Financial Times Article on Self-Archiving: 23 July 2001 Stevan Harnad 24 Jul 2001 18:30 UTC

Here are some comments on today's Financial Times article on

> Financial Times; Jul 24, 2001 By RICHARD POYNDER

> A mission to free scientific ideas:

Not just ideas, and not just scientific: There are at least 20,000
refereed journals published annually, in all disciplines and around the
world, publishing at least 2 million articles per year, in the sciences,
the humanities, the arts. All of these need to be freed from all access
barriers on-line, today. It could have been done yesterday, and
posterity will chide us. Researchers, and research itself, will be the
greatest beneficiaries.

The title should have been "... to free refereed research."

>   championing do-it-yourself scholarly publishing using the internet

Nothing of the sort. And calling it "self-publishing" instead of the
SELF-ARCHIVING (of refereed, PUBLISHED research) that it is simply
perpetuates one of those trivial but persisting misunderstandings that
is holding us back from the optimal and inevitable.

Self-publishing is vanity-press; it bypasses peer review; and it would
put the quality of the entire refereed research literature in question
and at risk. I am and always have been a staunch opponent of
self-publishing (of unrefereed research).

In this special area (refereed research publication), "publish" is
synonymous with "publish in a peer-reviewed journal." It is also this
sense of "publish" that is implicit in the slogan "publish or perish."
Scholarly/scientific publication is not, and never has been, a matter
of handing out pamphlets in Hyde Park -- or all of cyberspace, for that

   Harnad, S. (1998/2000) The invisible hand of peer review. Nature
   [online] (5 Nov. 1998) Longer
   version in Exploit Interactive 5 (2000):

This widespread but incorrect and misleading confusion of
self-archiving with self-publication stems from another persistent
error: the incorrect notion that the Los Alamos Physics archive is or
ever has been merely an archive for the self-archiving of unrefereed
preprints: Yes, this earlier embryological stage of research is often
self-archived there TOO, and if so it is invariably self-archived first
(because it comes earlier in time).

But virtually all of the self-archived preprints in arxiv are
submitted to refereed journals, revised in accordance with the referees'
recommendations, and if the author judges the changes substantive, the
corrected final draft is self-archived too; otherwise, the reference is
merely updated to make it the formal journal bibliographic citation.

Moreover: This preprint + corrigenda strategy is a very important
component of the legal strategy for circumventing any attempt by
vendors to prevent self-archiving by forcing authors to sign
over-restrictive copyright transfer agreements.

>   ... with a polemical style more reminiscent of a revolutionary than
>   a scientist, [Harnad] outlines why he has spent the past eight
>   years trying to persuade fellow researchers that, rather than
>   relying on scientific journals to disseminate their ideas, they
>   should "self-archive" their papers directly on to the internet.

I wish it were as easy to open researchers' eyes to the optimal and
inevitable, and the way to get there, as it is to report research
findings. Alas, trivial but persistent misunderstandings like the ones
being straightened out in this very posting make it necessary to adopt
an active advocacy stance rather than a merely passive expository one
-- if this generation of researchers is not to retire without tasting
the benefits of the optimal and inevitable:

>   Unlike most writers, he explains, scientists do not sell their
>   texts but hand them over to publishers for free. "The aim is to
>   report their findings to their peers and contribute to the ongoing
>   cycle of creating more knowledge. They don't want to make money
>   from their texts but to reach as many minds as possible."

Lest we cast these monkish researchers as even more disinterested than
they really are: We are not saints. We do want income.

But SELLING our refereed research articles is not only no way to make
money, but it in fact blocks the real source of our kudos and revenue,
which is research impact and uptake, by making our work inaccessible to
all potential users who are at institutions that cannot afford the
access tolls. That is why (unlike all other authors), we give them
away (and always have done, even in the form of on-paper offprints
mailed to all requesters at our own expense):

And make no mistake about it, it is a fact about ALL 2,000,000 annual
articles in all 20,000 of the refereed journals alluded to above, that
MOST of the potential users of those articles cannot access them. This
is as true of the Harvards as it is of the Have-Nots:

>   The problem is that publishers then sell those papers back to
>   academic institutions in the form of journal subscriptions. Aside
>   from discomfort with the principle that institutions should buy
>   back something their researchers have given away, Prof Harnad
>   worries that the "financial tolls" of journal subscriptions
>   represent an unacceptable barrier to the free flow of scientific
>   ideas.

I have no problem with "buy-back" (especially with "added-value"
add-ons) of whatever institutions can and wish to pay for. My only
problem is that a vanilla version of the refereed draft is not yet
available for free online IN ADDITION to those options.

And this is not just about the "free flow of scientific ideas": It is
about the free flow of refereed research, which is an author give-away.

There is plenty of "science" in books, for example, that authors still
seem to prefer to be paid for writing, rather than allowing it to "flow
freely" (although some of this may change, for the more esoteric work
currently reported in books rather than journals).

    Harnad, S., Varian, H. & Parks, R. (2000) Academic publishing in
    the online era: What Will Be For-Fee And What Will Be For-Free?
    Culture Machine 2 (Online Journal)

It is a mistake to cast the special case of the refereed research
literature, which is and always was and will be an author give-away,
with any general calls for "freedom of information" or even "freedom of
scientific information." Some information producers will always prefer
to be paid, otherwise they will prefer not to produce it at all.

>   Prof Harnad has become something of a bete noire for publishers,
>   who dismiss his call for do-it-yourself scholarly publishing as a
>   pipe dream. It could never, they argue, provide researchers with
>   the reach and punch of powerful branded journals.

This is a false opposition. All this is not about self-archiving VERSUS
refereed publication: It is about self-archiving OF refereed

I suppose anyone who points out (and even tries to accelerate) a very
natural technological development that may eventually dry up certain
classical revenue streams is a "bete noire" for the vendors whose
revenue streams are at risk. But there is no point cavilling at the

Either this development (the freeing of the refereed literature by
author/institution self-archiving) is indeed (1) attainable (now), (2)
optimal for research and researchers, and hence (3) inevitable (as I
and others have suggested), or it is not.

If it is not, then it is incumbent on the defenders of the current
revenue streams and the status quo to show how and why it is NOT
attainable or optimal (for I and others have already described how it

Perpetuating the non-sequiturs and misunderstandings like the above one
may be one way to delay the inevitable, but it will not stop it.

>   While the corrigenda strategy has never been tested in the courts,
>   Prof Harnad dismisses any suggestion that researchers could be
>   sued. "What would a publisher hope to gain from taking to court a
>   poor author who was never even paid a penny for their work in the
>   first place?"

What I said to Richard during this interview (and he left out) was that
10 years and 150,000 papers' worth of confirmation of this in the
Physics archive sound like evidence enough that vendors know there is
no point in trying to test taking authors to court for self-archiving
their own give-away work: All that would do would be to hasten the
inevitable, by making glaringly obvious the massive conflict of
interest in pitting what is best for research against what is best for
publishers' revenue streams, without being able to summon even the
slightest hint of the virtue of necessity in defence of perpetuating
this needless trade-off!

The above quote merely suggests WHY there would be no point in trying
to take giveaway authors to court. (And by the way, it should read:
"... never even BEEN paid a penny for their work...". At the root of
this is that all-important author give-away/non-give-away divide that
people keep forgetting, or failing to understand, and that profoundly
separates this anomalous literature from all the rest in the
PostGutenberg Era.

>   Founded in 1991, was created to enable physicists to
>   self-publish their pre-prints while waiting for the long-drawn-out
>   peer review process to be completed, with the aim of speeding up
>   the research cycle. Today the site hosts around 150,000 physics
>   papers, with another 2,500 added each month.

Just a correction: Physicists SELF-ARCHIVE their preprints; the
PUBLISHED version is as imperishably the refereed one today, as it was
before the advent of self-archiving, when physicists circulated (and
still did not thereby "publish") their preprints on paper.

>   The son of a social democrat member of the Hungarian government,
>   Prof Harnad was born in Budapest in 1945. When the Communists took
>   control in 1948 his family left the country illegally, settling in
>   Canada.

The "revolutionary" and "socialist" quips are good fun, but in all
seriousness, kids, this has nothing to do with socialism and revolution.
It's just a natural evolution in the dissemination of give-away
research findings...

>   Critics accuse Prof Harnad of being overly dogmatic and of
>   polarising the debate.  Nevertheless, his eight-year stint on the
>   soapbox has made him an influential voice in the debate. "I think
>   people will come round to my view," he says. "After all, what I am
>   advocating is optimal and inevitable."

Perhaps (some of?) what my critics have mistaken for, or misclassified as
dogma, were simply substantive points that I, for want of any
substantive rejoinder, simple persisted in supporting?

Most of the "compromises" that have been proposed by way of
alternatives to freeing the entire refereed corpus through
author/institution self-archiving immediately -- such as releasing
journal contents 6-12 months after publication -- have fallen so short
of what is already otherwise attainable and optimal that I find it
difficult to imagine any clear-sighted observer with the best interests
of research in mind wanting to settle for any of them.

On the other hand, for all its attainability and optimality, it has to
be admitted that so far the inevitable has been mighty slow in coming:
indeed, it is already quite overdue!

So if anyone has strategic suggestions that will hasten it (rather than
retard it), I'm sure I and my comrades-at-arms are held back by no
dogma from listening, hearing, and doing whatever it takes.

In point of fact, I am optimistic about the author/institution
self-archiving initiative, bolstered by the power of OAI-compliant
interoperability and university initiatives to implement it. Maybe that
will prove to be what gets us over the top at last:

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: