Question Do men need speech in order to think? In other words, can we do the act of thinking without "speaking" to ourselves consciously or unconsciously? For myself, I use colloquial English, the language I am most fluent in, when I think in my mind. Does it have to be the case that one would use his or her most developed language to think?

Accepted:
February 21, 2006

Comments

There are two competing views on this. The first view -- possibly the more popular view among philosophers -- is that thought and language are essentially tied together, so that there cannot be one without the other. (Leave aside all the evidence from casual observation that it's all too possible to talk without thinking). The argument for this view appeals to the evidence we generally need for attributing thoughts to others -- namely, verbal behavior. (You can read a good example of this sort of argument in an article by Donald Davidson called "Thought and Talk." ) There is also the consideration you raise, that thinking sort of "feels" like talking silently to oneself.

But the inconvenient thing about this view is that it requires us to deny that pre-verbal children and animals have thought. And while we don't have all the evidence for attributing thinking to such beings that we have in the case of fully verbal adults, we do nonetheless have lots of evidence that such beings think. Indeed, the best accounts available of how children acquire language requires us to attribute thoughts to them -- at a minimum, thoughts like "so that's what that thing is called...". Or again -- most philosophers believe that the emergence of language involves the emergence of conventions -- a sort of unstated mutual agreement that certain things will be done in certain ways, that we'll use "dog" to refer to dogs. But conventions require, not only thinking, but thinking about thinking -- I must not only grasp that a certain sound made by another human being is correlated with the presence of dogs, I must grasp that that person intends to communicate to me the presence of a dog.

There's also a lot of evidence from cases of stroke victims that loss of language does not entail loss of the ability to think -- such individuals who are fortunate enough to caring, creative, and patient people helping them through recovery can work out new systems of nonverbal communication, clearly showing that they are formulating thoughts that they no longer have the ability to express in a "public" language. (There was an account of one such case in The New Yorker a few months ago.)

A great deal depends on what one means by "thinking." According the second view, thinking involves having and manipulating representations of the world. One is justified in attributing thought to a creature if one has to assume the existence of such representations in order to explain the creature's behavior. We know, for example, that chimpanzees can find novel solutions to problems: left alone in a room with a box and a banana dangling above them, just out of their reach, they figure out to move the box below the banana and climb onto it. This cannot be simply a conditioned response; the chimps have had no experience with standing on boxes before. We have to assume that they somehow represented to themselves a situation that did not yet exist, or that they went through some process of reasoning, like: "if I were taller, I could reach the banana. If I stood on that box, I'd be taller. Hence, I should try to stand on that box." But now you're going to ask, "how could the chimp think that if it doesn't have the words to express it?" Answer: the chimp thinks in a kind of "private" language -- a built-in system of representation. If this seems implausible, try giving an alternative explanation of the chimp's accomplishment, one that doesn't end up making the same assumption.

Now philosophers in the first camp, who think there can be no thought without language, may just mean by "thought" something that is consciously accessible, something that one can reflect about. Perhaps this kind of "self-conscious" thought really is tied necessarily to language (although the experiences of the stroke victims suggest not). Some of these philosophers then allow that animals and pre-verbal human children can have "thought-like states" that play the same role in explaining behavior as do the "mental representations" posited by the cognitive scientists. But in that case, the issue becomes merely verbal, and merely verbal issues are boring.

A good defense of the first view (No language, no thought) is Thought Without Talk by Jose Bermudez. The canonical statement of the second view is The Language of Thought by Jerry Fodor, but you might find the explanation of the view a little more accessible in The Language Instinct by Steven Pinker.