Jonathan Haidt, or Is It Wrong to Eat Your Dog?

This post at BPS Research Digest reminded me that I had long meant to blog something about psychologist Jonathan Haidt, arguably one of the more interesting social scientists working today. He works mainly in the fields of morals and disgust, which he says are connected.

A classic paper of his is called "Affect, Culture, and Morality, or Is It Wrong to Eat Your Dog?" (low-quality pdf). The background of this paper is that in the study of moral reasoning, people are usually asked for their moral judgments on scenarios and then asked to explain their reasons. And sure enough, reasons they give. Haidt and his coauthors had the hypothesis that often, people come up with those reasons only after they've already made their decision. They used the same techniques as usual - show people scenarios, let them make a judgment and ask them to explain their reasons - but with a twist. The scenarios were carefully constructed so as to elicit disgust in the study participants, but didn't display any behaviour about which one can say why this should be considered wrong. An example is given in the title: This is a story about a family who eat their dog after the dog has already been killed by a car. Sure enough, many participants would express that the behaviours displayed in the scenarios were morally wrong but couldn't say why. (Also consider in this respect that incest is forbidden in many legal codes even if contraception is used; and many of those codes allow sex between disabled people.)

Here is a nice, relatively short overview of his and other people's work on the topic. Among other things, he mentions that he distinguishes between five categories that people may use to make moral judgments: Harm/Care, Fairness/Reciprocity, Ingroup/Loyalty, Authority/Respect and Purity/Sanctity. Now get this:

Those who described themselves as "very liberal" gave the highest relevance ratings to questions related to the Harm/Care and Fairness/Reciprocity foundations and gave the lowest ratings to questions about the Ingroup/Loyalty, Authority/Respect, and Purity/Sanctity foundations. The more conservative the participant, the more the first two foundations decrease in relevance and the last three increase

You may want to remember this for the next political discussion you have, lest you talk past each other.

Note: The first link contains a link to a bloggingheads discussion between Haidt and Will Wilkinson (which I have not watched). Many of his papers are downloadable here.