Retaining Models of Human Evolution After Repeated Falsifications—Why?

ABSTRACT Thomas Kuhn's book on the Structure of Scientific Revolutions profoundly influenced our thinking about how science changes over time with his proposal of paradigm shifts, but doubts have been raised about how general Kuhn's proposal is for all scientific disciplines. In biology, the problems are so complex that often parallel paradigms coexist based on different, non‐commeasurable methodologies without one replacing the other for long periods of time. An example is given about two models of human evolution that arose in the mid‐19th century: One favored by Darwin that human populations intergrade with no sharp boundaries due to interbreeding (gene flow), versus Haeckel's view that humans are separated into distinct lineages that define separate branches on an evolutionary tree of human populations. These two models have persisted into the 21st century with no resolution. This debate also touches on another aspect of the philosophy of science—the importance of falsification of a hypothesis. Karl Popper argued in the 1950's that falsification of a hypothesis is strong inference in science, and is a type of inference that distinguishes science from other realms of knowledge. Hypothesis testing is the methodology that favors Darwin's gene flow model, but hypothesis testing and other statistical considerations are ignored or trivialized by advocates of the evolutionary tree model of human evolution. Advocates of the tree model value the parsimonious and simple nature of an evolutionary tree, and the typological subdivision of humanity into distinct types or branches in that tree. Both groups are satisfied with their methodologies and feel that their paradigm is the superior one, allowing the two paradigms to coexist without resolution for over a 150 years. Key Points The problems within biology are so complex that parallel paradigms can co‐exist for long periods of time without invoking a Kuhnian crisis or paradigm shift due to non‐commeasurable criteria of validity. An example is the persistence since the 19th century of two views of human evolution as (1) an evolutionary tree of splits and isolation resulting in distinct branches of humans, versus (2) a model of gene flow that allowed all humans to evolve as a single lineage, with genetic differences between populations due mostly to isolation‐by‐distance. Hypothesis testing has strongly falsified the evolutionary tree model in favor of the gene flow model, but the advocates of the tree model place greater value on a more parsimonious explanation of human evolution and a typological view of human diversity.