Not all lives are easily narratable. As the author's subtitle indicates, Cora Du Bois's life breaks restlessly into a number of only faintly connected pieces. Her career as an anthropologist, from the 1920s to the 1970s, itself went through a number of quite distinct phases. In between came surprising periods as a secret agent and a State Department functionary. But Du Bois is probably most interesting today – and will probably be best known in future – as a woman and a lesbian in a period when academia was much less friendly to both than it is now.

Du Bois was roughly contemporary with the more celebrated Ruth Benedict and Margaret Mead, who shared, of course, her gender, and (sporadically in Mead's case) her sexual preference, and also her pedigree: all three were Boasians, Du Bois a “second-generation” Boasian, trained at Berkeley by Boas's students Alfred Kroeber and Robert Lowie. Like Benedict and Mead, too, she developed psychological interests that were at first rather unsettling to the older generation, and made a signal contribution to the “culture and personality” movement by means of fieldwork in Alor in present-day Indonesia in the late 1930s, relating the “primary institutions” of kinship, family structure, social structure and environment to “basic personality structure” (or, as she preferred, “modal personality structure”) and through it to affective behaviours and a wide range of cultural consequences. As this vocabulary indicates, Du Bois did the fieldwork under the intellectual aegis of (and partially funded by) the neo-Freudian psychoanalyst Abram Kardiner, alongside his anthropological collaborator Ralph Linton, Benedict's nemesis at Columbia. It's a bit of a conundrum why she chose to work with them rather than with the superficially more congenial Mead and Benedict. Her strongly analytical bent might have been offended by Benedict's aestheticism, and, while the three women were very different in personality, they all had strong egos that were hard to combine. Still, it is difficult to see how she could work at all with Kardiner and Linton, who treated her very badly, and whose rather clunky, mechanical understanding of culture and personality was at odds with her more supple, individualistic approach. Already by the time she left for Alor she had learned to mistrust Kardiner and accepted help from Benedict instead.

The Alor research marked virtually the end of her contribution to the “culture and personality” movement, however. The war broke out before she was able to finish her book and she found herself, like many anthropologists, swept up by authorities keen to exploit her regional expertise on Southeast Asia for war service. Again, like many academics, she joined the Research & Analysis Branch of the Office of Strategic Services, the “Chairborne Division,” but unlike most she escaped the tedious business of piling up reports in Washington. She got a posting instead to run the R & A Branch in Mountbatten's South East Asia Command, based in Kandy in what is now Sri Lanka – a practically unique position for a woman in OSS; most, like her close friend in Kandy and future celebrity chef Julia Child, held clerical positions. This frontline experience of secret intelligence, combined with her regional expertise, made it possible for her to stay in government service after the war, where she moved along with other OSS staff to the State Department. There she fought a lonely battle against the backdrop of the early Cold War to get government to recognize the strength of indigenous anticolonial and nationalist feeling, as yet extricable from Communism. She was under suspicion for both her politics and her sexuality. Ultimately she jumped before she was pushed.

After an abortive attempt by Berkeley to hire her – she refused to sign the loyalty oath – Du Bois finally settled down at Harvard, appointed in 1954 to the Zemurray-Stone Chair, which made her the first woman to be appointed to a tenured professorship there. Her research thereafter took a zigzag course. For a time she worked on “friendship.” Her interests subsequently drifted away from psychological anthropology towards what we would now recognize as an area-studies orientation. Her last major project – in which the author participated – was a decade-long multidisciplinary enquiry into social change in the Indian city of Bhubaneswar, which launched many careers, but on which she herself published only a single paper. She became a staunch critic of the growing influence of modernization theory on area studies, though even here she kept her profile low, and her finest statement on the subject, the Cooper Lecture at Swarthmore, “An Anthropologist Looks at Modernization,” wasn't published either.

Du Bois was a stern, private person, hard on herself as well as others, a highly conscientious teacher, and a loyal if not often very self-revealing friend. Seymour has mined her rich archive well to give as full a picture as can be given of someone so private, and of a life that was so mercurial. Du Bois did not make it easy for her. She was not a campaigner for gay or women's rights, constantly downplayed her own achievements (and difficulties), and went through her files every year and destroyed what she didn't want a future biographer to find, including all her private letters to her longtime partner Jeanne Taylor. Partly for these reasons, Seymour may have (perhaps unwittingly) toned down the impact on her of the prevailing misogyny and homophobia even, or especially, in “liberal” academic circles. Kardiner, for one, told an interviewer that he found her “repellent” – and that was even before he learned “[s]he wasn't female,” by which he meant she didn't want to sleep with him.