Julian Sanchez revisits his post on bombing Japan, torture, and morality:

I find striking about both the arguments over torture and the recent revival of the old Hiroshima dispute is that it sometimes seems as though the defense of necessity is regarded as  purely empirical one: We can argue whether bombing Hiroshima or waterboarding detainees saved American lives, but if it did, then that settles any question of justification.  And we seem to end up with these awfully binary framings”would you agree it was justified if it saved lives?”where the options are “saved lives” or “didn’t save lives,” as opposed to “saved X lives with probability A, saved Y lives with probability B, saved no net lives with probability C, cost Z net lives with probability D.”  This tendency isn’t unique to defenders of torture, mind you: Whether the same information could’ve been obtained by other means, or whether some particular attack would have occurred in the absence of that information, can only honestly be answered in probabilistic terms.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.