Posting Drafts of Papers on the New SSRN-Like Service

David Estlund (Brown) writes:

I recently received an email about an SSRN-like service for philosophy....I’m writing to you to propose some discussion on your blog of this sort of thing. A few issues: Do journals object to final drafts being posted? Do we think it’s a good idea to treat so much work-in-progress as effectively published? Is it fair to cite a posted work in one’s own published work even though the cited thing might not be in final form? Is the shape of the field going to be affected by the dynamics of who participates the most in this technology, even though there’s no referee system that applies to it? I’m not a technophobe, but I do think these are worth continually thinking about. Distributing work in progress is time-honored, but it’s a new world. That was nothing like publishing. Now the lines are getting pretty blurry.

Good questions; what do philosophers think?  Post only once; comments may take awhile to appear, and are reviewed for substance and relevance; signed comments strongly preferred as usual.

Comments

[Dean C. Rowan]

Blurry, indeed. This is, however, largely a matter of degree, rather than kind, because the concerns implicated in Prof. Estlund's questions have been longstanding, as librarians have understood (see, e.g., http://www.firstmonday.org/ISSUES/issue2/content/index.html ) since prior to and at the advent of the 'net's going public.

Not being a philosopher, I'm not aware of journals' preferences regarding pre- (and free) publication online, so I'll move on to the remaining questions taken together. On the one hand, there's nothing extraordinary about references to works-in-progress. The Chicago Manual of Style, 15th ed., provides examples of citations to unpublished working papers from 1980 and 1971, i.e., pre-Web, suggesting that even earlier editions recognized the problem. It also suggests citation styles for lectures(!) and paper presentations at meetings. (Entries 17.215 and 17.217) So the simple answer to the central two questions is "Yes."

On the other hand, all three questions raise issues of authority and reputation that have been complicated by widespread digital dissemination of scholarship. Traditional indicia of authority for scholarship--a publisher's imprint; an index; a big, fat volume plopped on the shelf--more or less disappear online, particularly respecting working papers. They have begun to be replaced with updated indicia: numbers of links, hits, or downloads; the reputation of the repository (SSRN, for instance). One factor of authority shared by both traditional and new media, however, continues to be word of mouth, either face-to-face or via blogs like this one.

Of course, the easy answer to the final question is also "Yes, of course, the dynamics will be affected." Participation in advances in technological modes of scholarly communication will be an important expectation of scholars who hope to influence their fields, at the very least, and in a trivial sense, much as most--but certainly not all, marvelously--have migrated from typewriters to word processors. But a crucial concern regarding technology ought to be, ironically, not so much how we deal with the consequences of the extreme relaxation of barriers to creation, publication, access, and dissemination of work, but how to fend off seemingly random ephemerality. This, I think, underlies much of Prof. Estlund's anxiety. He wants to cite to a working paper posted somewhere on the Web, but worries that the document will have changed or disappeared by the time his own work is (perhaps more traditionally) published. This circumstance, of course, was already an issue with respect to, e.g., references to lectures that hadn't otherwise been transcribed or recorded.

[P.D. Magnus]

"Do journals object to final drafts being posted?"

For most journals, this is explicitly addressed in the publication agreement. Details vary, but it is typical for journals to allow on-line preprints prior to acceptance. Final drafts after acceptance are often allowed only after some delay; e.g. twenty-four months after publication. Whether journals actually enforce such strictures, these are things we explicitly consent to when we sign the forms.

"Is it fair to cite a posted work in one’s own published work even though the cited thing might not be in final form?"

I don't see any problem with citing on-line preprints, just as one can cite unpublished manuscripts or personal communication. Preprints have the advantage that they aren't a resource available only to the authors' confidants; any reader can get their hands on the preprint. For papers that have subsequently been published, however, I always track down the final version.

[Kai von Fintel]

1. Do journals object to final drafts being posted?

Most journals do not object to preprint draft manuscripts being posted. Some even allow postprints. And yet others allow the final published version in all its typeset glory to be freely posted online (open access journals in particular, which include Philosophers' Imprint (http://quod.lib.umich.edu/p/phimp) and Semantics and Pragmatics (http://semprag.org).

2. Do we think it’s a good idea to treat so much work-in-progress as effectively published?

Why not? The system it replaces where preprint drafts were only shared within an exclusive old boys network is much more exclusionary.

3. Is it fair to cite a posted work in one’s own published work even though the cited thing might not be in final form?

What wouldn't be fair is to use such work and not cite it. On the other hand, I suppose if one is going to demolish such non-final work, that wouldn't be very nice. The feedback should have been given first in private, giving the author a chance to correct obvious flaws.

4. Is the shape of the field going to be affected by the dynamics of who participates the most in this technology, even though there’s no referee system that applies to it?

Yes, the field is democratized by making cutting edge work available to everyone with an internet connection, which is a much larger audience than that which has access to subscription journals in university libraries or has access to the old boys network.

[Dennis Whitcomb]

What Kai said.

[Alan Thomas]

I think we are all painfully aware that we supply the content of established journals, edit them, referee for them but then see our academic institution pay out many tens of thousands of pounds/dollars a year to buy all that content back in subscriptions to external companies. The established print journals used to be able to respond that they were a distribution business, but the internet has undercut that argument. Now that the established journals are migrating to the net, it highlights the expense of paying for content hosted on a publisher's server when it could be hosted on one's own university server.

The function of traditional journals is to establish quality control and allow a person to build an academic reputation by publishing in reputable places, but it is not clear to me that once a person has an established reputation that they need to keep re-proving themselves. If I know a person's work and their established reputation I am quite happy to work from drafts on their web pages without waiting for an editorial board to tell me that it is good stuff. I noted with interest Colin McGinn's recent post that he has plenty of stuff he wants to publish but does not want the hassle and long lead in times of journal publication. Quit frankly I wonder if the encouragement of networks like the SSRN won't serve to take the pressure away from the traditional journals, that are already showing signs of strain, and leave publication there to those who either need to establish themselves or those who want to get a job or change their job. I think David Estlund is right: some scholarly practices will have to change to acknowledge this shift. Perhaps like software producers we should all start adding version numbers to our papers - 'On Denoting, Version 8.1', that kind of thing...

[Stephen Latham]

My wife (an economist) informs me that in economics, pre-publication (whether on SSRN or as an NBER or university working-paper) is widely accepted and peer-reviewed journals are increasingly perceived mainly as providing a quality signal to people not already familiar with the work.

Kai von Fintel is right about the democratization. And the democratizing effect is not limited to increasing the number of people who have access to good work; it also increases the readership for good work published by obscure authors or in obscure journals. In law, for example, if you're writing in a given area, it's entirely natural at the early stages of research to use keyword-searches in Westlaw or Lexis or on SSRN to track down the full range of articles on your topic. In the first cut, you get everything from student notes to practitioner pieces to articles in the Journal of Law and the Dog at the Eastern University of Western Podunk. But guess what? Some of this stuff is good, and the broad inclusiveness of electronic topic-searching increases the likelihood that good pieces will be plucked from obscurity via discussion or citation in later publications.

[Brie Gertler]

I'm glad to see this discussion of the Philosophy Research Network. First, a point of clarification: the PRN is not only SSRN-like, but an expansion of SSRN, which has been hugely influential in law, economics, and other fields. Researchers in these fields routinely publish working papers and preprints on the site.

About permissions: SSRN has relationships with most academic publishers -- see http://papers.ssrn.com/sol3/DisplayPipPublishers.cfm -- to allow for publication of abstracts, and also for users to access the published paper, sometimes for a fee. This is from the site:

Publisher Rights: Some of the materials posted to SSRN are provided by publishers, who typically retain copyright to the posted materials. There is a charge to download some of these materials. SSRN policy is that there is never a charge to view an abstract, and that the charge to download a full paper from SSRN cannot exceed the lowest price charged by the publisher for downloads from other sources.

PRN allows for author and keyword searches. Users may also browse topics: Go to http://papers.ssrn.com/sol3/DisplayJournalBrowse.cfm , click to expand "Humanities Research Network", click to expand "HRN Philosophy Research Network", click to expand "HRN Philosophy Subject Matter Journals". (In SSRN lingo, a journal is, in effect, a bundle of entries on a topic.) Most of these can be further expanded to reveal subtopics. This process will be streamlined in the future.

I want to note that we're now in the process of cleaning up the PRN. Most of the current postings were cross-listed from existing SSRN networks before Larry Becker and I came on board as PRN's co-directors. Many of them are now being removed or reclassified.

To submit a paper, go to http://www.ssrn.com and hit 'submit' at the top of the page.

[J. David Velleman]

Although I post my work on PRN (thanks Brie and Larry!), I do not regard it as a substitute for publication, as some of the comments above suggest. The reason is that services like PRN are not in fact archives, because they are impermanent. I can revise or remove my posted work at any time. If others cite or quote material that I have posted on PRN, their readers may in future be unable to find the original source of those citations or quotations.

This problem matters less in some disciplines than in others. In the experimental sciences, an investigator's results are largely independent of the text in which he writes them up. His results advance the discipline by first being replicated or confirmed in other experiments and then being incorporated into ongoing practice. But philosophy is a textual discipline -- a genre of writing. A philosopher's ideas are not so readily separable from the way he expresses them, and so other philosophers must be able to consult the original text.

That's why philosophy is not separable from its history. In the case of experimental sciences, the people who need access to original texts are the historians of those disciplines, who are different from the practitioners. If Pasteur's writings were lost, his contribution to biology would live on in the experimental tradition; the loss would be felt only by historians of biology. But if a philosopher's writings are lost, philosophy itself suffers, because the philosopher's contribution is less easily separated from his words.

The recent trend toward conducting philosophy in ephemeral venues such as blogs and online postings, without entering it into the permanent published record, will have the result that future ages will view us the way we view some of the Presocratics. To the twenty-second century, twenty-first century philosophers will appear the way Heraclitus does to us -- as only barely accessible, through fragments gleaned from secondary sources.

If you think of publication primarily as a medium for communication among contemporaries, then of course you will find it an unbearable "hassle". But if you think of publication as creating the discipline's permanent record, then the screening and polishing that make the process so frustrating begin to make sense.

[Kai von Fintel]

Hmm, which comments were suggesting that posting manuscripts on the web is a substitute for publication? What I and others were saying is that posting manuscripts on the web is a good thing. But it doesn't need to supplant publication in peer-reviewed journals (although in some disciplines, things are possibly moving that way). The journal I edit is very much centered around high quality peer reviewing, but I don't see that at all in conflict with liberal prepublication posting on the web, thus increasing communication in the discipline.

[Thom Brooks]

Dave Estlund raises some excellent questions, some of which we discussed at the recent APA-Eastern. My own view is that the SSRN is absolutely terrific. I regularly post drafts of my own work and regularly receive helpful feedback (granted, on some papers rather than on others). In addition, there is a lot of interesting material readily available for free.

My own view is that one should never cite the drafts of other people's work, certainly not without their permission. However, it is well worth noting that several times I have seen my own drafts cited in law reviews. (And, at least for the present, it is being cited in law reviews --and not in philosophy or political science journals-- that is a potential worry, at least for now.) My trick has been to post drafts that are more or less complete. Thus, having this work cited has not caused me any problems. Nevertheless, others might join me in sharing a bit of caution on what is posted on the SSRN as it can find itself cited and even distributed.

[J. David Velleman]

To Kai von Fintel: I was thinking, for example, of this comment, above:

"I noted with interest Colin McGinn's recent post that he has plenty of stuff he wants to publish but does not want the hassle and long lead in times of journal publication. Quit frankly I wonder if the encouragement of networks like the SSRN won't serve to take the pressure away from the traditional journals, that are already showing signs of strain, and leave publication there to those who either need to establish themselves or those who want to get a job or change their job."

Or this:

"My wife (an economist) informs me that in economics, pre-publication (whether on SSRN or as an NBER or university working-paper) is widely accepted and peer-reviewed journals are increasingly perceived mainly as providing a quality signal to people not already familiar with the work."

Both of these comments strike me as ignoring the archival role of publication.

[Jeff Horty]

I'm all for sharing drafts, and also of course respect the archival and especially the quality-control work of carefully edited journals, but I have a more elementary question that I hope someone can help me with. What is the actual advantage of posting drafts and preprints and so on at one of these central sites, such as SSRN or PhOnline or the PhilSci Archive at Pitt, as opposed to simply maintaining a personal web site, where anyone with a search engine can find your work? There must be an advantage or these services wouldn't be so popular, but I'm having trouble seeing it. Help!

[Thom Brooks]

For Jeff:

The reason why sites like SSRN are as popular as they are is because when you post a paper on your website the only people who see it are people who have a peek at your site. The traffic that we all get is much less than, say, SSRN. When posted on SSRN, there a lot of people using the site and your chance of having your work noticed by more people is greater. Plus, with the increased traffic, comes the greater chance that someone may be able to offer some great advice to help with a paper in draft. (I found this out particularly with my publishing advice papers.) Is the mystery now revealed?