What Counts As Pre-Publication?
Dale Miller (ODU) noticed that Public Affairs Quarterly has the following “Pre-Publication Policy“:
Public Affairs Quarterly will not publish material that has already appeared elsewhere. This is not at odds with authors sharing their papers with selected individuals whose comments they would welcome or who they wish for other reasons to inform about their work. But it precludes pre-publication with broadcast dissemination, alike in print or electronically in open-access forums such as Academia.edu.
Presumably posting papers to SSRN and ResearchGate is ruled out by the policy, too. Is posting a paper on one’s own website in the same category, or doing so and making sure there’s a link to it from PhilPapers?
A quick scan of the other University of Illinois Press philosophy journals shows that History of Philosophy Quarterly has the same policy, while others, like American Philosophical Quarterly, The Journal of Animal Ethics, and The Pluralist do not, so it at least does not appear to be a publisher-mandated policy.
Professor Miller asks whether other journals have such a policy and comments: “It seems very restrictive for a journal to say that it won’t consider papers that have ever been posted online.”
Publishers and journals need to evolve on this point, particularly as the Pacific APA now requires authors submitting a paper to grant a worldwide right to post a pre-print online prior to the meeting. I think the rationale for this is good; every pre-read conference I’ve ever been to has been better in terms of quality of feedback for authors than every non-pre-read conference I’ve been too. But I would think journal editors wouldn’t want to systematically exclude such a huge number of potential papers from consideration.Report
I can think of one (and only one) good reason for the quoted policy. Papers posted “with broadcast dissemination” are easily retrieved by search engines with only a few phrases. For this reason, blind refereeing is dying, if not already stone cold dead. Referees simply cannot resist the temptation to search for a paper online. If we value blind refereeing, we should support this sort of policy. Or maybe the discipline would like to allow for “peep holes” in the veil of referees’ ignorance. Individual authors could still ensure that their work would be blind refereed by withholding it from searchable web sites, including their own personal pages.Report
The thought about blind refereeing occurred to me, but the policy doesn’t seem broad enough to solve that problem. All it takes to compromise blind review is listing the paper as work in progress on your CV and then posting THAT online. This the policy doesn’t prohibit. In fact, if you Google the title of the paper for which I was considering PAQ as a possible home you’ll find it listed on my university’s research office page, since I got a modest internal grant to work on it.
When I mentioned this on FB one of my friends pointed out that medical journals, where much of his work appears, have similar policies. When I looked into this, I found a story about someone’s having a journal retract an acceptance because one of his co-authors had posted their article as a working paper on the website of a center with which she was associated. The total number of downloads from that website was 21, but that didn’t matter. So the norms about this seem to be very discipline specific. For better or worse, PAQ seems to be importing norms from some other disciplines into philosophy.Report
Referees should be aware that the internet often allows authors to identify *them* as well. Normally, all I have to do to discover who my (likely) referees are is check the analytics on my webpage. Normally I will get a few hits on my site a week or so after sending an article out to a journal from very specific locations with the title (or a partial title) of the article under review as the search string. A cursory examination of the location will reveal a university with a philosophy department and typically only one (or two) people in that department who could be searching for my as-yet-unpublished work. Voila! I have discovered my likely referees. Is it fallible? Definitely. However, it is often the best explanation for the strange combination of the search term + the hit on my site.Report
Dale, you could just change the title of the paper, to restore anonymous review, if the only leak was a leak of the title.
I hope David Velleman is right, because that at least makes some sense. The policy otherwise seems foolish. Surely they miss out on a lot of great papers with that policy.Report
@ejrd For every philosopher who agrees to referee a paper there are a number who don’t. Even if you are correct that the hits are connected with refereeing, there’s no guarantee that the hit on your website is the philosopher who eventually agreed to do the refereeing.Report
That’s a good point, although the title may be the best part of this paper!Report
I presume — without any insider information and without any good claim to be a copyright expert with respect to academic publishing — that part of the motivation might be an intellectual property claim. If the paper has already been published (and posting even on a personal website presumably counts as publishing), then the journal’s claim to be _the_ home for the paper is cast into doubt. It is of course a possibility that the first publisher (sometimes presumably the author via her own website) could transfer any intellectual property claims to the journal. But if the first publication was not restricted — e.g., if during the period of the first publication, I a third party outsider could download a relatively unrestricted copy of the document — then my guess is that the journal is thereafter (legally?) burdened by that lack of restriction. Perhaps the journal can say, “You may not copy the paper from _here_,” but it cannot say “You may not copy the paper,” since you could copy it from the permissible, prior source.
For similar reasons, I’ve often wondered why journals are so lax about permitting authors to upload penultimate versions, though I’m certainly glad that they do.Report
I hope David V. is wrong to this extent – that most referees, in fact an overwhelming majority of referees manage to referee without Googling papers. I kind of think he’s at least wrong to say that anonymous refereeing is dead. (It may be wounded.) For all that it may be the reason for the policy, though my theory is that it has more to do with wanting intellectual property rights.Report
Another FB friend asked about the new “Sessions” capability on Academia.edu. It turns out that you can create a session for a paper without making the paper generally available. Only those whom you invite to participate in the session will have access. This, I think, should qualify as sharing only with “selected individuals” as opposed to “broadcast dissemination.”Report
I recently had an acceptance at the Australasian Journal of Philosophy nearly rescinded because I posted the paper to PhilPapers after the paper was accepted at AJP (but before it was published by them). The journal’s publisher wrote that they “would not be able to publish this paper based on the fact that it has already been published online on philpapers”. In the end the acceptance was not rescinded, but I can “post the accepted manuscript to philpapers [only] after the embargo period of 18 months post publication in print.” I believe others should be aware of this policy.Report
Yikes! PAQ accepted one of my papers last year, after which I uploaded it on my Academia page. Having just learned about this policy, I’ve promptly taken it down, as it hasn’t been published in an issue yet.Report
I don’t see how intellectual property rights can be at the bottom of this. Authors own the copyright in their work no matter how they disseminate it online. They can transfer the copyright to a publisher as they please (though they needn’t and shouldn’t do so in this day and age — see below). This transfer will be valid irrespective of how the work has previously been disseminated.
About transferring copyright: Most publishers will initially send a contract that transfers copyright to them, and that includes no permissions for the author to post the work on his or her own website or archive. Authors should resist signing such a contract: it’s just the publishers’ initial offer. The publishers also have — in their desk drawer, so to speak — a contract that gives them a license to publish the work while leaving the copyright in the hands of the author, and that grants the author the right to post the work online in venues of particular kinds. Authors should not be afraid to ask for this contract, or to edit the offered contract before signing it. A journal that has accepted a paper will not dump it just because the author pushes back against the publisher’s legal department — though the author may in the end have to settle for less than her or she deserves.Report
@Ben Bronner — that’s a dangerous turn for the AJP, given the ARC’s Open Access Policy:
Any publications arising from an ARC supported research Project must be deposited into an open access institutional repository within a twelve (12) month period from the date of publication.
In cases where researchers may not be able to meet the requirements because of current legal or contractual obligations, Final Reports must provide reasons why publications derived from a Project, Award, or Fellowship have not been deposited in an open access institutional repository within the twelve month period.
The Policy has been incorporated into all Funding Rules and Agreements released after 1 January 2013. The Policy does not apply retrospectively to pre-existing Funding Rules and Agreements.Report
Sam Baron: AJP’s publisher stated that PhilPapers was “effectively an abstracting and indexing service, not a pre-print server.” So perhaps the ARC requirements could be met by posting one’s papers to a pre-print server other than PhilPapers.Report
@Ben So it’s the abstracting and indexing aspect that worried them? How very strange. That seems to put paid to the idea that it’s intellectual property per se that’s the problem. At any rate, I imagine that posting on an archive that’s open access but not picked up by any of the indexing services would get around the problem. Still, that’s no easy matter, given how thorough some of the indexing algorithms are.Report
@Ben Bronner, as I understand things lots of journals object (in theory) to placing the accepted version of a paper in a repository, but allow an earlier draft to be archived. Was the version you uploaded the final version? Do you know if the AJP allows drafts to be posted?Report
Hi, Neil. The version I placed on PhilPapers was an extended version of my paper, which included the content in the final, accepted version. Relevant information from Taylor & Francis is here:
The “Author’s Original Manuscript” can be shared as much as you like — that appears to be the version of the paper prior to modification in light of referee comments.Report
I am not a copyright expert as compared to litigating copyright lawyers. But my understanding is that if a copyright owner publishes something in a way that allows third parties licenses, any later transfers of that original owner’s interest will be subject to those licenses. If those licenses could be retracted, of course, then that possibility of retraction can be exercised at the time of transfer. But that is not the normal way for licenses to work.
So, if I publish something, possessing copyright, but freely allowing download, transfer, and dissemination, and you download it, transfer it around, and disseminate it, then when I later transfer my remaining interest in the copyright to some third party, that should not affect your license to transfer and disseminate.
This is not to say that the transfer wouldn’t be valid. It would of course be valid. But you cannot transfer what you do not have. So if the original owner lacks the full _exclusive_ copyright, then the original owner cannot transfer the full _exclusive_ copyright.
All of this is meant to be descriptive. And I completely agree that authors should not just accept default terms. I’m just offering a bit of legal speculation as to why we’re seeing the behavior from journals.Report
This AJP policy also applies to academia.edu and other academic social networks (I inquired about this after I got an acceptance email that explicitly stated that the preprint paper couldn’t be posted on PhilPapers). As far I understood, a preprint version of the paper could only be posted on a personal or departmental website.Report
AJP has been mentioned above. Here is the relevant section from an AJP acceptance letter: “What is permitted at PhilPapers or academia.edu, as at your own personal or departmental website, is a mention of your paper as forthcoming or as published, plus a link to the paper at the AJP website. The paper is available there, under ‘Latest articles’, once it has been proofed and before it has been published in an issue.” That version of the paper is the final one, the one that will subsequently be assigned to an issue.
Hopefully, that clarification allays the main concern; for it is clear that, via any such link, an accepted AJP paper can be reached very easily. Neither I (as AJP Editor) nor the Australasian Association of Philosophy (who owns AJP) wishes to impede — among our peer philosophers, after all — the flow of those ideas that will be appearing in AJP. That would be in no one’s relevant interest.Report
Professor Hetherington: I don’t see why linking to the official version would allay the concern. It lets subscribers to the journal get the article (which they could do anyway) but it’s no use to non-subscribers. The reply, I suppose, is that professional philosophers can get access through their institutions’ libraries. But some interested readers won’t have such access, and even scholars with such access on-campus will sometimes be doing research from off campus.Report