fire This Time: US War Crimes in the Gulf (NY: Thunder’s Mouth Press 1992), 209Google Scholar. Greenpeace gives the higher estimate of 230,000 civilians killed. See Newsweek (January 20, 1992), 28.

24 See Clark, 209. Citing Pentagon and other sources, Newsweek gives the range as 70,000 to 115,000 (18). Former Navy Secretary John Lehman estimated that 200,000 Iraqi soldiers were killed (Clark, 43 and 208).

25 If a war has a JC but fails to satisfy P (particularly in its prospective versions), then soldiers who fight in it may not be altogether innocent. But that fact cannot figure in the calculation that reveals that P is not satisfied.

26 New York Times (January 29, 1991)

27 To be permissible, an alternative to war must be proportionate. But it need not be necessary in the sense defined by N. Suppose, for example, that there are two nonbelligerent means of achieving a JC and that both would have better consequences than war. While N rules out war, it does not require that, if one chooses to pursue the JC by means of one of the alternatives, one must pursue the better of the two.

28 Although this is clearer in the case in which a war is fought in defense of another state, it also applies to other cases, including wars fought in self-defense.

29 It is worth noting that this objection applies equally to the traditional LR, which forbids a state to go to war if peaceful alternative means of achieving the JC remain untried.

30 One response is that, if a JC is sufficiently important to tempt us to permit war rather than allow the JC to go unachieved, then the pursuit of the JC by the better means is after all required rather than optional. This, however, is surely inadequate as a general solution.

31 Thus ‘wrong’ here means pro tanto wrong, not wrong all things considered.

32 The main objection to assassination is that it would have been ineffective, as there were a number of younger officers who had achieved prominence during the Iran-Iraq war who were in line to succeed Saddam and would apparently have maintained continuity with his policies.

33 It might be argued that Iraq’s aims were principally political — that the invasion was the first step in a projected series of moves intended to establish its political dominance in the region. If that were true, however, it would be mysterious why Iraq did not wait to implement its plan until its program for developing nuclear weapons had succeeded. Only the urgency of its economic needs explains the timing of the invasion.

34 As early as the end of October, 1990, the director of the CIA reported that sanctions had stopped 98‥ of Iraq’s oil exports and blocked 95‥ of its imports. See Theodore Draper, ‘The True History of the Gulf War,’ The New York Review of Books January 30, 1992), 39.

35 For a summary of Iraq’s proposals, see Chomsky, Noam “’What We Say Goes“: The Middle East in the New World Order,’ Z Magazine (May 1991) 58-60Google Scholar.

36 Newsweek quotes a ‘senior US official’ as saying that the US’s aim in ‘keeping pressure on’ is to ensure ‘that the Iraqis understand that the future’s a bleak one as long as [Saddam] is around’ (Newsweek, 27).

37 See Draper, 39.

38 As early as August 9, 1990, Iraq proposed that it would withdraw from Kuwait in exchange for recognition of its long-standing claims to these disputed border areas.

39 See Chomsky, 61; and Draper, 39.

40 See ‘UN Says Iraq Was Building H-Bomb and A-Bomb,’, Bigger New York Times (October 15, 1991)Google Scholar; and ‘Iraq’s A-bomb Capability Overrated, UN Now Says,’ New York Times (May 20, 1992).

41 ‘Warning on Iraq And Bomb Bid Silenced in ‘89,’ New York Times (April 20, 1992)

42 New York Times (September 14, 1990)

43 Michael Walzer, ‘Perplexed,’ The New Republic January 28, 1991), 14

44 See Walzer, Michael ‘Justice and Injustice in the Gulf War,’ in Decosse, David E. ed., But Was It Just? Reflections on the Morality of the Persian Gulf War (New York: Doubleday 1992), 6Google Scholar.

45 This claim is defended at length in McMahan, Jeff ‘Revising the Doctrine of Double Effect,’ Journal of Applied Philosophy 11 (1994)CrossRefGoogle Scholar. The view defended there is indebted to that advanced by Warren Quinn in ‘Actions, Intentions, and Consequences: The Doctrine of Double Effect,’ Philosophy and Public Affairs 18 (1989) 334-51.

46 This article evolved from a paper that McMahan wrote during the summer of 1991 and presented at Indiana University, Cornell University, and the University of Illinois. We are grateful to Angela Blackburn, Hugh Chandler, Belden Fields, John Lynn, and two anonymous reviewers for the Canadian Journal of Philosophy for helpful comments and discussion. Our greatest debt is to Thomas Hurka, who gave us extremely detailed and perceptive comments on earlier drafts and also gave generously of his time in discussing our ideas with us. Finally, we are pleased to acknowledge support from the Program for the Study of Cultural Values and Ethics, the Program for Arms Control, Disarmament, and International Security, and the Center for Advanced Study — all at the University of Illinois — and from the US Institute of Peace and the John D. and Catherine T. MacArthur Foundation.