Six Impossible Things before Breakfast: Assumptions, Models, and Belief in Molecular Dating

Abstract

Estimating dates of divergence from DNA sequence data is a methodological and computational challenge as the amount of data explodes, and as models and methods become more statistically sophisticated. The focus on methodological advances distracts from deeper challenges associated with historical inference: how can we construct plausible narratives for events long-past and long-term processes that we cannot directly observe?

As molecular dating methods grow more sophisticated, the number of assumptions we must make about the evolutionary process increases, so that it becomes difficult to keep track of what we must believe in order to believe the molecular dates. Some disagreements between molecular dating studies should primarily be interpreted as a debate about assumptions and prior beliefs, such as the interpretation of fossil evidence.