Wednesday 21 January 2009

More on know-how

I’ve spent some of my commuting time to Preston and back catching up on some of the growing literature on the analysis of know-how stemming from Stanley and Williamson’s paper, which I mentioned a month back. In it, they argue both against Ryle’s regress argument for the priority of know-how over knowledge-that, and proffer an analysis of know-how as a particular sub-species of knowledge-that. Hannah knows how to ride a bicycle iff, relative to a context, there is some contextually relevant way ώ such that Hannah stands in the knowledge-that relation to the Russellian proposition that ώ is a way for Hannah to ride a bicycle, and Hannah entertains this proposition under a practical mode of presentation.

Of the papers I’ve just been looking at (mainly drafts on the web hence the lack of page references etc.), Alva Noë (pictured) attacks Stanley and Williamson’s criticism of Ryle’s argument and defends a close connection between know-how and ability. Ted Poston and Elia Zardini independently point out problems for know-how that it would inherit from a too close connection to knowledge-that. Phil Joyce argues for the existence of practical modes of presentation through a different route. And John Bengson and Mark Moffett attempt to explain why know-how sometimes implies ability and sometimes does not through a substantial theory (to which it will turn out I do not get on this train journey).

I’ll come back to Ryle’s regress argument another time. (No time for research, really, in my current role.) But, having attempted to defend something like it, Noë argues for a much closer association of know-how and ability than Stanley and Williamson allow. They use the failure of this connection to motivate a different ‘intellectualist’ account of what it comprises. But Noë argues that the counter examples they deploy can be explained away as perfectly ordinary cases where possession of a general ability is not sufficient for action because of contextual impediments. Absence of a piano or loss of one’s arms (!) need not undermine one’s ability to play and thus its connection to know-how.

Noë also says something that seems to me just right about Stanley and Williamson (and I said roughly like this in my previous post).

I don’t mean to suggest that that the fact that Stanley and Williamson analyse knowledge-how as a special kind of knowledge-that derogates from their claim that it is, for all that, a bona fide species of knowledge-that. They are right to defend themselves against that charge (433ff). What I would suggest is: (1) Stanley and Williamson’s analysis is merely technical – it presents a new notational or conceptual framework within which it is possible to make the same old distinction. (2) Whereas the distinction between knowing how and knowing that is pretty straightforward and is easily illustrated with examples, the account of the distinction that Stanley and Williamson offer is somewhat obscure. [Noë 287]

Although, in order to argue that know-how is a species of knowledge-that, Stanley and Williamson place most weight on some linguistic analysis, they also address a potential source of disanalogy (which might undermine the assimilation) which involves Gettier cases.

We can imagine cases of justified true belief that fail to be knowledge-that, because they fail to satisfy some extra condition. It may appear difficult, however, to formulate examples that fall short of being knowledge-how for a similar reason. That is, one might think it is difficult to conceive of Gettier-cases for knowledge-how. But if knowledge-how is really a kind of knowledge-that, there should be such cases. [Stanley and Williamson 435]

Ted Poston addresses this point. He argues that there is a conflict between the necessary requirement for something to count as a Gettier case of know-how and for it to count – contra to its Gettier status – as knowledge.

The first premise of the argument states that Gettier cases for know-how, if they exist, require that the subject intelligently and successfully φ, where φ ranges over actions. In general, Gettier cases for know-how, if they exist, would require that the intelligence condition and the success condition are satisfied. These conditions are analogous to the justified belief condition and the truth condition in Gettier cases of knowledge-that. In a Gettier case for know-that justified belief comes apart from the truth in a way that is incompatible with knowledge-that. So also, in a purported Gettier case for knowledge-how the intelligence base the subject uses doesn’t connect to success in the right way to yield knowledge-how…The second premise of the argument is that one knows how to φ, if one can intelligently and successfully φ. If, for instance, Sally intelligently moves this way and that way with the goal of riding a bike and she succeeds then Sally knows how to ride a bike. So given the first premise, the sufficient condition for knowing how laid down in the second premise is satisfied. Therefore any alleged Gettier case for knowing how will turn out not to be a Gettier case, for it will be a genuine case of knowing how. [Poston]

The important step of this argument is the second. It would not help intellectualists to challenge Poston’s specific suggestion for how Gettier cases of know-how are to be constructed. The key claim is that successful action (of the right intelligent sort) implies possession of know-how.

Poston goes on to suggest, interestingly, that the underlying disanalogy between knowledge-that and know-how is that the latter is not affected in the same way by intuitions of luck. Acquiring the right sort of skill though luck does not undermine the status of the know-how ascribed on the basis of that (right sort of) skill. ‘Knowledge-how isn’t constrained by the same anti-luck intuitions as propositional knowledge’ [ibid].

Elia Zardini connects his criticism to Stanley and Williamson’s need to appeal to a practical mode of presentation. Without it, knowing that some way is the way to play a piece of music is not sufficient to have practical know-how: to know how to play it. Adding this idea into the mix stops the account ‘overgenerating’ cases of know-how (in cases of mere spectatorship). But, Zardini argues, the resulting account ‘undergenerates’ the range of cases of know-how.

Although there is conceptual complexity in the connection between dispositions to act and know-how, still action can be evidence for know-how.

Now, as Carl Ginet (1975, pp. 8–9) first pointed out, much care is needed when considering any putative connection between knowing how to F and being disposed to F (or even being able to F): Segovia himself may know how to play the guitar without being disposed to play it, maybe because he has decided never to play it again, or maybe because, every time he touches one, he falls asleep (in the latter case it would even seem right to say that he cannot play the guitar, even though he knows how to play it). Yet, it seems that, however complex and mediated it may be in view of Ginet-like cases, some such connection between knowledge-how and disposition to intelligent and successful action must exist, on pains of jeopardizing what seem to be our best grounds for attributing and withholding knowledge-how. [Zardini]

Zardini offers the following biconditional: One knows how to F iff, under suitable conditions, one is disposed intelligently and successfully to F. So we have a couple of equivalences in play. There is one between intelligent and successful action and know-how; and one between know-how and knowledge that a certain way of acting is right under a practical mode of presentation (the equivalence that Stanley and Williamson set out).

But just as Poston suggested the link from practice to know-how undermined the connection between know-how and knowledge-that in Gettier cases, so Zardini argues that it blocks a distinction between knowledge-that and merely true beliefs that. This results from the intuition that if knowledge under a practical mode of presentation would be sufficient for practical ability then so would merely true belief under a practical mode of presentation. So, going the other way across the biconditional, practical ability may imply something less than knowledge and so, if Stanley and Williamson’s analysis of know-how as knowledge-that were correct, there would be much less knowledge about.

The very same materials employed by them to address the overgeneration problem (PMPs [practical modes of presentation]) can thus be turned against them to raise what seems to be a serious undergeneration problem. [ibid].

So Noë, Poston and Zardini all use the connection between practical ability and know-how to make trouble for the connection between know-how and knowledge-that. Of course the first connection is, at the very least, a complex one. But there is surely something right about it and thus forceful about their arguments.

In support of intellectualism, Phil Joyce has an independent argument for practical modes of presentation based on the ‘different mode of presentation’ response to Frank Jackson’s knowledge argument. Just as in the normal case, this reply argues that Mary acquires a familiar physical proposition under a new mode of presentation, so in a case of knowledge of the physics of riding a bicycle, Mary does not yet have those propositions under practical modes of presentation. But I do not think that this suggestion gets to the heart of my interest. I can’t seem the harm in talking about practical modes of presentation. But it seems to me that this is just a label for know-how, not a step towards understanding it in other terms and is thus a small part of the intellectualist case.

That leaves (for now) John Bengson and Mark Moffett’s intellectualist paper but the lights of Kendal are now visible through the train window and I must get off the train.