Re: Zen response to e-Archiving Challenge

From: Stevan Harnad <>
Date: Fri, 25 May 2001 13:50:11 +0100

The posting by Albert Henderson, rather than illustrating fallacies
on the part of others, is merely illustrating the errors to which one is
prone when one has a conflict of interest (between what is really the
case, and what one is advocating -- in Albert's case, as we all know,
the advocacy is on behalf of certain parties' interests...).

On Thu, 24 May 2001, Albert Henderson wrote:

> FALLACY 1. Scientists give away their reports and copyrights.
> Scientists (and scholars) exchange their reports for
> effective dissemination services and the unique recognition
> provided by publishers who organize new knowledge with
> authority. Obviously the exchange has value to both parties.
> Authors struggle to be accepted by the publishers of their
> choice. Publishers compete for authors but may reject work
> and require revisions.

Everything Albert says is spot-on for the Gutenberg era, where
print-on-paper was the only means of disseminating refereed research,
and print's true, sizeable costs had to be covered by access-blocking
tolls if the refereed research was to be disseminated at all.

That era is now over, but Albert does not seem to have noticed.

Why? Because he wants to continue to protect, at all costs, the revenue
streams of the Gutenberg era.

How does Albert manage to make it seem as if this still makes sense? By
conflating the gate-keeping function (peer review: refereeing) with the
toll-gating function (charging for access to the paper or the PDF).

Brief resolution of this pseudo-fallacy: Authors can give away their
give-away refereed research on-line by self-archiving it on-line in
interoperable eprint archives ( If and when the
paper and PDF sales of the publishers' version are no longer enough to
cover the only remaining essential cost -- peer review -- because of
subscription/license cancellations, then they can be covered by
authors' institutions, per outgoing paper published, rather than per
incoming paper subscribed/licensed, out of their annual windfall
subscription/license cancellation savings: 100% is saved, but the peer
review alone costs only 10% of it:

> FALLACY 2. Authors can legally leave the preprint version
> of an article up by adding "corrigenda" after transferring
> copyright to the publisher.
> No copyright attorney that I know would agree with this. It
> is a clear case of wilful infringement. Having transferred
> copyright, the author is obligated to delete the preprint
> or be a party to Napster-like infringement. Moreover, the
> preprint server is probably also liable.

Incorrect. See Charles Oppenheim's earlier reply to this in this Forum.
(He is not a copyright attorney but a digital copyright expert who is
writing and updating the book on the subject [the one the copyright
attorneys study]; and he is advisor on copyright to the European
Commission's Legal Advisory Board.)

    Oppenheim, C. (1992/2001) The legal and regulatory environment for
    electronic information. (4th edition) Infonortics.


> FALLACY 3. Use of the word "archive" to describe unreviewed preprints.
> "Archive" has long been associated with peer-reviewed journals
> such as Archives of Internal Medicine and with research libraries that
> are selective about what they keep. This usage is a pathetic plea for
> status, much like sewage processors claiming to be "water recovery plants."

This is a trivial semantic quibble. Besides, archives are NOT just -- or
even primarily -- unrefereed preprint archives. They include both the
pre-refereeing preprints and post-refereeing postprints.

> Speaking of sewage, the problem with the usage is that mixing unreviewed
> preprints with published papers will confuse readers. You wouldn't offer
> sewer water side by side with 7-up and Coke and offer it to your trusting
> children, would you? The misleading usage is an open invitation for
> fraudulent promotion of unsafe and ineffective products presented as
> "research." Freedom of speech issues do not excuse the reckless and
> uncaring mixing of dangerous material with original research.

Nonsense. One might as well say that the Web itself is unusable because
it mixes the reliable with the unreliable.

Eprint Archives have TAGS (exactly as library shelves and index cards do)
to guide the user. As a default option, a user could (and often should)
restrict search and retrieval to papers tagged "REFEREED," including
"JOURNAL NAME." This is simply a PostGutenberg version of precisely the
same navigational guides used in the Gutenberg era.

(I trust the reader has enough sense to see through the sensationalistic
rhetoric of "sewer water" without my having to make any comment...)

> Another problem is that the word "archive" is being used to market the
> displacement of libraries, librarians, editors, and publishers with the
> notion that a computer can replace them all, and very cheaply.

Online archives can and will replace some (perhaps many) Gutenberg
library functions, and very cheaply. The rhetoric here is in the
hyperbolic "all", and the trivial semantic quibbling.

> FALLACY 4. Peer review is certification of quality.
> But not of results. Most peer review of published articles is
> done in a few hours and without examining the authors' original
> data.

This is true, but UTTERLY irrelevant. None of the issues at hand
(toll-gated refereed research access vs. free online access) has
anything whatsoever to do with the defects of peer review. Implying
otherwise is merely diffusing diffuse misinformation.

> Because of the impoverishment of their libraries and the
> slowness of interlibrary loan, referees are unable to check
> unfamiliar works cited in a paper under review.

PostGutenberg, abstracts are on the Web, and more and more of the
full-text papers too. Is Albert attacking the problem here, or the

[Note Albert's familiar formula for the fix, lurking behind all this,
is, as always: "The real problem is that libraries are (conspiratorially)
underfunded. Give them all the money they need and deserve (which you
are wantonly withholding right now), and everything will be accessible
to everyone"...]

> referees
> are not likely to be experts on the topic they review according
> to one study published recently.

Can we be spared one-shot wonders about peer review, please? The
blemishes of peer review are not the subject of discussion here: The
freeing of access to the current peer-reviewed literature (such as it
is!) is.

> FALLACY 5. "Do-it-yourself" Self-archiving by authors will solve
> scientists' communication problems.
> Even Paul Ginsparg has noted that an intellectual non-automated
> approach is needed. The literature is too massive and chaotic,
> the scientific community is too broad and unruly, and the ability
> of any individual (no matter how large the ego) to keep on top
> is doubtful. Like the responsibility for workers' safety, the
> burden of intelligence must be supported from the top. Like
> the issue of safety, the people at the top would just rather
> save money. Sputnik, like the Shirtwaist factory fire, shook
> them out of this groove for a while.

What is all this about? We have, at the moment, a refereed research
literature (20,000 journals' worth, across disciplines). Currently most
of that refereed research is not accessible to most researchers. All
that is being proposed here is that this author give-away literature
(so different from royalty/fee-based books, etc.) be freed online, by
being self-archived by its own authors.

Now what on earth do any of these red herrings about peer review
imperfections and Sputnik have to do with this, other than as
obfuscations of the very real and very simple problem and its equal
real and simple solution?

> FALLACY 6. Authors' income results from the impact of their
> findings.
> While publication helps substantiate a scientists capacity for
> research, there is no provable relationship between authorship
> and income. The most prolific authors achieve tenure early on.
> Most gainfully employed authors produce only 1 or 2 papers in
> their lifetime. Many scientists and engineers publish nothing
> at all. Well-paid industrial technical consulting is more
> likely to involve trade secrets rather than open communications.

Again, this is not the place for data-free speculation about the
relationship between publishing and perishing, between impact and

Forget about income, if you like. Free the access to the research just
for the sake of the research itself then, and its impact, if you don't
think that has anything to do with the researcher's career!

    Harnad, S. (2001) Research access, impact and assessment. Times
    Higher Education Supplement 1487: p. 16.
    longer version:

> I recall that the Association of Research Libraries Serials Prices
> Project also stepped in this hold, accusing researchers of
> excessive publishing!

Can we let the publication glut, if any, take care of itself please,
rather than trying to use it as an excuse for continuing to hold
give-away research results hostage to obsolete Gutenberg toll-gates?

> FALLACY 7. Universities are too poor to maintain self-sufficient
> collections.
> In the United States, at any rate, it is clear that higher education
> institutions have increased their profitability at the expense of their
> libraries for over 30 years.

This is Albert's hobby-horse. Nolo contendere.

The only relevant point is that there is not remotely enough money
available worldwide to ensure that every researcher has access to every
paper in the annual 20,000 refereed journals I mentioned. Self-archiving
remedies that. Other library matters are outside the scope of this

    Harnad, S., Carr, L. & Brody, T. (2001) How and Why To Free All
    Refereed Research From Access- and Impact-Barriers Online, Now.

> FALLACY 8. Science budgets cannot afford dissemination.
> Science budgets seem to aim for high employment and a high rate of
> grant renewals. Why? Dwight Eisenhower (former president of
> Columbia University) pointed out that the government contract has
> replaced curiosity as a motive. Newt Gingrich once noted that science
> bureaucrats don't care about results.
> A science policy that cares about results cannot afford not to support
> libraries as the core of a private enterprise market that responds --
> far better than any bureaucrat -- to wants, needs, and demands. The
> 1960s were a golden age of science in the US partly because the growth
> of spending on science was matched by spending on libraries.

The exact same hobby horse as above, already replied to repeatedly in
this Forum, but nevertheless rolled in by Albert relentlessly on every
occasion, until I have to re-invoke cloture.

> Is 8 enough for now?

We have not heard 8 fallacies. We have heard variations on one endless
theme, mostly in the form of rhetoric, confusion and error, if not
outright misinformation...

New readers should consult the archives (sic) of this Forum to see where
Albert is coming from, and going (over and over and over). The
neophytes may find this reply harsh; the veterans will know why...

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
Received on Wed Jan 03 2001 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:06 GMT