Columns

David Sheffield ’11: Scientific misconduct

By
Opinions Columnist
Wednesday, September 1, 2010

Last week, a committee at Harvard ended a three-year inquiry into one of the school’s professors, Marc Hauser. It concluded that he engaged in scientific misconduct. Hauser is a psychologist who studies animal cognition, particularly moral behavior. Scientific misconduct is a serious offense that hampers the advancement of the entire endeavor.

Science is a data driven enterprise. No one scientist can collect all of the necessary data to formulate theories, so researchers must consequentially trust the results of others. This is not a blind trust — results are tested through peer review and replication. The punishments meted out by institutions and the community can be correspondingly devastating. This makes claims of fraud an appealing avenue for the antiscientific crowd when it does not want to appear antiscientific.

The investigation started when a graduate student and research assistant working with Hauser lodged a complaint with a dean. The group was working on an experiment to see if rhesus monkeys could determine specific patterns of syllables. Hauser and another researcher watched videos of the monkeys to determine whether they responded to changes in the pattern of syllables that the scientists played for them. Hauser’s results showed that the monkeys did notice the change and were therefore able to determine the specific patterns. Meanwhile, the other researcher saw no response from the monkeys.

The experiment relied on a subjective analysis of the monkeys’ behaviors, which is why the group used multiple people to analyze the recordings. The two observers inferred contradictory data, so the next logical step was to have other observers judge the results to settle the issue — something that the whistleblowers did want. However, Hauser refused and wanted to publish his results as the correct interpretation.

The graduate student and research assistant should be commended for what they did. They were both heavily reliant on Hauser for their careers and their decision to alert a dean could have backfired.

There are numerous other instances of fraud and misconduct in science: Piltdown Man, human clones and the measles vaccine–autism link were all fabricated. One of the seminal recent incidents is Jan Hendrik Schoen’s brief acclaim. In 2001, Schoen claimed that he created a transistor from organic molecules — a hugely consequential result. Other physicists were soon skeptical of his phenomenal work and started looking over them more carefully. Other scientists could not replicate Schoen’s results. Eventually, his colleagues discovered that Schoen was reusing data multiple times. The anomalies soon built, and, within a year, the first committee found Schoen guilty of scientific misconduct.

Such efforts by brave students and diligent scientists make ideologically driven claims of fraud disappointingly weak by comparison. Last year, a hacker broke into the Climate Research Unit’s computers. The hacker stole thousands of emails and distributed them on the World Wide Web. Climate change deniers and sympathetic news organizations mined the e-mails and misrepresented what was written to fabricate a conspiracy by scientists to deceive the public. Most other media, suffering from a lack of science journalists and desiring balance in place of objectivity, credulously reported the controversy.

“Climategate” was front-page news. The later exoneration of the scientists was buried further back.

Of course, a lack of misconduct has yet to stop Ken Cuccinelli, Virginia’s political hack of an attorney general, from investigating climate scientist Michael Mann, who wrote some of the e-mails. Mann formerly worked at the University of Virginia and held government grants. Cuccinelli is prone to interpreting the law in politically convenient ways and has easily turned Mann into a prop to further his ambitions. Luckily, a judge denied Cuccinelli’s request for a subpoena this week. The witch-hunt is not finished, but the decision is a hopeful sign that the legal system sees through the political pandering.

This does not mean that prosecutors should never investigate scientists. A U.S. attorney is currently examining Hauser’s misconduct because he received federal funding. However, it is a waste of government money to investigate a scientist without a rational justification. Science is effective at policing itself, and frauds are eventually detected. Prosecutors should rely on already established processes to determine frauds and not waste taxpayer money themselves.

All scientists want to make some important discovery. Overturning a well-accepted theory such as climate change or evolution would garner not only the praise of colleagues for correcting a major error but surely fame among the public. Researchers comb over studies for errors — especially ones that contradict their own pet theories. It is preposterous to claim that entire branches of science have entered into pacts intended to deceive. Eventually, someone would present the accurate and devastating contradictory evidence.

David Sheffield ’11 is a math-physics concentrator from New Jersey. He can be contacted at david (underscore) sheffield (at) brown.edu.