When is There a Group that Knows? Distributed Cognition, Scientific Knowledge, and the Social Epistemic Subject

I pose three questions. (i) When does a collection of individuals form an entity that is more than just the mereological sum of its constituent persons? (ii) Given that there is a group of this sort, under what conditions does it know (or believe etc.)? (iii) When we talk of, for example, ‘the growth of scientific knowledge’, can we regard this scientific knowledge as an epistemic state of some social entity? I draw upon ideas from distribution cognition and Durkheimian sociology to provide responses to the first and second questions, and thereby give a positive answer to the third.