Language and thought

Wow, lots of interesting stuff from you recently. There's actually something in my own reading that I'd be interested to hear your thoughts on, but first I should respond to some of what you said. Most pressingly, I suppose, I should answer for what I said about thoughts and poo.

First of all, you say you don't recognize yourself in my characterization of thoughts sitting in the head like poo in the bowels, there yet unformed, and then given shape as they leave the body. Maybe it was the provocative way I put it? What I meant is that you seem to have a model of the relationship between thought and language such that thought is something that exists in the head independently, and language is a tool for putting those thoughts across to others, and that this tool can shape or distort the original thought. At one point in the post I was responding to you describe language as a "vehicle for thought." If language is a vehicle for thought, that suggests that thought is a kind of substance, something that can be transmitted but that requires a medium in which to be transmitted. Later on, your write: "Thoughts have to be coded before they can be communicated, and when they are decoded by someone else, they never quite match what they started out as." Here the metaphor isn't a vehicle but a code. A code is something that can be transmitted between two people, but what the code encodes is a message that exists independently of the code. Unpacking the metaphor, you seem to think of thought as a something that isn't quite language but enough like language that it can be "encoded" in language. It's these sorts of things that made me associate you with my poo analogy.

Actually, I think the code metaphor is quite telling: a literal code is a language that is translated from another language. If we follow the metaphor out, thought itself must be a kind of language in order to be encoded. And if so, do we have any better model for the kind of language that thought is than language itself?

It occurs to me that I don't have to reach as far as Wittgenstein to object to this model of the relationship between language and thought. Isn't this really one of the main targets of Derrida's work in the 60's and 70's? In "Signature Event Context," Derrida tries to deconstruct the notion that communication consists essentially of transmitting encoded thoughts from one mind to another, and in particular that writing is simply an extension of this transmission that allows us to communicate with people who aren't present. You know Derrida better than I do, so I'll let you judge whether you think he'd find your characterization of language as "something that couldn't possibly cause any objection."

In your defense, you insist that we can't simply say that thought = language and condemn children and others who don't think to thoughtlessness. I don't think I want to say anything as crude as that. But I do want to say that our concept of thought is closely tied to our concept of language, so that we're only able to ascribe thoughts to infants and animals by virtue of giving words to those thoughts. I think it's unproblematic to say that the infant wants its mommy even if it can't yet cry "mama." The mistake, I think, comes in then reifying this "thought," and thinking that there's a something in the infant's head which is the thought-that-I-want-mommy, and which we're identifying when we say the infant wants its mommy. That leap--from saying the child wants its mommy to saying that there's a definite thought in the child's head to the effect that it wants its mommy--then leads us to suppose that when the child learns to speak and says "I want mommy," it then too has a thought in its head, only now it's able to put that thought into words.

What I'm saying when I question your characterization of language is that I question this leap from how we talk about thoughts to the ontology of thoughts. Ascribing thoughts to infants, animals, and even other talkative people, is simply a way of talking, a way of describing behaviour that we find useful and illuminating. But if you think about how we use this language of thought ascription--and especially how we first come to learn to use phrases like "he's thinking..."--you'll see that this language doesn't really consist of descriptions of inner states, but descriptions of behaviour. Which isn't to say I'm a behaviourist--I'm not saying that thought is behaviour--but just that we learn how to talk about people's mental lives by seeing how they behave, not by seeing into their souls.

This might also explain what I said about thoughts on the tips of tongues. I (and Wittgenstein) wasn't denying that we can often have that experience, and that often we then find the words we were looking for. The point, rather, is that that expression ("it's on the tip of my tongue") isn't a description of my consciousness, where I can as it were see the thought sitting there, but through a thick enough haze that I can't accurately find words for it. Rather, it's a way of expressing the fact that I don't quite have the words I want, but I hope to come up with them soon. As Wittgenstein suggests, if no one ever then came up with the words, we wouldn't have use for this expression.

So yes, I agree with you that thought and language aren't the same thing, and that it's perfectly legitimate to talk about thoughts I can't find words for or what have you. But I want to resist the further inference that these thoughts are things of a sort, mental things, or even mental processes as you suggest.