PREPARE to be shocked. In Germany, officials are busy unraveling a case in
which accusations of fraud could taint more than 500 scientific papers. In the
US, a graduate student commits suicide and leaves a note blaming the advising
system at an elite academic institution. And an American biotech company has
been involved in a case in which one of its scientists allegedly stole a gene
from an academic lab during a midnight robbery.
These scenarios may sound like the plot lines for a TV miniseries, but they
are all true. The accusations of fraud stem from research performed in the early
1990s by scientists at the Max Delbr眉ck Centre for Molecular Medicine in
Berlin. The elite institution at which the suicide took place is Harvard
University and the American biotech company is Genentech, one of the world鈥檚
most successful. And these are only a few of the crimes and misdemeanours that
haunt science and scientists the world over. It would be just as easy to pick
cases of plagiarism, falsification of data or discrimination.
A growing number of scientists are worried by these and other cases. They say
that the pace of science has outstripped researchers鈥� ability to tackle the
ethical issues it raises, that the 鈥渢ribal鈥� culture of science is preventing
proper discussion, and that science and scientists must change or face a gradual
but certain moral decline. Currently, rules of conduct are passed on by a sort
of intellectual osmosis in which scientists are supposed to learn by watching
other scientists, not by discussing their values. 鈥淢any scientists realise there
is something fundamentally wrong,鈥� says Carl Djerassi, a chemist at Stanford
University in California who is world famous as the father of the birth control
pill. 鈥淭hey just don鈥檛 want to open a Pandora鈥檚 box.鈥�
Advertisement
Djerassi and like-minded reformers say it is time to do just that. Their goal
is to break the taboo surrounding ethical issues and to educate and empower
scientists to make changes. At Stanford, Djerassi is experimenting with an
ingenious and unconventional way of tackling the problem. He hopes that his
experiment will not only raise awareness of ethical issues but improve the way
scientists treat each other. It might even save a few lives.
Interest in scientific ethics underwent a major growth spurt about a decade
ago. The interest was triggered by two widely reported fraud investigations.
First came allegations that Robert Gallo, a virologist then at the US National
Institutes of Health in Bethesda, Maryland, had lied about the role his lab
played in isolating the AIDS virus. These were followed by allegations that
Nobel laureate David Baltimore, then at the Massachusetts Institute of
Technology (MIT) in Cambridge, had helped cover up the falsification of data by
one of his colleagues in an immunology paper.
While both investigations fizzled out without proving any wrongdoing, the
episodes spurred many scientific organisations into action. Professional
scientific societies wrote or revised their codes of ethics, the US National
Institutes of Health introduced a requirement for the ethical training of all
the graduate students it funded. Ethics courses sprung up across the world.
False data
But some researchers say that these reactions underestimated the extent of
the problem. In 1993, Judith Swazey of the Acadia Institute in Bar Harbor,
Maine, and her colleagues published survey results from more than 2600 senior
researchers and graduate students in the US. It found that 8 per cent knew of
instances of other researchers plagiarising or falsifying data, while more than
18 per cent said such behaviour occurred among graduate students. And both the
scientists and students agreed that ethics classes and guidelines in place at
the time were having little influence on their behaviour (American
快猫短视频, vol 81, p 542). 鈥淥ften this was because they didn鈥檛 even know
there were any,鈥� says Swazey.
Ironically, beneath this ignorance runs a deep concern for moral issues, says
Hugh Gusterson, an anthropologist at MIT. In his 1996 book Nuclear Rites: A
Weapons Laboratory at the End of the Cold War, he described his study of
weapons scientists at the Lawrence Livermore National Laboratory near San
Francisco. Among themselves, they talked about science and little else, least of
all ethics. But to the outsider Gusterson, they were willing to talk about their
ethical concerns in such detail that it was clear they had spent a great deal of
time thinking through the issues.
The study also uncovered why scientists usually remained reticent. One
physicist who tried to start ethical discussions was ostracised. 鈥淗e found that
the seat next to him at a conference table was always the last one to fill,鈥�
says Gusterson.
Part of the problem, says Gusterson, is that many scientists see striking up
a conversation about ethics either as an admission of guilt, or an attack. His
colleague at the MIT provost鈥檚 office, Stephanie Bird, who is also an editor of
the journal Science and Engineering Ethics, discovered the same problem
when she began inviting scientists to ethics workshops on campus. 鈥淲e were just
trying to get a wide range of perspectives,鈥� she remembers. 鈥淏ut people called
back and asked why they were being targeted.鈥�
Paranoia may be widespread, but Robert Park, a physicist at the University of
Maryland at College Park, says that few scientists ever set out to commit
crimes. 鈥淭hey simply reach a fork in the road and they take the wrong turn,鈥� he
says. He gives the example of a researcher who once copied most of one of Park鈥檚
own papers and published it in a foreign journal. Park has reconstructed the
slippery moral slope that he believes this researcher slid down. At first, he
had simply tried to save time by using some of Park鈥檚 figures in a presentation
at a meeting, but didn鈥檛 properly attribute them. When he was then asked to
contribute a paper for the meeting report, he was reluctant to admit his error
and ended up submitting much of Park鈥檚 original paper to back up the figures.
When Park accidentally stumbled upon the paper and pointed out the
transgression, the researcher apologised and Park let the matter rest.
Even the best of intentions can sometimes persuade scientists to misbehave. A
study led by Kenneth Schulz, an epidemiologist with the Centers for Disease
Control and Prevention in Atlanta, Georgia, found that more than half of 400
medical researchers interviewed knew of cases where their colleagues had cheated
in clinical trials. Schulz has even found statistical evidence of these abuses
which indicate that researchers decided which volunteers were to receive
experimental and which control treatments, rather than assigning them randomly
as strict scientific protocol demands. Their major motivation seemed to be that
they believed they already knew the best treatment for the patients. But this
innocent motive caused a major skewing of results. Preferential assignments made
experimental treatments appear 30 per cent more effective than they actually
were鈥攖hat鈥檚 enough to make a marginal drug seem miraculous or a
detrimental drug seem beneficial (Journal of the American Medical
Association, vol 274, p 1456).
Professional relationships are also a thorny issue, particularly those
between lab leaders and the students who work for them. The problem was thrown
into stark relief by the case of Jason Altom, a graduate student in chemistry at
Harvard University. Last summer, Altom left a note on his bed reading 鈥淒o Not
Resuscitate, Danger: Potassium Cyanide 鈥� and drank a fatal dose of poison.
It wasn鈥檛 the first suicide at Harvard Graduate School (there have been eight
since 1980), but what made it stand out was another note that Altom left,
implying that the advising system at Harvard had let him down and leaving him
unable to cope. The suicide was the second within two years in that lab and the
third student since 1980 although there is nothing to suggest any particular
adviser was to blame in these cases. 鈥淧rofessors here have too much power over
the lives of their grad students,鈥� Altom wrote in one note. He suggested that an
advisory committee with several members would 鈥減rovide protection for graduate
students from abusive research advisers鈥�.
Whether Altom himself actually suffered abuse has been debated. Some say
Harvard has been unjustly tarred in coverage of the tragedy, but it is clear
that Altom believed his relationship with his adviser was in trouble and he had
nowhere to turn. After Altom鈥檚 death, the chemistry department enacted some
reforms. Last year, in an effort to dilute the power of any one professor,
graduate students established committees comprising three faculty members with
whom students can discuss progress towards their degree. Students have also been
given confidential access to psychological counselling and the department now
sponsors dinners to encourage informal interaction between researchers and the
people who work in their labs.
But many people, including Djerassi, feel that these measures are not enough.
In a commentary in the journal Nature (vol 397, p 291), he questioned
whether a student like Altom who couldn鈥檛 face one member of the faculty would
be any more likely to be able to face three. He suggested instead a system of
anonymous evaluation of lab leaders by all members of their labs. These would be
passed to an ombudsman who could look for early warning signs that trouble is
brewing and provide some sort of counselling to the lab leader. 鈥淚f half the
research group says there鈥檚 trouble, it鈥檚 time to take that professor aside,鈥� he
says. Nature, which rarely prints anonymous letters, published two from
students who agreed strongly with Djerassi that an anonymous evaluation
programme would be invaluable.
But Djerassi is the first to admit that implementing such evaluations would
be a tricky business. Safeguards would need to be put in place to mask the
identities of the lab members and protect the rights of the lab chiefs. And he
knows from experience that most scientists aren鈥檛 interested in the idea of
evaluation.
This is where Djerassi鈥檚 experiment at Stanford could help. Consider the case
of the biochemist Alan Simpson. Among the beakers and burners of Simpson鈥檚 lab,
big trouble is brewing. One member of his team is leaking information to another
research group. Meanwhile, Simpson can鈥檛 reproduce a crucial experiment
performed by his star student, Sheryl, so he strips her name from their latest
paper. To top it all, he is being forced to cover up his father鈥檚 extramarital
affair with a member of his team. Luckily for Simpson, neither he nor his
colleagues actually exists.
Simpson is a fictional character in a short story, one of several written by
a group of students and researchers in the experimental class given by Djerassi
at Stanford. In his office, over the remains of chocolate and pizza, they
discuss the issues the Simpson story inspires: secrecy, irreproducible results,
sex in the lab. 鈥淪ome of the discussions have been absolute fireworks,鈥� says
Djerassi.
He is well qualified to chair the debate. He oversaw the synthesis of
19-nor-17(a)-ethynyltestosterone, the active ingredient of the world鈥檚 first
oral contraceptive. But after four decades in science, he has turned his skills
to writing fictional accounts of the moral and ethical problems that science
raises. So far, he has cranked out five novels and a play about the complexities
of laboratory life and he believes that this genre鈥攊t is called
science-in-fiction鈥攊s the perfect forum in which scientists can raise
their concerns about ethical issues. Science-in-fiction is the focus of
Djerassi鈥檚 class.
He asks his students to write fictional accounts of situations like Simpson鈥檚
to raise topical ethical issues. One of the important elements of the class, he
believes, is the anonymity that the students enjoy when they submit their
stories. This protection is crucial to alleviate concerns that the issues they
raise will implicate themselves, embarrass their laboratory or invite
retribution. One researcher who has taken Djerassi鈥檚 course says that the class
provided the only forum to discuss openly a transgression that had been
committed against them. 鈥淚 wanted to use the class discussions to learn how such
situations could have been handled differently, avoided or prevented.鈥� The other
element that Djerassi believes makes the course work is the power of fiction to
flesh out detail and make the presentations more compelling.
True to life
At MIT, Stephanie Bird agrees that the way the information is presented can
make a huge difference. She says that ethical dilemmas are typically presented
almost like mathematical problems鈥斺�漃rof S asks D to put J on a paper; D
doesn鈥檛 want to鈥�. The problem with this, says Bird, is that once the scenario
has been stripped of life-like detail, the students find it difficult to relate
to or learn from.
So in Bird鈥檚 workshops she has participants perform short 鈥渕orality plays鈥�.
The narrative above would be dramatised when the fictitious Professor Barbara
Steel tells her student Sandra Dunn: 鈥淚 can鈥檛 force you to put John鈥檚 name on
the paper, but I think you should consider it again. I like to think we all work
together in this lab.鈥� As the students act out the parts, they are forced to
think how they would react in these situations.
Another goal of Bird鈥檚 class is to remind scientists that behaviour that
starts out as a bad habit can end up as serious misconduct. For example, Bird
points out that it is common practice for scientists to discard bad data points
from experiments they have good reason to believe were technically flawed. But
start ignoring data for other reasons鈥攂ecause they refute a certain
theory, for example鈥攁nd this behaviour edges toward fraud. Another example
of normal behaviour that can be corrupted is advisers who push their students to
work hard. Problems can set in if the advisor begins to ignore valid protests
since this can turn into abuse.
Swazey thinks scientists are beginning to get that message. The trouble is,
they rarely bother to show up at workshops. 鈥淪tudents are extremely receptive,鈥�
she says. 鈥淢any faculty members need to be bludgeoned into attending.鈥� She
believes this is a major problem with the current system, because the lack of
involvement from lab leaders sends the message to students that ethics is not
really important.
Moreover, she says it is dangerous to continually encourage students to
adhere to certain standards, while doing nothing to make it easy for them to
deal with behaviour in others that they know is wrong. 鈥淭he reality is that if
you do something to anger the head of the lab and get kicked out, then you are
screwed,鈥� she says.
Michael Zigmond, a neuroscientist at the University of Pittsburgh in
Pennsylvania, has found a carrot that appeals to senior researchers: money. He
holds the purse strings of a large training grant that pays the salaries of
graduate students and young researchers in the neuroscience programme. He
insists that any lab leaders who get money from the grant to pay their staff
attend ethics workshops. 鈥淚t鈥檚 a very strong incentive,鈥� he says.
But are the reformers鈥� efforts paying off? Djerassi says that since he
described the science-in-fiction programme鈥攁gain in Nature (vol
393, p 511)鈥攈e has received dozens of e-mails, mostly from students eager
to seed grassroots science-in-fiction groups on their own campuses.
Other ethics programme leaders often give lectures at other universities and
say that institutions are warming to the idea of expanding ethics instruction.
In the end, they admit, there is no simple way to gauge whether there has been
any long-lasting benefit.
Nonetheless, Djerassi believes all researchers must learn to consider the
tough ethical challenges that science raises. 鈥淭hen they can make up their minds
if they are satisfied with the present situation,鈥� he says, 鈥渙r whether they are
willing to try and change it.鈥�
