² Walzer, 53, quoting Westlake, John, Collected Papers, Oppenheim, L., ed. (Cambridge, U.K.: Cambridge University Press, 1914), 78Google Scholar.

³ The stress here on something like “consent” is consistent with an interpretation of Walzer's theory as a version of consent theory, at least in part. James Childress notes the centrality of consent in his review of Walzer's book. The exact relation between “consent” and losing immunity from attack is central to what I wish to exploreGoogle Scholar.

⁴ In these summary statements I use the word “attack” to signify direct and intentional attack, not indirect, unintentional killing. I also normally use the term that way elsewhere, but may not always be consistent in doing so, partly because Walzer sometimes uses the term differentlyGoogle Scholar.

⁵ Walzer, 162, quoting Cheney, CharlesHyde, International Law, 2nd rev. ed. (1922; Boston: Little, Brown, and Company, 1945), III, 1802Google Scholar.

⁶ Walzer, 162, citing The Works of Josephus, trans. Lodge, Tho. (London, 1620), 722Google Scholar.

⁷ Walzer's discussion of those forced to stay in the city by the besieged army (p. 168) seems to assume that all those who are forced to stay are performing essential services. Exactly how his discussion of other civilians forced to stay (p. 163) relates to this is not entirely clear. It seems from the earlier discussion that there would be a category of civilians who might be forced to stay by the besieged army, but who are not performing essential services and are therefore not threatening. The fate of this group is not addressed directlyGoogle Scholar.

⁸ Walzer here uses terms that focus attention on how civilians might be killed (directly/indirectly), while my language of “threat/nonthreat” focuses attention on the grounds for saying they may be killed. This is a significant difference in some respects, but the crucial issue for my purposes is the shift from a focus on either threat/nonthreat or direct/indirect killing to a focus on coercion and consentGoogle Scholar.

⁹ Here, incidentally, is another case where in Walzer's account persons lose their civilian immunity through no act of their ownGoogle Scholar.

¹⁰ The military itself did not justify the shootings, but rather denied that it had done themGoogle Scholar.

¹¹ It is also true that blame can be divided. The case of the civilian who chooses to stay in an exposed position and is killed directly by the attackers may be such a case. It is true, whatever the exact mathematics of blame apportioning in this situation, that the attacker is less blameworthy than when no exit is permittedGoogle Scholar.

¹² There has been a great deal of writing about the ethics of guerrilla warfare, some of it dealing specifically with Walzer's treatment of it, though most of what I have found does not approach the issue from the perspective of the right of civilians to immunity. Treatments which I find especially helpful in reflecting some of my concerns about the responsibility of guerrillas and which deal with Walzer are Phillips, Robert L., War and Justice (Norman: University of Oklahoma Press, 1989), 85–100Google Scholar, and Cohen, Sheldon, Arms and Judgment (Boulder: Westview Press, 1989), 153–63Google Scholar. Ramsey, Paul, The Just War (New York: Charles Scribner's Sons, 1968), 427–512Google Scholar, also raises important issues pointing to the responsibility of guerrillas, though I do not share many of his judgments and conclusions. Other helpful sources here are Tom J. Farer, The Laws of War 25 Years After Nuremberg, International Conciliation series, no, 583 (New York: Carnegie Endowment for International Peace, 1971); Franchise Hampson, “Winning by the Rules: Law and Warfare in the 1980s,”Third World Quarterly (April 1989), 1–62 on international law; and Campbell, Courtney S., “Moral Responsibility and Irregular War,” in Johnson, James Turner and Kelsay, John, eds., Cross, Crescent, and Sword (New York: Greenwood Press, 1990), 103–28Google Scholar.