Right from the beginning, the term "bureaucracy" was a sneer or an accusation, to be contrasted with organized wisdom. Officials concerned with productive enterprises, territories and resources, public services, and financial and social regulation (inter alia) were more vulnerable to the charge of being bureaucrats than were (seemingly) independent scholars and scientists. Yet knowledge-making of all sorts depends on social organizations. Mines, for example, were classic early modern sites of organized knowledge. If anyone thinks science can be confined to academic-type institutions, they are deluding themselves. Those who generated knowledge and managed information within organizations, however, had to work harder to defend their professional dignity. The divide between effective administration and dull routine was defined on the basis of presumed moral character and experience. Especially after 1789, high officials presented themselves as gentlemen and insisted on the continuity of family traditions; knowledge passed along as a kind of inheritance. The push for human sciences was grounded at first in aspirations of this kind. Specialist expertise, grounded perhaps in (neutral, statistical) information, was one of the claims of social science, but mostly social knowledge was closely linked to practical issues rather than trying to rise above them. Science appeared as one of the key resources to prevent elite knowledge from appearing as nothing more than bureaucratic. Claims to wisdom by state officials do not yet seem ridiculous. But could such an ideal be reconciled with the ideal of data-driven knowledge?