Memorandum from Science
1. The integrity of Science's peer review
process is assured by the familiarity our in-house editors have
with those experts in the discipline and to the history of their
participation in our process. The performance of outside reviewers
is evaluated ex-post facto; on the basis of our experiences with
reviewers, we cease using those who fail to approach the task
conscientiously, or who regularly return opinions that are at
odds with those of other, equally capable reviewers on a consistent
2. No matter how it is implemented, I cannot
see how a "pay-to-publish" business model (your term)
or "open-access model" (their term) would have any impact
on our peer review process. We undertake peer review because our
readers and those who submit to us expect it. With regard to peer
review, it costs us approximately as much to reject a paper as
to accept one, so there are questions about the scalability of
the process for any "pay-to-publish" journal as the
ratio of rejections to acceptances rises.
3. Our responsibility for papers we publish
after their publication is limited, as it is for any journal.
When we publish a paper, we make a decision about its newsworthiness.
We carefully vet the press releases prepared by the American Association
for the Advancement of Science's News Office, and we will occasionally
hold press briefings for especially important papers in order
to explain their content more carefully to members of the media.
If papers have a special appeal to "certain campaigners",
we might under some circumstances send letters of correction to
media outlets that contain distortions or misrepresentations.
But it would be plainly impossible for Science, or any other journal,
to systematically monitor all kinds of press outlets to search
for inaccuracies in the representation of our papers, and we would
consider it a misallocation of resources were we to do so. We
do supply Perspective articles and articles in our News section
that help to place the results in the appropriate context.
4. We require authors to submit statements
of possible real or perceived conflicts of interest, and publish
these if they might influence a reader's perception of the objectivity
of the report. In Science's News section we ask our journalists
to be certain to report possible conflicts on the part of persons
who are quoted or to whom attributions are made. The COPE guidelines
are among the many we have reviewed in establishing and refining
our publishing policies over the years. Others include the "Uniform
Requirements for Manuscripts Submitted to Biomedical Journals"
from the International Committee of Medical Journal Editors and
the "Ethical Guidelines to Publication of Chemical Research"
developed by the American Chemical Society. Our publication policies
are outlined in our Information For Authors/Contributors. As Editor-in-Chief,
it is my duty to see that these policies are followed. I work
closely with the Executive Editor and Deputy Editors to review
our compliance with these policies. In addition, we routinely
review these policies with our Senior Editorial Board to assure
that they evolve as circumstances change.
5. The retraction issue arises only when
one or more of the authors declare that they are (a) aware that
they are unable to replicate the result they found in the publication;
or (b) when one or more of the authors is found to have made a
mistake, or to have used fraudulently obtained data. In this last
case, we may be informed of irregularities by the authors' institution
or by the Office of Research Integrity at the National Institutes
of Health. We require that all authors must agree on a retraction.
In the event that an agreement is not forthcoming, we attempt
to persuade the reluctant author or authors to participate. We
will allow the authors' institution to retract the paper under
certain circumstances. In the event of a disagreement among authors,
we may issue an Editorial Retraction, announcing that Science
no longer regards the findings published in the report as valid.
In rare cases, authors may retract a particular interpretation
but not the main results of the paper.
You should perhaps know that Science has an
office with Editorial and News staff in Cambridge, UK Processes
of peer review are carried out both in our Washington office and
in Cambridge. Where the work is conducted has nothing to do with
country of origin; rather, it relates to the distribution of particular
areas of expertise among our Editorial staff, of whom seven are
resident in Cambridge.