Re: Financial Times Article on Self-Archiving: 23 July 2001

From: Stevan Harnad <>
Date: Wed, 25 Jul 2001 14:09:06 +0100

For the most part, I find that Albert Henderson's postings are their
own best rebuttals, hence best left to undo themselves unanswered.

But I am also continually stunned by the sluggishness with which the
trivial token -- that the REFEREED literature can and should be freed
of all access/impact barriers right now through author/institution
self-archiving -- is dropping in the minds of researchers. It remains a
profound puzzle to me, because all the reasons so far adduced against it
have been so equally obviously either elementary misunderstandings or

But given that it is all not nearly as trivial or obvious to most
researchers as it seems to some of us, I shall spell out the fallacies
inherent in Albert's comments:

On Tue, 24 Jul 2001, Albert Henderson wrote:

> Harnad wants to have his cake and eat it too ...

Correct. The self-archiving of all REFEREED research amounts to having
one's (peer-reviewed) cake and eating (freeing it from impact/access
barriers) too. That is precisely why the original proposal was
(self)-labelled "subversive," but we can all be grateful to Albert now for
noticing and pointing this out for those who may not have noticed:

One small correction, though: The ones who are to have their cake and
eat it too are the researchers of the world -- the authors of the
annual 2,000,000+ articles in the world's 20,000+ refereed journals. (I
happen to be one of them, but apart from that, I have no special
interest; in particular, unlike Albert, I am trying neither to protect
nor to garner any revenue streams. It is all only in the interest of
research itself.)

> sh> But virtually all of the self-archived preprints in arxiv are
> sh> submitted to refereed journals, revised in accordance with the referees'
> sh> recommendations, and if the author judges the changes substantive, the
> sh> corrected final draft is self-archived too; otherwise, the reference is
> sh> merely updated to make it the formal journal bibliographic citation.
> "virtually" doesn't make it so. Garvey and Griffith found
> "two thirds of the technical reports produced in 1962 had
> not achieved journal publication by 1965, and, apparently,
> the contents of the vast majority of these reports were
> never submitted for journal publication. Many authors of
> such reports indicated that 'no further dissemination of
> the information was necessary.' ... This raises some
> questions about the ultimate value of the information in
> these reports and its relevance to the established body
> of scientific knowledge." [from Garvey, COMMUNICATION:
> THE ESSENCE OF SCIENCE. Pergamon 1979]

Now what on earth could this spectacular non-sequitur imply for the
validity of the thesis that the refereed literature can and should be
freed as of the 1990's (since online self-archiving has become

(1) that the authors of, say, the 150,000 eprints (= pre-referring
preprints + refereed postprints) self-archived in the Physics Arxiv
have been doing something irrelevant with material of questionable
ultimate value? (That will come as a surprise to the 60,000+ daily
users of the Arxiv...)

(2) that the proportion of the PREprints self-archived in Arxiv that
are then submitted to refereed journals is not at least as high as the
proportion of preprints NOT self-archived in Arxiv that are then
submitted to refereed journals?

(3) that the proportion of the PREprints self-archived in Arxiv that
are then accepted by refereed journals is not at least as high as the
proportion of preprints NOT self-archived in Arxiv that are then
accepted by refereed journals?

(4) that the proportion of articles accepted by journals that have been
previously self-archived as unrefereed preprints or subsequently
self-archived as refereed postprints is of lower value or relevance
than the proportion that have not?

In other words, just what point does Albert think is made by these
old G & G data on overall proportions of submissions and acceptances
in the 60's? And what on earth do these pre-archiving statistics have
to do with the face-valid benefits of freeing the refereed literature
from all impact/access barriers online through self-archiving, now that
it has at last become possible?

> A recent study produced similarities to this data and
> also called into question the value of citing informal
> papers as if they were a part of the formal literature.
> [Callaham, M.L., et al. J A M A 1998. 280:254-257]

Again, what has this to do with anything? And in particular with the
benefits of freeing the refereed literature through self-archiving?

The value of pre-refereeing preprints themselves, and of citing them,
is a scholarly/scientific matter, and probably differs from discipline
to discipline. But we are talking here about the REFEREED literature,
and the only way unrefereed preprints even got into the discussion was
by way of a reminder that most of the unrefereed preprints
self-archived in Arxiv are submitted to and eventually accepted by
journals (with no implication that that proportion will be either higher
or lower than the overall baseline proportions for preprints, whether
self-archived or not: that irrelevant baseline is the only datum the
1960's G&G data estimates; ditto for the virtues/vices of citing
unrefereed papers.

In fact, forget about those papers that are NOT eventually submitted to
and accepted by refereed journals. THEY are irrelevant, and this
discussion is not about them.

> I don't think it is fair to the student, the young
> researcher (and research sponsors) to encourage the study
> of preprints that may never be submitted, much less
> published, in the same breath that demands preparation
> carefully rooted in the formal literature.

And who, may I ask, is doing this reprehensible thing (encouraging the
study of unrefereed papers)? This Forum is about freeing refereed
papers online. The unrefereed papers are merely a bonus. Caveat

Stevan Harnad
Professor of Cognitive Science
Department of Electronics and phone: +44 23-80 592-582
             Computer Science fax: +44 23-80 592-865
University of Southampton
Highfield, Southampton

NOTE: A complete archive of the ongoing discussion of providing free
access to the refereed journal literature online is available at the
American Scientist September Forum (98 & 99 & 00 & 01):

You may join the list at the site above.

Discussion can be posted to:
Received on Wed Jan 03 2001 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:11 GMT