Either DA-RT Works, or It Does Not

8 April 2016, 1701 EDT

This is a guest post by Theo McLauchin (@TheoMcLauchlin), Assistant Professor of Political Science at the Université de Montréal

When is a norm not a norm? I ask this question when I read Colin Elman and Arthur Lupia’s vigorous defense of the Data Access & Research Transparency (DA-RT) initiative in APSA’s Comparative Politics section newsletter. I think Elman and Lupia try to have it both ways. Their piece argues, first, that journals need to adopt norms of openness. It then argues, in defense of DA-RT against a series of concerns that it will bring the editorial hammer down on many different forms of work, that DA-RT doesn’t change anything. Editors always could implement whatever policies they wanted to. But of course if the norms change, then the content of that editorial discretion – what decisions are actually made where the submission meets the desk – changes with it. Either DA-RT has an effect or it doesn’t; either the norms change or they don’t; either some articles become newly unpublishable at some journals or they don’t. If they don’t, then DA-RT cannot have the effect its creators hope for it.

And speaking of norms, Elman and Lupia are awfully sanguine about the impact of DA-RT on the multiplicity of political science. They put a lot of stock in different “research communities,” each of whom can happily pursue its own norms of openness, interpreting DA-RT each sees fit. This is their big blanket reassurance: whatever your fears about DA-RT, you and your research community can basically choose how you implement what is simply (to Elman and Lupia) an uncontroversial call for transparency. Implicit in this is that it’s OK if there is little dialogue across those research communities. I’m not OK with that, and I don’t think many readers of, say, APSR – whose new, DA-RT-compliant guidelines you can see here, and judge for yourself how they accommodate different traditions – should be OK with it either. I happily read things with wildly different norms of inquiry, I think that’s an irreducible richness of political science (and of sociology, the field which I read second most). The impact of DA-RT, as a norm, may well be to crystallize implicit suspicions of research traditions not one’s own, to reinforce concerns about the scholarly value of others’ work with concerns about its honesty, and to push editorial boards to decide which side they’re on. DA-RT’s creators should not be judged responsible for all this; everybody in question is capable of deciding how to react. But again: for DA-RT to have an effect, some articles must become unpublishable at some venues. If we all divide up and adopt different norms, the result would be to fragment the discipline in a manifest way, journal by journal.


Let’s take an example to make it more concrete. I’m proud to have published a qualitative piece at Journal of Conflict Resolution as part of a great special issue made up of pieces that use very different methodologies. Depending on how JCR’s editors react to DA-RT and what they require of case studies based mainly on secondary literature – as it happens, the journal links to DA-RT but does not appear to have new guidelines just yet – it might become too burdensome (in terms of a detailed, “active-citation” documentation of each step in the argument) to publish a similar piece. Future grad students (as I was at the time) with that kind of opportunity might well find it too onerous to comply with whatever standards JCR decides to adopt. That would be a real shame. It would reduce the impact of journals as sites of cross-pollination and curiosity. To be clear, I’m sure Elman and Lupia don’t wish for this, but their approach implies that journal editors must draw the line differently than they have done. If not me and my co-author, then someone.

In any event, Elman and Lupia show their hands when they lament that DA-RT’s critics risk keeping Poli Sci behind in a world where it’s so easy to share data. It is indeed, mechanically, much easier than ever. That ease has in no degree reduced the ethical quandary around whether, for a host of research questions, it is appropriate to do so. Indeed, I think that quandary may now be harder to the extent that technology makes the mechanics easier, for the field notes read by a scholar in South Bend are now just as easily read in a police post in Cairo or Quantico (on the real risks here, read especially Marc Lynch’s comments in the same CP newsletter).

And for the qualitative scholar who may be made[1] to document each of a hundred tiny inferences that add up to a big one, the mechanics aren’t much help. It is true that in picking up the dead-tree edition of World Politics in 1978, I could not have reached through some sort of magical portal in one of Robert Jervis’ footnotes to pull out the dusty manuscript about the obscure international dispute he references, alongside The Jerv’s own marginal note explaining why he thought the particular fact about it was really cool. The fact that I could now do something equivalent online is helpful, but it also creates for Jervis the labour involved in putting that online — a burden much more onerous than my uploading my latest dataset and do-file. (And I say this as someone who takes a lot of time to comment his do-files, to make them as accessible as possible.) Where Elman and Lupia see opportunity in the web, other scholars may quite justifiably see constraint.

We therefore cannot do what Elman and Lupia hope to do in a manner that falls evenly across research traditions. Their purported flexibility, their claim to avoid one-size-fits-all standards, cannot actually work in practice; or, if it does, it will be at the cost of further fragmenting a discipline that ought to be eclectic at its heart and had finally been, in my view, moving in that direction in the last fifteen years or so. That would be an immense shame.