Peer review-Faseb J. editorial

Allan Frey afrey at uunet.uu.net
Sat Oct 28 19:11:04 EST 1995


In view of the continuing discussion of peer review, I thought the view
of an editor of the Journal of the Federation of American Societies for
Experimental Biology (FASEB J.) might be of interest.  The following is
an editorial that I wrote and which recently appeared in the FASEB J.

Allan

Allan H. Frey                           email afrey at uunet.uu.net
11049 Seven Hill Lane                   voice 301.299.5181
Potomac, MD 20854, USA


The courts are considering whether the identity of peer reviewers should
be secret.  Should we consider it?

Allan H. Frey
11049 Seven Hill Ln.
Potomac, MD 20854  USA
afrey at uunet.uu.net

It has been a fundamental tenet of science that the names of grant
application and manuscript reviewers must be kept secret in order for the
system to work.  But is that true?  The courts may soon force the
scientific community to find out.  The first round concerning the question
of secrecy has been fought in Federal District court; the second round is
about to begin in a Federal Appeals Court (1).

Suit was brought in the US District Court for the District of Columbia to
force the NSF to reveal the names of the reviewers who wrote the
evaluations of a rejected grant application (Henke vs US Dept of Commerce
and NSF).  The plaintiffs believe that some of the reviewers may have had
a conflict-of-interest.  In the first round, the District Court issued a
summary dismissal order based on the technical point that the reviewers
were promised secrecy; consequently, the names are exempted from
disclosure under the Privacy Act.  Thus, the real issue for science was not
addressed.  The plaintiffs are appealing to a Federal Appeals Court.

Given the social changes that have been occurring in recent decades, it
would not be surprising if a higher court found that the names should not
be secret; if not in this case, in a subsequent case that will surely follow.
Some people believe that the recent US Supreme Court ruling on Daubert vs
Merrill Dow Pharmaceuticals will be used to subpoena secret records of
granting agencies and journals.  Thus, a full debate on this secrecy issue
within the scientific community at this time would be timely.  It would be
much better for the scientific community to actively debate the issue, and
explore and test alternatives in an orderly fashion, than to wait and find
itself suddenly forced to make a precipitous change.

Through such an open debate and exploration, the scientific community
may actually come up with something better.  The present system of
secrecy has certainly not been without its problems. With reason, it has
been argued that secrecy in peer review shields unqualified reviewers and
biased editors; and invites abuse by reviewers who may have conflicts of
interest, who may steal ideas and who may deliberately delay publication
(2-4).

On the other hand, an apparently reasonable argument can be made that the
identity of people who do peer reviews should not be revealed.  But
implicit in this latter argument are assumptions that are not true in real
life.  One implicit assumption is  that scientists  (reviewers) are as the
public often images them,  objective, dispassionate, all-knowing beings.
We know they are not.  A "peer" is  often a direct competitor for fame and
funds and this influences his  actions in science as much as if he were in
business or sports.  Scientists (reviewers) are  human beings with all the
faults and motivations that you find in every  other field.  Once upon a
time, when the body of knowledge was small, an editor could  filter out
such abuses or could recognize the validity of an argument in  rebuttal.
But now, the body of knowledge is so large and science is  divided into so
many sub-specialities, that no editor or program manager  can prevent
these abuses.

Further as Judson has pointed out "...Although peer review and refereeing
seem rational, indispensable, and immutable, the histories demonstrate
that they are social constructs of recent date.  They are not laws of
nature, nor of epistemology."  He also points out that "... as over the years
an ever-smaller proportion of grants gets funded and as the applications
themselves ... are more difficult to put in any reliable rank order, politics
become overt in the review process.  Rivalries between scientists,
laboratories, and schools of thought emerge as palpable factors." (5).
Given these facts, is secrecy wise?  Is it necessary for peer review to
work?

It can, and has been argued, that a primary reason secrecy is required is
that scientists will be afraid to review if their identity were revealed.
There is something to this argument; but they may also feel constrained to
give a fair review.  And consider the fact that the reviewer's name is
always published with his review of a scientific book. Consider, also, that
we already have other indications that secrecy is not necessary for peer
review to work.  Scientists post mss on Internet newsgroups now and
receive lively criticism from readers (reviewers) whose identity is not
secret.   I put on the Internet for criticism sections of a book ms which
was just published (6).  I didn't perceive anyone pulling punches by any
means, and they knew I would know their identity.  Maybe the expanding
use of the Internet will resolve the question for us.

As a scientist who has worked full-time in  research,  I have come to the
conclusion that the identity of reviewers  should not be kept secret.  I
personally tell editors and granting agencies  that they may reveal my
identity to authors.  I think science and scientists are ill-served by the
practice of secrecy and, ultimately, Society which benefits from and pays
for scientific  research is harmed.

Consider also that this matter of secrecy is not a new problem for
Society.  Society has struggled with it for a millennium in other forms and
in other contexts.  This  question in science is merely its latest
incarnation.  Its a matter of  balancing the pros and cons, and the balance
that our Society has arrived at is incorporated in our Constitution as the
clause concerned with due process.  In a criminal trial, witnesses are
sometimes in fear for their life, but our Society through long experience
has decided that, on balance, it best serves Society to allow the defendant
to know the identity of and to cross-examine the accuser.  Why should we
reviewers for journals and granting agencies be an exception to these
rules that were incorporated in the Constitution after a millennium of
effort to find the right balance.  In my judgment as a scientist, science
and Society would be better off without the veil of secrecy.  Certainly
there will be problems, just as there are now with the veil of secrecy.
But on balance, with due process, science will find ways to manage as the
courts have and will be better off.

References
1.  Henke vs US Dept of Commerce and the National Science Foundation.  US
District Court for the District of Columbia, Civil action 94-0189
2.  Rennie, D., (1993) More peering into peer review.  JAMA  270,  2856-
2858
3.  Osmond, D. H., (1983) Malice's Wonderland: research funding and peer
review.   J. Neurobiol. 14, 95-112.
4.  Broad, W. J., (1980) Imbroglio at Yale, I and II,  Science. 210, 38-41,
171-173.
5.  Judson, H. F., (1994) Structural transformations of the Sciences and the
end of peer review.  JAMA. 272, 92-94.
6.  Frey, A. H., ed (1994) In "On the Nature of Electromagnetic Field
Interactions with Biological Systems". R. G. Landes Co. Austin.



More information about the Bioforum mailing list