Re: STM Talk: Open Access by Peaceful Evolution

From: Stevan Harnad <>
Date: Fri, 28 Feb 2003 23:03:41 +0000

On Mon, 24 Feb 2003, Jean-Claude Guédon wrote:

> Collectively, the universities and research centres produce the literature
> published in refereed journals (forget about books as I am not dealing with
> that category of literature). And the same institutions buy some of the
> articles that some among them have produced. They are buying collectively
> what they have given away collectively.

I will try to show more explicitly why this collective "we" does not
make sense here at all: Individual authors and their own institutions are
indeed giving away their refereed research papers. That is indisputable
(and different from books, which they are *not* giving away).

But that has nothing to do with the fact that what each institution is
buying *in* is not what *it* (that institution) has given away, but
what *other* institutions have given away.

Now it is all very well for me to say "But I do not wish that to be the
case: I do not wish for other institutions to have to buy-in my give-away
research!" The researcher can say that, and can do something about it:
He can self-archive it, ensuring that the give-away reaches everyone
this time.

But this is definitely *not* the same as saying that there has been a
"collective give-away and a collective buy-back"! No such thing. The
give-away has been individual, and the buy-in has likewise been
individual. And the individual buy-in has *not* been a buy-back of what
has been individually given away, because it was not given away by that
same individual (or institution).

I think this buy-back lament not only formulates the problem (and the
solution) incorrectly, but it is a formula for yet another decade of
unproductive bickering with publishers (who are merely doing what they
have always done: remind yourself that *nothing* has changed in this
give-away/buy-in scenario except the advent of a new medium!) when
the real solution is elsewhere. The solution is with the researchers,
and with the new medium, which they are in the unique position of being
able to use to self-archive their refereed publications, thereby making
what has always been a give-away (to the publisher) into a Give-Away
(to every other would-be user).

The "collective we" misconstrual of the give-away/buy-back lament makes it
into an Escher impossible-figure from which there is no escape!

> The give-away is access. Refereed research seeks to be placed in free access.

Correct. And self-archiving places it in that free access!

> Librarians create the illusion of free access by supporting the whole
> structure financially. So, if the give-away is access, why should we give
> away copyrights? Copyright and access are entirely different and I have
> treated those two entities as completely diffrent from the very beginning.

Another formula for a decade's fruitless bickering is to get drawn into
copyright questions: They are irrelevant! They are an empty
distraction. The only sensible, effective thing to do is *exactly* what
the physicists have had the good sense to do for over a decade (and what
more and more individual researchers have been doing on their own
websites at the same time): Self-archive your own give-away research,
using the new possibility opened up by the new medium.

Anyone who prefers instead to fight with publishers over tolls -- or to
lock horns with clueless lawyers, blindly leading the blind around the
uncharted regions of cyberspace while clutching the old, incongruous
paper-based categories for dear life -- must be interested in something
else more than open access! Those who want open access, now, should just
go ahead and do the tried, tested and proven thing that will provide it
immediately: "Self-archive unto others as ye would have them self-archive
unto you." *That* is a sensible use of the notion of "collective" and
"we": not the incoherent buy-back lament.

> It can also lie with the publshers. Biomed Central is a good example of free
> access.

It certainly is, but this is hardly an example of the fast-track to
universal free access! Count the toll-access journals: 20,000.
Count the open-access journals like BMC (XXX?) and the rate at which
they are materializing. Then extrapolate: How long will it take to reach
free-access (20,000 journals) by that route?

We are preparing some figures now to compare tempos. Self-archiving is
certainly not setting any speed-records either, yet, but it is a fact that
it is in a position to do it all overnight. If all the authors of all
the 2,000,000 annual papers in those 20,000 journals simply
self-archived them tonight, tomorrow we would have universal open

By contrast, having to found 20,000 new open-access journals (and persuade
authors to submit to them rather than to their current journals) or trying
to persuade 20,000 toll-access journals to convert to open-access strikes
me as a recipe for losing another decade.

Self-archiving may lose that decade too, but certainly not because the
open-access was not there, within one evening's reach, for anyone and
everyone! That is already a historical fact, has been for a decade; and
since OAI in 1999, it has become a glaring, even embarrassing fact. (If
I were a journal publisher and either librarians or authors started to
pester me about open-access I would tell them: "Why on earth should we
believe -- and make a sacrifice -- for open-access, this 'benefit' you
allegedly want so much, when you don't bother to get it for yourselves,
by self-archiving your 'give-away' work? And even when we go to the
trouble and risk of formally stating that we support it -- as 50% of us
have done -- you still don't bother doing it! Do your part to prove you
really want open-access -- apart from wanting to pester us! -- and *then*
we will begin giving open-access some more serious thought.")

> ...tell me a reason why an individual should want to go the next
> step and do the self-archiving.

To provide immediate open access to one's refereed research, thereby
maximising its research impact, as tens of thousands of authors have
already done. (The actual is always more real than the hypothetical,
and I am talking about all the actual authors who have actually
self-archived. You are talking about those who have not, because of
hypothetical obstacles, and I think only neuroimagery will reveal to us
what the difference is between them and those who have self-archived,
undaunted by the hypothetical obstacles.)

> This is still viewed as unusual

So was reading and writing, I suppose, but we got used to it. It's less
aversive than mobile phones...

> and, as I
> pointed out in earlier messages, when journals refuse the self-archiving, the
> legal status of your strategy is in question at best.

When journals refuse the self-archiving? But who asks the journals to
self-archive for me? And what does an author wish to hear from journals:
encouragement? That is rather much to ask. Why ask at all? The
physicists did not ask; the tens of thousands of others who have
self-archived did not ask. Fifty percent of journals already formally
and explicitly support self-archiving, presumably because they are pro-
rather than anti-research; --
Shouldn't that at least result in 50% of the literature being in
open-access already? But no, as I also wrote, there are 25 other
worries leaving those researchers in that neurocognitive state I've
called "Zeno's Paralysis":

Will this continue for yet another decade? But by then the Web-bred
generation would already have come of age and cleaned up. Do we really
want the historical finger pointed at us, as having simply been too
hidebound to be able to reach for open-access for ourselves?

> In other words, at this
> moment, in most cases, self-archiving is a risky proposition both in terms of
> its legality and in terms of its efficacy.

On legality, as I said, the worst risk one runs is eventually being asked
to remove the paper from the Web. On efficacy, you'll pardon me if I don't
give the pre-emptive compunctions of those who *haven't* self-archived
quite as much weight as the empirical successes of those who have!

> Self-archiving is not a tried, tested and proven road to immediate
> open-access. It still depends on disciplinary circumstances (e.g. the
> favourable circumstances of high-energy physics),

Jean-Claude, the physicists didn't *have* any "favorable circumstances"!
They *created* them, by self-archiving. (I have no idea what you have in
mind! Are you imagining that they asked their publishers, and received
encouragement to go ahead? *Now* the physics publishers like the APS are
among the most progressive: but that is *after* the fait-accompli of
self-archiving. It did not predate it; nor is it a precondition for it.)

> on institution behaviour (especially when it comes to evaluation procedures)

Institutions reward publishing in prestigious, high-quality,
high-impact journals. There is no change whatsoever in this, wrought
by self-archiving. All that happens is that the research impact of the
self-archived work is enhanced, which is very congenial to the evaluators.

You must be confusing self-archiving with some other road to open
access, one involving giving up the journals that the evaluators
value... (why?)

> and on the degree of
> enlightenment, if it may be described as such, of journal editors.

Journal editors have even less to do with this than journal
publishers. Self-archiving is between the research community and itself.

> So, while
> the research community must obviously be engaged, institutions and publishers
> must also face the issues squarely and react positively.

We can squander yet another decade's potential research impact while
being very "engaged," and "facing the issues squarely" and "reacting
positively." -- Or we can just go ahead and self-archive, and claim that
decade's worth of impact and access after all...

>h> So the reason more and more refereed journal publishers are supporting
>h> self-archiving, despite the element of risk for them, is that it is so
>h> obviously beneficial to research and researchers.
> My, my! These publisher folks are the kindest, most generous people I've ever
> seen. :-))) Stevan, for Pete's sake... But there again, I would distinguish
> between truly learned associations and commercial publishers or even
> professional associations.

My dear friend, 1682 of the 3501 journals (out of the total 7169
covered) that are on the side of the angels on this are Elsevier journals:

"Truly" learned associations indeed! I think we are truly fighting
different battles!

>h> Intuitions differ on this question, but here are the facts: Since the
>h> early 90's, hundreds of thousands of peer-reviewed papers have been
>h> self-archived without a single court challenge. Moreover, talk of a
>h> "court challenge" is rather alarmist, as what would really happen,
>h> if anything ever did, would merely be that an author receives a
>h> notice from the legal department of the journal requesting that he
>h> remove a particular self-archived article. At that point the author would
>h> have the choice of complying with the request and removing the file in
>h> question. That is all that is at issue. And it has not even happened once
>h> in ten years of growing self-archiving -- nor is it likely to happen,
>h> for during that decade publisher support for self-archiving was likewise
>h> growing, because of its obvious benefits to research and researchers.
> Can anyone confirm what Stevan claims? I believe there were some stiff
> standoffs between Paul Ginsparg and some publishers and it is through real
> threats that no action was taken.

Nope. The standoff was about NSF funding of the Physics Archive, not
about copyright issues or publishers.

> In any case, I was locating myself squarely
> on the ground of copyright or authors' rights laws and plagiarism is not
> allowed under either kind of system.

Locating squarely where, on what? And what on earth has plagiarism to do
with any of this?

> But it would be interesting to document what kinds of reactions, if any, have
> occurred in response to self-archiving.

The trouble with the absence of evidence is that it is hard to document
its absence. It's like recording the sound of silence...

> And if the benefits to research are so obvious, and if publishers are so nice
> because of this realization, why are some (quite a few actually) still
> holding out on this issue?

It is not publishers who are "holding out", it is researchers!

> You ask me for an empirical demonstration of
> X and I ask you the same of ~X. Here, empirical research is really needed
> because, if you are right, then you could prove it is groundless.

But the empirical research has been done: 10+ years and hundreds of
thousands of uncontested self-archivings. The only ones who think that the
only evidence can come from a courtroom test are those who are either
in zeno's paralysis or prefer pugilism.

[There is something in this that sounds like the logic of Pascal's
Wager: It is better to act in conformity with Belief, even if it is
false, because the cost of being wrong if the Belief is true (burning
forever in hell) outweighs the cost of being wrong if the Belief is
false (leading a somewhat more constrained lifetime on earth). But of
course this logic is all wrong, for Pascal simply presupposed the
received Credo, the Christian one, C, with its myth of heaven, earth,
and eternal damnation as the default option -- the null hypothesis, so to
speak. But I could always raise the stakes, with another arbitrary Credo,
even more vengeful than the first, promising that if you believe C rather
than Q, your punishment will not just be eternity in damnation, but your
soul will split into an infinity of souls, each in its own hell, and
each will feel not only its own suffering but that of all the others too!)

So one can reverse Pascal's Wager by simply arbitrarily changing
the pay-off matrix, by fiat. One can do exactly the same with self-archiving,
prophecying a horrible fate from which no self-archiver is safe (from
unless some brave soul actually faces and survives it in a court
challenge) and one can discount all negative evidence to the contrary:
The fact that legions of self-archivers are happily self-archiving
oblivious to all this metaphysics is not evidence... (And not even
50% of journals formally stating that there is no hellfire makes any

>h> Plagiarism is completely irrelevant to the self-archiving of refereed
>h> research!
> Not in the case of an article published in a refereed journal with transfer
> of copyright and a journal policy against self-archiving. If you put your
> article in its pre-refereed form, there is presumably enough resemblance
> between that and the printed version to allow the publisher to run after you
> for plagiarism.

Plagiarism? But whom have I plagiarized? Myself? Is that plagiarism? (I
think you mean "copyright infringement" or possibly "duplicate publication,"
but not plagiarism.) In any case, it is not a worry for self-archivers.
(And, a fortiori, not for those who publish in the 50% of journals that
are already on the side of the angels formally!)

> As you know, in logic, saying that black swans do not exist simply on the
> basis of having seen only hundreds of thousands of white swans does not work.
> I must confess being mystified by your logic.

This is not logic, it is empirical evidence. You say "It is not safe to
go out, because there may be toxic rain." I say "But I and hundreds of
thousands of others have been going out safely for 10 years, and not a
single case of toxic rain!" (Don't reply with black swans; this is just
ordinary inductive statistics, with rejection of the null hypothesis;
if there is any pertinent philosophical example, it is the inductive
fallacy of the chicken who believed there would always be a tomorrow --
but I think Chicken Little is more relevant here....)

> Incidentally, about 20 000 journals, Alice Lefler Primack, in Journal
> Literature of the Physical Sciences: A Manual (Metuchen, N.J., Scarecrow
> Press, 1992), states that in 1990, about 90,000 journals that specialize in
> some areas of science. The number has grown since by a fair percentage. You
> will object that they are not all refereed, but, following yur lead, I will
> ask you to prove that... :-)

But nothing is changed by whether there are 20,000 refereed journals or
200,000. Their contents should all be self-archived, yesterday!

> between self-archiving as you understand it and
> pre-prints, there is a difference that you have (rightly) insisted upon in
> the past. Why does it not matter so much now?

Preprints are merely an earlier embryological stage of what eventually
becomes the refereed postprint. Both are important, and both should be
self-archived and openly accessible. But the postprint is the essential
one, because (1) not everyone wants to trust or even waste time on
unrefereed preprints and (2) the preprint enthusiasts have incorrectly
suggested that the usefulness of preprints is evidence for the
uselessness of peer review (so we need not bother with the postprint at

All of this is, of course, just a ritual dance. The data remain the
same: Self-archivers continue to submit all their work to peer-reviewed
journals, and they self-archive both preprints and postprints. The only
thing that has changed is that both are openly accessible now.

That is self-archivers. Non-self-archivers are even more hypothetical,
for what we are talking about here is freeing access to the peer-reviewed
literature, not about freeing the literature from peer review through
preprint self-archiving.

>h> Influence [young researchers] to do what, insofar as self-archiving
>h> is concerned?
> For example that this may not be the best way to get tenured, or that it may
> not be the best way to get accepted in the best journals, etc.

But how does enhancing impact by self-archiving one's refereed
publications diminish rather than enhance one's tenure chances? And what
evidence is there for the contingency that self-archiving diminishes
one's probability of being accepted by any journal (best or worst) by
even one iota? (The vanishing "Ingelfinger Rule"? )

> If you accept the possibility of some undeniable influence of gatekeepers
> on young researchers, and if this gatekeeper does not see self-archiving with
> kind eyes...

This is an "if" conditional on yet another "if," for neither of
which there is any evidence at all. (Of course editors influence young
researchers -- it is to a great extent in their hands whether their work
is accepted, but that has nothing to do with self-archiving.)

Many editors do not look at all unkindly on self-archiving (they are,
after all, us!), but either way, they are as irrelevant as publishers.
It is only those who have contracted zeno's paralysis for whom (worries
about) editors exert any causal influence at all in this respect (but
these worries are rather like the worries about toxic rain, when
self-archivers are happily going about it without any worries, or any
ground for them.)

> But why is the research community so slow? Why is the tempo picking up?

About the first, I don't really know (though I have mentioned 26
possible reasons!). About the second: Could it be that good sense (and
10 years of uncontested success) is prevailing at last?

Best wishes,
Stevan Harnad
Received on Fri Feb 28 2003 - 23:03:41 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:53 GMT