Hausergate: Scientific Misconduct and What We Know We Don't Know
Confusion still lingers after the recent news that Harvard University has found that noted cognitive scientist Marc Hauser engaged in scientific misconduct. Researchers don’t know whether to consider all of Hauser's work suspect, or just some of it.
Last Friday, a Harvard dean issued a statement acknowledging for the first time that an investigative committee found "eight specific instances of scientific misconduct," and that Hauser was "solely responsible" for them. Later, Hauser put out his own statement that stopped short of admitting any wrongdoing but said, "I am deeply sorry for the problems this case has caused." Neither Hauser nor Harvard explained exactly what he had done to be sorry about.
Questions about the specific problems in his work could remain unanswered for months—or possibly even years, barring leaks from those on the inside who know what happened, like the accusation in this report, in which a former research assistant alleged that Hauser reported bogus data in behavioral experiments with monkeys. What follows is a rundown of the “known unknowns,” as former U.S. Secretary of Defense Donald Rumsfeld would call them—the things we know we don't know.
What exactly did Hauser do?
We don't know. Harvard uses the same definition of research misconduct used by the Office of Research Integrity (ORI) of the U.S. Department of Health and Human Services and other federal agencies. It includes fabrication (making up data or results), falsification (manipulating experiments or experimental findings such that the research is misrepresented in the scientific record), and plagiarism.
It explicitly excludes "honest error or difference of opinion."
Why won't Harvard release its findings?
There are some good reasons—and maybe some not so good ones, says Nicholas Steneck, who directs the Research Ethics and Integrity Program of the University of Michigan Institute for Clinical and Health Research in Ann Arbor. Hauser received funding from both the National Institutes of Health (NIH) and the National Science Foundation (NSF). Both have rules on how to investigate research misconduct, and both request that universities keep their own internal findings confidential until any federal investigations are complete. That's born partly from a philosophy of innocent until proven guilty (institutions' findings of misconduct aren't always substantiated by federal investigators). It's also intended to protect the individuals who came forward with allegations, Steneck says. In addition, going public too soon can make it harder for federal investigators to collect more evidence, Steneck says. For example, an investigator accused of misconduct might decide to destroy evidence or intimidate whistleblowers.
At the same time, institutions do have some leeway. Steneck notes that the names of accusers can be redacted and details that might compromise the investigation can be withheld, enabling some information to be released. "Although our regulations request continued confidentiality, it's not absolute," says John Dahlberg, director of the division of investigative oversight at ORI. When Luk Van Parijs, a young RNAi researcher at the Massachusetts Institute of Technology admitted to falsifying data in grant applications in 2005, the university released a public statement before ORI's investigation was complete. "We did not seriously object," Dahlberg says.
In Hauser's case, Steneck thinks Harvard could be more forthcoming than it has been so far. "You're talking about a very important area of science and raising doubts about what can be trusted and what can't," he says. "For the sake of the integrity of the science, Harvard does have some obligation to release more information so that the community can make judgments about it."
Why won't the people who actually know what he did come forward with what they know?
We can only speculate, but there are likely several reasons:
- They may be worried that going public with details about the case could undermine the federal investigation.
- They may be worried about incurring legal problems of their own. In one notorious misconduct case, Kimon Angelides, a molecular physiologist at Baylor College of Medicine in Houston, was found guilty of fabricating data by ORI in 1997 and sued the university and 14 employees, including two former lab members, for slander and denial of due process (he lost).
- The people closest to the investigation are likely to be Hauser's students and postdocs, young scientists with a long future to worry about who don't want to make a powerful enemy.
- Finally, despite the allegations, several of Hauser's former collaborators still admire his intellect and personality. "He needs to stop doing science, but I still like the guy," says one researcher.
What happens now?
Presumably federal investigators are already working on the case. Both ORI and at NSF's Office of the Inspector General (OIG) have a policy of not commenting on ongoing cases—even to confirm or deny their existence. But universities are required to notify federal agencies at the beginning of a misconduct investigation (the investigation of Hauser reportedly began in 2007) and keep them posted throughout.
Once the university's investigation is complete, investigators at ORI begin their own review to determine whether the university's findings are soundly based on the supporting evidence, says Dahlberg. Investigators may ask the university for more information—transcripts or videos of testimony, for example—and may call in experts in the accused investigator's field of research to evaluate lab notes and raw data. Depending on the complexity of the case, the investigation can take anywhere from a few weeks to several years, but Dahlberg says most take about 6 to 12 months.
ORI also pulls up the investigator's grant applications—the ones that got rejected as well as the ones that were funded. "That's one of our primary areas of interest," Dahlberg says. "Legally, scientists are far more vulnerable to serious consequences if they cheat in grant applications than in papers." (In response to requests for an interview about investigative procedures at OIG, a spokesperson sent this link to a brief summary).
What consequences could Hauser face?
The sanctions imposed on a researcher found guilty of misconduct in a federal investigation range from fairly minor, such as having a monitor assigned to keep tabs on their work, to being barred from receiving federal funding. The length of the restrictions varies to fit the situation, says Dahlberg. He notes that in cases where an investigator receives funding from multiple federal agencies, a misconduct finding and "debarment" by any one agency would prohibit the person from receiving funding from any other. In other words, if ORI finds Hauser is guilty of misconduct and bars him from receiving NIH funding, he would be ineligible for NSF funding as well (and vice versa).
The dean's statement says the university is also cooperating with the U.S. Attorney's Office in Massachusetts. A spokesperson there would neither confirm nor deny that the office is investigating Hauser. It's somewhat unusual for the U.S. Attorney's office to get involved in scientific misconduct cases, and historically they've done so in cases in which the misconduct was particularly egregious and significant amounts of money were involved. One recent example is the case of Eric Poehlman, a researcher formerly at the University of Vermont in Burlington found guilty of falsifying data in more than a dozen federal grant applications that landed him nearly $3 million. In 2006, Poehlman was sentenced to a year in prison for defrauding the government.
So far, there's no evidence that Hauser—if he in fact engaged in any misconduct at all—did anything bad enough to merit jail time. But of course there could be unknown unknowns—things we don't know we don't know. And as Rumsfeld said, those are the most worrisome unknowns of all.