Some economists say the president of Harvard talks just like one of them.

The New York Times, "Economic Scene" , February 24, 2005

Before he went into government at the World Bank and Treasury Department, Lawrence H. Summers was a star economist. The habits of mind that made him a successful researcher -- including the style and rhetoric that economists use when they talk to each other -- help explain why he is now embroiled in controversy as president of Harvard.

On Tuesday, the Harvard arts and sciences faculty met for the second time in two weeks to debate his leadership of the university. This week's meeting was less contentious and hostile to Dr. Summers than last week's, according to several of those attending. But some critics have threatened to call for a no-confidence vote at the next regular faculty meeting on March 15. Such a resolution would be nonbinding, since the Harvard Corporation, the board with the sole power to hire and fire the president, has said it supports his continued presidency.

Although many faculty complaints arise from Dr. Summers's managerial style, the immediate source of the dispute is a speech he gave Jan. 14 at a National Bureau of Economic Research conference on women in science and engineering. In that talk, Dr. Summers laid out three hypotheses for why women might be underrepresented in academic science and engineering, particularly at the highest levels.

First and by far most important, he argued, was what he called the "high-powered job hypothesis" -- that women are, for whatever reason, less likely to enter jobs that demand enormous time commitments. The second was that, while average ability may be the same, women may be represented in smaller numbers at the very highest (and very lowest) levels of scientific aptitude. In statistical terms, more men than women may be at the top and bottom levels of the distribution of scientific aptitude compared with the middle; in other words, the bell curve mapping men's scientific aptitude may have "fat tails." The third, and least significant, he suggested, was that women are discriminated against or socialized as children not to go into science.

He prefaced his discussion by saying he would "adopt an entirely positive, rather than normative approach," looking at how the world is, rather than how it ought to be. He ended with a call for "marshaling of evidence to contradict what I have said."

Despite his caveats, some women's advocates were outraged that Dr. Summers entertained, let alone endorsed the hypotheses that family arrangements or innate abilities might have anything to do with the success of women in science. Nancy H. Hopkins, a biology professor at the Massachusetts Institute of Technology who has campaigned for more hiring and promotion of female professors, walked out on his remarks and later told reporters, "I felt I was going to be sick."

In lambasting his nonjudgmental, empirical approach to the question, opponents are not merely challenging Dr. Summers's brash manner or his evaluation of the data. They are attacking the very method economists use to address social policy questions. And, not surprisingly, some of his most outspoken supporters are fellow economists.

"He speaks the language that we speak," said Claudia Goldin, one of two Harvard economists who circulated a letter supporting Dr. Summers's "continued role as president of Harvard." "Economists," she said in an interview, "understand what it means to 'offer some hypotheses."' (She added that she believed Dr. Summers misread the evidence on aptitudes.)

Dispassionate hypothesis testing is particularly important for practical questions, because different explanations may imply different solutions. Take Dr. Summers's argument that the primary barrier to women in science, as in other high-powered jobs, is that employers demand single-minded dedication to work. "They expect a large number of hours in the office, they expect a flexibility of schedules to respond to contingency, they expect a continuity of effort through the life cycle, and they expect -- and this is harder to measure -- but they expect that the mind is always working on the problems that are in the job, even when the job is not taking place."

Married women, he argued, especially those with children, are far less likely than married men to put up with such demands.

For universities, this suggests the tenure clock for junior professors may hurt women who have young children in the same years they are expected to "publish or perish."

Last May, Dr. Summers invited Professor Goldin and a colleague, Lawrence F. Katz, to discuss what changes Harvard might make to tenure to retain promising young female scholars. Professor Goldin is a longtime advocate of women in the economics profession, and much of her research concerns women in the labor market.

The two economists' conversation with Dr. Summers assumed that balancing family and work commitments is a significant problem for women in academia. If, however, as many of Dr. Summers's opponents insist, the only legitimate explanations are childhood socialization and discrimination, changing the tenure process would not help.

Not all Harvard economists support Dr. Summers's presidency, of course. At Tuesday's meeting, Caroline M. Hoxby, a leading empirical economist specializing in education research, criticized his abrasive management style. Others, however, are happy to tolerate some personal abrasiveness to get Dr. Summers's scholarly enthusiasm.

The week after the furor over Dr. Summers's January remarks broke out, Edward L. Glaeser sat next to him on an airplane. Instead of talking university politics, Dr. Summers immediately began discussing Professor Glaeser's research on why politicians make appeals to the extremes rather than to the "median voter," as much political science theory predicts. He asked questions and probed whether there might be alternative explanations, like the role of primaries.

"To have a president who loves research that much is really, really valuable," Professor Glaeser said in an interview. "At least for us, you can forgive a lot."