Who Will Speak for Hauser?
The recent controversy over the influential cognitive psychological work of Harvard Professor Marc D. Hauser has performed the valuable public service of raising issues of scientific method and accountability. Unfortunately, it has done so by employing irresponsible and inaccurate reporting and argumentation in an inexplicable attempt to undermine legitimate research by one of our generation’s leading scholars.
Newspapers generally pride themselves on reporting at least two competing perspectives on hot-point issues, but in the Hauser case the only voices presented are those of scholars known to be virulently opposed to his research program. One such scholar, Michael Tomasello, has assumed the role of muckraking journalist—claiming his own anonymous sources within the Hauser lab—reporting on unsubstantiated rumors of staff “insurrections” and university “raids” to seize hard drives and videos. Then there are the insinuations and warnings of Hauser’s former supervisor, Robert M. Seyfarth: “Marc’s way of doing things and ours were not really the same”; co-authors and students “run the risk of being tarred with the same brush” of misconduct by Hauser’s compliance with the confidentiality protocol of his investigation. Most recently, “Cognition” editor Gerry Altmann has announced his “conjecture” (his word) that Hauser fabricated data; this is exceedingly improper, for Altmann is not privy to the totality of the evidence, and thus any opinion he expresses is by definition speculative and thereby a contravention of the due process (subsuming the presumption of innocence) that Hauser deserves. Aren’t journal editors supposed to be champions of disinterested formal processes?
Gossip, innuendo, and speculation are unbecoming of professional scientists; furnishing a bully pulpit for such defamation is unbecoming of professional journalists. Journalists and readers ought not to heed the hearsay and hunches that have thus far characterized the coverage of the Hauser case, but rather consider all sides in the case when possible.
Even the most meticulous of scientists makes mistakes, and Hauser has done so: “I acknowledge that I made some significant mistakes and I am deeply disappointed that this had led to a retraction (i.e., of a 2002 “Cognition” paper) and two corrections (i.e., to a 2007 “Science” paper and a 2007/2010 “Proceedings of the Royal Society B” paper).” However, it is not justified to see “a shadow [cast] over the several different fields in which Dr. Hauser and his students published papers,” or a “ripple effect” so violent that “all of Dr. Hauser’s results are suspect.”
In a 2009 paper in “Biology Letters,” Watumull—a member of Hauser's lab—collaborated with Hauser on a study conducted with the experimental design of the 2002 Cognition paper; this design is also implicated in some allegedly dubious 2007 research. Watumull can testify to Hauser's unimpeachable scientific integrity within the design, which obviously—as with any experiment—cannot be error-proof.
The experiment required the researcher to determine whether a monkey discriminated particular sound patterns; the evidence for discrimination was whether the monkey oriented toward the sound source for particular patterns. Researchers in some labs are irresponsibly subjective in their analyses: any turn of the head is counted as a positive datum. But as the most conservative and reasoned of scientists, Hauser insisted upon applying a Cartesian grid to the animal's head (in the computer analysis), defining orientation with respect to the transcendence of planes and positions in quadrants. This methodological meticulousness in our personal experience typifies all of Hauser's work as surely as it did in the 2002 research. Any mistakes made, therefore, ought not to “besmirch” (to borrow a detractor’s word) Hauser and his work, but humble us to concede that even the best science—one of our noblest pursuits—is only as infallible as the best scientists.
Watumull did not conduct any experiments for the 2007 and 2007/2010 papers but was present with many lab members to review the data of videotaped animal performances on which they were built. The data were collected with demonstrable rigor. To test action-perception in nonhuman primates, the researcher must methodically perform his actions to a focused, undistracted subject. Hauser choreographed researchers’ actions meticulously—from the angles of body parts to the speeds of movements—and selected subjects judiciously—aborting any trials wherein the attention of the animal was compromised. Such well-designed experiments are the stuff of good science. The findings reported in the papers were manifestly consistent with these data. It is a further testament to Hauser's integrity that he insists upon as many lab members as possible peer-reviewing data.
In our experience, Marc Hauser is the consummate scientist—the most disinterested, the most rational, the most ethical. We are proud to be his colleagues. However, we are less than proud of those in the cognitive sciences reacting publicly to Hauser's case with irresponsible impatience (disrespect for due process), unjustified slurs, and half-baked conjectures. All are interested in the truth, but as scientists we ought to consider the case reasonably and measuredly, with objectivity and fairness.
Bert Vaux is a former Professor of Linguistics at Harvard University, Jeffrey Watumull is a graduate student in Linguistics at the University of Cambridge and a member of Hauser’s lab.
CORRECTION: September 23, 2010
An earlier version of the Sept. 21 op-ed "Who Will Speak for Hauser?" stated that Bert Vaux is a former Professor of Linguistics at Harvard University. In fact, Bert Vaux is a former Associate Professor of Linguistics at Harvard University.