UK Select Committee Inquiry into Scientific Publication

Stevan Harnad harnad at ecs.soton.ac.uk
Wed May 5 14:32:30 EST 2004

On Wed, 5 May 2004, Richard Poynder wrote:

> I agree that self-archiving and OA publishing are frequently and
> inaccurately conflated, and that I am sometimes guilty of this myself. 

No, dear Richard, it is OA and OA publishing that are conflated (i.e., treated as
if they were the same thing). OA self-archiving is thereby *overlooked* as a means
of providing OA, let alone the far faster, surer means that it in reality is,
and the means that is already providing far more OA -- and could easily provide it
all: 100%


> There is, however, a good reason why this happens: self-archiving cannot be a
> solution in itself, since it feeds (today at least) off traditional
> publishers. 

I can only repeat: Cannot be a solution to *what*? To the library serials
budget crisis and the journal affordability problem? Agreed. But that
is not what OA is about. 

OA is about the access/impact problem: about the needless loss of research
usage and impact that would have come from those countless would-be
users who were unable to access my paper because their universities
could not afford the access-tolls to the journal in which my paper was
published. Self-archiving it certainly solves *my* access/impact problem
-- and so it does for every author who self-archives.


Moreover, it is *supplementing* toll-access, not *supplanting* it. It is done on
an article by article basis, and in 83% of papers it is now even done with the
blessing ("green light") of the journal in which it appears!


So on whose behalf are you worrying about doing this? The 83% of journals who have
already given it their official green light? Or the 17% who have not yet done so?

> Nor is it sustainable, since the more researchers self-archive,
> and the more effective the OAIsters become at connecting researchers with
> relevant research material outside the financial firewalls of commercial
> publishers, the quicker the point arrives where traditional publishing stops
> working. 

Again, you seem to be determined to speculate about hypothetical
contingencies that don't even appear to be worrying at least 83% of
journals. Perhaps the reason it does not worry them is because it is not
a matter of some catastrophic "point where traditional publishing stops
working" at all, but a matter of a gradual process of adaptation and
co-evolution -- which has already been going on since 1991, and has not
yet even made itself felt in the form of downsizing pressure in those
fields where it has already reached 100%.

You keep returning to your catastrophic conjectures, but you seem to be
ignoring the actual objective data. In one of the fields (High Energy
Physics) where OA has been at 100% for several years now, a born-OA
("gold") journal, JHEP, has since successfully reverted to the
"traditional" toll-access (TA) cost-recovery model, even though all of
its contents have been and continue to be OA through self-archiving
ab ovo. 


Is it not a bit Cassandra-like to cleave to one's doomsday scenarios
even in the face of nothing but contrary evidence? (Of course, Cassandra
happened to be right, but that was only a myth, whereas we are presumably
speaking about reality, in a time when there are no oracles, merely
objective indicators of empirical trends, plus the aid of logical

> Self-archiving, therefore, can only be a preparatory phase. 

A preparatory phase for *what*? This is not about restructuring journal
publishing, nor about solving the serials crisis. It is about providing
open access to each researcher's own research, now! Can you not see that
you are talking to Peter about the problems of Paul?

> While there is a distinction to be made between self-archiving and OA publishing,
> conflation is inevitable since beneath that conflation lies a reality that
> will not go away: self-archiving is not a solution in itself, but a catalyst
> to OA publishing. 

The second conflation, as I said (after the conflation of OA with
OA publishing, overlooking OA self-archiving) is the conflation of
the journal-affordability problem with the access/impact problem:
And you are doing it again! 

The solution I have in my sights is 100% OA. That means no would-be user
is denied access any longer because his institution cannot afford the
access-tolls. No more usage and impact is needlessly lost. That outcome
is a *supplement* to the present system, not a *substitute* for it. It
may or may not prove to be a means of catalyzing OA publishing. That is
not what matters, because catalyzing OA publishing is not the end for
the sake of which self-archiving is done: 100% OA provision is. And the
problem it solves is the research access/impact problem, not the journal
prcising problem, or the restructuring of journal publication.

> As for conflating OA with the journal price problem: the more librarians
> become reluctant/unable to pay for the journals that sustain self-archiving
> the quicker the day of reckoning will come.

Don't you see that you are mixing apples and oranges? It is the journal
pricing/affordability problem/crisis that is making librarians more and
more reluctant/unable to pay for the journals. That supply/demand/pricing
problem needs to be fought out in the arena of the market (with whatever
consortial help the consumer-institutions can give one another).

But that has nothing *whatsoever* to do with self-archiving, which is done
by authors and their institutions in order to provide access to their own
research output for all its would-be users (regardless of whether the user's
own institution can afford the access-tolls for the journal in which it
appears) so as to maximize the uptake, usage, and impact of that research,
and the rewards (in salaries, research-funding, prestige and prizes)
that that research impact brings to the author, his institution,
his research-funder, and the tax-payer who funds the research-funder!

In other words: what is this "day of reckoning"? Librarians are reckoning
what they can and can't afford *now*. Self-archiving does not increase
or decrease libraries' cash-flow problems. What, then, is the causal
connection between this putative "day of reckoning" and self-archiving?

Here's a (hypothetical) possibility, but it goes in the exact opposite
direction from the one you are apparently envisioning: 

I am a hard-pressed librarian. Every year, I can afford fewer and
fewer journals because of their price-rises. So I have to decide what
to cancel. Normally, I consult my user logs and users, about which journals
they use and need most, and I cancel according to a multiple constraint
satisfaction analysis of the costs/benefits (including journal impact

I also have a table of the "green" journals -- the 83% that have given
their green light to self-archiving. So consider the following scenarios:

(1) I now try to add the "green/gray" factor to my constraint
matrix. Should I tilt toward or against green journals? The prima facie
instinct might be to tilt against: After all, these are the ones that are
more likely to be self-archived. But would that be wise? Do I really want
to reward the gray journals? And would favoring the gray journals not
in fact disfavor OA provision (inasmuch as it may be in part encouraged
by journal policy)?

(2) What about it I start to do OA surveys, and confirm that OA is
indeed growing faster for the articles in the green journals? Is *that*
now a better reason for cancelling green and favoring gray?

I think it is obvious that for a cash-strapped librarian, I am better either not
taking green/gray into account at all, or, if anything, tilting in favor of
*retaining* rather than cancelling green journals, in competition with gray.

Either way, library cancellations can only either be a countervailing
effect (slowing the progress of OA, through a selective bias against
green) or a facilitating effect (hastening the progress of OA, through
a selective bias against gray), but neither way does it look as if
self-archiving is creating or hastening any cancellation catastrophe point!

> While insisting that self-archiving and journal price inflation are
> separate issues may help encourage researchers to self-archive, the two
> issues are nevertheless locked together.

I would be interested to know how: It's certain that the countless physicists
and others who have been self-archiving since the early 1990's were not doing
it because of any thoughts about journal price inflation. Why should the two
be interlocked now?

It is true that *librarians* and *administrators* have embraced OA in part
because of their journal pricing problems. But they are not the ones who
are doing, or can do, the self-archiving: And so far they have been
singularly ineffective not only in persuading their researchers to
self-archive, but even in coming up with a coherent rationale for
self-archiving and OA for their researchers! 

Could this be because librarians and administrators, too, keep conflating
the pricing problem and the access/impact problem, and imagining that
researchers will want to switch to OA journals or will self-archive merely
in order to help the library and administration with their budgetary
crises? It might be a good time to rethink this strategy, and recast
the rationale for OA and self-archiving in the access/impact terms that
are the only ones that matter to the only constituency that matters --
and is in a position to provide the OA -- namely, the research community.

To avoid eliciting or encouraging the default "sitting pretty" response
from researchers


librarians and administrators will need to defer the prospect of
gratification for their serials budgetary problems and focus instead on
presenting the objective data demonstrating the dramatic impact-enhancing,
and hence income-enhancing implications of OA:


> Steven's 'just do it' approach to OA is a positive force. If people hadn't
> just got on with things in the past many positive changes may never have
> happened. But to discourage speculation about possible outcomes and likely
> scenarios is not helpful if we want to plan for the future. It is not
> dissimilar to saying: "Let's do away with our present form of government as
> it no longer suits our purposes, but let's not waste time speculating about
> how society will be organised in the future until the future arrives."

There is a rather prominent hidden premise here, and it is around this hidden
premise that all this speculative discussion is based: that self-archiving
is analogous to throwing out our present form of government! It is
not. It's much more like buying medical insurance because there is
no state medicare -- except that self-archiving is free! It *might*
eventually lead to a change to universal medicare -- if/when medical
insurance reaches 100% -- or it might not. Either way, I and my family
am protected from the moment I begin self-providing.

By the same token, 100% OA self-archiving *might* eventually lead to a change to
universal OA publishing -- or it might not. But that's certainly no reason not
to start doing it right now!

> One possible scenario that Stevan has raised in the past is that publishers
> will gradually downsize and convert to OA publishing; but is that really
> likely, and would any such transition be smooth? Companies like Reed
> Elsevier do not downsize, they exit as soon as they believe that the
> margins their shareholders expect cannot be sustained into the future.

Who cares? As OA self-archiving grows, there may or may not be downsizing
pressure. If/when there ever is, some publishers/journals will want to
downsize, others may not. If not, their titles will migrate to other
ready/waiting publishers (e.g., PLoS or BioMed Central) who are ready
and willing to take them over. 

Why on earth should anyone fret about this now, when the only *sure*
thing to fret about is that OA -- already optimal, inevitable, and
immediately feasible -- is also already long overdue!


> So companies like this don't downsize, they don't wait for the ship to go
> down, they get out. This, after all, is what Wolters Kluwer did in 2002,
> when it sold KAP and announced it would focus on the better margins
> available in other information markets. This is surely also why Bertelsmann
> sold BertelsmannSpringer last year.

And notice what happened to Kluwer journal titles? They migrated!

> The difficulty next time round could be that if self-archiving has become as
> successful as OA advocates hope it will be, there might be a shortage of
> buyers. Certainly VC companies like Cinven and Candover (who bought KAP, and
> BertelsmannSpringer) are not going to buy again if the industry is no longer
> the licence to print money it was.  They too will be rushing for the exit.

Fine. Leaner, hungrier buyers will be waiting in their place. Indeed,
universities themselves may get more involved in becoming journal
publishers (not house journals for their own output, of course, but
proper 3rd party publishers to take over migrating established titles,
their editorial boards, refereeships and authorships). And there are
still the Learned Societies too.

> What other traditional publisher would want to get into the scholarly
> publishing business if there was always a free version of everything it
> sold? 

The ones that are prepared to occupy the downsized niche of peer-review service
provision/certification. But why on earth do we need to speculate about this now?
Surely attaining OA is far more important than how journal publishing may
restructure and redistribute itself to accommodate it if/when that should ever
become necessary.

> Let's face it: what self-archiving threatens to do is to destroy most
> of the value that traditional commercial publishers (and VC companies)
> require to breathe. Of course there are new (and certainly better and more
> desirable) business models to be found, and organisations that will adopt
> them (and some are already trying them out) but we should not assume that
> the process of change will be gradual and smooth. Maybe it will be, but
> maybe it won't.

I think it is an utter waste of time -- time that we have wasted far,
far too much of already -- to persist in this state of speculative
paralysis about hypothetical (but by no means certain or even probable)
contingencies -- when what is certain and actual is so clear: OA is
both *optimal* for research, researchers, their institutions and their
funders and it is *attainable* (and being attained, but not yet at a
sufficient rate -- in part because of this kind of Zeno's Paralysis
about counterfactual conditionals).


What researchers need to do is to provide OA, now, not keep doing empty
and idle speculation.

> That is the potential transition problem. Who better to puzzle over it than
> the UK Select Committee?

MPs may be professional puzzlers, but it is the research community that
can and must act. I had hoped that parliament, as the purveyer of the
research purse-strings, might help persuade researchers to provide OA. But
to do so, the Committee would first have to understand what OA is really
for, and about. It is about access/impact, but they have instead focussed
on journal affordability. And the "green" road of OA self-archiving is a
far faster, surer and wider road (with already a good deal more traffic)
than the "gold" road of OA publishing, but they have instead focussed
on OA publishing.

It will be interesting to find out whether anything useful came out of
the parliamentary exercise just the same.

Stevan Harnad

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online (1998-2004)
is available at the American Scientist Open Access Forum:
        To join the Forum: 
        Post discussion to: 
    american-scientist-open-access-forum at amsci.org
        Hypermail Archive: 

Unified Dual Open-Access-Provision Policy:
    BOAI-2 ("gold"): Publish your article in a suitable open-access
            journal whenever one exists.
    BOAI-1 ("green"): Otherwise, publish your article in a suitable
            toll-access journal and also self-archive it.

More information about the Jrnlnote mailing list

Send comments to us at biosci-help [At] net.bio.net