When Intuitions Lose Authority

Since intuitions are expert opinions, they have authority. They are not to be contradicted lightly. When are they on less-secure ground?

Since ordinary language users are only experts w.r.t a particular fragment of a language, their expert authority is limited to their area of expertise.

Ordinary language is suited to ordinary purposes: safely banging about in the world of medium-sized dry goods. Since ordinary terms acquire their meaning in ordinary circumstances, they lose their authority in unusual circumstances.

The worry is that almost any serious metaphysics involves extra-ordinary situations where competent ordinary language speakers are not authorities. One thinks of van Inwagen's modal skepticism here. I think we have to take such things on a case-by-case basis.

On a final note, it seems to me that a degree-theory of truth can help here since natural language terms acquire their extensions by application to paradigm cases--think of how we acquire proficiency in applying the term "heap". Thus extra-ordinary situations fall outside the rules of application of most terms.

We can merely supervaluate, but I think a better use of a supervaluation space would be to use it to form proportions of the generated space to yield degrees of applicability which we might as well call "degrees of truth". This works well for the Sorites Paradox and it would be interesting to try to extend it to other issues in metaphysics. Perhaps only their epistemic counterparts will be useful, but you can't win if you don't play.

This should sufficiently piss off those who weren't pissed off by the other post.

Intuitions as Expert Opinions

Posting recently on Larry Horn's lecture on pragmatics reminded me about some thoughts I had about intuitions.

The following notion of intuitions will be pretty simple and have the result, I think, that we ought to be pretty conservative of intuitions (but not slaves to them).

The most accurate way I can think of to think of intuitions is as follows:

(TDI1) Intuitions are judgements that competent L-language users make about the truth-value of sentences in L.

Competence comes in degrees so the "weight" of an intuition as well as it's "strength" will vary. By "strength" I intend to indicate the confidence the target speaker has in the target judgement. By "weight" I intend to indicate the force such testimony would/should have for a 3rd party. Since the 3rd party in most cases will also be an L-speaker, she must weigh the judgements together (gives rise to interesting problems in the epistemology of disagreement, but that's another post).

I now take fluency in a language to be a species of expertise. It is not at all clear to me that professional lexicographers have more expertise in the meaning of terms than the ordinary fluent English speaker, since I afford quite a strong role of use in meaning (but that's another post).

So my second thesis is a strengthening of the first:

(TDI2) Intuitions are the expert opinions of fluent language users.

From this thesis I draw the following conclusion:

(TDI3) Intuitions have the warrant which accrues to expert opinions.

I think that's enough to piss off quite a few people so I'll stop there for now.

New IEP Entry on Coherentism

The Internet Encyclopedia of Philosophy has a new article up on Coherentism here.

It's a pretty good article though, of course, I don't think it gives enough attention to the "emerging market" of non-doxastic coherentism as developed by Kvanvig and Riggs.

This is, of course, not true of Jon's article in the SEP.

Why read the IEP entry when there's the SEP? For one thing, I think it is often good to read material from multiple sources on the same topic. For me that's sometimes the only way I get something is by reading several pieces differently expressing almost exactly the same material.

Also, who knows how long the SEP will remain free. So it's really good to have the IEP around. Many of the articles are very good, and the IEP covers items the SEP doesn't, for example our own Dan Mittag's entry on Evidentialism.

A Report on Horn's Field Report on Pragmatics

Philosopher of language Larry Horn was here at Rochester last week (he's an alumnus by the way) and he's pretty much converted me to Maxim Monism (albeit unintentionally).

I was aware that since Grice there have been theorists who've generated different taxonomies for his conversational maxims. I'd never looked into it much because what do I care how they are sliced up? They all seem like good maxims, so they should all be theorems regardless of which are axioms and which are derived. Still, it's an interesting question when it gets your attention and my attention was drawn to it by one of Horn's lectures.

Apparently the regnant positions are a "Trinitarian" view and "Unitarian" view (my terms, but perhaps others have used them (Horn himself is apt to use religious terms as we'll see)). I can't recall the Trinitarian view and can't reconstruct it from the notes or handout (if you know or have a guess, please tell me), but as I take it one prominent view has it that there are three fundamental maxims.

The Unitarian view--called the Relevance Theory (Carston 1995) holds that relevance is the sole driving force is relevance:

"Human cognitive activity is driven by the goal of maximizing relevance: that is...to derive as great a range of contextual effects as possible for the least expenditure of effort" (231).

Now I'm just lifting that quote from the handout, but it seems clearly over-stated. First, it's a statement about "human cognitive activity" which is quite a broad topic indeed. It's not clear to me that there is any single goal of human cognitive activity (surely one is to have accurate and comprehensive beliefs about matters that are of interest to us).

Now if you just take the part that starts with "derive..." then it does seem to me that you end up with a plausible maxim. I suppose I'm confused as to why it's called "Relevance Theory" because it seems only tangentially related to the Maxim of Relation.

Horn's preferred view he calls "Manicheanism" because it is dualistic: there are just two maxims:

1. Avoid synonymy (two forms with one meaning);

2. Avoid Homonymy (two meanings with one form).

He criticizes Relevance Theory as being "covertly dualist" but I think this is not right. The maxim seems to be truly unitary:

(*) Be efficient.

It so happens that the concept of efficiency entails two goals: 1. maximum effect, 2. minimum effort. That it has these twin entailments does not make it not one single maxim any more than the number 2 is not one single number.

I find the Efficiency Maxim preferable to the Manichean picture for two reasons:

1. It is more generalizable: In poetic or rhetorical speech there is a great deal of utility in synonymy and homonymity.

2. Even in normal communication-oriented speech synonymy could be quite efficient, since it requires fewer resources and we can usually rely on context to disambiguate.

This is a non-expert opinion to put it mildly, but I calls 'em like I sees 'em.

Key Words: Larry Horn, Lawrence Horn, linguistics, pragmatics, Grice, implicature, conversational implicature, conventional implicature.