Terrence Deacon: UC Berkeley Department of Anthropology

Although biologists have long recognized examples of regressive processes in evolution as well as a role for regressive processes in the development of brains, research interest tends to focus on presumably constructive and progressive processes under the influence of natural selection. Particularly in the case of human brains and their evolution, it is generally assumed that the neurological differences underlying the complexity of human language abilities must have arisen due to progressive improvements of function via selection favoring these traits. In this talk I will explore an alternative possibility: that devolutionary loss-of-function due to reduced selection,
including degradation of developmental-genetic specificity, may contribute to the evolution of novel complex neural functions, such as language. The general logic of this argument originates from a critique of a commonly cited evolutionary mechanism: the Baldwin Effect. Although this theoretical “effect” is often invoked as an evolutionary mechanism leading from functional plasticity to increased specificity of genetic control, biological examples and simulation studies will be presented that show that the opposite effect is more likely, and also that there are other surprising correlates of this process. An animal example—song production in a domesticated species of finch—illustrates this effect and its paradoxical consequence. In this breed, increasingly complex song structure, expanded involvement of forebrain mechanisms, greater flexibility of behavior, and a larger contribution from social learning evolved without positive selection for these traits. Instead, these birds were bred for feather coloration. Absence of selection on song appears to have led to evolutionary degradation of song control specificity and with it unmasking of otherwise hidden synergies among diverse brain systems able to play some role in song structure. These results suggest some informative parallels with features of human language functions, and the possibility that regressive evolutionary processes might play an important role in the evolution of biological complexity more generally.