Idols, Paradigms and Specters in Data Sciences
摘要
This chapter begins by clarifying the principle behind most false reasoning about abstraction. This misleading principle is called here inebriation by abstraction, which leads the explorer in the field to always want to rise to a more abstract level. The best way to get rid of this spontaneous philosophy in the study of abstraction is to assume the concrete conditions of its production and circulation. It was handwritten inscriptions in the sixteenth to seventeenth centuries, the proliferation of printed collections in the nineteenth century, and electronic media today. The three regimes are characterized by comparing their respective know-how and presuppositions. The question then arises as to the conditions of their contemporary presence on the same scientific horizon for instance today, from both an epistemological and a sociological or historical point of view.