Inner Speech and “Thoughts” (I)
Fernando Martínez-Manrique and Agustín Vicente have an interesting article in a recent issue of The Journal of Consciousness Studies. (No link here because it’s a subscription journal. Full reference appears below.) The title is “ ‘What the . . .?’ The role of inner speech in conscious thought”. By “inner speech” they mean the silent ‘talking to ourselves’ that most people report as something they do quite a lot of.
Major sections of this article are devoted to raising problems for the views of other writers on inner speech, but I’m not going to go into those. Nor am I going to try to summarize all of the informative points made in this rich and well considered article. Instead, I’m going to confine myself to comments on a picture that is popular in discussions of brains and minds. (“Pictures” in philosophy is a metaphor for metaphors.) A follow up posting will comment on another popular picture that’s suggested by this article.
The picture I have in mind today is that of “broadcasting”. In the background of this picture is the assumption that there are many processes that go on unconsciously in our brains, and that proceed at the same time and largely independently of each other. Against this background, consciousness is often thought of as having a special role – namely, that of broadcasting information to all parts of (or, processes in) the brain. What gets into consciousness is thereafter available at many places in the brain and thus may influence many brain processes.
The broadcasting idea seems plausible. After all, if something gets into consciousness, I know about it. If I think of myself as a unified self, it seems that what I know should be available to my whole self, and therefore to any process that’s relevant to my thinking that may be going on in me.
Martínez-Manrique and Vicente, however, make a striking observation that should lead us to think carefully about this “broadcasting” picture. Inner speech is, evidently, linguistic – it’s composed of words and their order makes a difference. (“John loves Jane” is different from “Jane loves John” whether it’s said inwardly or out loud.) So, if inner speech were “broadcast”, what is said in it could have a useful effect only on a system that works on linguistic inputs – a language system that could understand the words as words and their order as making a distinctive contribution. Whatever effect a “broadcast” bit of inner speech might have on other kinds of systems could not be an effect that made use of the linguistically encoded information.
This point seems to have quite general application. If any neural event has informational content of any kind, and it causes some effect in other neural events, those latter neural events will “receive” that information only if they are structured so that they can use it. Otherwise, there may still be an effect, but it wouldn’t be one that depends on the informational content. The voice that shatters glass does so because of its pitch and loudness – not because of the meaning of the word being sung, even if that word happens to be “shatter”.
Sometimes we learn something on one occasion and then recall it and put it to use in some different context. Cases like this may suggest that once we have been conscious of some fact, it is then generally available for further use. But it would be risky to draw much of a conclusion from such happy occasions. That’s because we do not know now many occasions there may have been where something we learned would have been helpful, but it did not come to mind.
Sometimes we have evidence for unhappy cases of this kind. For example, after losing some game, a friend might ask why we didn’t make a certain move. The best answer we can give might be “I just didn’t think of it. I know the rules, so I would have known I could make that move if I’d thought of it, but it just never crossed my mind”. That would be a case of something I knew that was not available to my cognitive processes when it was needed. But while we know there are some such cases, it would take an ingenious experiment to figure out how often that kind of failure happens. If there had been no friend there to ask why we didn’t make a certain move, we would likely never have realized that we’d failed to use something that in some sense we knew, and that might become available for use on some other occasion.
A radio or TV broadcast is received by many receivers that are designed to process the incoming signal in a way that preserves information. And the message is received by listeners, each of whom can understand it. We won’t be making progress in understanding how our minds work if we populate our brain with a lot of understanders. (After all, understanding is something we want our cognitive science to explain.) The brain events that cause our consciousness undoubtedly have many other effects, in many parts of the brain. But it may be more helpful to think of these simply as neural effects rather than as receptions of broadcast messages.
[The article referred to is Fernando Martínez-Manrique and Agustín Vicente, “ ‘What the . . .?’ The role of inner speech in conscious thought”, The Journal of Consciousness Studies 17(9-10):141-167 (2010).
There are several interesting posts about inner speech on Eric Schwitzgebel’s web site: go to http://schwitzsplintersinnerspeech.blogspot.com .
I discuss inner speech in Your Brain and You. There’s more about it in my “Thoughts Without Distinctive, Non-Imagistic Phenomenology”, Philosophy and Phenomenological Research, 70:534-561 (2005) and “A Frugal View of Cognitive Phenomenology”, soon to appear in T. Bayne and M. Montague, eds., Cognitive Phenomenology (Oxford: Oxford University Press). ]
Your Brain and You 
I think that the idea that we shouldn’t populate our brain with a lot of understanders is a sensible one. However, I’m not sure about the idea that “the brain events that cause our consciousness” should be understood “simply as neural effects”. I agree that the broadcasting picture can’t be correct as it stands, and I agree that whatever happens in mind is a neural effect. Yet I’d still say that what matters for a neural event to have a mental effect are its informational properties.
Perhaps there is a different moral to draw from the fact that inner speech has to reenter the system through the linguistic processor in order to have such an informational impact, i.e., the idea that the relevant informational properties are not strictly those of the linguistic input –the internally rehearsed sentence– but of the conceptual structure that interfaces with the rest of the systems. This structure is reached by means of the linguistic input but it doesn’t mimick its content.