Asking is good, theory is better

In discussions with friends who from other social science disciplines, we often get into disagreements about the relative importance, and role, of theory and quantitative or qualitative data in finding the answer to a question. Most discussions come to the conclusion that anecdotal or informally gathered data (from surveys, observations, conversations etc.) can serve as the basis for developing new hypotheses. In the next step, theoretical models from psychology, economics, anthropology etc. are useful to develop a strong hypothesis for the mechanism underlying the observed phenomenon and identify parameters and predictions that can be tested to verify the applicability of the model. Finally, well-designed empirical studies (eg randomized controlled trials) should test the validity of the model in contexts other than the one that originally inspired the model.

As an economist with some interest in theory and empirical methods, I get especially frustrated by some ethnographic studies, often several decades old, where the researchers simply asked the individuals involved about why they exhibit strange social behavior X. Then, tallying up the answers, they proclaimed that the “reason” for the behavior must be X, because the people said so. Often the “reason” would involve religious prescriptions or “tradition” or some other question-begging concept.  As the severity and nature  of religious prescription is usually strongly adapted to local needs (just ask any Christian why they don’t think women have to cover their hair in a religious service, as Paul clearly demands in his letters in the bible) and thus endogenous to social mechanisms, the answer is basically useless, because it just says that locally observed behavior is due to local inclinations to promote the locally observed behavior.

One good example of this rule never to trust individuals to know the reasons for common behaviors or rules of their group is given by Robert Wyman in an online lecture:

There’s many taboos on sexual relations, especially after birth. In many cultures, as I said, a prescribed period of nursing. There’s also a postpartum taboo against having sexual relations, and again, [this is] another obvious mechanism for birth spacing…If you ask a member of a society that has a taboo they report for instance that sex at that time is very dangerous, a life and death matter.

It is dangerous to mix the man’s blood with the woman’s, and the man’s blood is transmitted through semen…If man’s blood gets into the woman through his semen then it also gets into her milk, and then the man’s blood goes back into the baby through the mother’s milk and this is poison for the baby. That’s their version of why they shouldn’t–why they have this taboo.

Note that the issue with this justification of a fertility practice is not that it is biologically inaccurate, but rather that it should quickly have become obvious that the predicted outcome (death of the child) was rarely associated with intercourse beforehand. The persistence of such an empirically untenable belief suggests that it is merely a conscious rationalization of a more deeply rooted mechanism, that is, the need to limit group fertility in order ensure survival.

But inventing “reasons” to justify deep-seated beliefs is not limited to developing countries. I think we can all find instances in our own lives, especially related to issues of identity, where there is a group norm and we make the “reason” for it up on the fly. Why is it bad to have a government-regulated market for kidneys? Well, you know, it is dangerous, a life and death matter, to mix the man’s money with…

  1. Leave a comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: