More often than I would have expected, I encounter this observation from seemingly intelligent, well-informed people: Evolutionary biology applied to human behavior simply lacks empirical validation. All hat, no horse; all theory, no data.
The reality is dramatically different. There are several full-scale refereed journals that regularly publish data-heavy manuscripts (Human Behavior and Evolution, Evolutionary Psychology, Human Nature, plus several others), also a proliferation of textbooks, monographs, and so forth. Here is one example of this hard-headed empiricism, a recent research report that I find especially intriguing because it yields such precise correspondence between theory-based expectation and results.
At issue is the straight-forward evolutionary prediction that parental concern for offspring should track confidence (albeit not conscious awareness, mind you) of genetic relatedness, because all things being equal, selection would act strongly against investing in unrelated individuals (with certain notable exceptions, including reciprocity and adoption). Mothers are guaranteed relatedness to their offspring whereas fathers are not; hence, a paternal link in the chain of genetic relatedness is necessarily less certain than is a maternal link.
Moreover, if we consider grandparents, the opportunity arises to test this prediction with remarkable specificity. Although all grandparents are separated from their grandchildren by two generational links, those of maternal grandmothers (mother’s mother, or MoMo) are of highest reliability, those of paternal grandfathers (FaFa) are the lowest (because there are two uncertain, paternal links), whereas those of maternal grandfathers (MoFa) and paternal grandmothers (FaMo) are in between, one certain and one uncertain.
W.T DeKay therefore predicted that grandparental investment would be highest among MoMo’s, lowest in FaFa’s, and in-between among MoFa’s and FaMo’s. Looking at four different measures—subjective feelings of closeness, time spent together, knowledge of each other’s lives, and actual expenditure of resources from grandparents to grandchildren—it turns out that this was exactly the case each time.
It seems unlikely, incidentally, that this finding is simply a result of women being more grandparental than men, because MoFas are slightly more investment prone (although not significantly so) that are FaMos.
It also turns out that a similar pattern of evolutionarily predicted preferential investment occurs via the aunt/uncle connection, since according to the above logic, paternal aunts (father’s sisters) should be less inclined to invest than are maternal aunts (sisters of the mother). And ditto for paternal uncles versus maternal uncles. Bingo, in both cases!
Moreover, the pattern persists regarding cousins: FaBro’s children self-report the least closeness; MoSis’s children, the most; with FaSis’s and MoBro’s children precisely in between
It is very rare in the annals of social science research that such precise predictions have ever been made, never mind obtained. And I’d bet that it’s no coincidence that the intellectual backbone of these findings rests not on social science but on something much more secure: biology.