Saturday, 20 December 2008

Making up the Mind

And here's an early version of a review recently written, pulling together some of my bad-tempered Wittgensteinian criticisms, in earlier blog posts, of Chris Frith's latest book...

Imagine this: waking one morning we discover that it is not we who are in direct contact with the bed; not we who are open to our lover’s caress; the movements our bodies make are not really our movements. Such privileges of direct expressive and receptive contact with our world and companions have, we discern, been afforded not to us, but—to our brains! We must rest content, on the side of action, with mere illusions of free will and, on the side of perception, with inspecting mere models which present themselves as reality but which are really just illusions thrown together by the brain.

The consolation prize is that, were we actually in direct contact with the world, the task of making sense of its complexities would just be overwhelming. So thankfully our clever brains perform these tasks “off-stage,” supplying us with outputs in the form of simple “pictures” or “messages” clear or intelligible enough for us poor cognitive beings to grasp. As Frith says in the conclusion of Making Up the Mind: How the Brain Creates Our Mental World — a well-written and accessible book which notwithstanding fully embraces and endorses the above-described theorization of, and some might say nightmarish predicament for, the self, mind and body — all “this complex activity is hidden from us. So there is no need to be embarrassed. Just go back to the party and have fun” (p. 193). Whether this is consolation enough may be questioned. As Malcolm (1986) once wrote regarding Searle’s notion that he was the brain stuck inside his own skull: “Searle says that we can receive messages. But in that predicament, who wants messages?” (p. 186).

Unlike Searle the philosopher, Frith the neuroscientist aims to substantiate his claims not with conceptual argumentation but with empirical evidence drawn from cognitive neuropsychology. In his own words, here are the key theses Frith takes the neuropsychological evidence to support: The “distinction between the mental and the physical is … an illusion created by the brain” (p. 17). “By hiding from us all the unconscious inferences that it makes, our brain creates the illusion[s] that we have direct contact with objects in the physical world [and that] our own mental world is isolated and private” (p. 17). These unconscious “inferences can be wrong,” even in “an ordinary, healthy brain” (p. 60). Furthermore, we have no “direct contact… even with our own bodies”; this is another “illusion” created by the brain (p. 81). Our “perception of the world is a fantasy that coincides with reality” (p. 111) arising when “our brains discover what is out there in the world by constructing models and making predictions” (p. 138).

Our knowledge of the “minds of others” is created by our brains “in the same way” (p. 159). And whilst “we experience ourselves as agents with minds of our own,” this too is an “illusion created by our brains” (p. 184). Frith acknowledges that our experience of freedom, individuality and responsibility is a cornerstone of societal stability and morality, but this is simply the “final illusion created by our brains” (p. 193).

Such claims are prima facie extraordinary, and if the neuropsychological data Frith presents could substantiate just one of them, his book might cause a major revision in human self-understanding. Yet what struck this reviewer again and again was the way in which the content of the hypotheses these data supposedly evidenced, and the theoretical unity of the text, derived principally from unargued and tacit metapsychological commitments which radically constrained the way the data were interpreted.

Concerning perception, Frith cites three sorts of evidence for his claim that “even if all our senses are intact and our brain is functioning normally, [the feeling that] we have direct access [to the physical world] is an illusion created by our brain” (p. 40). First, in chapter 2, he provides evidence from various malfunctions of, and curiosities regarding, visual experience – change blindness, subliminal perception, visual illusions, synesthesia, dreams, visual hallucinations, etc. Second, in chapter 4, he cites the fact that there is no direct mapping to be had of sensory (e.g. retinal) stimulation onto the contents of consciousness. Third, in chapter 5, he notes that we are normally unaware of the vast amount of complex neurophysiological processing (the activation of motor programs, say) that subtends everyday experience, and infers that “my perception [cannot be] of the world, but of my brain’s model of the world” (p. 132).

Whilst the data are fascinating, they are also incapable of motivating Frith’s theoretical claims, which instead appear to be consistently driven by a ‘homuncular’ conception of the self constantly invoked in the data’s interpretation. By ‘homuncular’ I mean a conception of the subject’s relation to its brain which harnesses a) a mentalistic conception of the immediate contents of perceptual consciousness as ‘inner images’ or ‘internal representations’ occurring ‘in our minds’ to b) a causalist construal of such immediate contents as the final products, delivered to the mind, by a CNS which has worked over information originally received by the sense organs (Kenny, 1984). (Conceptions of consciousness as an inner stage (or ‘Cartesian theatre’; Dennett, 1991) populated by inner visibilia may not explicitly posit an actual homunculus as an audience. The philosophical concern is however not ontological but methodological (Kenny, 1984): that theories deploying the conception do not neglect to demonstrate how, rather than simply assert that, they do not reduplicate the very phenomenon (perceptual consciousness) they aim to explain.)

The following are representative examples taken from Making up the Mind. Sense organs are said to work ‘just like a video recorder [transmitting] information about the physical world … to our minds’ (p. 21). The brain is described as “showing us false information” (p. 49); as not “telling us everything it knows” (p. 42); as “not simply transmit[ting] knowledge to us like a passive TV set … [but as] actively creating pictures of the world…from the very limited and imperfect signals provided by the senses.” (p. 85). My “brain manages to create for me the experience of a constant, unchanging world through which I move” (p. 110). It also “constructs models” (p. 138) of both the physical and the “mental worlds” of others (p. 159).

Accordingly, when dreams or illusions are offered (in ch. 2) as evidence that we have no direct visual access to the world, the conception which constrains the interpretation of the data already presupposes that, if we are not witnessing the world accurately, then we must (with some kind of further and as-yet-unexplained perceptual system) be accurately witnessing inaccurate mental images of the world. Or when (in ch. 4) the facts that retinal images are inverted or two-dimensional or duplicated are cited – or when movements of these images are as it were ambiguous between movements of the perceived objects and movements of the eye or head – it is simply presupposed that, since perception is construed as input to consciousness, the work of the visual system must be understood as one of ‘undoing’ the infelicities introduced at the sensory surfaces. Or when it is pointed out (in ch. 5) that the vast complexities of the CNS’s information processing are completely unknown to us, the inference is straightway drawn that therefore what we are aware of must be neither the world around us, nor our neurological processes, but their supposed illusory upshots.

Perhaps I should confess that I am convinced that what Kenny calls the “homunculus fallacy” is indeed a fallacy, and that Dennett is right to deconstruct the “Cartesian theatre”. Whilst in confessional mode I might also relate that Frith’s description of the mere brain as engaged in personal-level activities (knowing, believing, interpreting, deploying Bayesian inferences, etc.) strikes me as implicating him in another (‘mereological’) fallacy – that of ascribing to a part what can only coherently be ascribed to the whole (Bennett & Hacker, 2003).1 Yet my intent is not to foist my Wittgensteinian sensibilities onto the reader, but merely to relate that Frith’s striking theses regarding the allegedly illusory nature of our experience of the world are quite simply not a function of the data he presents, but rather of the homuncular framework used to interpret them – whatever we make of that framework. Perhaps it is a set of harmless metaphors – and if so this may also be the best way to take Frith’s theories.

Similar presentations of interesting data recruited by tacitly homuncular theorizations of the self arise throughout the book, whether we are considering perception (ch. 1, 2, 5), interpersonal understanding (ch. 6, 7), planning (ch. 4), or action (ch. 3, 6). For example, chapter 6 relates that an alleged everyday “experience of agency”— of being in control of our actions, making decisions to act, and acting on these decisions — is actually an illusion created by the brain. In truth, we are told, the brain distinguishes between intentional and non-intentional movement by measuring sensorimotor timing differences. These differential responses to the timings of causes and effects in perception and action are, it is said, translated for us into experiences of agency, providing an illusion of free will.

The experimental data (pp. 151-155) are again fascinating. But it is instructive that Frith appears to take his phenomenology of intentional action from cases such as (that which he quotes:) Ian McEwan’s marvelous description, in his novel Atonement, of Briony’s contemplation of her relation to her moving body:

She raised one hand and flexed its fingers and wondered… how this thing, this machine for gripping, this fleshy spider on the end of her arm, came to be hers, entirely at her command. Or did it have some little life of its own? She bent her finger and straightened it. The mystery was in the instance before it moved, the dividing moment between not moving and moving, when her intention took effect. It was like a wave breaking. If she could only find herself at the crest, she thought, she might find the secret of herself, that part of her that was really in charge.

What Frith seems to miss is that such descriptions are precisely not of everyday intentional action — but rather of an extremely alienated state of mind. Briony has dissociated from her lived bodily experience, becoming a disembodied homuncular spectator consciousness experiencing the body as merely a distant mechanism or “fleshy spider.” Our actual everyday experience of agency is rather characterized by the immanence of intention in action. Accordingly, the striking conclusion Frith draws – that the timing experiments reveal a genuine aspect of our self-conception to be illusory – is misplaced, for the conception of agency on offer here is drawn not from everyday experience but from an alienated theorization of it presupposed by his interpretation of the data.

The very idea that we have control over our actions is taken by Frith, in a curious Epilogue, to entail that there is supposed to be an inner homunculus enjoying a direct causal impact on a merely mechanical outer body:

For me it seems as if I am fully in control of my actions. This is why it is so hard to get rid of the idea of a homunculus. It is the dominant part of my experience that I am in control. … This is the brain’s final illusion: to hide all those ties to the physical and social world and create an [illusion of an] autonomous self. (pp. 188-9)

But what I suggest is the real reason for Frith’s struggle to rid himself of the homunculus is the unacknowledged homuncular conception of the relation between subject and body constantly inscribed within his theories.

Early in the book Frith tells us that he is “not a philosopher”, that he does “not expect to persuade people of truth by the power of argument”, and that the “only arguments [he] accepts] come from practical experiments” (p. 15). What Making up the Mind reveals, however, is one of the principal risks of eschewing philosophical reflection: that one’s theories will then be even more driven, and potentially vitiated, by tacit philosophical commitments which no amount of experimental data can evidence, challenge or extirpate.


  • Bennett, M. R. & Hacker, P. M. S. (2003). Philosophical foundations of neuroscience. Oxford: Blackwell.
  • Dennett, D. (1991). Consciousness explained. London: Penguin.
  • Kenny, A. (1984). The legacy of Wittgenstein. Oxford: Blackwell.
  • Malcolm, N. (1986). Nothing is hidden: Wittgenstein’s criticism of his early thought. Oxford: Basil Blackwell.

1 Both Kenny (1984) and Bennett & Hacker (2003) run together two conceptually distinct alleged ‘fallacies’: the ‘homunculus fallacy’ of tacitly reduplicating our relation to perceptibilia on a mental stage, and the ‘mereological fallacy’ of ascribing psychological properties to the brain.