Friday, 29 February 2008

Making up the Mind 10: Sharing Minds

Chapter 7 of Chris Frith's latest book Making up the Mind: How the brain creates our mental world starts with a consideration of 'The Problem of Translation'. The hoary problem of other minds threatens to rear its ugly head almost immediately. That problem, to recap for the non-philosophers, is the epistemological difficulty one arrives at if one starts from the metaphysical supposition that (Frith, p. 163) "each mind is a private place" accessible only to the person to whom it belongs. How then can we ever know what is going on in the minds of others? Even if they tell us, how would we know that their words meant the same as ours? And so on. (The answer - or better, the way to dissolve rather than solve the problem - as has been sketched out now countless times by the existential phenomenologists and Wittgensteinians, but perhaps most clearly by Anita Avramides, is that 'the mind' is not 'a private place', and that we can directly see or hear the thoughts and intentions avowed in one another's discourse, or the sensations and feelings expressed directly in our vital expressions, as these occur in their natural contexts.)

In line with the traditional epistemological problem (but with apparent unawareness of its several proposed dissolutions), Frith opines (p. 165):

I have in my mind some idea I want to communicate with you. I do this by turning my meaning into spoken words. You hear my words and turn them back into an idea in your mind. But how can you ever know that the idea in your mind is the same as the idea in my mind? There is no way you can get into my mind and compare the ideas directly. Communication is impossible. ... And yet even at this moment we are having this vigorous interchange about the problem of meaning. Our brains have solved this impossible problem about communication.

The answer to this putative problem emerges over the next few pages: We, or rather our brains, make 'guesses' or 'prejudge' what is in people's minds. We, or rather our brains (the two are used fairly interchangably in the text), make a guess, then make a prediction about what will come next if our guess is right, and then refine our guesses according to such guesses. The problem of course is that (p. 168) 'I predict what you are going to do on the basis of what I would do if I were in the same situation. So if you are different from me, my prediction may be wrong.'

Let's pause and take stock. First the extraordinary phenomenology of what it is to speak meaningfully must be noted. For the truth, obvious as soon as we stop for even a second to consider it, is that when I want to tell you something, I precisely don't have to 'convert a (wordless?) idea in my mind into words', 'transmit' these to you. Nor do you have to then 'decode' them, through guessing, back into their original ideational form. Nothing in my experience suggests to me that I have wordless ideas. Nothing in my experience provides evidence of a translating process. This just appears to be the old empiricist 'theory of ideas' in cognitive neuroscientific garb.

Second, what evidence is there that either we or the brain is involved in making predictions as to what someone will do next? The entire discussion makes it seem as if we must posit unconscious or preconscious processes of guessing, Bayesian probabilistic reasoning going on in the brain, etc. What this (mereologically fallacious) idea overlooks is that there is no more reason to suppose that the brain is involved in such 'reasoning' than there is to suppose that, say, a falling body is busy unconsciously calculating where it should next move, according to the laws proposed by Newton or Einstein, before it moves there. As disengaged scientists we may be in the business of making predictions. As engaged cognisers, we simply have a range of expectations (not predictions!) about what people will say, expectations which are phenomenologically manifest only in terms of an absence of surprise when what happens is what is expected. When what is expected doesn't happen, then perhaps some reasoning is required of us. But to go round making predictions about what our friends do and mean would be not simply unnecessary, but also somewhat, well, rude.

Why has Frith become embroiled in this epistemological problematic? The obvious answer is that he seems to take it that what is proximally available in experience is not the actions of the other, but rather only their movements - or, not what the other says, but rather only certain word sounds. Because of this, we are forced to guess as to the underlying intentions which cause these movements or talking. Hence (on p. 166) the putative problem of discerning the meaning of movements or talk is described as akin to an engineering problem. This is what is called the inverse problem, which is the question as to what are the causal forces that must have been applied to an object such as a mechanical arm in order for it reach a desired location. (i.e. it is the inverse of the forward problem - which is the question as to where (e.g.) an arm will go given the forces we know are to be applied to it). The conflation of the intentional and the causal orders is complete.

Here, then, is how the conceptual confusions endemic within the picture of the mind on offer create the epistemological pseudo-problematic driving the supposedly empirical problem of how we understand one another. First we must imagine that we are fundamentally estranged from ourselves. As such, as a kind of inner homunculus, if I wish to act or speak then I must convert the meanings in my own private mind space into actions or words, and I do this by causally impacting on my own body. Intentions then become inner causes of actions. Naturally, since this conception of mind is to govern not just my own case but that of the others who I encounter, I must engage in a similar decoding process in order to reach even a semblance of understanding them. Accordingly I unconsciously reason my way from surface body movements to the underlying inner causes, making best guesses and eliminating hypotheses that aren't confirmed by later observations. Or perhaps it is my brain that does this reasoning for me - the mereologically fallacious attribution of psychological functions to the brain supporting the naturalistic story about the need for the brain to solve what are essentially psychological or epistemological problems for us.

That the same estranged conception of our relation to the world is also at work in Frith's conception of our relations to one another is evident from the following (p. 170):

Remember, there is nothing special about the problem of minds. When I look at a tree in the garden, I don't have the tree in my mind. What I have in my mind is a model (or representation) of the tree constructed by my brain. This model is built up through a series of guesses and predictions. In the same way, when I am trying to tell you something, I can't have your idea in my mind, but my brain, again through guesses and predictions, can construct a model (a representation) of your idea in my mind. Now I have two things in my mind: (1) my idea and (2) my model of your idea. I can compare them directly. If they are similar, then I have probably communicated my idea to you successfully. If they are different, then I certainly haven't.

By this stage in the book Frith is not even attempting to supply empirical evidence for his principal assertions (the scientific evidence is given in the rather nice appendix called 'The Evidence', pp. 212-215). We are told some interesting facts such as the following: Children as young as three years have formed expectations about the kinds of toys boys and girls will, respectively, play with. That people who are presented with words associated with old age in a psychology experiment will tend to leave the building more slowly afterwards. (This is evidence of the 'contagious' quality of other people's states of mind.) That mothers speaking 'motherese' to their children accentuate the differences in the sounds of vowels far more than when speaking in a similarly high-pitched and babyish way to their pets. That gorillas have a special way of folding nettles so that they don't get stung in the mouth when they eat them - and that young gorillas can learn this trick by observing the adults. That autism and paranoid schizophrenia are in a sense opposites and in a sense very similar: in the first meaning is seen nowhere, in the latter everywhere; yet both result in experiencing a world not shared by others. None of these substantiate the underlying framework idea, the idea that our brains construct and reconstruct models for us of the minds of others. It is not even clear what evidence could substantiate this.

Here is how the book ends (excepting the Epilogue): (p. 183):

By making models of the minds of others (in the same way that it makes models of the physical world), my brain enables me to enter a shared mental world. By sharing my mental world with others, I can also learn from their experiences and adopt the models of others that are better than my own. From this process, truth and progress can emerge, but so can deception and mass delusions.

This comes just after a description of the Jonestown massacre and the delusional beliefs of Jim Jones. But in what way has Frith really explained how this sharing of true and false ideas is possible? All we have been told are the ubsubstantiated claims that the brain makes 'models' or 'representations', that it makes 'predictions'. Along with the supposition that we are somehow located on the receiving end of these. Yet when we think about it, it becomes unclear what there really is that needs explaining. If we were in the perilous epistemological predicament that Frith implies - having to inspect mere behaviour and make inferences, create models or representations, and so on - then we would need all the help from a 'mereologically' boosted brain that we could get. And more, as it happens, since within this framework it would be entirely unclear (as Wittgenstein argued) that we could ever really comprehend what it would even be for someone else to have an idea, and certainly unclear what would license the belief that it is the possession of particular ideas by others which putatively causes the movements of their bodies. Put aside the hoary problem of other minds and the conceptual confusions implicit within it, and we emerge once again out in the open. Now we can understand our friends without having to make models of their putative inner goings on. Their meanings, feelings, and states of mind are once again immanent within their gestures which themselves are no longer mere movements but are rather replete with living intentionality. And we too are no longer on the receiving end of messages from our brains, but are rather active agents in the world which we directly perceive, our understanding of which and of the fellow inhabitants of which is constituted by the embodied dispositions and expectations causally inscribed on our brain through our years of interactive experience.