[Journal-notes] Re: Leading academics back UK Research Councils on self-archiving

Stevan Harnad harnad at ecs.soton.ac.uk
Tue Aug 23 12:26:48 EST 2005

On Tue, 23 Aug 2005, J.F.B.Rowland wrote:

> I think Sally Morris is on somewhat stronger ground than Stevan alleges -

It would be useful if Fytton made it clear in precisely what this
"somewhat stronger ground consists." It is not clear whether he has read
the two rebuttals in question:


A quick summary is this:

    Sally hypothesises that the RCUK Self-Archiving Policy would lead to
    the (strong version) "destruction of journals" and/or (weak version)
    "negative effect on subscriptions."

    Sally provides no evidence whatsoever in support of this (either version). 

    (She cites 5 examples, 3 of them having nothing at all to do with
    self-archiving -- concerning only journals that make their contents free
    online; plus 2 examples having to do with author citation and usage
    statistics, both of which can and will be easily and naturally adapted
    to the new medium, hence have no implications one way or the other.)

    All actual evidence is contrary to both the strong and weak versions
    of Sally's hypothesis: Self-archiving has been co-existing peacefully
    with journal publication for 15 years now. And even in areas where
    it has been practised the longest (physics) and approaches 100%
    in some fields, the journals report no cancellations associated
    with self-archiving.

Fytton is not providing any further evidence here, for or against the hypothesis
(either the strong or the weak version). He is merely stating that he too holds
the hypothesis.

But there's no more accounting for hypotheses than for tastes, in the absence of
any supporting evidence, and in the presence of nothing but contrary evidence.

> although the suggestion that widespread use of OA repositories will
> ultimately harm the subscription sales of journals is only a prediction, it
> is a fairly logical one.  If an item can be obtained free of charge, for how
> long will people go on buying it?  

If every prediction that was not in contradiction with logic were provisionally
taken to be true, Doomsday Prophecies would indeed rule.

The question is not whether the prediction is contrary to logic but whether it is
contrary to the evidence: And it is contrary to all the evidence to date.

The rest is speculation: Why do libraries still subscribe? Here are a couple
of logical speculations:

    (1) They still want the print edition

    (2) They want the publisher's value-added online edition, not just
    the author's self-archived final draft.

Probably there are more one can think of. But note that they are all speculations
about the reasons why the destruction/cancellation speculation is *not* supported
by any evidence. In other words, they are merely counter-speculations.

But why are we speculating and counter-speculating, when one body of evidence is
substantial and irrefutable: Self-archiving increases research usage and impact
dramatically. That is extremely good for research. And there is no sign of its being
bad for publication either.

So RCUK is taking the logical step of increasing self-archiving, so as
to increase research usage and impact, and Sally and Fytton are instead
just speculating.

> On the other hand,  it seems likely that any such effect will occur gradually 
> over a period of years. 

Which effect? We have different effects in mind. I am thinking of the
*demonstrated* effect of self-archiving: increased usage and impact,
a face-valid benefit for research and researchers.

Sally and Fytton are instead thinking about an *undemonstrated* negative
effect of self-archiving: increased journal cancellations,

And the net result is that a hypothetical, undemonstrated negative effect
(for publishers) is being taken (by Sally, perhaps not Fytton) as grounds
for delaying or derailing a real, demonstrated positive effect (for researchers).

Let us hope that the RCUK will not be persuaded by such logic.

> This gives all parties concerned time to adapt.  

The RCUK immediate-self-archiving policy needs to be adopted immediately. Whether
or not there is something that publishers will need to adapt will be seen if and
when there are any signs of it. Right now, there are none.

> OUP and Springer are each starting to do so

Fytton is again changing the subject: OUP and Springer are experimenting with
making their journals free online, or with giving their authors the option to pay
them to make the journal version of their articles free online for them. That is
an adaptation, to be sure, but it is not an adaptation to the effects of
self-archiving (of which there are none, insofar as journal renewals and
economics are concerned). 

There are also logical things one could say about these experiments, but never
mind: let 1000 flowers bloom. The only thing about OUP and Springer policy that is
remotely pertinent to self-archiving is that Springer is full-green (green light
to self-archive both postprint and preprint) and OUP is only pale-green (green
light only for preprint). http://romeo.eprints.org/ That's not tragic:
There's plenty of wiggle room in what counts as the "preprint" -- and,
at bottom, authors don't really need their publishers' blessing to
self-archive their own drafts; it's just a sop for the timorous and
the pedantic.

But the point is that self-archiving is the green road to OA, and what OUP and
Springer are experimenting with is the golden road, which is perfectly fine
(though probably premature).

> and Bo-Christer Bjork from Finland has also recently made a proposal for
> transitional arrangements that look as if they could work (see
> http://www.jisc.ac.uk/index.cfm?name=event_international_0605#Spks).  

One can speculate about hypothetical transition scenarios -- and I have not been
un-guilty of doing a spot of that myself, in my more naive past -- but it is now
clear that among the many things that have been needlessly delaying the optimal
and inevitable -- 100% OA -- was this constant predilection for counterfactual
speculation while ignoring and failing to act upon the actual facts on the ground.

So, for now, I declare, with Newton (and for the sake of research progress):
Hypotheses non fingo.

> There are potentially greater problems for learned society publishers, for whom
> Sally speaks, than for larger publishers. 

I think the research community would do better to deal with its actual problem of
needless research impact loss, rather than subordinating it to publishers'
hypothetical/potential/maybe problems -- whether the publishers be commercial ones
or those that are *nominally* closer to the research community, the learned
society publishers (though one wonders, sometimes!).

> A current JISC-funded project being undertaken by Mary Waltham is 
> investigating possible future business models for them; I look forward 
> with interest to reading her report.

It is splendid to be working on possible future business models for publishers,
but you will forgive me for being far more concerned about the actual impact loss
for research and researchers, today...

Stevan Harnad

More information about the Jrnlnote mailing list