Questions To Ask Scientific Authority
by Wendy McElroy
of the media and the scientific and medical community has recently
been focused on the fraud committed by South Korean scientist Hwang
Woo-suk, who claimed to have created human cells from cloned
has centered on Miodrag
Stojkovic, the scientist who was instrumental in cloning the
first human embryo in Britain. Stojkovic recently resigned from
his celebrated post at Newcastle University. Now doing research
in Spain, he is leveling accusations of scientific impropriety toward
Stem cell research
may be the most sensational and explosive scientific development
of recent years. But the most heralded breakthrough was a massive
fraud; another has prompted a pioneering scientist to cry "foul."
It becomes important to remember the admonition, "Question Authority."
is an arcane mystery to lay people, like me, who must rely to an
uncomfortable degree upon expert opinions. We live under the medical
and political policies that can often proceed from research.
Our hopes hinge
on technologies like stem cell research which may point to
cures for diabetes or Alzheimer's Disease. Our fear creates opportunities
for modern versions of snake-oil remedies: for example, the "genetic
on the Internet which allegedly measure the risk of developing
diseases but which scientists call a "waste of money."
educated media asks the skeptical questions that protect public
interest. That ideal is rare. And, so, "Question Authority" becomes
personal responsibility. But which questions should be asked?
The facts of
the two controversies provide some guidance.
Hwang was a
case of outright fraud but Stojkovic's research has not been discredited.
Rather, his allegations concern misconduct in how research has been
credited and presented. One criticism: the university ignored sound
scientific practice by announcing a breakthrough to the press before
peer review had been conducted. The timing seemed designed to steal
publicity from the then-lauded Hwang whose paper hit the press at
the same moment.
weekly journal of science Nature responded with a June 2,
entitled "Too much, too soon: How not to promote your latest research
findings in the media." Key to Nature's condemnation was
the fact that a "full research paper is kept confidential until
it is accepted and published" which meant that the media did not
have access either to the Newcastle team's data or to an informed
review. They were merely given conclusions with which to run.
who co-operated with the press announcement, now claims to have
been blindsided by the university.
truth, common sense questions should be asked Hwang, the Newcastle
team and every other researcher.
The first question
pertains to the nature of any claim. Are the results "statistical,"
or do they proceed from an unambiguous "yes/no" experiment?
results generally involve observing data from which correlations
can be drawn to indicate possible cause-and-effect. An example is
the much-acclaimed research on mouth cancer for which Dr.
Jon Sudbo of the Norwegian Radium Hospital observed a database
of 908 participants. Sudbo has admitted
to fabricating his database. Many questions addressed to statistical
studies involve little more than closely analyzing the specifics
of the data. For example, when 250 of the 908 people studied by
Sudbo shared the same birth date, a red flag should have fluttered.
as those claimed by Hwang and the Newcastle team, are "yes/no."
That is to say, the cells and embryos were either cloned in the
manner indicated, or not. The questions addressed to "yes/no" experiments
may be more fundamental than those addressed to statistical claims
but all research should be able to answer them. Those questions
- Is the report,
including all data and methodology, available for examination?
If not, then the researcher is asking you to accept his word for
- What is
the researcher's reputation? More credibility should be accorded
to the claims of a scientist with a sound track record than to
an unknown factor who comes out of nowhere.
- Who funds
the research? A questionable source of money does not invalidate
research but public skepticism should sharpen if the funder stands
to profit from a specific finding and, indeed, that finding results.
- Have the
findings been independently verified? Claims should be sufficiently
documented to allow replication. (Unfortunately non-scientific
concerns, like patents, sometimes interfere with disclosure.)
- Does the
claim contradict previous data? A breakthrough that achieves a
difficult result is qualitatively different than one that achieves
a result previously believed impossible. A "paradigm shift"
demands a high degree of proof because it involves invalidating
- Does the
claim include policy recommendations or changes in law? Research
that includes a political agenda is more likely to express the
researcher's personal beliefs than work that merely states data
- What is
the response of the scientific community?
- Where was
the research published? The differing levels of prestige for scientific
journals has been quantified in terms of their "impact
factor." If a researcher publishes in a low-impact journal,
then asking "why" becomes appropriate.
questions do not guarantee that fraudulent or incompetent work will
be detected. For example, Hwang's work was heralded by the prestigious
Sudbo's work was published in both the New England Journal of
Medicine, in April 2004, and the Journal of Clinical Oncology,
in March 2005. The scientific community, like the media, is simply
not doing its job.
these questions becomes more necessary. The claims of scientific
authority should receive the same skepticism that usually greets
similarly bold claims of political authority. Both impact your life
and are your business.
McElroy [send her mail]
is the editor of ifeminists.com
and a research fellow for The
Independent Institute in Oakland, Calif. She is the author and
editor of many books and articles, including the new book, Liberty
for Women: Freedom and Feminism in the 21st Century
(Ivan R. Dee/Independent Institute, 2002).
© 2006 Wendy McElroy