Re: UK Select Committee Inquiry into Scientific Publication

From: Stevan Harnad <>
Date: Tue, 4 May 2004 14:39:52 +0100

What appears below is an exchange with Richard Poynder in
Information Today (IT). It begins with my own comments on RP's April 1
article. Richard's response co-appeared with these in IT May 3, along
with a new article. My subsequent comments (unpublished) on both his
response and his new article then follow. Richard is an OA-friendly,
knowledgeable journalist. Some of my comments are about points where
I think he is in error; some of them are about points where others
whose views he is reporting are (in my view) in error:


    Richard Poynder, The Inevitable and the Optimal
    Information Today, April 1, 2004.


            Stevan Harnad

In the Internet age, open access (OA) has indeed become optimal and
inevitable. It remains only to make it actual! In an article that is
largely on target, Richard Poynder notes that there are two ways to
provide OA: (1) publishing articles in OA journals and (2) publishing
them in conventional journals but self-archiving them publicly on the web
as well. The UK Select Committee has so far ignored (2), even though (2)
is providing and can provide far more immediate OA.

But then Richard adds:

    RP: "[I]f governments truly want to help, they need to also ensure
    that scholarly communication does not break down in the process
    of transition... Self-archiving... is the fastest growing form of
    [OA]... with or without publisher approval. At the same time... the
    library community is voting with its feet by aggressively cutting
    journal subscriptions... The danger is that these growing acts of
    civil protest could, in the short term, exacerbate the crisis. For if
    research institutions and universities cancel more and more journal
    subscriptions and open access publishing cannot immediately fill
    the gap, those in need of research may find themselves having to
    sift through a hodgepodge of (frequently unrefereed) self-archived

This is unfortunately a non sequitur! There is no civil protest and no
prospect of a breakdown! Self-archiving one's own articles is perfectly
legal, has been growing since at least 1991, and already has the
official "green light" from close to 60% of publishers, all eager to
demonstrate that although they may not wish to lower their prices, nor
to take the risk of converting to OA publishing, they have no wish to be
seen as blocking what is optimal and inevitable for research and
researchers. But the optimal and inevitable is OA, not necessarily lower
journal prices or OA publishing!

Although it was the library community and its journal budget crisis that
first brought the research-access problem to the research community's
attention, the journal-pricing problem and the research-access problem
are not the same problem! Libraries cannot cancel journals unless their
users no longer need access to them. OA publishing (1) grows journal by
journal. But OA self-archiving (2) grows anarchically, article by article.
So it is not at all clear whether and when libraries can cancel any
particular journal; yet the research community's access problem keeps
shrinking as OA grows.

Nor is there a hodgepodge to worry about: OA means open access to the
article; that is what authors self-archive. They may also self-archive
the unrefereed preprint, and later revisions, and other things too,
but the measure of the amount of OA there is at any given point is the
percentage of the annual 2.5 million articles (published in the world's
24,000 peer-reviewed journals) that are openly accessible on the web:
currently 5% through (1) being published in OA journals and about 20% (2) through being self-archived, much
of it accessible through the "google" of the OA literature, OAIster:

One can speculate about hypothetical transition scenarios, and I and
others have (Harnad 1997, 2001), but there is nothing either speculative
or hypothetical about what is needed now, which is a systematic policy
on the part of universities and research institutions worldwide to provide
OA for all their journal article output. A JISC survey (Swan & Brown 2004)

    "asked authors to say how they would feel if their employer or funding
    body required them to deposit copies of their published articles in
    one or more... repositories. The vast majority... said they would do
    so willingly."

That is what the UK Select Committee should be worrying about building up
(not about a counterfactual breakdown), if we are all to reach the optimal
and inevitable while we are still compos mentis and able to benefit from


    On May 4 Richard Poynder replied. [My unpublished
    comments are interpolated with his reply below:]

    RP: I thank Stevan for his comments. My understanding is that Stevan
    believes OA could be adequately achieved if researchers self-archived
    their published papers and publishers downsized to a peer-review
    role only. I doubt this is a likely scenario.

[No, I think OA would be *fully* achieved if researchers self-archived
all their published articles, right now. (The downsizing is just
speculation about the possible sequel.) Hence it is not clear what it
is that RP is doubting here. The first prediction is a certainty: (1)
100% self-archiving *will certainly* provide 100% OA. The second is
a speculation: (2) 100% OA *might* lead to publisher downsizing:

The important question is *when* will the research community at last
understand that it is needlessly losing precious research impact daily
until it goes ahead and provides 100% OA to its own articles? Speculations
about how the publication system might adjust to this afterward are far,
far less important and urgent then getting it to happen sooner rather than

    RP: Firstly, today's commercial journal publishers will surely exit the
    market if their profits fall dramatically, which the above scenario

[To repeat: OA is fully achieved if all articles are made OA by self-archiving.
That is the only part that is *sure* (i.e., that 100% OA = 100% OA.)

That journal profits will fall dramatically is *hypothetical*. (There
is no evidence for it from self-archiving so far, even in the areas of
physics where OA has already reached 100% years ago.) I could speculate
that profit reduction would be likely after 100% OA is reached, but that
is certainly not sure; even less sure is when OA from self-archiving
will reach 100%! and hence whether/when downsizing pressure would begin
to grow.

That journal publishers would exit the market if profits fell dramatically
is also speculation. (How dramatically? How soon? Which publishers? Some?
Many? Most? All?)

Only two things are certain: That (1) 100% OA is optimal and inevitable
for research and researchers, and (2) that 100% OA is 100% attainable
through self-archiving, now.

Apart from that: "Hypothesis non fingo"!]

    RP: Presumably, they would be replaced by new OA publishers, but
    in a disjointed fashion.

[Hypothesis non fingo. The only two certainties are (1) and (2). Why
speculate about the possible future evolution of the peer-reviewed journal
publication system in response to OA provision when (1) OA provision is
a certain benefit to research and (2) providing OA is certainly in the
hands of researchers?]

    RP: Secondly, given the significant budgetary pressures that
    librarians face, they will meanwhile continue to cancel journal

[Journal cancellations have nothing to do with OA or self-archiving! They
are happening because journal prices are too high and libraries can't
afford them. Please don't conflate the journal affordability problem
with the access/impact problem. They are not the same problem. And it
is certainly not OA that is causing journal cancellations today!]

    RP: Self-archiving, therefore, will likely prove a temporary phenomenon,
    as we undergo a transition from conventional to OA publishing.

[This is now purely RP's speculation, though I would agree that if/when
100% OA prevails through self-archiving, it is likely that journal
publishers will eventually convert to OA publishing. Let us focus now
on getting this "temporary phenomenon" to become an *actual phenomenon*!
The goal, after all, is 100% OA, not hypothetical publisher conversion!]

    RP: During the transition, it will become more difficult for
    researchers to find the papers they need, since increasingly they
    will find themselves behind a subscription firewall.

[Here not only do I disagree (100%), but I am even puzzled that RP is
saying this. Papers are behind a subscription firewall *now* (just as
cancellations are going on *now*). As OA increases with self-archiving
papers will be *decreasingly* behind a subscription firewall and
*increasingly* easy to find (thanks in part to OAI interoperability). I
have no idea where RP came up with these predictions, both contrary to
fact and to logic!]

    RP: The papers may also be "out there somewhere" on the Web, but
    finding them could be challenging.

[It is not at all challenging to find OAI-compliant papers on the web today
(see OAIster ) and in some cases it
is easy to find the non-OAI-compliant ones too (see citeseer ). And surely RP has seen the announcements
of how google and its collaborators are working to make this still easier?
And surely as OA content increases to 100% this can only get still easier!
So what is this "challenge" RP has in mind? (other than the challenge of getting
more articles self-archived, hence OA)]

    RP: Valuable as services like OAIster are, they cannot (yet) match
    products like ScienceDirect -- particularly given the varied nature of
    the content. OAIster today covers just 277 institutions and publishes
    a warning about duplicate items and dead links.

[All true, but OAIster's main difference from ScienceDirect is that it
has as yet far less than 20% of the articles published yearly! The duplicate
items, dead links and variegated contents are *not* the problem! It is the
absence of 80% of the articles that is the problem that needs to be remedied!
Self-archiving them, now, is the solution.]

    RP: The challenge, therefore, will lie in managing the transition, which
    is why the Select Committee would do well to discuss self-archiving.

[Well, I can't disagree about the need to discuss self-archiving, but
the "transition" that needs to be managed is the actual transition from
c. 20% OA to 100% OA (i.e., the optimal/inevitable), not the hypothetical
transition from non-OA to OA publishing!]

    RP: With regard to the legality of researchers self-archiving papers
    where copyright has been assigned to the publisher and permission to
    self-archive unforthcoming from that publisher, Stevan is probably
    referring to the "pre-print/corrigenda" strategy. I doubt any
    publisher would sue, but I am not aware that this has been tested
    in the courts.

[No, I am referring to the Romeo Policy Table of publishers/journals that
have already given their official green light to self-archiving: The
percentage of green (self-archiving-friendly) publishers rose from 42%
to 58% from 2003 to 2004, and the percentage of green journals rose
from 55% to 83%! So let's not worry about hypothetical lawsuits and
instead applaud the publishers, who are demonstrating their support
for the research community's expressed need and desire for OA. It is
now time for the research community to stop insisting on publishers'
immediate conversion to "gold" (OA) publishing, and to take them up on
going green! And it is time for the library community to stop conflating
the journal affordability problem with the research access/impact
problem, and to support OA self-archiving as fervently as they support
OA publishing! Time to stop fussing about counterfactual speculations
and start acting upon the facts!]

    RP: I think many researchers do view self-archiving as a form of civil
    protest. Based on public statements from libraries like Harvard,
    Cornell, and Stanford, many librarians clearly see journal
    subscription cancellation in that light.

[The researchers who do not *do* self-archiving may see it as civil protest, but
the ones who are *doing* it (some for over a decade now) see it simply as
the natural way to maximize the impact of their research in the web age,
by maximizing access to it.

Librarians are certainly cancelling journals to protest high prices, but that is
ordinary market pressure -- supply and demand -- not "civil protest." Nor does it
have anything whatsoever to do with OA!]

    RP: Indeed, librarians will be puzzled by Stevan's assertion that they
    cannot cancel journals unless their users no longer need access. They
    may also resist his claim that the journal-pricing problem is separate
    from the research-access problem. I fear that the greatest casualty
    of the scholarly publishing crisis will be the traditional goodwill
    between librarians and academics.

[Librarians can certainly cancel journals if they can't afford them or
don't want to, and they are doing so, and have been for years. So that
clearly cannot be what I meant above! What I meant was that if librarians
wish to cancel journals *because their contents have become 100% openly
accessible*, they will not *know* when that threshold has been reached for
some time to come because self-archiving grows anarchically, article by
article, not journal by journal.

    "The Green Road to Open Access: A Leveraged Transition"

Librarians will undoubtedly (and deservedly) get the historic credit for
having alerted the academic community to both the journal affordability
problem *and* the access/impact problem (the two are not entirely
unrelated). Many, many librarians have also been invaluable allies --
and in some cases the prime movers -- in the OA movement.

I hope it does not betoken an absence of appreciation or of good will
to *also* point out that *some* librarians have been so preoccupied
with problems other than OA (such as journal affordability, electronic
preservation, intellectual property rights) that they have mistakenly
conflated them with OA -- to the disadvantage of both.

We need to disentangle these crossed wires -- but I hope we can do it in
a good-natured way, preserving the good will that should prevail between
these two gentle and interdependent communities (librarians and academics)!]


Now a few excerpts from Richard's latest article. In it, some confusion
arose because of the (common, regrettable) conflation between OA itself
and OA publishing (the "golden road to OA") in particular.

    "U.K. Academics and Librarians Disagree Over
    Open Access Publishing" Information Today May 3

    RP: "Much of the discussion revolved around OA (Open Access)
    publishing. OA, said [Fred] Friend, is an appropriate response to
    the current monopolistic environment."

[Here we once again see the familiar double-conflation: (a) OA with OA
publishing, and (b) OA with the journal pricing problem.

The access/impact problem is *not* the same as the journal pricing
problem, and it would *not* solve the access/impact problem even if all
24,000 journals were being sold *at cost*! Even then, most universities
could still not afford access to most journals, not all would-be users
could access all articles, and their potential research impact would
continue being needlessly lost. Lower ("nonmonopolistic") journal prices
would solve the journal pricing problem, but only 100% OA will solve
the access/impact problem.

It is this double conflation that sometimes prevents even OA-savvy
librarians like FF from giving their full support to OA self-archiving
too (often subsuming it instead, if not under the pricing agenda, then
under the preservation agenda, with which it likewise has almost nothing
to do!) or even from quite understanding what OA self-archiving is about,
or for!]

    RP: "The problem, [FF] later explained, is that those who pay for the
    journals (librarians) are not the people who make the buying decisions
    (academics). Since OA requires that researchers pay to be published"

[Again, the OA/OA-publishing conflation: It is not *OA* that requires
researchers to pay, it is *OA publishing*! OA is about access and impact,
not about journal pricing.]

    RP:" When academics were called it was apparent they viewed things
    differently. David Williams, professor of tissue engineering at the
    University of Liverpool, was critical of OA, and disputed there was
    a crisis."

[There is no *crisis* for academics! To claim there is a crisis is again
to conflate the access/impact problem with the journal affordability
problem (aka the "serials budget crisis"). They are not the same problem
(and though there are some causal connections, they are weak ones,
not substantial ones).

For academics the problem is that a newfound, web-based opportunity
to maximize the impact of their research by maximising access to it
is being *lost*, even though it is fully within reach. Most academics
are not yet even aware of this new opportunity. They think they are
"sitting pretty," indeed better off, not worse, since the advent of the
Web (as indeed they are):

They haven't the faintest idea that they are needlessly and avoidably
losing dramatic quantities of potential research impact.

So what the OA community needs to do (and is doing) is to provide hard
empirical evidence of how dramatically OA enhances impact:

And this evidence must also be applied to university publish-or-perish
policies, updating them to mandate OA provision for all journal article
publications, so as to maximizing their impact by ensuring that all
would-be users worldwide have access to them:

    RP: "While supportive of self-archiving, Nigel Hitchin, professor of
    mathematics at the University of Oxford, was also doubtful about
    OA publishing... The only academic to support OA was professor
    James Crabbe"

[Yet again, the conflation of OA and OA publishing: Nigel too was
supporting OA, in supporting self-archiving!

In fact, OA self-archiving provides far more OA than OA publishing --
at least 3-10 times as much, according to various estimates. But far more
important, it could be providing at least 83% (Romeo) and closer to 100% OA --
once academics and their institutions become apprised of the empirical
facts (about OA and impact) and the immediate technical possibilities.

    RP: In short, academics were skeptical about OA publishing. Outside
    the committee room, critics complained that, in their choice of whom
    to call, politicians had shown a bias towards publishers.

[Academics are mostly not skeptical but *uninformed* about OA, access and
impact. There was not so much a bias toward conventional publishers on
the part of the UK Committee (the baseline proportions -- 23,000 non-OA
to 1000 OA journals -- already see to that) as there was a bias toward
seeing OA as OA publishing and seeing the OA problem as the journal
affordability problem. This too is a matter of being uninformed: The
remedy is again facts, the objective data, plus a little logical

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:
        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.
Received on Tue May 04 2004 - 14:39:52 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:27 GMT