The advent of techniques like PET scans and functional MRI has enabled researchers to observe the brain in action with a precision that is unprecedented. One of the interesting aspects of these studies is that we can now actually perform a limited version of what might be called mind reading: identifying what's going on in the brain without having the owner of said brain describe it. In the latest development in the field of neuroimaging, researchers have watched the brain of someone watching an image, and were actually able to perform reasonable reconstructions of the image.
Pretty impressive. But does it amount to a kind of “mindreading,” as the author says? Does it show (what the author does not say, but which many readers will no doubt infer) that the having of a mental image can be identified with a certain kind of brain process? Not so fast. Here is a good example of how empirical discoveries which might seem to provide answers to philosophical questions actually presuppose such answers.
Notice first that no one who used fancy scientific instruments to observe the image on someone’s retina would regard this as an instance of “mindreading,” even though such an observation would under normal circumstances allow the observer to infer what sort of visual experience the subject was having. So why should observing some brain process associated with a certain sort of visual experience count as “mindreading”? The answer, of course, is that researchers assume there to be a connection between mental events and neural events that is more direct and intimate than that which exists between (say) mental events and events occurring within the eye.
Now that assumption may be correct – and I think it is in fact correct – but it is an assumption rather than something observed in the data, and more to the point it is an assumption with philosophical, and not just empirical, content. Recall my earlier post about Karl Popper’s critique of causal theories of the mind: Before we can identify any causal relation between the brain and the external world as having any sort of explanatory force vis-à-vis the mind, we have to be able to identify some particular external event as “the beginning” of the relevant causal chain, and some event within the brain as “the end.” Yet, as Popper argues, there is nothing in the bare empirical facts that can justify such identifications, nothing that determines this particular point as a “beginning” or that particular point as an “end.” (In the case at hand, there is nothing in the bare empirical facts that determines that it is such-and-such brain processes, rather than the image on the retina, or rather than some different event altogether, that counts as the terminus of the relevant causal chain.) Such identifications are interest-relative and thus non-objective – or at any rate they are, I would add, if we assume a “mechanistic” conception of the natural world.
To avoid the conclusion that they are interest-relative and non-objective (and the idealism or anti-realism this would seem to imply), we have to return to the Aristotelian idea that natural objects and processes have essences, that these essences entail inherent powers, and that these powers are defined by ends or goals toward which the objects or processes “point” as a final cause. In the case at hand, we have to assume (among other things) that the perceptual process inherently “points to” and thus of its nature terminates in, not the generation of a retinal image, not some set of neural events further down the line from the ones the researchers cited in the article were focusing on, but rather in those specific neural processes themselves. Here as elsewhere (as I argue at length in The Last Superstition), despite the “mechanistic” conception of nature officially and unreflectively endorsed by most scientists, the actual practice of empirical science, and certainly the intelligibility of its results, presupposes something like Aristotelian essentialism.
Secondly, it is also worth emphasizing that the researchers in question have (quite obviously) not literally been looking at any subject’s mental images or sensations. They have instead merely inferred from certain brain processes that the image must have such-and-such a character. Hence they have not discovered anything that need trouble any Cartesian dualist, since such dualists would (going back to Descartes himself) happily concede, and indeed emphasize, that there are neural processes causally correlated with various mental events, and insist only that the correlation does not and cannot entail that the relevant mental events are either identical with or metaphysically supervenient upon any neural events.
In any event, from an Aristotelian-Thomistic point of view, results of the sort cited in the article are not only no evidence against dualism, but indeed just the sort of thing we should expect. For unlike Cartesians, Aristotelians and Thomists regard sensation and imagination as entirely material processes, not immaterial ones. Hence even an outright identification of (and not just correlation between) the having of a perceptual experience or mental image and a certain neural event would be no evidence at all against the claim that the mind is immaterial. For the defining aspect of the mind is intellect, and intellect is irreducible to sensation, imagination, or indeed any other material process.
The main reason has to do with the differences between the objects of intellect on the one hand, and the objects of sensation and imagination on the other. For one thing, sensations and mental images are always concrete and particular, while the concepts grasped by the intellect are abstract and universal. A given mental image of a man, for example, is always going to have features that apply at most to many men, but never to men universally; it will be of either a fat or a thin man, of a bald man or a man with hair, of a tall man or a short man, etc., and thus be limited in its application in a way that the universal idea of man is not. Secondly, images are always vague and indistinct when their objects are more complex, while the ideas grasped by the intellect are clear and distinct however complex their objects. To take a famous example from Descartes, no mental image you can form of a chiliagon (a thousand-sided polygon) is clearly different from your mental image of a 997-sided figure or a 1001-sided figure, but the concept of a chiliagon grasped by your intellect is clearly different from the concept of a 997-sided figure or a 1001-sided figure. Thirdly, there are things we simply cannot form a mental image of which the intellect can nevertheless grasp the idea of: abstractions like economics, law, or love; temporal relationships; logical relationships like entailment, conjunction, disjunction, and negation; and so on.
Since the objects of the intellect differ in kind from mental images and sensations, it follows that to detect neural patterns of the sort described in the article, whether or not it amounts to “reading” what someone is sensing or imagining, does not and cannot amount to “reading” what is going on in their intellects, and thus does not and cannot amount to reading their thoughts, if we confine “thought” to what the intellect does when it makes judgments and inferences, which involve the grasp of concepts.
Might the detection of some other kind of neural pattern amount to “reading” someone’s thoughts? No, for (among other things) the reasons outlined in my series of posts on short arguments for dualism. In particular (as I argued here), given a mechanistic (i.e. final causality-denying) conception of the material world, any material process must be devoid of intentionality. But thoughts are inherently intentional. Hence nothing detectable in any purely material processes (again, where “material” is understood in mechanistic terms) could possibly reveal the content of any thought.
Now if we reject a mechanistic conception of the material world and acknowledge the existence of final causes, then a kind of intentionality does become detectable within the material world after all. But now another consideration comes into play. For (as I argued here) the specific kind of intentionality involved in conceptual thought still cannot be accounted for in material terms, because material processes are always inherently indeterminate while at least some of our thoughts are not. Conceptual thought, the characteristic activity of the intellect, is (unlike sensation and imagination) thus essentially immaterial. Hence its presence can never even in principle be detected merely by examining someone’s brain.
(As indicated in that earlier post, this is, as James Ross has pointed out, the upshot of arguments like Quine’s famous argument concerning the inscrutability of reference and Kripke’s “quus” argument, though materialist writers like Quine conclude, not that thoughts are therefore immaterial, but rather – and incoherently – that none of our thoughts has any determinate meaning. I say “incoherently” because – again, as Ross points out and as I said in the earlier post – to deny that we have any thoughts with determinate meaning is in effect to deny among other things that we ever reason in accordance with valid forms of inference, which undermines any argument anyone, including Quine himself, has ever given.)
This leaves it open that, at least given certain background assumptions, we might guess with some measure of probability what someone is thinking. Indeed, we can do that already, just by observing a person’s behavior and interpreting it in light of what we know about him in particular, his circumstances, human nature in general, and so forth. And of course, further knowledge of the brain might give us even further, and more refined, resources for making inferences of this sort. But what it cannot do even in principle is fix a single determinate interpretation of those thoughts, or reduce them entirely to neural activity. So, no entirely empirical methods could, even in principle, allow us to “read” someone’s thoughts in anything more than the loose and familiar sense in which we can already do so.