Martin Seligman, who assisted the U.S. government in the torture of detainees, was made famous by his dog torture experiments. Graeme Wood wonders about the similarities:

The parallels between the logic of human and canine torture -- sacrificing the well-being of the few (terrorists and dogs) for the well-being of the many (innocents and depressives) -- are worrisomely obvious. What's less obvious is which way the argument cuts. Seligman, a morally thoughtful man and a self-professed dog lover, condemns torture, yet his experiments suggest a moral calculus that might allow it. If torturing a terrorist to save actual human lives isn't permissible, then by what logic could he torture dozens of dogs for a smaller -- and perhaps less certain -- payoff? Today, many universities would, I suspect, reject his experiments (and a lot of other fascinating research) on ethical grounds.

These comparisons are notoriously tricky to deploy in an ethically scrupulous way. In the novel Elizabeth Costello, a poet objects to a comparison between abattoirs and WWII death camps. "If Jews were treated like cattle, it does not follow that cattle are treated like Jews." Indeed: caution against glib moral comparison seems extremely wise right now, as do hasty condemnation or absolution of anyone involved in the torture controversy. Seligman has already caught undeserved blog-flak because of the reports about Mayer's book, even though he almost certainly never knowingly abetted torture. Our ethical approaches to each question -- whether to torture dogs, and whether to torture people -- do seem like they should be related, though. And whatever our conclusions, it seems worth noting, again with worry, that we appear to be doing to people what many have already decided it is wrong to do to dogs.

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