Microbiologist Eckard Wimmer never conducted the next big experiment he had planned: Not after the public drubbing he took in 2002 when he reported building a live polio virus from scratch.
A firestorm erupted over the ethics of making the pathogen, and Wimmer - although convinced that the experiment and its publication in the journal Science were ethical - decided he didn't want to go through anything like it again.
Today, another paper in the same prestigious journal raises questions about how common it is for researchers to do what Wimmer did: censor their own work to avoid controversy.
The answer has serious implications for scientific research and whether it is widely biased by the fear of backlash.
Citizens and interest groups can influence the debate about a proposed ban on human cloning by contacting their legislators. But a worried researcher's decision to avoid an experiment or leave controversial data unpublished can affect the outcome of science without a word of public debate, the authors say.
"The problem is, we don't know what's being censored and why," said Jon F. Merz, an assistant medical ethics professor at the University of Pennsylvania and a co-author of the study.
Co-author Joanna Kempner, a sociologist and University of Michigan research fellow, added that "controversy may be shaping scientific agendas."
The paper's findings are based on 10 pilot interviews and 41 in-depth interviews with researchers from prestigious U.S. academic departments in seven fields - neuroscience, sociology, molecular and cellular biology, genetics, industrial psychology, drug and alcohol abuse, and computer science.
Some 42 percent described how their work had been targeted for public censure in one form or another, the paper said.
But the authors were surprised to find that scientists were most affected by so-called "informal constraints," such as the fear of breaching an unspoken rule against inquiry in a particular area.
The paper avoids specifics to protect the identities of those interviewed, and the limited sample means findings can't be generalized. But Merz and Kempner said in interviews that a number of the scientists they surveyed avoided topics considered politically incorrect, such as those - like the controversial book The Bell Curve - that explore race and intelligence.
Others said they have given up experiments on dogs - traditional objects of medical research - to avoid the wrath of animal rights activists. And a psychologist told the authors he quashed his own idea for a study of "subtle cues" that may be misinterpreted as sexual harassment, fearing it would be too controversial.
The paper's release comes amid criticism of Harvard University President Lawrence H. Summers, who riled scientists and women's rights groups in January with comments that some viewed as an endorsement of research into the innate differences between the sexes to explain why fewer women have succeeded in math and science than men. He has since apologized.
Though not based on a particular study, Summers' experience exemplifies the controversy that can result when academics cross an "unspoken" line - as well as the beneficial debate that follows when their ideas are subjected to rigorous scientific scrutiny, Kempner said.
Kempner conducted audiotaped interviews in 2002 and 2003, a period marked by controversies over published or publicly funded research.
During that period, for example, the National Institutes of Health reviewed more than 200 scientific projects reportedly targeted by the conservative Traditional Values Coalition. Meanwhile, editors of research journals were scrambling to develop policies on publication of findings that conceivably could be used by bioterrorists.
The editors were reacting to criticism of Wimmer's 2002 paper, which described how to construct poliovirus from DNA samples that can be ordered through the mail.
Wimmer, a professor at the State University of New York-Stony Brook, said his intent was to raise awareness. "The time had come where somebody would have to show that viruses have to be looked at as chemicals," he said, adding that any virus can be synthesized if its structure is known. Publicizing that issue, he said, forced society to be "aware of that and cope."
Still, he found the experience - which included a lambasting by genomics luminary J. Craig Venter - so unappealing that he scrapped the experiment he had planned to do next.
That one, already funded, would have attempted to turn a similar, benign virus into polio by tweaking its genetic code. If it had worked, Wimmer said, it might have proved how easily the related virus could naturally morph into real polio. That would challenge the wisdom of ending polio vaccinations because the disease has been virtually eradicated here.
Uproar over controversial science is nothing new, and Merz and Kempner acknowledge it can sometimes be positive.