The Excommunication of a Heretic

26 Nov 2007

The article below is translated from an original article in the Swiss German-language newspaper WOZ

by Roland Fischer

An unusual article was published in the September printed edition
of the science magazine "Nature Biotechnology". The editor of the
magazine had arranged a sort of "triangular" interview. In one
corner he had invited a Russian scientist to answer a few critical
questions about her feeding study of GM soybeans. That was
researcher Irina Ermakova, who had already presented her initial
results at conferences, and who now gladly agreed to help. The study
had created considerable controversy, since Ermakova had reported
toxic effects on the offspring of laboratory rats, leading to stunted
growth and low survival rates.

In another corner, prior to publication, Ermakova's responses were
submitted to four other researchers, allowing them total freedom to
demonstrate to their satisfaction the shortcomings of her study.
Their criticisms were printed, and even became the main part of the
article. Ermakova was not given the opportunity to respond to their
damning comments, and actually saw them in the final version of the
text for the first time on the day when the issue was published. The
editor was kind enough, on publication day, to send a PDF of the
finished article to Ermakova.

A scientific journal's publication route

Andrew Marshall, the editor of "Nature Biotechnology," argues that
there were "logistical reasons" for the manner in which Ermakova was
treated. He claims that there would have been an endless back-and-
forth dialogue if her criticisms before printing should have been
permitted. That would have involved adjustments to the text, which in
turn would have involved changes in the comments of her critics.
"This has to stop somewhere," said Marshall in justifying his action.

Where this "somewhere" lies, however, is entirely within the
discretion of the editor. In the case of the Ermakova article he
opted for the simplest variant and allowed no editorial exchange
whatsoever. This is strange, especially for a scientific publication.

Usually scientific journals follow a meticulous process. A scientist
who thinks he/she has discovered something remarkable follows a
submission procedure according to strict formal rules. Experts are
selected for the evaluation of a submitted paper. A referee can
either flatly refuse to comment or, most commonly, make suggestions
to the author for improvement. An author can prepare a revised
version, which is then re-assessed and (possibly after further
additional alterations) may be published or not, depending on
referees' recommendations. This so-called peer review system has its
flaws (some promising results from direct competitors may be slowed
down or rejected by partisan referees), but at least the mechanisms
are transparent. The rules of the game are clear for all concerned.

The treatment of the Ermakova journal article was not remotely like
this. It was a strange mix of interview and written examination.
Indeed, in many scientific publications in recent years, the
"journalistic part" has been enlarged. Because, for outsiders,
journal articles are often about as exciting to read as meeting
protocols, this is a move by "Nature Biotechnology" away from
specialist science, with a view to enabling wider audience access.
Marshall himself says that the Ermakova article presented him with a
challenge. "We have never before published material with this
format," he says. Nevertheless, in conversation he repeatedly
refers to "normal procedure" in order to justify his actions.

Speared by the critics

The fact that "Nature Biotechnology" has been in "uncharted waters"
with this article is confirmed by Harvey Marcovitch, former editor of
a scientific journal and now director of COPE (the Committee on
Publication Ethics), an organ of journal editors working mainly in
the medical field. "This is a type of publication which I have never
encountered," says Marcovitch. In fact, while reading it he was
struck by "some surprising things." He is unwilling to speculate as
to what exactly happened: "Either the Editor was experimenting with a
new journalistic format, in which not everything went according to
plan, or there was indeed something more sinister, a conspiracy or
whatever one wants to call it." As long as nothing could clearly
indicate the latter, however, he prefers to think the former.

If you look more closely at the background to the publication of
this article by "Nature Biotechnology", however, doubts are raised
about an innocent journalistic experiment. One thing is obvious: the
article is anything but balanced. The supposed experts who reviewed
Ermakova's work hardly had a good word to say about her. And they
were so intent upon "shooting the messenger" that they criticized
aspects of her work on which they themselves had no expert
knowledge. Marshall himself is forced to admit this. When asked
whether the four would be acceptable as referees in a peer-review
process, he replies evasively that for "some aspects" they might be
included. But in practical questions about feeding studies or
regarding animal physiology and toxicology all four referees should
have had professional expertise. They had sought additional
expertise, says Marshall. One can imagine where. Because the four
men are not impartial or unknown. They are all well known as GM
advocates, with a variety of relationships with industry.

How come that a publisher of a supposedly independent magazine
managed to select four experts who were not exactly impartial? The
answer is simple: he did not need to select them and did not even
need to look, since the whole thing was the idea of the critics
themselves. They had sent Marshall a message in the summer, and even
proposed that they should attack Ermakova. Marshall tried to give a
somewhat more balanced appearance to the feature article by not
leaving the stage entirely to the critics; but he did not regard it
as necessary to inform Ermakova about what was going on. To
understand things from the perspective of an editor, this hot topic
was too good to miss: but Ermakova has said herself that if
everything had been transparent she would never have agreed to
participate in the game .

"Nature Biotechnology" is now allowing the Russian researcher the
possibility of replying to her critics in a subsequent number of the
journal. Marcovitch finds this to be an unsatisfactory solution: "An
author must always have the opportunity to respond to criticism,
preferably in the same number." Indeed, the publisher must accept
the question why the Ermakova study results, which might not stand up
to the rigorous scientific requirements of a peer review process and
which might therefore not be published, were not simply ignored.
Ermakova has never made a secret of the shortcomings of her studies,
saying that she has always been open to suggestions for improvement.
Due to her good faith, she was an appreciative and innocent victim
for a "show trial".

The response from the industry came immediately. In the newsletter
"Inter Nutrition", syndicated by the Swiss Federation of Genetic
Engineering, the case of Ermakova was presented as an exemplary belly-
flop by a research scientist who dared, without respectable results,
to report something that might be damaging to GM food. The GM
lobbyists couldn't pass up this opportunity of destroying the
credibility of all of their critics at a single stroke.