Learned friends

by Henry on July 22, 2003

Larry Solum and Dan Drezner are having a minor contretemps that touches on one of my pet peeves; the lack of a one-stop-shop for working papers in political science. Drezner takes issue with a recent Slate essay by Steven Johnson which argues that Google is pushing academics towards writing (PDF-able) articles rather than books.

Johnson sets this up as an either/or question—online papers or books. In point of fact, for most academics this is a progression. First you circulate your ideas in draft form, then as a conference paper, then as an article, and then—maybe—publish a book.

But Solum gently dissents, suggesting that legal academics are going through a gentle revolution, in which online sources (and means of sorting through them) are becoming more and more important. The SSRN (Social Science Resource Network), a free online repository of working papers and published papers, is playing an increasingly important role in shunting ideas around the legal academy. Furthermore, bloggers provide an important filter, separating out the gold from the dross. I can vouch for this claim – I read the summaries of new SSRN papers in Solum’s Legal Theory Blog every day (lawyers write a lot on the issues I’m working on).

There isn’t much real disagreement between Solum and Drezner; as both acknowledge, their arguments are shaped by discipline-specific experiences. And that’s the point that I want to develop. Frankly, I’m jealous of the lawyers. To the best of my knowledge, political scientists like Drezner and I have no equivalent to the SSRN. Columbia University have set up a sort-of-similar service called CIAO, but it’s pretty spotty, and is only available by subscription anyway. There are websites with papers for the big annual conferences, and a few subject specific sites but there’s nothing that really brings the discipline together. And it’s not only political science that has this problem. My impression is that the only social science that has really gotten its act together is economics; I don’t think that sociology, geography, anthropology and so on have SSRN equivalents either.

This dearth of a clearing house goes hand-in-hand with a disciplinary culture in political science that de-emphasizes working papers. Lawyers usually seem happy to put out draft journal articles as working papers for everyone to read. Political scientists tend, in my experience, to be much more suspicious of the idea. By and large, they don’t want to put out their big journal articles as working papers before they’re accepted, in case someone rips off their big idea and beats them to publication. And there are other problems; some journal editors are rather sniffy about accepting articles that have previously appeared in working paper series.

So why is it that lawyers, b-school types and economists have SSRN, and we have nothing? I’m not sure I have an answer, but here are some suggestions (all of which have flaws).

  • Differential exposure to technology. As Solum points out, lawyers have been using Westlaw and Lexis-Nexis for a long, long time – they’re used to having everything on line. Political scientists haven’t. Flaw: political scientists aren’t Luddites; why haven’t they set up an equivalent (or jumped onto the SSRN bandwagon, as soon as they realized what a good wheeze it was?
  • Path dependence. Chance events in the development of legal academics and political science have created deep-rooted disciplinary structures that shape publishing patterns in both fields, which would be difficult and costly to reform, even if political scientists recognize that they’re in a suboptimal situation. Flaw: plausible, but unfalsifiable without more specific arguments about why these patterns have persisted and reproduced themselves.
  • External resources. SSRN’s financial backers seem to have used the series as a way to get street-cred with financial analysts and the like. They’ve probably poured a substantial amount of money into setting SSRN up. Nobody is likely to do the same for other social science disciplines. Flaw: political science may not be as commercially attractive to outsiders; but there are a lot of foundations and other punters who might well be prepared to put money into a SSRN equivalent. And we’re not talking about a huge fortune – enough to ire a couple of organizers, rent some server space, and direct punters to the various working paper series across the web.

I’m sure there are other possible explanations.

Addendum: This is probably a good place to note that Chris Lawrence is trying to get a pol-sci disciplinary newsblog up and running; it sounds like a very worthwhile project.

{ 1 trackback }

Crooked Timber » » SSRN on Law and Politics
11.10.05 at 10:30 am

{ 13 comments }

1

Brian Weatherson 07.22.03 at 5:15 am

If you aren’t too concerned about the need for a screening process, it isn’t that hard to simply roll your own clearing house.

In semantics, which admittedly is a pretty small field relative to political science, Chris Barker and Peter Laserholn set up the Semantics Archive without too much trouble. (Though it did need a bit of funding, and a bit of advertising to get everyone to use it. And the latter is much easier in a small field than a large field.)

Or, if people are in the habit of posting draft papers to their own websites, you can set up a database that keeps track of what’s available where. I haven’t tried doing that in its entirity (though it would be an interesting challenge), but I did try something almost as useful for philosophy. With the help of some useful software I scan all of the pages (that I know about) that contain online philosophy papers and keep a blog where I report the changes in the last 24 hours. (The blog, in case you’re interested, is here.)

I haven’t advertised this blog much, but it is growing steadily, and I think its led to people who post papers getting more feedback than they otherwise would. (Not that I’ve done any actual research to check this.)

The trick is to (a) have a culture where people are at least prepared to post papers to their own pages, (b) have a database of all the places where people do post their own papers, and (c’) have someone prepared to scan all those pages daily (or less frequently, but daily is ideal) and report back. If you don’t have (a) then a central clearing house probably won’t make much difference. Obviously (b) is hard work, though it’s largely a startup cost. And (c’) is a running cost, though when I don’t bother to stop and check the quality of the papers I’m linking to, it’s only about 15 minutes a day for the areas I cover. That would go up if you tried to cover something larger, but perhaps not dramatically.

Basically if you have a culture of people posting papers to their websites, then all the labour you need to maintain a blog that lists all the new work could be provided by a grad student doing less work than they’d normally put in teaching. If it’s a useful project, I’m sure some school somewhere in the world will spare one grad student a year from teaching to maintain the blog.

2

kb 07.22.03 at 5:17 am

I think the cred for SSRN came from having a top-flight group of academics signing on from the start. I recall getting several emails at the outset from Michael Jensen and Martin Feldstein advertising the service, which certainly helped make me think they were legit.

I’ve found SSRN to be quite democratizing. In the past, to get good working papers you either had to be in a network or subscribe to something like NBER (and I think NBER’s working paper series, which Feldstein helped run, has always been a model for disseminating ideas pre-publication). Just as the Journal of Economic Literature was the one place for published articles, SSRN, RePEc and WoPEc have become search engines for working papers.

3

Henry 07.22.03 at 5:52 am

I’d actually meant to include the Feldstein effect as a possible explanation and forgot. I think you’re right – it’s quite important that you have key figures in the field who are willing to associate themselves with the project to get the ball rolling.

Brian – v. interesting. Another issue that scares some pol. scientists (although they’re getting increasingly bolder), is copyright issues, and whether the journal in question will get annoyed if they put copies of their mss on-line pre- or post-publication. Although some journals take an enlightened attitude – _International Organization_, which is one of the biggies, specifically grants you rights in the contract to put articles it publishes on your own website.

4

Jacques Distler 07.22.03 at 6:25 am

In most branches of physics, large parts of mathematics, and most recently, in computer science, the primary means of scholarly publication are the arXivs formerly at Los Alamos, now hosted by Cornell University.

The oldest section of the archives (hep-th — high energy physics, theoretical) will be 12 years old this August. For those of us who watched it develop, the least expected, but most welcome side-effect of this shift to online publication was a radical democratization and internationalization of the flow of information.

The most puzzling aspect, to me at least, is the widely varying rates of adoption in different subfields. Some converted rapidly; other fields dragged on for years before the bulk of active researchers would routinely submit their papers to the archives.

Maybe it’s a cultural thing. Maybe it’s that people in some field were wedded to antiquated or proprietary document preparation formats. Hard to tell. But they all come around in the end…

5

Eszter 07.22.03 at 6:50 am

This is a very interesting discussion and does raise some hard questions. I don’t have any better suggestions than what’s already been posted re why it is that working papers are more common in some fields than others. (And it was correctly stated, sociology does not have much of a culture of working papers.)

I just thought I’d add an additional point, because I wonder if it’s related. My impression is that traveling across the country to give seminar presentations on works-in-progress is much more common in econ than in soc (these are two areas with which I’m fairly familiar so I thought I’d use them as examples). I don’t know if it’s partly simply that econ depts have bigger budgets for seminar series, but I have certainly noticed that while an economist may give half a dozen if not more seminar talks during a year, a sociologist is lucky to have one invitation. My impression is that most sociology depts don’t even have a regular speaker series. I don’t have any data to back up this claim and I don’t even know how many econ depts have a regular speaker series. It would be interesting to see a compilation of depts by discipline on this.

The reason I think this relates to the topic of working papers is because seminars are another way of disseminating one’s work informally before publication. And if a field has the culture of frequent seminar series then it seems like a reasonable next step to value working papers in contrast to a field where works in progress rarely get discussed by peers.

Of course, you could make the opposite argument whereby a field that does not have a seminar series would be more inclined to pursue a working paper culture because it is an important way for authors to get feedback (lacking the option of seminar presentations), but this doesn’t seem to be the case.

This was just a thought I had that I think in some ways may relate to this discussion. Unfortunately, I don’t know enough about the seminar culture in cs and other science fields to consider how this works out in those areas.

6

dsquared 07.22.03 at 7:33 am

In economics, working papers are the primary means of communication of the central banks, the World Bank and the IMF, each of which has a research budget for economics which is around five times the size of a normal university …

7

William Sjostrom 07.22.03 at 1:27 pm

It is worth noting that SSRN was originally started by financial economists, and other parts of economics were then added. Law was originally added by the law and economics crowd, but that has grown into a full scale legal section. If you look at their home page, they now have sections on accounting, economics, financial economics, law, management, information systems, marketing, and negotiations research. Each one has a group of editors, all major figures in their fields. SSRN says it is devoted to social science. I do not know why fields such as sociology and political science do not simply join up. It would turn SSRN from the value asset that it currently is into a multidisciplinary gold mine.

8

Chad Orzel 07.22.03 at 1:35 pm

The copyright issue is one that does come up. I know that, as of early 2001, at least, the journals Science and Nature had policies that regarded posting on the Los Alamos preprint server as “prior publication,” which would disqualify an article from the journal. As these are the very top science journals, work that was headed for one or the other wouldn’t get put on the preprint server.

Particle physics has gotten around this by just not dealing with actual journals any more, a move which strikes me as foolish, but that leads inevitably to the “crappy papers I’ve refereed for Physical Review Letters” rant, and this comment section isn’t large enough to contain it…

9

Loren King 07.22.03 at 7:44 pm

Henry: some thoughts on the idea of a discipline-wide “clearing house” for working papers in political science. I think the big conferences you refer to (APSA and CPSA both post conference papers on the web) do bring the discipline together in the sense you seeem to mean. Also, several departments (MIT, Chicago, and NYU that I know of) have papers available from some of their seminar series. And at least one subfield association within political science, the methodology division, runs a superb working paper archive. Personally, I’m not sure I want much more “bringing together” than that: I like some decentralization, some cacophony. Or at least, it isn’t clear to me that the scholarly gains of disciplinewide distribution of prepublication material is worth the effort of establishing and maintaining such a node (and I accept that the costs probably aren’t all that high).

Also, I don’t yet share your optimism that scholarly blogs can be trusted to separate the gold from the dross. Maybe, but there is a distinct “echo chamber” quality to such mediating efforts. Sometimes I worry that the so-called blogosphere is merely a refined and articulate variation on how they “elected” leaders in Sparta.

10

Russell L. Carter 07.22.03 at 8:28 pm

“In most branches of physics, large parts of mathematics, and most recently, in computer science, the primary means of scholarly publication are the arXivs formerly at Los Alamos, now hosted by Cornell University.”

citeseer is basically a google for computer science/computational mathematics preprints and published papers, with an extremely convenient interface for exploring citation and bibliographic reference trees. Links to the documents in pdf/ps formats are provided, and for the most part are cached as well. I’d guess out of the thousands of papers I’ve looked up, maybe 2 or 3 were jailed behind paywalls. It has freed me from living within commuting distance of research libraries. It’s free, but I’d pay a couple hundred bucks a year for it without blinking. Along with the arxiv, this sort of thing is hugely democratic and an example of real progress.

11

Henry 07.23.03 at 12:26 am

Loren

interesting points – my quick responses. First, you’re right that the APSA meeting websites etc do bring the discipline together – but in a rather scattershot fashion. Not everyone presents there, and not every presenter uploads papers. And that is even more true of the departmental seminar series etc. What I like about the SSRN is precisely that it’s a one-stop-shop, where everyone who does research in law can go if they want to find something. I think that this encourages a working paper culture, which is a good thing imo, for two reasons. First, people in law really know that others are going to have access to working papers, and read them and be influenced, and thus they have an incentive to make them available. Second, I suspect that it actually makes plagiarism less attractive; if I’m a plagiarist, and I rip off a paper from a less well known working paper series and submit it for journal publication, I have some chance of getting away with it, if I’m careful to remove the obvious tell-tale signs, because anonymous reviewers are unlikely to have read the original piece. But becomes a lot trickier if working papers are centralized, so that most people in a subfield or specialization can be expected to have read the original. Or at least, this seems plausible to me.

On the bloggers-as-filters, I’m talking specifically to what Larry does on legal theory blog. He casts his net quite widely, but because he shares some of my interests (e-commerce and Internet law), I can be pretty sure that he’s going to pick up on anything half way interesting that comes out. Of course, the quality of the filtering would depend on the quality of the filterer – but that’s true of all forms of filtering, whether they be journal editors, smaller working paper series & c.

12

Russell L. Carter 07.23.03 at 4:01 am

“Second, I suspect that it actually makes plagiarism less attractive; if I’m a plagiarist, and I rip off a paper from a less well known working paper series and submit it for journal publication, I have some chance of getting away with it, if I’m careful to remove the obvious tell-tale signs, because anonymous reviewers are unlikely to have read the original piece. But becomes a lot trickier if working papers are centralized, so that most people in a subfield or specialization can be expected to have read the original. Or at least, this seems plausible to me.”

Select any paper from here.

Note the “Similar documents (at the sentence level)” and the “Similar documents based on text:” fields. Click on any of those. Wherefore art thou, plagiarism?

Cool eh?

13

Henry 07.23.03 at 4:22 am

>Cool eh?

And then some.

Comments on this entry are closed.