IRBs

Your blood runs cold at just reading those letters, eh? And seeing that scary graphic doesn't help, does it?

The New York Times has a front-page piece this morning about Institutional Review Boards, the ethics folks who make sure that we don't do bad things to people we gather data from. The hook is this:

faculty and graduate students across the country increasingly complain that these panels have spun out of control, curtailing academic freedom and interfering with research in history, English and other subjects that poses virtually no danger to anyone.

Syntax footnote: it's not entirely automatic for me to get the last that to refer to research rather than subjects — I thought poses was a typo until I finished the sentence. Maybe it's just a convoluted sentence. Anyhow, the examples are:

Among the incidents cited in recent report by the American Association of University Professors are a review board asking a linguist studying a preliterate tribe to “have the subjects read and sign a consent form,” and a board forbidding a white student studying ethnicity to interview African-American Ph.D. students “because it might be traumatic for them.”

The article never returns to the linguist, but we all have our horror stories. Fortunately, they normally get resolved without too much grief. At Wisconsin and surely most major institutions, we have explicit instructions for obtaining oral consent from subjects who may not be able to read, might have political or cultural reasons for not signing something, and so on. But people are talking lawsuit …

Philip Hamburger, a professor at Columbia Law School … argues that prior approval violates the First Amendment. “There are potentially hundreds of thousands of plaintiffs.”

Actually, research with human subjects needs to be regulated: We train students in all aspects of doing research, and having a clear plan for what we're going to do and what impact it might have on subjects / consultants is something every researcher needs to consider carefully. I see reasonable IRB stuff as quality control and research methods material.

The problem, I think, is the profound lack of explicit, established standards in this realm. The Linguistic Society of America has been talking about laying out guidelines for linguistic research. I hope they do, and I'll immediately take them to our social and behavioral sciences IRB.