A number no greater than the sum of its parts: the use and abuse of heritability.

Though widely used in quantitative genetics, in the study of human variation perhaps no statistic is more easily misinterpreted than heritability. While the contribution of genetic heritage to complex biological and behavioral phenotypes cannot be lightly dismissed, nonetheless we remain profoundly ignorant of how that legacy plays out in any environmental context. To be sure, it is not reducible to a single number. Nor does the preference among anthropologists for analyzing biological rather than behavioral phenotypes improve what heritability can reasonably say about the sources of human variation. This paper discusses the meaning of heritability, the methods for its estimation, the fallacies underlying its misuse, and its utility for inquiries in evolutionary anthropology and epidemiology. Progress in anthropological genetics will be realized through greater sophistication in study designs, including the measurement of environmental (physical and sociocultural) variation and the judicious choice of phenotypes for study. Elucidating the ontogenetic processes underlying adaptive plasticity is particularly critical to understanding the evolution of human biological variation. Such advancements will also shed light on the feasibility of genotype-targeted biomedical treatments. Failure to appreciate the limits of such approaches can divert resources from demonstrably effective environmentally based health interventions that benefit entire populations. Simplistic notions of genetic determinism should be challenged for the sake of our theories and the well-being of larger communities. As exemplified by the work of Frank B. Livingstone, anthropological genetics is at its best when incorporating anthropology into the study of human phenotypic variation.