A Nice Derangement of Epitaphs To riff a bit more on thoughts inspired by the paper: I think the distinction between 'passing theory' and 'prior theory' is really interesting, and they are even more so when thought together with some other super interesting issues regarding langauge. Two, related issues, in particular: that of the status of examples, as well as the issue of learning. In both cases of language-use (which are usually brought up in tandem, i.e. we use examples in order to teach) what's at stake is a kind of spontaneous generation of a "passing theory". A passing theory, moreover, that can be generalized to become a 'prior theory' (i.e. when you're teaching someone, you say 'you do it like *this*', and the student is expected to figure out how to do the same for other cases of *this* - c.f. Witty's comments on 'learning how to go on', PI§151).
In some sense, the 'prior theory' is misnamed: 'prior' theories are not 'prior'; they are, instead, after the fact. They are ratiocinations of what are instead generated in situ and then projected backward in time: effects mistaken for causes. So-called 'prior' theories function, at best then, as sets of heuristics, resources to look to in some cases of trying to figure out novelty, but not at all as distributing the grammatical shape of words or phrases.
Anyway, examples are so interesting because they effect a kind of convergence between 'passing' and 'prior' theories: they enact a passing theory whose status is to be taken for a 'prior' theory ("this is how things ought to be done"). Or to put it otherwise, they effect a kind of short-circuit between saying and showing: examples show how one is to do something as much as what one is to do. Examples show what they say. (to quote Girogio Agamben: "Neither particular nor universal, the example is a singular object that presents itself as such, that shows its singularity. Hence the pregnancy of the Greek term, for example: para-deigma, that which is shown alongside". (The Coming Community)).
And this in turn sheds light on the notion of 'use', and helps to show why 'use' does not in any way mean 'use among a community'. There's a great remark in Davidson: "Someone who grasps the fact that Mrs Malaprop means ‘epithet’ when she says ‘epitaph’ must give ‘epithet’ all the powers ‘epitaph’ has for many other people... These remarks do not depend on supposing Mrs Malaprop will always make this ‘mistake’; once is enough to summon up a passing theory assigning a new role to ‘epitaph’" - a single instance is all that is needed generate a use: it might even be a 'one-off. Malapropisms function very similarly: they are 'one-offs' that generate their own passing theory that can be recognized as such. And I can't think of any off the top of my head, but I'm pretty sure that there are malapropisms that have become, through common use, accepted as terms of their own.