According to Popper (1968, 174), ‘every language incorporates and preserves countless myths and theories, even in its grammatical structure.’ Science begins when myths and theories become self-modifying. ‘There is much less accumulation of knowledge in science than there is revolutionary changing of scientific theories.… science advances by the tradition of changing its traditional myths.’
Kuhn on the other hand says that ‘normal science’ consists of filling in the details of established theories, that revolutions are exceptional. Vague as they are, these are quantitative estimates, and perhaps we can account for the difference between them by considering what their respective authors chose to count. Kuhn is looking primarily at the everyday lives of ordinary working scientists: Popper might agree that their work may not advance science very much, and Kuhn would certainly agree that normal science prepares the ground for the revolutions which really advance it. The disagreement here is similar to the debate about “gradualism” vs. “saltations” or “punctuated equilibrium” in evolution (see for instance Dennett 1995); or the debate among Buddhists about whether “enlightenment” is sudden or gradual.