Critique of AAP/PSP Critique of FRPAA Proposal

From: Stevan Harnad <>
Date: Sat, 10 Jun 2006 18:12:35 +0100

The latest AAP/PSP critique of the latest US Public Access Bill (FPRAA)

makes much the same points that were rebutted two years ago,
in their critique of the NIH Public Access Proposal:

    "Critique of PSP/AAP Critique of NIH Proposal"

Peter Suber has already sounded the right overall note by way of reply
in OA News (see also his 10 detailed points, much the same as mine):

    (a) There is zero evidence that mandating self-archiving reduces
    subscription revenue.

    (b) But even if self-archiving were ever to reduce subscription
    revenue, surely what is in the best interests of publishers' current
    revenue streams should not dictate what is in the best interests of
    research and of the public that funds it.

> "The proposed legislation would require the majority of
> recipients of U.S. federal research agency funds to make their
> findings free within six months of publication. Publishers argue that
> the legislation, if passed, will seriously jeopardize the integrity
> of the scientific publishing process, and is a duplicative effort
> that places an unwarranted burden on research investigators."

AAP provides no evidence of how making research findings accessible for
free to would-be users who cannot afford it would "seriously jeopardize
the integrity of the scientific publishing process." They merely stipulate
that it would.

Nor is it clear why AAP is speaking on behalf of researchers about
"unwarranted burdens". Surely enhanced research usage and impact
is not an unwarranted burden for research and researchers?

> "According to the publishers, the provisions of S.2695 threaten
> to undermine the essential value of peer review by removing the
> publishers' incentive and ability to sustain investments in a range
> of scientific, technical, and medical publishing activities."

Translation: "Self-archive and I may not want to publish journals
any more."

Peer review is done by researchers, for free. Whoever funds the
management of peer review and the certification of its outcome is a
journal publisher. There is no evidence that self-archiving reduces
subscription revenue but even if there should ever be such evidence it
certainly does not follow that research and researchers should renounce
the demonstrated benefits of self-archiving. If/when some publishers
should ever become dissatisfied with reduced subscription revenues, their
journal titles can migrate to other publishers who are not dissatisfied,
or to Open Access ("gold") Publishers.

Surely demonstrated benefits -- in terms of increased research impact
-- to research, researchers and the public that funds them are not
to be sacrificed in order to insulate publishers from undemonstrated
hypothetical risk to their current subscription revenues.

    Bibliography of Findings on the Open Access Impact Advantage

> "The proposed legislation comes at a time when increased public access
> to government-funded research is already occurring in a voluntary
> and highly effective manner through a variety of publisher-initiated
> mechanisms and cooperative approaches."

"Highly effective" for whom? The fact is that many researchers cannot
afford access to much needed research, and the proof of this is the fact that
when subscription access is supplemented by author self-archiving, research
usage and impact increase dramatically.

(Note that the issue is not primarily public access to research, but
researcher access to research, in order to maximize the benefits of
research to the public that funds it.)

> "Americans have easy access to scientific and medical literature
> through public libraries, state universities, existing private-sector
> online database, as well as through their professional, academic,
> or business affiliations, low-cost online individual article sales,
> and innovative health literacy initiatives such as patientINFORM."

The primary objective of Open Access is to provide access to researchers,
worldwide, for the sake of research uptake, usage, applications,
and progress, by way of a return on the public's investment in the
research. Researchers do not now have nearly as much access as they need,
because no research institution can afford all or most of the journals
in which the research appears. The demonstrated impact advantage
of self-archived research is the direct evidence of the substantial access
shortfall there is for research that is not self-archived.

Paid or library access is certainly not what OA is about or for. OA
means online access, free for all would-be users.

> "The Cornyn-Lieberman bill would create unnecessary costs for
> taxpayers"

This is complete nonsense. Self-archiving costs are negligibly small:

    "Institutional Cost of Creating/Maintaining an OA Repository"

    "[it would] place an unwarranted burden on research investigators,"

Again complete nonsense. Self-archiving takes a few keystrokes:

    Carr, L. and Harnad, S. (2005) Keystroke Economy: A Study of the
    Time and Effort Involved in Self-Archiving.

> "[it would] expropriate the value-added investments made by
> scientific publishers-many of them not-for-profit associations who
> depend on publishing income to support pursuit of their scholarly
> missions, including education and outreach for the next generation
> of U.S. scientists"

Nothing whatsoever is "expropriated": Publishers can continue to sell
subscriptions and licenses for their paper and online editions, exactly
as before. The author's self-archived final draft is not a substitute
but a supplement, online only, for all would-be users who cannot afford
the publisher's version. And so far there is no evidence whatsoever that
self-archiving reduces subscription revenues at all, even in the areas
that have been doing self-archiving the longest (15 years in high energy
physics, even longer in computer science) and that are already at or
near 100% self-archiving for years now.

     "[W]e asked the American Physical Society (APS) and the Institute of
     Physics Publishing Ltd (IOPP) what their experiences have been over
     the 14 years that arXiv has been in existence. How many subscriptions
     have been lost as a result of arXiv? Both societies said they could
     not identify any losses of subscriptions for this reason and that
     they do not view arXiv as a threat to their business (rather the
     opposite -- in fact the APS helped establish an arXiv mirror site
     at the Brookhaven National Laboratory [and shortly the IOP will
     host one too]).

    Swan, A. (2005) Open access self-archiving: An Introduction.
    Technical Report, JISC Survey.

Not-for-profit publishers (e.g. Learned Societies) do not differ in any way
insofar as any of these considerations are concerned: There is abundant
evidence that self-archiving increases research usage and impact and no
evidence that it reduces subscription revenue. And research is not funded,
conducted and published in order to generate revenue for publishers,
let alone in order to guarantee their current revenue streams and insulate
them from any risk. In particular, what has already been demonstrated
to be in the best interests of research outweighs what has not even been
demonstrated to have any negative effects on the interests of publishers.

> "If enacted, S.2695 could well have the unintended consequence of
> compromising or destroying the independent system of peer review
> that ensures the integrity of the very research the U.S. Government
> is trying to support and disseminate."

Pure nonsense. See prior reply about peer review, done for free by
researchers (the peer reviewers); publishers merely administer it,
and for any publishers who may no longer wish to administer it, other
publishers will be happy to do so in their place.

> "publishers invest hundreds of millions of dollars each year in
> publishing and disseminating peer-reviewed journals. These investments
> ensure the quality of U.S. taxpayer-supported scientific research by
> subjecting all articles to a rigorous technical review by experts in
> specialized fields prior to publication and pay for the development
> of technological innovations that enable broad web dissemination."

Quality is ensured through peer review done by the research community
itself; and the peers review for free. Publishers merely *administer* the peer
review, and in exchange they get to charge for the paper edition as well
as the online edition. There is no evidence whatsoever that self-archiving
diminishes their revenues from any of this, and if/when it should ever do
so, the solution is certainly not to *not* self-archive, and thereby deny
research of self-archiving's substantial benefits in terms of research
uptake, usage, applications, impact and progress.

The solution -- if/when subscription cancellation pressure were ever
to happen -- would be to cut costs and adapt, scaling down to the new,
smaller but still essential niche of peer-review service provision that
will remain for peer-reviewed research journals in the PostGutenberg
age even if no one wants to pay for the paper edition or the publisher's
official online version any more because the author's self-archived draft
is enough. The solution is certainly not to deny research, researchers
and the public that funds them the benefits of the research impact and
progress that self-archiving brings them.

> "Mandating that journal articles be made freely available on
> government websites so soon after their publication will be a
> powerful disincentive for publishers to continue these substantial
> investments."

At the moment, over 90% of journals have given immediate author
self-archiving their green light. If some publishers are not happy with
conferring this benefit on their authors' research, there are plenty of
other publishers for their journal titles to migrate to (including the
new breed of Open Access "gold" publishers).

> "publishers are concerned that S.2695 would result in a significant
> loss of revenue from subscriptions, licensing, and individual article
> sales, thereby making it difficult for them to sustain and recoup
> the investments they make in support of scientific communication."

There is no evidence whatsoever that self-archiving has reduced
subscription revenue in the very fields that have been doing it the
longest and the most (see above). So this publisher concern is purely
hypothetical; and the actual effects to date contradict the hypothesis.

But if/when there should ever be a subscription revenue decline, the
remedy is to adapt, cut costs, drop inessentials, and downsize to the
new PostGutenberg niche for peer-reviewed journal publishing. The remedy
is certainly not to sacrifice research impact in order to sustain
current publishing revenues instead of adapting to the new technological
contingencies opened up by the newfound possibility of providing Open
Access to all research.

> The proposed bill was introduced on the first anniversary of the
> National Institutes of Health's (NIH) adoption of its Public Access
> policy, which encourages the posting of journal articles based
> on NIH-funded research within 12 months of publication on its
> existing PubMedCentral database -- a policy that gained PSP/AAP
> member publisher support and yet remains in its early stages
> of government-led implementation. A departure from the NIH's
> voluntary approach, the Cornyn/Lieberman bill would mandate that 11
> federal agencies create new systems and data repositories to enforce
> internet posting of government funded research within six months of
> publication. As the NIH's implementation of the policy has not yet
> progressed to the point where its impact can be assessed, publishers
> view the introduction of the Cornyn-Lieberman proposal as premature."

(1) The NIH policy can be and has been assessed, and it is a failure:
The level of compliance with its non-mandatory "invitation" to
self-archive is below 4% after a year. The spontaneous self-archiving
baseline worldwide and across disciplines is 15%!

(2) Meantime, self-archiving mandates (such as those of the Wellcome
Trust, CERN, and several universities) have been tried, tested, and
shown to be successful in generating high compliance rates, exactly as
the JISC author surveys had reported they would be:

     "The vast majority of authors (81%) would willingly comply with a
     mandate from their employer or research funder to deposit copies of
     their articles in an institutional or subject-based repository. A
     further 13% would comply reluctantly; 5% would not comply with such
     a mandate."

    Swan, A. and Brown, S. (2005) Open access self-archiving: An author
    study. JISC Technical Report, Key Perspectives Inc.

(3) Research is funded, conducted and published in order to be taken up,
used, and applied as soon as it has been validated by peer review.
Research is not funded, conducted and published to be embargoed so as to
guarantee publishers' current revenue streams.

> "No evidentiary record exists, and no impact studies have been
> conducted, to document the long-term cost to tax payers of government
> agencies developing yet another system to promote public access.

Surely it is not the business of American Association of Publishers to
concern itself with the cost to tax payers of providing open access to
government-funded research. But studies have indeed been done, across
disciplines, and they have found that self-archived research has
substantially higher research impact (25% - 250+%), and this translates
into substantially higher return on the tax payers' investment in
research than what they are getting for their research money today.

Competitively speaking, it also means higher salaries and more research
income for the early self-archivers. And all, as noted, at a negligibly
tiny cost per paper in terms of either author keystrokes or distributed
institutional self-archiving costs.

So it is a self-serving red herring for publishers (in reality fretting
about their own current revenue streams) to portray this as a "tax payer"

> "Moreover, no consideration has been given to what the impact
> of this government mandate will be on publishers and scholarly
> societies ability to maintain their broad base of library and
> other customers worldwide and invest in independent peer review
> systems."

The purpose of research and research funding is not to ensure publisher
revenue streams, but to conduct, use and apply research, to the benefit
of the tax payers that fund it. Peer-reviewers (researchers) review
their peers' research for free. Journal editors merely *manage* the peer
review process, and the true costs of managing peer review can and will
certainly be paid out of just a small portion of institutions' own annual
windfall subscription cancellation savings -- if and when subscription
revenues were ever to collapse catastrophically as a consequence of
universal self-archiving.

But at the moment there is not even the slightest sign of a subscription
decline: just speculations about doomsday scenarios, intended to
hold self-archiving, with all its demonstrated benefits to research,
researchers and tax-payers, at bay, so as to protect publishers' current
revenue streams from a hypothetical risk.

Surely the rational thing to do is to mandate the self-archiving now,
and then review its effects on publishers' revenues yearly, rather than to
deny its certain benefits to research on the grounds of its hypothetical
risks for publishers. (The delay has already been unconscionably long
and wasteful of research impact and progress, and will be all the more
embarrassing in historic hindsight.)

> "Responsible major U.S. government policy revisions must be based on
> a solid, researched understanding of the long-range impact of any
> policy changes. This perspective is conspicuously absent from the
> proposed legislation, which would cause severe harm to the publishing
> community, scientific societies, and taxpayers."

The long-range effects should be investigated empirically. The positive
effects of OA self-archiving for research, researchers and the
tax-payers that fund them have already been empirically tested and found
to be substantial. Meanwhile, there have been no detectable effects of
self-archiving on subscription revenues at all so far, even for the two
publishers (American Physical Society and Institute of Physics) in the
fields that have been doing it the longest and most (15+ years).

The way to test the long-range effect of the FRPAA self-archiving
mandate on subscriptions objectively and empirically is to adopt the
mandate and monitor its effects annually, not to deny or keep delaying
its already demonstrated positive effects on research impact on the
basis of undemonstrated hypothetical negative effects on publisher revenues.

> "publishers and scholarly societies urge that an independent study
> be conducted to measure the potential impact that any changes to the
> existing NIH policy or the adoption of the proposed Cornyn-Lieberman
> legislation would have on scientific quality, the peer review process,
> and the viability of numerous journals and societies--as well as
> the additional costs that would need to be shouldered by taxpayers."

The study in question is to adopt the self-archiving mandate and test
and review its empirical outcome annually. All else is mere delaying

Stevan Harnad
American Scientist Open Access Forum

Chaire de recherche du Canada Professor of Cognitive Science
Ctr. de neuroscience de la cognition Dpt. Electronics & Computer Science
Université du Québec à Montréal University of Southampton
Montréal, Québec Highfield, Southampton
Canada H3C 3P8 SO17 1BJ United Kingdom
Received on Sat Jun 10 2006 - 19:27:53 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:48:21 GMT