Re: Self-Archiving and the reaction of publishers

From: Stevan Harnad <>
Date: Tue, 7 Nov 2000 13:22:08 +0000

On Tue, 7 Nov 2000, Bernard Naylor wrote:

> We need to be clear that a communication with deliberately
> limited circulation (category 1 as defined by Stevan
> Harnad) is not published in any sense of the term. It is
> emphatically not "published" in the legal sense - though it
> is protected by copyright, as is any unpublished
> manuscript. Limited circulation of one's thoughts is a
> well understood device for testing one's views on people
> who (one judges) might have useful comments to make towards
> the process of refining them, before they are "published",
> that is, offered to the world at large.

Category (1) included, but was by no means limited to "deliberately
limited circulation." Writing a text down on 1 single copy of paper
falls in category (1) (copyright can be asserted, and protected, and
yes, indeed, this falls under a formal legal sense of "publishing"

Circulating that paper to a small chosen number of colleagues is also
in category (1), including circulating it via email. In the case of a
text that is a report of scholarly or scientific research that is
eventually destined to appear in a refereed journal (the only kind of
text that is at issue in this Forum), the name of such a pre-refereeing,
prepublication (2) publication (1, sic) is a "preprint."

It remains only to add that placing a text in a departmental (paper)
preprint archive, where any visitor can browse it, is also in category
(1). [I don't know about orally presenting it at a conference; I
suspect it is, but we stumble on the print-based connotations of
"publication" and prefer to call it something like "publicising"
instead.) And placing it in a public online archive is also in category

What makes all of these (1) rather than (2) is not the number of people
that do see it (that is always finite), nor even the number of people
that COULD in principle see it (I suppose that is infinite, even with
one piece of paper). The only separator (logically, legally, morally,
scientifically, practically, etc,) between category (1) and category
(2) is the "vanity-press" criterion that Bernard finds so arbitrary
(perhaps it is!):

For this specific literature, and for this literature only (and I hope
it is will be clear that it is not tautological to remind everyone that
the literature is the REFEREED RESEARCH literature), the dividing line
between publication (1) (unrefereed preprints) and publication (2)
(refereed reprints) is indeed refereeing, and certification thereof.

Do not waste time trying to apply this distinction (1/2) to the rest of
the literature (e.g., books); the rest of the literature is
non-author-give-away: Potential revenues (royalties, fees) and their
potential loss must be taken into account for that literature, and
assume a much greater weight in deciding what to call what. Hence what
is defined as "publication" (priority, vanity, etc.) there will have
little to do with what we are discussing here, and everything to do
with saleability.

> With respect to Stevan Harnad's category 2, refereeing is
> not an intrinsic requirement, except in so far as Stevan
> Harnad seeks to make it so. Many intellectual statements
> have been, and are proclaimed to the world, on paper,
> without ever being refereed.

Correct. But what is the point?

First, we are talking here only about the refereed literature (and its
embryological precursors). Yes, unrefereed preprints can report great
things, and be right, and establish priority, and change the world. But
that is the exception rather than the rule. To be taken seriously, most
scholarly and scientific research has to be peer-certified. (The
"refereeing," may vary from field to field, sometimes formal peer
review, sometimes editorial review, but in any case "vetted" rather
than "anything goes," and the vetting is normally certified by a known
and reliable "quality-control tag, e.g., Proceedings of the National
Academy of Science.)

The norm for reporting scientific and scholarly research is to publish
it in a peer reviewed journal.

> It is done without thought of
> reward and to characterise it as "vanity publishing" seems
> to me to be simply "name calling".

To "publish" only the pre-refereeing preprint, be it ever so correct,
would indeed be "vanity publishing," but the rightness (if it was
indeed right, and we had a way to know it was right) would completely
outweigh the vanity.

But such cases, I repeat, are the exception, not the rule. In an ideal
world, we would not need peer-vetting for quality: The right results
would just shine out of their own accord.

> The important and
> distinguishing feature is not that a statement is refereed,
> though I do believe firmly that refereeing does serve a very
> useful purpose, albeit not an intrinsic or absolutely
> essential purpose, in the process of scholarly
> communication. The fundamental feature of scholarly
> publication, which has been well recognised for centuries,
> is that an intellectual statement is offered to the whole
> world so that anyone can test it and judge it.

Correct. But at the scale at which people are doing and reporting
research today (and probably even at smaller scales), the watchword is
"caveat emptor" until it has received a reliable QC tag. (Yes, there
are exceptions, but peer review is a system built to handle the bulk of
the literature, and not to wait and hope for the self-validating,
self-evident exceptions.)

To a first approximation, is a world where so many "intellectual
statements" are being "offered to the whole world so that anyone can
test it and judge it," the researchers of the world, with their finite
time and resources, have to have a basis for deciding which of it is
worth trying to read, judge, and test. One searches in vain for the
reliability of a statement on its sleeve. (Perhaps that's why it's
called vanity press...)

> E-print repositories are a relatively new feature on the
> scene and we do need to settle, without too much delay,
> what they imply. I shall need a good deal of persuading
> that an intellectual statement, refereed or not, which is
> deposited in an e-repository, is not "offered to the whole
> world so that anyone can test it and judge it".

But it is, in both cases! The question is: which statements to risk
taking at their word, in all this mass of statements? That's were the
refereed stage of the embryological transition from (1) to (2) comes

> There is
> therefore very little doubt in my mind how this will be
> resolved. I think we shall in due course have to accept
> that an intellectual statement which is deposited in an
> e-repository is published.

We already do. Published in sense (1).

> Of course, this could well have
> serious implications for the way we view certain existing
> elements in the chain of scholarly communication, in
> particular, print-on-paper journals which publish
> intellectual statements which have previously been
> offered to the world, by deposit in an e-repository.
> Sooner or later, we shall have to get round to facing up to
> that. The sooner the better in my view.

Physics has already faced up to it. Unrefereed preprints are preprints
and peer-reviewed postprints are postprints. When only the preprint is
available, you make do with that, guided perhaps by the author's name
and reputation (based on prior peer-reviewed work!); when the postprint
is available, that is the locus classicus, and the reference point for

    Harnad, S. & Carr, L. (2000) Integrating, Navigating and Analyzing
    Eprint Archives Through Open Citation Linking (the OpCit Project).
    Current Science 79(5) 629-638.

I think Bernard is conflating book publication with
refereed-journal-paper publication, and rare, successful unvetted
exceptions with the vetted rule, in the special case of the refereed
research literature.

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):

You may join the list at the site above.

Discussion can be posted to:
Received on Mon Jan 24 2000 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:56 GMT