“I said that raised a whole set of questions about how job expectations were defined and how family responsibilities were defined,” recalled Summers in an interview Monday. “But I said it didn’t explain the differences [in the representation of females] between the sciences and mathematics and other fields.”
Goldin, who prepared a memo that Summers cited in his speech Friday, said the president “had mountains of research” on the subject, although he spoke extemporaneously.
Summers referred repeatedly to the work of University of Michigan sociologist Yu Xie and his University of California-Davis colleague Kimberlee A. Shauman, whose analysis of achievement test results shows a higher degree of variance in scores among men than among women. While males and females posted similar average scores, males were more likely to fall at the higher and lower ends of the distribution. (Please see related article, page 1.)
Summers suggested that behavioral genetics could partially explain this phenomenon. He stressed that this hypothesis required further research, and he hoped it would turn out to be incorrect. But by that point Hopkins, who last year was inducted into the prestigious National Academy of Sciences, had left the conference room. She said she was concerned that it would be “rude” to get up midway through Summers’ speech, but “it was just too upsetting” for her to stay.
FULL COURT PRESS
Freeman said that he invited Summers to the NBER event “to come and be provocative.”
“We didn’t invite Larry as a Harvard president per se,” Freeman said. “I think if we had invited him as Harvard president, he would have given us the same type of babble that university presidents give. And thank God we have a president who doesn’t say that.”
Freeman said that Hopkins’ decision to take her concerns to the press was “very bizarre in my view.” Summers said he had not expected that the comments would be published.
“If I disagree with you, I should tell you why I disagree with you and what the evidence for my point is,” Freeman said.
Hopkins said she mentioned the Summers speech in an e-mail exchange relating to another matter with Boston Globe reporter Marcella Bombardieri on Friday—but that she did not intend for her sentiments to spark the media circus that is now underway. By late last night, Summers’ comments had been reported by hundreds of newspapers from Britain to Beijing.
Hopkins dismissed the idea that Summers’ remarks were meant to be kept private among conference attendees: “The notion that Larry Summers’ position should be kept a secret on issues like this—that’s just wrong.”
Summers said that universities must address discrimination head-on, but that academics must also engage in “careful, honest and rigorous research” to understand the factors fueling the underrepresentation of females. “My speculations were intended to contribute to that process,” he said.
—Staff writer Daniel J. Hemel can be reached at hemel@fas.harvard.edu.