This talk analyzes the use and abuse of utilitarianism in "the torture debate", arguing that the latter might turn out to be utilitarianism's nemesis. For what the debate lays bare is that, if we are to take utilitarianism seriously, then we must be prepared to torture the alleged terrorist's child, or indeed anyone at all, to prevent the "imminent catastrophe". Furthermore, if that conclusion is unpalatable on rule-utilitarian grounds - in terms of the institutional and long-term consequences of such a practice - then those same sorts of consideration rule out torturing the alleged terrorist themselves. That this is systematically obscured by those who would purport to justify interrogational torture by their being highly selective about the consequences they consider, and/or by arbitrarily “modifying” the scope of utilitarianism when it generates inconvenient conclusions, again suggests that utilitarianism may be fundamentally flawed; and that its use to defend interrogational torture shows this. Finally, it should be a matter of concern to utilitarians that those among them who defend interrogational torture remain remarkably reluctant actually to acquaint themselves with the available empirical evidence.