Verified by 
A note in The Atlantic earlier this week caught my attention: Apparently, in an on-air discussion of new data showing that women now are the primary source of income for 40 percent of households in the United States, a commentator dragged "science" in to support what I can only characterize as a wildly inaccurate argument, saying that people ...
... who defend this and say it is not a bad thing are very anti-science. When you look at biology — when you look at the natural world — the roles of a male and a female in society and in other animals, the male typically is the dominant role.
Well. I tend to cringe whenever I see a citation of "the natural world," as if society is by definition unnatural. The author of The Atlantic commentary, Derek Thompson, does note that the current situation is not "unnatural," albeit using a different argument.
But he doesn't take on the central claim made here — which is that science (my bailiwick) shows that "the male is typically the dominant role."
To deal with that, we need to get a little precision into the murkiness of common wisdom, which is what the original comment draws on — no actual cited sources, just a kind of "everybody knows that" claim.
Starting with "the dominant role," that's imprecise. What scientists talk about are "dominance hierarchies," which can be defined as
... a social situation in which one organism dominates all below it, the next all below it, and so on down to the organism dominated by all; e.g., the pecking order in apes, seals, barnyard hens, and other species.
Think this is a difference that makes no difference? The shift from "dominant role" (ascribed automatically to a male in a presumed male-female pairing) to "dominance hierarchies" (plural, and with no assumptions about how many individuals are involved, nor whether there is a diversity of sexes) points toward what in fact is a very, very complicated literature about what we mean when we identify dominance hierarchies.
The basic idea is clear: a "pecking order," based on the behavior of barnyard chickens, where one animal can peck all the others, and each of the others in turn can peck at some, but not others, so that you can create a single sequence from dominant chicken who pecks others but cannot be pecked, to the loser at the bottom who cannot peck any others and can be pecked by every other.
In barnyard flocks, we are usually dealing with one rooster and a bunch of hens, so that the "pecking order" has a dominant male at the top — but most of the hierarchy is exhibited between females. The question is — what determines the place you have in the hierarchy?
In 2002, an article published in the Proceedings of the National Academy of Sciences undertook an assessment of two possible explanations for relative rankings like this:
... they are predetermined by differences in the attributes of animals, or they are produced by the dynamics of social interaction.
The "attributes" argument is the one that says males are naturally dominant over females — the attribute in that case being sex. The researchers publishing the 2002 article noted that "age, sex, physical size and strength, physiology, and level of aggressiveness" are the most common attributes cited as bases of dominance hierarchies.
What they actually found, though, was different. Their study supported the formation of dominance hierarchies based on sequences of social interactions, resulting in what they called "winner, loser, and bystander effects."
Basically, in one-on-one confrontations, the animal that backs down tends to continue to back down; the one that wins a confrontation tends to continue to win; and at least part of this continued tendency may be due to other animals witnessing the one-on-one confrontation, and adapting their behavior — assuming the animal they saw win would win in a contest with them, and deferring, but taking on the animal they saw lose.
The authors conclude:
One implication of our results is that current models in sociobiology are either too simple or too concerned with individual differences to account adequately for the evolution of behaviors leading to dominance hierarchies.
Individual attributes do not, it seems, automatically lead to particular dominance hierarchies: Social interactions do. Dominance hierarchies are not simple. They involve numbers of individuals and the outcomes of confrontations (or avoidance of confrontations) between them.
Most of the recent literature on dominance hierarchies examines more complex models that emphasize social interactions, and the outcomes of this research do not support the simple idea that nature dictates male domination of females.
In 2009, a group of scientists published an article in Behavioral Ecology that examined variation in dominance hierarchies. Their work is particularly interesting since they were centrally concerned with understanding dominance hierarchies in non-human primates, where, as they note:
... nonhuman primate females ... are unusual because ranks can depend on kin support or follow an inverse age-graded pattern independent of kin.
The authors summarize previous research as showing that:
... at its simplest, rank may be determined by differences in resource holding potential...expressed via relative difference in size, strength, or fighting abilities.
But other factors can be equally important:
... increasing dominance with age, experience, or tenure...decreasing dominance with age... [and] nepotistic hierarchies with kin ranking closely together.
Their own study shows that:
... patterns of hierarchies are not just the outcome of the strength of competition or relatedness, but contingent on several factors including the variance of within-group relatedness (affecting the indirect benefits for an alliance partner), costs and likelihood of coalitions (affecting the costs for an alliance partner), resource needs of contenders, and value of ranks for the contenders (both affecting the “willingness” to participate in a contest).
While these authors were not primarily interested in assessing the role of sex in dominance hierarchy formation, the recent literature on primates does directly address that topic — and the results are not good for the commentators at Fox News.
Many primate species actually have patterns of female dominance over males. Researchers have been able to use the variability in primate species to confirm that dominance hierarchies strongly reflect social interactions and their outcomes, not pre-existing attributes like sex.
One article directly addresses the characteristic of maleness that might be expected to contribute most to relative dominance: body size. Even here, the results do not support any argument for natural domination of (bigger) males over (smaller) females.
In 2009, the journal PLOS ONE published "Female Dominance over Males in Primates: Self-Organisation and Sexual Dimorphism." The authors examined data from 22 primate species, ranging from those in which females were completely dominant over males, to those in which males were dominant over females. They conclude:
... we expected a negative correlation between female dominance over males and species-specific sexual dimorphism in body mass. However, to our surprise we found none.
I wouldn't necessarily go as far as the National Geographic did, titling a 2003 story "Females are Dominant Sex, Primate Study Suggests". But even a cursory review of primate studies shows that male dominance isn't a given.
And even those chickens and their linear pecking order have a surprise in store for those who think nature has a sexually determined hierarchy.
The rooster may be top of the dominance hierarchy but some roosters start out as hens.
Nature as it really is — not as we might imagine it — is never so simple.
