Re: Draft Policy for Self-Archiving University Research Output

From: Stevan Harnad <>
Date: Sun, 12 Jan 2003 01:16:16 +0000

On Sat, 11 Jan 2003, Picciotto, Sol wrote:

> A response to Stevan Harnad's reply.
> Unless I have seriously misunderstood Open Archiving, it entails making
> a separate copy of a work available for access without charge, and not
> merely providing a google-type link to the article on the commercial
> publisher's website (which would be subject to access tolls).

Correct. (But what you mean is "Open Access Archiving," not "Open
Archiving," which refers only to a metadata harvesting and
interoperability standard).

> It is therefore a form of publication, both legally and in practical
> terms. That is why commercial publishers are reluctant to allow authors
> to retain the right to self-archive, especially in an eprint archive
> which would be fully searchable, and hence would directly compete with
> their journal. That is why Nature says it would be a breach of their
> licence (i.e. the rights they require authors to transfer to them)
> to publish/archive a paper in an institutional eprints archive.

The formal legal definition of "publication" is hardly relevant here,
because according to that definition, any nonsense I scrawl on paper
and make public is a publication. That sort of thing has no CV value,
so it is definitely not "publishing" in the (relevant) publish-or-perish
sense of the word, as it is used by academics. Nor is it of any relevance
to the publisher's commercial sense of publication either. Publicizing
my doodles, in other words, is a sense of "publication" that may be of
(theoretical) interest only to copyright attorneys!

This theme was last discussed in this Forum on this thread:

    Garfield: "Acknowledged Self-Archiving is Not Prior Publication"

As to why Nature is saying what it's saying: By the above (irrelevant)
definition of "publication," self-archiving on one's own website is
already "publication," and that is precisely what Nature is saying
would *not* be a breach of their license. So that can't be the point

(In my recent open letter to the Editor of Nature
inquiring about the new Nature Licensing policy -- -- I have
described, informally, what it is that I believe Nature is concerned
about, in trying to specify as explicitly as possible what does and does
not count as self-archiving under its new license: And placing a paper
(for which the exclusive right to publish has been licensed by the
author to Nature) into a commercial website that is gathering Nature's
contents together so as to sell or otherwise derive profit from them
would not count as self-archiving by the author. Nature wants to reserve
its right to go after such a re-publisher [sic] of Nature's contents in
court by making it clear that that would not count as the kind of author
self-archiving right that the license returns to Nature authors. This
does not apply, however, to the author's own institutional eprint
archive, which is not gathering Nature's contents so as to re-sell them,
but gathering only its own institutional research output, so as to make
it openly accessible and thereby maximize its research impact.)

> That is why open [access] archiving inevitably comes into conflict with
> existing commercial publishing models, and there is a debate here and
> elsewhere about alternative business models. I think a good case can be
> made that open archiving would not seriously damage commercial publishing,
> but I can see why commercial publishers are reluctant to take the risk.

No one knows for sure what the long-term effects of open-access through
author/institution self-archiving will be on the market for the
toll-access version, but what is certain is that the online era and the
new possibilities it has opened for maximizing research impact have given
rise to an unprecedented conflict of interest between what is best for
minimizing potential risk to publishers' future revenues and what is
best for maximizing potential research impact.

It is also very clear in what direction this conflict of interest
must be resolved (namely, in favor of maximizing potential research
impact rather than minimizing potential risk to publishers' future
revenues). The reason it must be resolved in that direction is not
only that research impact (and not royalty revenue) is the sole reason
researchers (anomalous among all other authors) publish (refereed
research) at all, but, equally important, it is also simply because open
access is indeed within their reach.

It is no longer possible to close that door. For where journal
publishers explicitly refuse to allow self-archiving, thereby
broadcacting that they do not share their authors' goal of maximizing
their research impact, the preprint-plus-corrigenda strategy -- -- is still
available to all authors for attaining almost exactly the same end --
while implicitly naming-and-shaming, each time that strategy needs to be
used, those publishers who thereby advertise that for them maximizing
their potential revenue streams is more important than maximizing the
potential impact of the research they publish.

It is in order to confirm that they stand with the angels on this
that Nature has adopted its new self-archiving policy.

> It is no doubt easier to make a case for open archiving of works for which
> the author receives no payment, which Stevan characterises as Giveaway,
> but the publishers merely respond that they bear all the remaining costs,
> which are substantial.

As long as toll-access revenues are covering those costs, that is no
reason for trying to prevent self-archiving. If and when toll-revenues
fall, because of competition from the open-access versions, to levels
where they can no longer cover all essential costs [and please note
that it is not at all clear when or whether this will ever happen]
there are ways that costs can be cut and cost-recovery models changed:

The issue now is whether or not it is good public relations for publishers
to protect themselves against such a possibility by trying to prevent
self-archiving (i.e., protect themselves at the expense of their authors'
impact). It seems it is bad public relations in any case, because forcing
authors to resort to the preprints-plus-corrigenda strategy leads to the
same outcome (open access) and only leaves those publishers with egg on
their face.

    Lawrence, S. (2001a) Online or Invisible? Nature 411 (6837): 521.

    Lawrence, S. (2001b) Free online availability substantially increases
    a paper's impact. Nature Web Debates.

> Stevan is wrong to characterise some authors as
> giveaway and others as non-giveaway. Many academic authors in many fields
> list among their research publications items for which they have received
> some form of payment, as well as others where this is not so (both in
> journals and not infrequently books). I assume we support moves such as
> the decision by UC Press to make many of its books available online?

But what is the problem here? What I am saying is not only
true but tautological: The Budapest Open Access Initiative -- -- is targeting only those
writings that really are author give-aways. This is true, without
exception, of the refereed journal literature. And it is almost without
exception UNtrue of (say) the royalty-bearing book literature (though the
occasional author or publisher sometimes chooses to make a give-away, or
a partial give-away, of some books, for various reasons, including that
they have no market). No one is trying to get authors to self-archive
and place into open-access writings from which they wish to make royalty
or fee revenue.

    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

One size does not fit all. Self-archiving and open access are not the
solution for the non-give-away literature. (I do predict, though, that
once the quantitative evidence of the causal link between access and
impact begins to be reckoned for scholarly monographs too, many of them
will prefer to forego royalty revenues in favor of maximizing impact,
just as refereed-journal authors do. Impact, after all, translates
into revenue, in the publish-or-perish academic world.)

> An important reason open archiving has been slow to take off is that
> academic authors are generally reluctant to oppose publishers who ask for
> exclusive publication rights.

That is true, and that is why we have been actively informing authors
about the preprint-plus-corrigenda option when all else fails. As more
journals emulate Nature's progressive policy, this option will become
less and less necessary.

> It would be easier to do this collectively
> and with institutional support or leadership. That is the importance
> of the claim by universities to retain the right to authorise free
> publication in archives. I'm glad that Stevan accepts that this can be
> done in parallel, but sorry he is so reluctant to concede that it is an
> important complementary step.

I not only concede but loudly proclaim that if/when it succeeds, it
will be an important complementary step (as welcome as news that the
rest of the journals have followed Nature's lead -- though that would
also make it somewhat supererogatory!).

But any implication that it is necessary in advance, or indeed necessary
at all, would simply be another reinforcement of researchers' (groundless)
worries about whether or not they can safely self-archive NOW. My own
reluctance is hence entirely based on researchers' current reluctance
to self-archive because they incorrectly believe that it would be illegal.

Stevan Harnad
Received on Sun Jan 12 2003 - 01:16:16 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:49 GMT