Great punctuations: Prediction, randomness, and the evolution of comparative political science

I argue that comparative politics has been shaped by two "Great Punctuations" that, on each occasion, transformed our conceptions of what the subfield is and what we do. Just before a punctuation occurs, the subfield seems especially coherent, united by a set of common assumptions, methods, theories, and so on, which are then punctuated by a series of events that destroys faith in them. The subfield then reconstitutes itself around new assumptions, until, just as coherence is achieved, the next punctuation occurs. To demonstrate why the subfield has evolved in this way, I draw on probability theory to argue that the desire to be a predictive science causes us to imagine the world to be far more predictable than it actually is. This results in the development of theories that are surprised by events; hence the peculiar trajectory of the subfield.