So how is it that, in spite of all this evidence of polyandry accumulating steadily in the literature, anthropologists for so long passed along the "it's virtually non-existent" story? Starkweather and Hames suggest anthropology has been accidentally playing a scholarly version of the Telephone Game.
In 1957, George Murdock defined polyandry in a seminal text as "unions of one woman with two or more husbands where these [types of union] are culturally favored and involve residential as well as sexual cohabitation." Using such a strict definition, Murdock could accurately say polyandry was extremely rare; almost no cultures have polyandry as the dominant and most preferred form of family life.
Then subsequent scholars mis-repeated Murdock's remark; polyandry went from being understood as "rarely culturally favored" to "rarely permitted." Thus mating diversity that was known to exist became relatively invisible in the big story told by anthropology about human mating. (If you write off every exception to a supposed rule, you will never think to challenge the rule.)
In an email interview with me, Starkweather remarked, "I don't think that anyone, including Murdock, was operating from an explicitly sexist standpoint. However, I do think that the definitions of polyandry, and thus perceptions about its rarity, may have been due at least in part to the fact that an overwhelming percentage of anthropologists collecting data and shaping theory at the time were men." During Murdock's time, "there seemed to be a fairly pervasive belief that polyandry didn't make any sense from a male's perspective."
That explanation -- that Western male anthropologists had a hard time "believing" in polyandry—makes sense. Humans appear prone, on average, to sexual jealousy, and so it would not be unreasonable for many of us—men and women alike—to project an assumption that sexual jealousy would make poly-unions untenable. Indeed, anthropologists have found that in both polyandry (one woman, multiple husbands) and polygyny (one husband, multiple wives), sexual jealousy often functions as a stressor in families around the world.
Yet certain environmental circumstances do seem to increase the odds of a culture accepting some form of polyandry. In particular, Starkweather and Hames find that polyandry is often found in societies with highly skewed "operational sex ratios." Translation: When fertile women are scarce, men are more likely to be found openly sharing women. Indeed, fully three-quarters of the 53 societies identified by Starkweather and Hames involve skewed sex ratios, with more adult males than females.
This led me to wonder, in our exchange, whether in places where sex ratios are becoming highly skewed—in places like India and China—is polyandry likely to emerge? Starkweather and Hames guess not. First, most of the cultures in which polyandry is found look very different from modern India and China; polyandry shows up mostly in relatively egalitarian societies (i.e., societies with very simple social structures, without massive governmental bureaucracies and elaborate class structures). So, for example, polyandry is regularly found among the South American Yanomamö, the people Hames studied in the field in the 1970s and 1980s.