Re: Mandating OA around the corner?

From: Stevan Harnad <>
Date: Sun, 1 Aug 2004 11:52:45 +0100

On Saturday, July 31, 2004 [identity deleted] wrote:

> I am apprehensive about strident Open Access advocacy.

There is still too little that is concrete happening yet, but with UK
and US governments committees recommending mandating OA self-archiving,
this cannot be described as strident OA advocacy. (We had strident OA
advocacy 10 years ago. This is sober contemplation of policy.)

> The report in the Scientist is guarded and NIH is certainly
> unlikely to be coercive.

The report in the Scientist, and most of the other press reports, stress
OA publishing, which it would indeed be unrealistic to imagine that
the US House Appropriations Committee, NIH, or the UK Select Committee
could or would mandate. Nor is that what they are recommending be
mandated. They are recommending to mandate that research funded by them
should be made OA by being self-archived by its authors, as a condition
for receiving funding.

If this sounds coercive, consider that there is already a far bigger
component of "coercion" in all research funding -- and indeed in
employment by all research institutions: The implicit mandate is that the
fundee's or employee's research findings must be made public through
publication. This mandate is so pervasive that it is often not even
formulated explicitly, though its existence is known as "publish or

Researchers are not free to do their research and then put it into
a desk-drawer and move on to the next piece of research. Or, rather,
if they do elect to do only that, chances are that their next piece
of research will not be funded, and they may have difficulties staying
employed at their research institution.

The expectation that research findings must be made accessible to
would-be users is so entrenched in the very concept of research that we
forget that the "publish or perish" contingency is necessary in order to
ensure that researchers do the right thing -- in the interests of both
themselves and research itself. For otherwise they are indeed prone by
(human) nature to want to move from one piece of learned inquiry to the
next without troubling either to answer to the scrutiny of peer review
or to make their results known.

(These days, in our online age, if there were no publish-or-perish
incentive, researchers' natural inclination would no doubt be simply to
put their unrefereed findings on the Web instead of a desk-drawer, for
it is not so much a disinclination to make them public that causes the
inertia, but the fact that being answerable to peer review is something
of a chore, and one that a busy -- or lazy -- researcher would just
as soon skip, if he could.)

So coercion there already is, universally. What is being recommended is
simply to update the existing publish-or-perish mandate to fit the
newfound potential of the online age: Maximise the usage and impact of
your peer-reviewed publications by making them openly accessible to all
would-be users, not just those whose institutions can afford to pay for
the journal in which it happens to appear -- by self-archiving them all
on the web.

(The same thing would have been said to researchers at the onset of the
print age if, counterfactually, refereed journals had predated print, in
the hand-written and hand-circulated manuscript era: Maximize the usage
and impact of your peer-reviewed publications by making them publicly
accessible to all would-be users who can afford it, by publishing them in
print, not just circulating the hand-written version. Indeed, there
is a deeper analogy here, if we consider the potential role of the
web in updating and optimising the existing practise of requesting and
mailing of author reprints!)

Moreover, the already "coercive" element by which publish-or-perish can
and will be generalised to "publish and maximise access webwide" is also
already implicit in how publish-or-perish has evolved in the past few
decades: For it is no longer just "publish and we will count the
publications" when it comes to promotion, salary, or research refunding.
Evaluation already takes the *impact* of each publication into account
too, weighting quantity with importance. This weighting used to
be derived from the established peer-review quality-standards and
impact-factor of the journal by which the article was accepted, but
these days it is also increasingly based on the specific individual author
and article citation counts.

So the instruments of "coercion" -- which are better thought of as
incentives for performance and productivity -- are already in place:
Promotion, salary, and research funding are already conditional on
research impact. OA maximises research impact. So these rewards (chief
among them being the magnitude of one's contribution to knowledge)
are already conditional on OA. It is just that most researchers don't
realize it yet, because it is still too new, and the causal
contingencies are not yet known and understood.

Maximising research access maximises research impact. This is the
principle that can and will (and should) induce research employers
and funders to mandate that their researchers do the right thing, for
themselves, for their employers and funders, and for research
itself: Maximise the usage and impact of your research both by
peer-reviewing/publishing it, and by providing OA to that publication,
by self-archiving it publicly for all its would-be users webwide.

This is the only upgrade of the existing "coercion" that is being
recommended, for the sake of research productivity and progress itself.

> Government diktats can become difficult.

This is not a government diktat. It is merely one of the (many)
conditions on receiving research funding from the government.

> The correct approach is to enhance the open archives initiative
> and persuade scientists to join the movement.

There is a decade of history (and centuries of human nature) to
suggest that waiting for OA to be provided voluntarily, without any
cost/benefit contingencies, would be like employing or funding researchers
unconditionally, without any cost/benefit contingencies based on outcome.

If we want research productivity and progress to be maximised now, rather
than whenever researchers elect to get around to it, the cost/benefit
contingencies must be implemented now. Otherwise research impact
continues to be lost, needlessly, in an online era that can at last
remedy this longstanding access problem, just as print remedied
its precursor, in its day.

> There are technical issues involved in archive generation and maintenance
> which will be hard to solve at all institutions.

The technical part is not hard, but if there are unaffiliated researchers or
researchers whose institutions cannot manage it, there already exist several
central archives to take care of such overflow, and more of them can be
created cheaply and easily. Surely the substantial benefits of maximising impact
for us all are not to be held up by the slight added complexity for a
small minority of cases? (Moreover, it is mandated self-archiving that
is most likely to induce Necessity to mother Invention, generating still
more of the requisite resources and incentives for archive generation
and maintenance.)

> Open Access is not an issue which seems to bother practicing scientists
> since the amount of accessible literature is already too vast to be
> read and digested in most fields.

This is unfortunately the expression of two familiar misunderstandings
about both the online medium and OA: "Info-Glut" and "Sitting Pretty"

The two can be encapsulated as: "There is already too much out there to
access, and I can already access more than enough of it."

There are two profound problems with this:

    (1) It implies that access-tolls are the optimal way to decide
    who accesses what. (Does the web not offer more rational ways to
    direct and filter navigation than merely to base it on the current
    ability to pay for access to a given journal?)

    (2) The user view is not the only pertinent one: Do I want my
    research output's usage and impact to be directed and filtered -- and
    restricted -- by would-be users' institutions' current ability or
    inability to pay for access to a given journal?

There are 24,000 peer-reviewed journals, publishing 2.5 million articles
per year. Current online navigational power is more than enough to
rationally navigate all that. It is already the way we navigate it,
through online indexes, but those contain only each article's metadata:
Access to the full-texts depends on our institution's ability to pay
the tolls. There is no reason the full-texts should not be accessible
toll-free too. Both authors and users would be far better off if they
were -- and so would research itself.

> I do not subscribe to the view that our collective impact will
> improve in an open access utopia.

Fortunately, that is not a utopic view, but an already demonstrated
empirical fact:

We have so far reported the OA impact-enhancement results in Physics
(same results, reported in 3 places):

    Brody, T., Stamerjohanns, H., Vallieres, F., Harnad, S. Gingras,
    Y., & Oppenheim, C. (2004) The effect of Open Access on Citation
    Impact. Presented at: National Policies on Open Access (OA)
    Provision for University Research Output: an International meeting,
    Southampton, 19 February 2004.

    Harnad, S., Brody, T., Vallieres, F., Carr, L., Hitchcock, S.,
    Gingras, Y, Oppenheim, C., Stamerjohanns, H., & Hilf, E. (2004)
    The green and the gold roads to Open Access. Nature Web Focus. Longer
    version: The Access/Impact Problem and the Green and Gold Roads to
    Open Access

    Harnad, S. & Brody, T. (2004) Comparing the Impact of Open Access
    (OA) vs. Non-OA Articles in the Same Journals, D-Lib Magazine 10
    (6) June

Others have reported similar findings in computer science, astrophysics,
and mathematics:

    Kurtz, Michael J.; Eichhorn, Guenther; Accomazzi, Alberto; Grant,
    Carolyn S.; Demleitner, Markus; Murray, Stephen S.; Martimbeau,
    Nathalie; Elwell, Barbara. (2003) The NASA Astrophysics Data
    System: Sociology, Bibliometrics, and Impact. Journal of
    the American Society for Information Science and Technology

    Kurtz, M.J. (2004) Restrictive access policies cut readership of
    electronic research journal articles by a factor of two, Michael J.
    Kurtz, Harvard-Smithsonian Centre for Astrophysics, Cambridge,

    Lawrence, S. (2001) Online or Invisible? Nature 411 (6837): 521.

    Odlyzko, A.M. (2002) The rapid evolution of
    scholarly communication." Learned Publishing 15: 7-19

> I think the issue is that high journal costs have resulted in a fight
> between publishers and customers (scientists and librarians) in which the
> latter are using the "politically correct" position that publicly funded
> research results must be freely available.

There is an element of truth in this, because the journal affordability
problem and the research access/impact problem have often been conflated
(i.e., treated as if they were the same problem, or as if the solution to
one was also the solution to the other). This is one of the reasons so many
people wrongly think that OA = OA Journal Publishing.

    Harnad, S., Brody, T., Vallieres, F., Carr, L., Hitchcock, S.,
    Gingras, Y, Oppenheim, C., Stamerjohanns, H., & Hilf, E. (2004)
    The green and the gold roads to Open Access. Nature Web Focus.

But the US and UK mandates in question are not about OA publishing
[the golden road to OA] but about OA provision, by authors, through
self-archiving [the green road to OA].

    5% percent of journals are gold (exact):

    84% of journals are green (sampled):

The green road is not taken in order to make journals more affordable;
it is taken in order to make all articles more accessible to their
would-be users. It has nothing to do with the fight between publishers
and customers about journal pricing.

> It is surprising that this high moral ground was not thought of
> earlier. There is great danger in soliciting government diktats.

This is not about moral ground but about maximising research usage and
impact. Nor is it about government diktats but about maximising the
usage and impact of government-funded research -- maximising the return
on tax-payers' investments.

Stevan Harnad

UNIVERSITIES: If you have adopted or plan to adopt an institutional
policy of providing Open Access to your own research article output,
please describe your policy at:

    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.

A complete Hypermail archive of the ongoing discussion of providing
open access to the peer-reviewed research literature online (1998-2004)
is available at:
        To join the Forum:
        Post discussion to:
Received on Sun Aug 01 2004 - 11:52:45 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:32 GMT