A recent Crooked Timber thread, attached to this post, got into both empirical and moral questions about Allied bombing in WW2, including the atomic bombings. In connection with this I've been urged by Anderson to read part of Richard Frank's Downfall (which I'm planning to do).
The CT discussion led me to take another look at something which, unlike Downfall, I already have on the shelf: Michael Walzer's chapter in Just and Unjust Wars on "supreme emergency," the phrase Churchill used to describe Britain's situation in 1939 (see p.251). Walzer's argument, in brief, about the bombing of German cities is that early in the war, when "Bomber Command was the only offensive weapon available to the British" (258), the real possibility of an imminent German victory constituted a supreme emergency, i.e., the sort of rare situation which "might well" (259) have justified overriding the norms/rules of war (which he calls 'the war convention') and engaging in city bombing, which was the only thing, at that point, that the bombers could do, given their crude navigational equipment and consequent lack of precision. However, despite improvements in navigation etc. and, even more importantly, the changing military situation, bombing of cities continued until almost the end of the war. "[T]he supreme emergency passed long before the British bombing reached its crescendo. The greater number by far of the German civilians killed by terror bombing were killed without moral (and probably also without military) reason," Walzer writes (261).
Walzer rejects, on moral grounds, the defense of city bombing given at the time by Arthur Harris (head of Bomber Command) and others to the effect that it would hasten the end of the war and thus, on balance, save lives. The passage in which Walzer explains his view is worth quoting (albeit in abridged form):
The argument used between 1942 and 1945 in defense of terror bombing was utilitarian in character, its emphasis not on victory itself but on the time and price of victory. The city raids, it was claimed by men such as Harris, would end the war sooner than it would otherwise end and, despite the large number of civilian casualties they inflicted, at a lower cost in human life. Assuming this claim to be true (I have already indicated that precisely opposite claims are made by some historians and strategists), it is nevertheless not sufficient to justify the bombing. It is not sufficient, I think, even if we do nothing more than calculate utilities. For such calculations need not be concerned only with the preservation of life. There is much else that we might plausibly want to preserve:...for example,...our collective abhorrence of murder.... To kill 278,966 civilians (the number is made up) in order to avoid the deaths of an unknown but probably larger number of civilians and soldiers is surely a fantastic, godlike, frightening, and horrendous act. (261-62; textual footnote omitted)He goes on to say that though "such acts can probably be ruled out on utilitarian grounds," it is only when "the acknowledgment of rights" comes into the picture that we are compelled "to realize that the destruction of the innocent, whatever its purposes, is a kind of blasphemy against our deepest moral commitments" (262).
The amendment I'd make here would be to replace the word "innocent" with "non-combatant." Why? Because if an 'average' civilian is innocent, so an 'average' soldier, one who has not committed atrocities but simply participated in battles or worked behind the lines, may also be, in some relevant sense, innocent. What is he guilty of, other than doing what soldiers are expected to do? The appropriate distinction, it seems to me, is not between innocent and not-innocent but between non-combatant and combatant. Putting aside the word "innocent" means that one doesn't have to inquire into any particular non-combatant's actions or, in the case of Nazi Germany, awareness of genocide, which a fair number of German civilians probably had. It was their status as non-combatants, not their "innocence," which made deliberately killing them, especially after the supreme emergency no longer existed, unjustifiable.
Notes: (1) Just and Unjust Wars has gone through several editions, with new prefaces, though I believe the main body of the text hasn't changed. I'm quoting in this post from the first edition (1977). (2) There are at least several journal articles specifically about "supreme emergency" as Walzer uses it. One is here [abstract; full text is gated].