A wide range of paradigms is often taken as a sign of the prescientific status of a discipline. That’s why this paper focuses on the reduction of sociological paradigms. It is shown that, beyond the differences, Simmel’s formal sociology and Weber’s comprehensive sociology are reduceable to the common paradigm of nominalist sociology, the axiomatisation of which is introduced here. Moreover, the paradigms’ reduction accentuates the continuity between all the programmes that claim one or the other of those legacies (actionnism, strategical analysis, relationism, practical interactionnism, network sociology). Such differences don’t result from incommensurability; they are due to the fact that these research programmes focuse on different facets of the same paradigm.