I’m very grateful to have been sent a draft paper (a manifesto of SEESHOP3, indeed) by Harry Collins which modifies his (and Rob Evans’) previous views on the relation between contributory and interactional expertise. Having raced through it, I think the new story has three main modifications of what went before (but you will have to wait for the proper publication to see whether this is accurate; I will add a link).
First, interactional expertise is now seen as fundamental. On the previous account it was true that contributory expertise implied interactional expertise but not vice versa but the suggestion was - I think - that the former explained the latter. Because one had practical contributory expertise, one should be able to put it into words (although for those who were inarticulate this remained a merely latent ability). Now the story is that contributory expertise is explained through the possession of linguistically structured interactional expertise.
I am not sure just how pure this priority thesis is, whether, eg, everyone who has a practical skill acquires it through acquiring a necessarily linguistically or conceptually structured ability. There is a contrasting suggestion in the draft paper that bits of language are themselves shaped by bits of practical expertise. If so then the language of a scientific area depends on practices (so practice is prior, for coal face experts) but then shapes the interactional expertise that in most cases is not paired up with corresponding contributory expertise (so language is prior, for everyone else). Still, there’s quite a bit of reference to the idea that ‘language is central to practice’ which rather implies the other, ie first priority. Indeed, that this is the key innovation is firmly stated:
What is new now is that language plays as big a part in learning a practical expertise as it does in learning a non-practical expertise and that language can carry understanding of practical expertise even in the absence of practice.
Second, having interactional expertise without having contributory expertise is the nature not only of sociologists and science managers but also of most practitioners, most of whom master only a very small fraction of the practical skills of any area of science. Interactional without corresponding contributory expertise is the norm.
Third, the difference between interactional and contributory expertise is ‘a difference in social roles rather than in the grounds of knowledge’. This is because there is almost no difference in epistemic terms between the two and so the difference has instead to be carried elsewhere.
The change seems quite plausible and goes some way to explain my confusion about just what the status was of the original distinction. Further, it can be accommodated within a broadly McDowellian picture of conceptualised practical judgement (or phronesis). Obviously I like the idea of putting language at the centre of things (on a suitably extended notion of language that borrows the McDowellian idea of conceptual capture in demonstrives such as that shade!.) But it does seem to have undermined what seemed interesting to me in the original account: that there was a fundamental kind of non-practical but still tacit knowledge. I’m not sure that I’d buy a book based on the new picture. (But to repeat my original thought: the content will very much depend on the use the ideas are put to by others.)
PS: Later I had an interesting email conversation with Harry Collins about this post.