Torture: When the Unthinkable Is Morally Permissible (review)

In one of the most recent installations in a long-line of utilitarian defenses of torture, Mirko Bagaric and Julie Clarke contend that,

torture is morally permissible where it is the only means available to save innocent lives. Torture should only be used where the threat is imminent, there are no other means of alleviating the threat, and the suspect is known to have the relevant information. Torture is justifiable in these circumstances because it is less bad to inflict physical harm on a person than to allow large numbers (or in some cases a single person) to die. When rights clash and only one right can be protected we should opt for the higher-order right. To this end, the right to life is more important than the right to physical integrity.¹

The overall argument of the book is utilitarian. Despite the claim that the deontological view of torture will also be considered, the authors focus their attention almost exclusively on the relatively classical, Benthamite version of the hedonic calculus, even offering a quick outline of an actual equation at one point.² When the deontological tradition is discussed, the authors quickly dismiss it as relying on an implausible conception of rights. The book very succinctly spells out what the argument for torture involves, pointing out that it is an argument that does not justify torture as it is currently practiced, and would likely justify very few instances of torture in the future. The remainder of the book is dedicated to deflecting typical objections to proposals like those offered by the authors. The final few pages of the book then present a brief meta-analysis of the torture debate.

The primary merit of this book is its succinct overview of a utilitarian defense of the use of torture in very specific circumstances, with equally succinct accounts of the primary objections to this view and responses to these objections. In this respect, the book constitutes a fine introduction to one line of argument for those unfamiliar with this literature. Likewise, it serves much the purpose of a manifesto: a short tract outlining the core elements of a view.

But it is largely because of these merits that the book suffers a number of serious flaws—not the least of which is a thin argument that will convince no one not already committed to the sort of utilitarian analysis that accepts without comment the acceptability of torture. To effectively advance the debate concerning the possible moral and political permissibility of torture, the authors of this book need to spend a great deal more time developing their respective responses to the classical objections and spend an equal amount of energy considering some of the significant recent objections to torture that do not appear in the text.

While there is much that merits attention and critical analysis in this text, I want to examine in more detail what I regard as a reductio of Bagaric and Clarke's proposal that can be found in the text (surprisingly offered by the authors themselves!), and to demonstrate that it leads directly to a claim that even theauthors do not accept: namely, that we ought to torture the innocent in order to better understand how to torture the guilty.

In responding to the claim that torture never works, Bagaric and Clarke present some compelling evidence to the contrary. They then claim that, "the way forward here is to obtain more pointed data regarding the circumstances in which torture has been effective and when it has failed."³ Surprisingly, however, they go on to claim that, "the study could only be retrospective—no one would seriously contemplate actually torturing people for experimental purposes."⁴

It seems that the authors here provide a reductio of the very view they are attempting to defend. As the authors see it, certain conditions must be met for torture to occur. We must consider the following five things:

(1) the number...