The historian Paul Veyne (1984) expresses an idea that exists in many social scientists' minds: "For a historical explanation to be admissible, it must not present any interruption of continuity in causal relations that connect the agents involved in the plot." Invoking remote causes would then always give a bad explanation, or at least an explanation less good then the one referring to proximate or immediate causes. I will argue that both kinds of explanations - invoking immediate causes or invoking remote causes - can be good explanations. Which kind is the most appropriate depends on your motivation and the related questions that have to be answered.