War and utility maximization  

Matthew Yglesias writes:

Probably the strongest point the anti-war bloggers have made in the past couple days is pointing to this "Salam Pax" post explaining that he is less-than-enthusiastic about the prospect of having his country bombed by the United States.

This raises, I think, an important normative issue. Normally, one would think that one of the strongest justifications country A could have for waging war against country B is that the population of B favors such a course of action. Now it may or may not be the case that Salam is speaking for the majority of Iraqis on this point, but even if he is not, his post reminds us that some of the civilians killed in war will be civilians who were against the war. If you're comfortable applying a brute utilitarian calculus to these sorts of decisions, it doesn't really make a difference whether the dead civilians were pro- or anti-war, but if you're not this seems to matter.

You could argue that including death (the most extreme possible outcome for an individual) in the analysis skews the equation so much that it's impossible to make any kind of rational choices. This is similar to the argument I made against creating a market for kidneys - that allowing death as a potential outcome puts entire fortunes in play, as the relevant utility function approaches infinity. Weighting death according to this kind of skewed utility - in the context of war or otherwise - just leads to the point where any risk of fatality is enough to paralyze us. If individuals consciously assume responsbility by supporting the war, their preferences don't exhibit the same distortion; but especially with a large population, won't a single naysayer will be enough to make the system impractical?