Ramesh, I don’t read Stephens as saying the intentional killing of civilians was not evil. He expressly says that the bombing was evil (framing the question as “whether the good that flowed from the bombing outweighed the unmistakable evil of the act itself“) (emphasis added). In fact, he calls this the “classic ‘choice of evils’ dilemma.” If it was “the right thing to do,” it is in the Pauline calculus of whether the consequence outweighed the evil, not whether evil in fact was done. And on this, I think he’s more nuanced than you suggest, conceding that, as compared with Hiroshima/Nagasaki, “a somewhat better argument can be made that the bombing of Germany failed to justify its price in human suffering, particularly the bombing of non-strategic targets.” (Granted, he provides the utilitarian counter-argument as well.)
I agree that there’s a danger on my side of this debate that we get too utilitarian, at least rhetorically. When you argue (as I’ve done myself) that waterboarding (or even some forms of torture) are worth employing if it is absolutely necessary to save thousands of lives, the questions naturally arise: How about ten lives, or one? Are we saying anything goes? How about something a lot more painful and enduring than waterboarding? How about threatening to harm or actually harming a guilty terrorist’s innocent loved ones in order to get him to talk? I don’t think it’s a numbers game and I do believe there are things we should never and, I hope, would never do. (The “shock the conscience” test is elusive, but it least it has the benefit of evaluating situations as they arise, mindful that we cannot anticipate and make rules for every possible contingency.)
I think Bret’s main point here is that, on the scale of bad things we’ve done to achieve results whose good far outweighs the bad, waterboarding (in the rare instances it’s apparently been used) is pretty benign compared to deliberately targeting civilians in bombings, and to intensive bombings of military targets knowing full well that collateral civilan deaths would be massive.
I wonder how far the other side of the debate takes the logic of its position. It is the rare person — one whom I respect but disagree with — who says we should never engage in any evil even to prevent a worse evil, and therefore that it is better, say, to lose a city than to waterboard. But as a prosecutor, I’ve seen testimony purchased for money or by the effective dismissal of very serious charges. The calculation is the same: we don’t like doing it, but it is for the greater good. If we didn’t do it, people wouldn’t cooperate and the worst criminals couldn’t be reached. Should we not do it?