The "theory" in theory-theory isn't really a theory

First a little bit of background for anyone who isn't familiar with the theory of mind debate.The question is how we are able to understand the mental states of other people, and whether, broadly speaking, we perceive them directly or rely on some kind of inferential process. There's also a subsidiary debate about whether that inferential process might involve "simulating" the mental states of others within our own mind, but I'm not going to discuss that here. What I want to focus on is the dialectic between Shaun Gallagher (probably the strongest proponent of the direct perception camp) and theory-theory, which is the theory that we rely on a theory in order to understand other minds.

As you may have noticed, philosophers are not particularly imaginative when it comes to naming their theories. Theory-theory is really quite simple, though. It basically says that we possess a theoretical model that we use to interpret other people's actions, allowing us to attribute folk psychological states such as pain and belief to other people, despite having no access to their mental life. So when I see you crying, I am able to understand that you are sad by checking my perceptual evidence ("crying") against my theory of mind ("crying = sad").

This infant, lacking a fully developed theory of mind, is unaware that she is sad.

In contrast, Gallagher appeals to phenomenological evidence to argue that we are actually able to perceive to the states of other minds directly, without any appeal to a theory. I was initially sceptical of this position, not because of the evidence that it appeals to (it certainly feels like I perceive your sadness directly), but rather because it seems to lack any account of the actual process that goes on when we perceive mental states. As soon as we try to give an account of this process, we seem to reintroduce a (limited) kind of theory, one that may not be explicit but nonetheless underlies so called "direct" perception.

My undergraduate supervisor Suilin Lavelle makes a similar point in her paper "Theory-theory and the Direct Perception of Mental States" (2012), and I won't deny that her view has undoubtedly influenced my own. However I do feel that Gallagher is right to deny that we explicitly theorise about other minds, at least under usual circumstances. It's still possible to reconcile his position with theory-theory, but not without putting pressure on our common-sense understanding of what constitutes a "theory".

A quick aside: I am focusing here on what is sometimes called "innate" or "modular" theory-theory. This is the theory that we are born with a theoretical understanding of other minds, one that develops in predictable ways as we pass through infancy. It can be contrasted with "scientific" theory-theory, which says that as infants we form a theory about other minds, based on our inter-personal experiences. I find the former theory a lot more plausible, for reasons that I won't go into here.

Back to direct perception. Lavelle argues that theoretical entities can be thought of as direct objects of our perception, provided that we are equipped with the correct theory (Lavelle 2012: 227-9). If mental states are like this, then theory-theory can claim that when I infer from your crying that you are sad, I am in a sense "directly perceiving" your sadness. This might not be enough to satisfy Gallagher, but it is certainly beginning to look a lot more like the kind of intersubjective experience of mental states that he advocates. In fact, I'm inclined to say that the dispute, from this angle, is little more than an aesthetic one. Gallagher doesn't want to call whatever underlies this process a "theory", whilst Lavelle (and others) do.

So why should we think that the tacit processes underlying our perception of mental states are theoretical? Theory-theorists tend to fall back on experimental evidence at this point, arguing that the kinds of systematic errors we find infants performing when they attribute mental states suggest that a certain kind of theoretical structure is at work. They also claim that in order to support inferential reasoning, our understanding of other minds must come in the form of a theory, with propositions and syntax. On the other hand, if this theory is relatively innate and non-explicit, it's unclear to what extent it could really be a "theory". Perhaps it is best described as a theory, just as we might want to say that I have a "theory of depth" that allows me to perceive a far-away cow as being normal sized, despite it appearing to be small. This doesn't mean that I literally understand depth theoretically though.

I think some theory-theorists would actually agree with this interpretation, which is why I said that the dispute is mostly aesthetic. Some people are happy to call a tacit "theory" a theory, others aren't, but this doesn't mean that they actually disagree about anything significant - which sadly is often the case in philosophy. There may be something more significant and fundamental resting on the distinction between a theory and a non-theoretical perceptual process, but I'll happily admit that I'm not quite seeing it yet.

Some credit is due to everyone in the Cognition, Culture, and Context seminar class at Edinburgh University, with whom I discussed this yesterday. Any mistakes that I've made are my own.

• Lavelle, J.S. 2012. "Theory-Theory and the Direct Perception of Mental States." Review of Philosophy and Psychology 3: 212-30.
• Image sourced from  http://flickr.com/photos/crimfants/327861820/, published under Creative Commons Share Alike.