Reciprocal Causation and Statistical Reciprocity

Ernst Mayr famously distinguished between proximate and ultimate causal explanations. His view was decisive in the fact that development was not included in the theory of evolution. Nevertheless, the explanatory role of developmental processes in evolution is a central theme of current theoretical biology, which has led to several revisions of Mayr’s distinction. One of the reasons for this is that the interactions between organisms and the environment influence evolution. This is a cornerstone of the Extended Evolutionary Synthesis. How are these reciprocal interactions integrated into evolutionary theory? I will argue that this question can sometimes be answered by adopting an interactivist view based on a reinterpretation of reciprocal causation: developmental processes (such as niche construction or developmental plasticity) interact reciprocally with natural selection to produce evolved traits. In the following, I will critically analyze Mayr’s critics to determine whether the integration of development and evolution based on the reinterpretation of reciprocal causation is an appropriate alternative. I will argue that this is not the case, since this interactivist framework raises several explanatory problems. Instead, I should reconsider Mayr’s distinction by adopting an alternative view of evolutionary causation, known as the statisticalist view of natural selection, which posits that the only level of causation is the individual level and that ultimate explanations are statistical in nature. I argue that this framework avoids the explanatory problems identified earlier. To explore the relationship between organisms–environment reciprocal causation and ultimate explanations, I introduce the concept of ‘statistical reciprocity’ to measure the statistical effects of reciprocal causation in population changes. I situate this proposal within a general framework I call ‘population ontogenetics’, which is seen as an attempt to unify development and evolution beyond interactivist positions.