Self-Archiving in a Repository is a Supplement, not a Substitute, for Publishing in a Peer-Reviewed Journal

From: Stevan Harnad <amsciforum_at_GMAIL.COM>
Date: Wed, 4 Mar 2009 07:48:02 -0500

[The following is a reply to an anonymous query to CogPrints
inquiring whether a deposit in CogPrints counts as a citable
publication.]

      Do academics refer, in papers, to what is in Cogprints,
      as they do to journals? Is it a functional substitute for
      a journal in the sense of establishing priority and being
      referrable?


Neither a Central Repository (CR) like Cogprints -- nor the preferred
option, which is your own university's Institutional Repository (IR)
 -- is a functional substitute for publishing in a journal. It is a
supplement to publishing in a journal, providing free access (Open
Access, OA) for those who do not have paid access to the journal.
Academics refer, as always, to the journal article, but they can add
the URL of the OA Repository version for access purposes.
In addition, before publication, even before submission, one can
deposit the unrefereed "preprint: of the article in an OA Repository,
in order to elicit feedback as well as to establish priority. The
preprint too can be cited, as always, as "unpublished manuscript",
but its repository URL can be added for access purposes.

      I am wondering because we have a paper which we wish to
      place in a journal but it looks as if it will take some
      time (two journals have called it "interesting and
      worthwhile" or words to that effect, but said that they
      don't publish that kind of paper).


Repository deposit is definitely not for papers that cannot meet the
peer-review standards of journals; the "preprint" is not a preprint
if it will never be acceptable to a journal. (However, an OA
Repository is an excellent way to make the research data on which a
published article is based openly accessible too; normally journals
do not have the space or the means to publish the data along with the
article. An OA Repository is also a good way to provide supplementary
information about a published articles; it can also provide access to
postpublication revisions, and updates, and even unpublished
commentaries on other articles and commentaries -- but the rather is
more like blogging than formal publication.)

      Does publication in Cogprints lead to plagiarism of the
      article? Or does it lead to protection of the article,
      through establishing priority?


For published articles, making them accessible online does make it
somewhat easier to plagiarize them than if their text is only
accessible on paper -- but it also makes it easier to detect, expose,
and punish plagiarism. 
This is also true for not-yet-published preprints, because the
repository exposure and time-stamp establish priority publicly and
demonstrate that the plagiarized version came later then the original
preprint. This is true for the verbatim text. But if someone simply
steals the ideas (i.e., the content, not the verbatim text), and then
publishes them, using other words, while the preprint just remains
unpublished, then claims of priority by the author of the unpublished
preprint relative to the published article are far less effective.

The moral is: If you have ideas you think someone else might steal,
submit them for formal publication in a journal, and circulate them
only informally to known peers until the paper is accepted. Do not
make the draft publicly accessible until it is accepted. (That said,
most published articles are little read and less cited, so paranoia
about plagiarism is perhaps over-optimistic!)

       Please let us know what you can of the status of
      Cogprints compared to journals.


Comparing OA repositories to peer-reviewed journals is comparing
apples and oranges (or copies to originals): Journals referee
preprints, require revisions (or reject), certify successful revised
postprints as having successfully met their established quality
standards, and provide access to the refereed, accepted, certified
draft. Repositories only provide access. That is not the same thing
as journal publication; an unpublished draft is an unpublished draft
whether or not it is openly accessible online.

Stevan Harnad
Received on Wed Mar 04 2009 - 12:48:34 GMT