I’m enjoying the conversation with Thiago and others over at Long Sunday about Austin, in the comments on a post on a piece by Michael Berube. This post follows up on some thoughts I had in an email exchange with Thiago.
One of the things I like about Austin is the he opens the door to all kinds of negative or relativizing moves. To my mind it’s just a hop, skip, and a jump from Austin to Grice, who I like quite a bit (from what I’ve read of the guy).
I like that it’s about what people do, practices not principles. And I think it opens up the following relativizing move, by making the criteria success or failure (understanding or misfire) of an utterance. It makes context, including extralinguistic context, important. And, well, I’m not sure I can put this cleanly.
Let’s call how a person talks (their total set of speech habits, say) an idiolect. Everyone’s idiolect is different (idiosyncratic). Like fingerprints. No two people talk the same. Success or misfire of an utterance is, for Austin, largely conventional. (Either one meets the procedures for marrying or one doesn’t.) Where things get super interesting is around the question of what is a convention and who decides. To my mind, conventions have many levels, like localities (there’s different levels of places - planets, hemispheres, contents, countries, regions, towns, blocks, rooms, etc). In one context one might use a non-English word (’Dasein’) without misfire. In that same context one might use
an English word (’baleen’) with misfire. Which is to say, some word’s being in/of a natural language is not a particularly relevant question when it comes to approaching speech acts. And, in a sense, claims about proper or correct examples of a natural language are legislations, prescriptions about what the language ought to be, and valuations of certain types of speech vs other types.
Here’s how I see it. Languages are, in one sense at least, names for relative similiarities across idiolects. If idiolects are particles, languages are clouds of particles, wherein the distance between particles is relatively small. (Distance here is a metaphor for ‘likelihood of misfire’, ie, a speaker of Spanish will misfire more often with a speaker of Italian than another of Spanish, and probably even more often with a speaker of German.) This means that the borders between languages are fuzzy, and only visible within a certain frame of reference (in the sense in which at certain levels of magnification we can distinguish the border of a cloud, tell cloud from not-cloud, but at other levels of magnification we can’t tell, or more, those distinctions don’t make sense).
One still might be able to derive a certain version of a standard case for a language based on a hypothetical test. If we were able, hypothetically, to measure the quantity of misfires to successes between each idiolect and every other idiolect in a language we might be able to find an idiolect that misfires least with every other idiolect. Using the dustcloud analogy, this would be finding the particle in the center of the cloud, the one that is closest to every other particle in the cloud.
The same procedure might be used to determine the boundaries of languages from each other (to tell the borders of dust clouds). This would involve setting some minimum standard of misfire to success ratio, below which two different idiolects can’t be said to be of the same language. So, if our minimum standard is any success at all, this would mean that idiolects that only ever misfire would be of different languages.
Thinking of it now, this actually doesn’t serve us much at all. Aside from the impossibility of the test, and the assumption that languages (dust clouds) are static entities such that the idiolect (particles) don’t move in a way that makes the measurements indicative of little after the time of their measure, there’s also the question as to whether or not it might not be possible to find idiolects at the borders of languages (dust clouds) that are always close enough to other idiolects such that there’d be no borders to the languages, just one big language (dust cloud) with varying clumpiness. Spanglish in the US, and micro-dialects like Academic Anglo-Heideggerian might serve as examples of border cases close to other languages, close enough to make it a rather arbitrary matter of legislating what’s in and what’s out. All of which is to say that “is it of/in this language” is not the most useful of questions. Or, to put this differently, any time there is a claim made as to something being in or out of a language we might ask three questions: “why do you say that (what’s at stake in the in/out distinction)? what are the criteria for distinguishing in or out? and is there always going to misfire in every case of the use of this utterance? (if not then there’s likely another agenda).” All of which is to say pragmatics is much more interesting.
I’d particularly like to think more about how different illocutionary acts (and even deliberate misfires) can be used in service of perlocutionary acts, particularly in the context of the academic labor market process (I tried to address this a bit in a post a while ago about what I called lexical rigidity).