Two distinct issues have hitherto been run together in the debate between “statisticalists” and “causalists” concerning the causal status of evolutionary theory. One is what I will call the contingency issue, and this concerns selection and drift specifically. The other, for now the level of causation issue, concerns the status of the explanations of population genetics and so targets the factors that occur in its equations.

One aim of this paper is to show that the issues are indeed distinct and that this has not been fully acknowledged hitherto; while the contingency issue is present in statisticalists’ arguments it has not been separated from the level of causation issue.

Another aim is to show that focus on the contingency issue establishes that selection and drift are not causes. The problem with the causalist position regarding these factors is that it amounts to positing causal factors to “ensure” non-contingent relations. This would then be to promote a “counter-possibilistic” account of causation. However, this verdict about selection and drift is consistent with thinking that explanations in population genetics are causal. While selection and drift can be ruled out as causes by appeal to non-contingency this is not an available option as for fitness, population size, or the other factors that appear essentially in the equations of population genetics.