I still haven’t had an opportunity to write a proper and considered review of Collins’ clearly important book on this subject but I can gesture at some of the developments in the second half and thus contextualise my own initial comments from a few weeks ago.
One of the features that marks out this book from most of the rest of the literature on tacit knowledge is that it builds on Collins’ recent work on forms of expertise that are carried in language. (Recall talk of interactional expertise etc. from Rethinking Expertise.) Thus whilst chapter 5 concerns bodily or somatic tacit knowledge, familiar from the work of Dreyfus for example, chapter 6 is called ‘Collective tacit knowledge and social cartesianism’. It is this latter chapter that brings in the kind of distinction which I find so important between linguistic agents and mere animals or trees.
This reflects Collins’ fairly explicit initial approach to tacit knowledge. As he explained to me in a recent email:
I implicitly define ‘knowledge’ as meaning that which you have when you can do the thing and that which you don’t have when you can’t do the thing. This approach has nothing to do with self-consciousness, or intentions, or actions as opposed to behaviours. Those distinctions are discussed at length in Collins and Kusch’s Shape of Actions. At least, in so far as it does have anything to do with those distinctions, it is that Collective Tacit Knowledge is still mysterious because it is the knowledge required to carry out polimorphic actions (the subject of the earlier book). Given this much broader implicit definition of knowledge, then there are no distinctions between humans, cats, trees and sieves (I even treat the later as having the ‘knowledge’ to sort stones so that looks even dafter if you want to go there). Obviously, and this was the stated intention at the beginning, the differences between humans and other entities re-emerge at the end of the book with ‘Social Cartesianism’, which turns on Collective Tacit Knowledge.
So the starting point is very minimal indeed. ‘Knowledge’ can be possessed even by sieves. The distinction between sieves, trees and cats, on the one hand, and linguistic humans, on the other, is then earned in that later chapter from which, among other things, distinctions between mere behaviour on the one hand and action on the other can then be constructed with conceptually clean hands. Once one realises that then, radical though the minimal starting point remains, one can understand why the book begins that way. It is a deliberate methodological strategy.
There is a view of philosophy according to which philosophy does not so much tell one what to think but the cost (by way of implications and supporting accounts) of thinking what one may wish to think. Against this idea, Tacit & Explicit Knowledge, sets itself the task of earning rather than merely borrowing or stealing distinctions which seem important for characterising tacit knowledge in its various species. I remain unsure, however, whether the further cost of such a reconstruction has been fully met. But you’ll have to take a look.