Davidson: Private Language

Posted on December 10, 2008
Filed Under Davidson, Philosophy | 3 Comments

Yeah, I’m still doing Davidson. Last time was Davidson and Rational Animals. This time I’m taking a common theme from several essays (so I don’t get repetitive). The question is the possibility of a private language. Roughly, as I see it, Davidson is taking a Peircean like angle.

To have a belief one must have the verifiable properties of a belief. That is we must be able to measure that someone has a belief because beliefs have practical consequences. To believe is to have a proposition one holds as true or false. That demands we understand truth. The only way to have truth is to be able to judge an interpretation as true or not. The only way that is possible is for there to be a second person we communicate with.

Now of course we know a private language as an empirical fact is impossible. But that’s really more to do with the physiology of brain development than a philosophical point. (The parts of the brain dealing with language don’t develop correct if you aren’t interacting with language users — as we find with feral child who often can thereafter never learn a language)

The philosophical point is a little more abstract. The fundamental question ends up being whether “communicating” with oneself is sufficient or not. That is can’t I “speak” to my future self who can tell if what I say is correct or not.

Davidson argues in many places (such as “Second Person”) that it can’t because a certain kind of triangulation is necessary. That is the possibility of error of a certain sort is necessary in order to learn truth. Without that kind of error in language language itself isn’t possible.

Maybe I’m misreading him but it seems really what he’s after is that two people must point at the same object in a way a single person separated by time can’t.

Why does this matter? (Beyond the fact it’s a long running philosophical controversy) I think it’s because language is seen by many (including Heidegger and Davidson) as grounding so much philosophically. It becomes a medium and the nature of that medium has huge philosophical implications.

Related posts:

  1. Davidson & Rational Animals
  2. Language, Externalism and Meaning
  3. Davidson: First Person Authority
  4. What is a Concept?
  5. Davidson: Knowing Ones Own Mind 2
  6. Davidson: What Is Present to the Mind

Comments

3 Responses to “Davidson: Private Language”

Suppose I invent a private nomenclature for a personal project, document it on paper, think about the project in those terms, and evaluate the results. Have I not just invented a legitimate private language that is known to me alone?

If the point rather that a language must knowable in principle (but not necessarily in actuality) by a second party, then I trivially agree. But then the point is rather that there is no such thing as an instrinsically private language, rather than no such thing as a private language.

Philosophically, I would say that all legitimate concepts, private or otherwise, must describe something “that one can be wrong about” in some context, or rather must describe something than can be real in some possible world (however abstract). And further that any construction that doesn’t meet that criterion is nonsense, i.e. properly speaking not language at all. Given that perspective, I fail to see why language needs to be intrinisically public, any more than thought needs to be.

The way Davidson is taking it is a bit more basic. That is the idea of a language without someone you talk to. Since you already know public languages any “private” language you derive is in theory translatable since you know both.

As I said as an empirical matter there really isn’t any question. The brain simply won’t develop in such a way a private language is possible of the sort Davidson is talking about. The question is the more philosophical issue.

I was doing some more reading and I found one reformulation of “The Second Person” which makes it a bit stronger. The idea is that truth qua truth is objective truth. The relationship between you at two times is such that it’s still contingent on a single subjective stance. That is objectivity to be objectivity requires multiple subjective stances which a single person can’t provide.

I think that’s still weak, (I don’t think time unifies the subjective self sufficient to make this claim), but it is stronger.

I suspect a stronger yet version is tied to science. That is objective truth requires multiple sufficiently separated stances. (i.e. the scientific sense of objectivity) This would invalidate the Davidsonian sense of triangulation since you’d need multiple triangles.

I am glad to hear you say that, Clark. I was debating the point until I was blue in the face, but a few months ago I could hardly convince anyone over at FPR that the sine qua non of any conception of truth worthy of the name was objectivity.

One thing that became obvious is that the correspondence theory of truth has technical problems. Truth (with a capital T) is not an artifact of propositions – it is rather an abstract, language independent meta-description of reality itself. A proposition only relates to Truth in the context of some schema. If you do not fix the schema, the truth of a proposition is a pretty ambiguous concept.

I would say communication requires more than a shared language, it requires a shared schema – namely shared semantics. The suggestion that there can be such a thing as a non-foundationalist semantics drives me up the wall.

Leave a Reply