On being a discussant

Chris Blattman offers up some generally good advice on how to be a good discussant.

I heartily agree with most of what he has to say. The main exception is that I think it is usually a bad idea to summarize the paper. The audience has just heard the paper. Hearing it again is boring. The only exceptions are when the presenter is either incoherent (it happens) or when you have a different way of framing the meaning of the paper (rare, but happens too). Otherwise, when a discussant spends time summarizing the paper I assume it was because they could not think of anything original to say.

I particularly like the bits about saving the minor comments to share with the author via email or by handing him/her a marked up copy of the paper. I once had a discussant remark to the audience at a conference about the size of the font in one of my papers. This is the sort of comment, along with minor details about citations and such like, that the audience should be spared.

I would add one other caution. The role of a discussant is to discuss the paper that was presented, not an opportunity for a short infomercial on the discussant's own research. I've seen that too.