Re: ePrint Repositories

From: Stevan Harnad <>
Date: Fri, 26 Jan 2001 10:28:21 +0000

Many thanks for the kind words about, but here I'd like to
make some comments on a passage not related to eprints:

    On Wed, 24 Jan 2001,
            Catherine Candee,
            Director, Scholarly Communication Initiatives,
            California Digital Library, University of California,

> We believe that discipline-based archives which encourage
> self-publishing hold great promise for stimulating open dissemination
> of scholarship, for focusing and simplifying the problem of persistent
> access to that scholarship, and for expanding the possibilities for
> integration within and across disciplines.

As you note elsewhere, the CDL projects, like other similar projects,
are experimental; they will evolve with feedback. As a consequence, I
wonder whether it is a good idea to prejudge and align them so
explicitly with "self-publishing".

This is a controversial matter, and, more important, an empirical one,
that cannot be pre-judged in advance. At the present time, scholarly and
scientific research is not self-published (except for a tiny minority of
vanity-press literature of doubtful value): It is published either in
peer-reviewed journals or in monographs that have been reviewed and
vetted by publishers' readers. (The latter is not really peer review,
and as I have no expertise in monographs, I will not say anything more
about it; I am speaking now only of refereed-journal research.)

There are several variables at issue, in the new PostGutenberg world.

The first is on-paper vs. on-line. (1)

The second is for-fee vs. for-free. (2)

And the third is refereed vs. unrefereed. (3)

No further data are needed to demonstrate the advantages of (1) on-line
access over on-paper access.

The advantages of (2) for-free over for-fee are also amply demonstrated by
the very heavy use of the few free eprint archives that exist so far, e.g.:

But (3) the refereed/nonrefereed issue is far from clear. Do not be
misled by the fact that Eprint archives have often been (mis)named
"Preprint" archives. The pre-refereeing preprint is an earlier
embryological stage of a paper, to be sure; hence it is naturally the
first one to be self-archived:

But from the very outset, and to the present moment, virtually every
paper in (for example) the Los Alamos Physics Archive, has also been
submitted for refereeing and eventual publication in the peer-reviewed
journals of physics. Nothing whatsoever has changed in this fundamental

The pre-refereeing preprints are useful to have in advance, to be sure,
and are heavily used and cited. But it would be highly misleading to
describe this process as "self-publication" (any more than the pre-Net
practise in Physics of circulating on-paper preprints constituted
"self-publication"). What the Physicists are doing, in the interests of
speeding and facilitating research progress, is making all their
findings -- pre- and post-publication -- available as soon as possible;
but the publication and authentication process is still proceeding
apace, exactly as it always has; and those findings all continue to be
answerable to classical peer review (whose "invisible hand" is surely
also keeping up the quality of the preprints themselves):

There are some ongoing experiments with "open review," where it is
self-appointed commentators' feedback that is being used to validate
preprints, but these experiments are far too small, and it is far too
early, to judge whether they have been successful in generating papers
of quality and useability anywhere comparable to those generated by
undergoing classical peer review.

Yet it is this large, important, and unanswered empirical question
whose answer you are in effect prejudging, in aligning your own
experiment with "self-publication" -- rather than the much more neutral
"self-archiving" that (I think) it really is, and ought to be.

Self-archiving has face-validity, for it covers, value-neutrally, both
the self-archiving of pre-refereeing preprints and the self-archiving of
refereed, published postprints. Moreover, it is the exact, theory-free
description of what it is that most people are currently doing in, say,
the Physics Archive or CogPrints

"Self-publication" is an entirely different matter, hypothetical, and
based on an (in my opinion) incorrect interpretation of what
eprint-archivers are actually doing today, when they self-archive their

It is not out of the question that self-archiving could turn out to
lead to self-publication, and that peer review will accordingly be
phased out. But that is far from the case now (and, in my opinion, it
is unlikely to be the actual outcome, either, in the long run). So, at
the very least, archiving one's pre- and post-publication work on the
Web today should not be pre-emptively labelled or treated as
"self-publication" until and unless it actually turns out to be that.
"Self-publication" should be characterized as a hypothetical
possibility, not as the description of what is actually going on now.

The reason I think this is so important is that self-archiving in the
first and second sense above [i.e., the on-line dissemination of one's
own pre- and post-publication papers (1), for free for all (2)], though
optimal and inevitable, is vastly overdue, and progressing far too

There are many possible explanations for why the literature-freeing is
growing so slowly, but one strong candidate is that it is being
confused with self-publication! And people simply do not wish to switch
their intellectual and practical allegiance from their current,
classical, quality-controlled, peer-reviewed journal-certified
publication to an on-line vanity press.

So portraying and seeing it as that sort of trade-off creates a false
opposition in people's minds -- and hence they don't self-archive!

In contrast, my own "subversive" proposal has always been cast in the
theory-neutral terms of self-archiving: We can all self-archive both
our prepublication preprints and our postpublication postprints,
freeing them for one and all, without giving anything up, and without
committing ourselves to any untested speculations about quality-control,
peer-review, or self-publication.

< The Subversive Proposal>

Well, that's about it, for the cautionary note I wanted to sound about
the a-priori (and perhaps prejudicial) language of "self-publication."
It remains only to add that we should also not get stuck on the legal,
technical definition of "publishing," which of course has nothing to do
with our formal and professional standards of what counts as a
publication: In the technical, legal sense, even writing something down
by hand on one piece of paper is "publication" (and accordingly
protected by copyright). Hence, a fortiori, so is broadcasting it on
the Web.

But that is not what "publication" means in the academic/scholarly,
"publish-or-perish" sense: There, the established quality-control
filter and certification is assumed, and "self-publication" would
rightly be seen as mere vanity press.

Last, having lamented the slowness with which the scholarly/scientific
world is moving toward the optimal and inevitable, I could be
upbraided, in rebuttal, to the effect that, after all, my own call for
"subversive" self-archiving has so far been just as unavailing as this
alleged misnomer and deterrent, "self-publishing"!

But that is precisely what I am now hoping that the availability of
OAI-compliant, interoperable, distributed institution-based
eprint-archive-creating software will at last remedy,
hastening us, finally, along the road to the optimal/inevitable.
Central, discipline-based archives obviously were not enough. (Even in
Physics, at the present linear growth rate rate, it will take a decade
before the full annual literature is being self-archived; the rest of
the disciplines are way behind even that.) So let's hope distributed,
institution-based self-archiving will now put us over the top at last
(and let's not put that at risk by calling it "self-publication"!).

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 - 19:46:00 GMT