The mind is what the brain does?

I recently ran across this post by neurologist Dr. Steve Novella over at his blog Neurologica. The post is basically a rebuttal to neurosurgeon Michael Egnor’s views regarding how memory is encoded in the brain; however, my agenda here is not to comment on this exchange. Rather, this current post of mine was sparked by comments made by Dr. Novella in his post regarding the nature of the mind. Here’s some of what he had to say:

As I have pointed out numerous times myself – mental phenomena are functional active things. They are based in the physical substance of the brain, but they are not just the physical substance – they are what results from the function of the physical substance. The mind is what the brain does.

I’ve heard this theory of mind put forward many times; that is, the theory that the mind just is what the brain does. Note that this is not the same as saying that the mind is identical to the brain, rather it simply says that psychological states just are the products of neurophysiological processes. This theory is simply a type of identity theory. Identity theory is a type of reductive materialism and is defined, by the Stanford Encyclopedia of Philosophy, as the theory that “states and processes of the mind are identical to states and processes of the brain.” And this seems to be exactly what Dr. Novella’s position is, for he says that “mental phenomena (including memory) do not simply correlate with the firing of neurons in the brain – they are the firing of neurons in the brain.”

The first step that most materialists take towards justifying identity theory—or really any variation of a materialist philosophy of mind—is to provide evidence from neuroscience that demonstrates that the brain and mind are intimately related. And needless to say this evidence is not lacking at all. We know quite well that altering brain chemistry and brain structure can subsequently alter personality, can create a “divide” in consciousness, and can even cause individuals to fail to recognize familiar objects or even loved ones. There is no doubt that this says much about the correlation between the mind and the brain, but does is say any more than that by itself? Not really. Philosopher of mind James Madden, in his book Mind, Matter and Nature, articulates:

Has neuroscience revealed an empirical identity between psychological states and brain states? It clearly has not, nor does it seem that it ever could. What neuroscience does is to provide us with ever more precise descriptions of the goings-on in our brains that are correlated with our psychological states[…] Correlation is a far cry from identity; two phenomena can be perfectly correlated, without being identical. (p. 108)

The point then is that identity theory cannot be justified solely by the correlation between the mental and the neurological. As is commonly said, correlation does not imply causation. Now I maintain that not only can identity theory not be justified by pure findings from neuroscience, but I maintain that it is rife with ontological problems regarding actually accounting for how it is even possible that psychological states just are, and are reducible to, brain processes.

Intentionality and Necessary conditions

One of the first problems that identity theory runs into is accounting for intentionality. In philosophy of mind, intentionality refers to the ability of something to be about, or refer to, or represent something beyond itself. When I think of my wife, my thought is about, and refers to her. Now the problem here is that physical things only exhibit derived or secondary intentionality. The written word “cat” only refers to a cat if a mind so interprets it that way. It could just as easily have been the case that we endowed the word and utterance “cat” with a completely different meaning, or that another group of individuals takes the word or utterance “cat” to mean love. This means that physical symbols only have extrinsic intentionality and derive their intentionality from a mind that already exists. Now note that psychological states that exhibit intentionality do not borrow this intentionality from somewhere else. That is to say, our thoughts are not intentional because something external to us interprets them, rather they exhibit intrinsic intentionality, and thus our mental states constitute intentionality that is completely opposite the physical.

Now this posits problems for any philosophy of mind that is primarily materialistic. For how can a physical process be that which grounds intentionality if the physical first needs a mind in order to have intentionality at all? The materialist, then, needs a mind in the first place before he can start talking of intentionality being present in the physical. But the mind is the very thing the materialist is attempting to account for!

Furthermore, things get even worse for the identity theorist than this. Let us imagine a scenario where I raise my arm to a certain trajectory T1. Now let us say that I intended to raise my arm for some specific reason—perhaps to greet a co-worker. Now an identity theorist would say that my act of raising my arm can be accounted for in specifically neurophysiological events which terminated in certain muscular events which caused the raising of my arm. We can call this physical sequence that leads to the raising of my arm X1. Note that X1, by itself, is a sufficient condition for the raising of my arm to trajectory T1. So, the raising of my arm is therefore accounted for in completely materialist terms right? Not necessarily. For now imagine that instead of the physical sequence X1 causing T1, we say that this physical sequence is altered by only one neuron to cause a slightly different trajectory in my arm raising. So we can call this different physical sequence X2, and this new trajectory T2. So now we can say that X2 is a sufficient condition for the raising of my arm to trajectory T2.

But this is where we run into problems. While X1 might be sufficient for raising my arm to T1, and X2 is sufficient for raising my arm to T2, X1 and X2 are not necessary conditions for raising my arm as an intentional act. That is, in order to raise my arm as an intentional act, I do not need X1 or X2, but could have some other physical sequence obtain. But this means that no specific physical act is necessary for explaining the intentional act of raising my arm! Rather, while physical processes are sufficient for explaining my arm being raised in a certain specific manner, they are not sufficient for explaining my arm raising as an intentional act. Now note that if psychological states just are neurophysiological processes (i.e. if the mind is what the brain does), then certain physical sequences would be sufficient and necessary conditions for raising my arm intentionally. And since this is not the case then it cannot be true that our psychological states just are processes in the brain.

Rationality

A second problem that confronts identity theory is that of rationality. When certain neurophysiological states in the brain are followed by other subsequent neurophysiological states, neurologists understand exactly how these states follow from one another. If S1 (a neurophysiological process) is followed by S2, which is then followed by S3, it is the electrochemical properties of the brain, ultimately grounded in physical laws, that cause this transition. Now imagine that I entertain the following line of thought: 1) John is usually late to class on Mondays, 2) It’s Monday, 3) Therefore, John will most likely be late to class. Now notice that, on identity theory, each thought (each psychological state) will correspond to a certain brain state (neurophysiological process). But this means that if identity theory is true, then the cause of me inferring 3) from 1) and 2) is simply the electrochemical properties associated with those brain states.

This obviously runs into problems. For my act of inferring 3) was due to the semantics associated with 1) and 2), and I simply saw logically that 3) follows. And, in fact, for any inference or conclusion to be considered rational, it needs to be caused by the meaning and semantics of previous thoughts, or propositions. But, again, this is not true on identity theory. On identity theory it is not the semantics of 1) and 2) that cause the mental state of 3), rather what causes 3) is simply the blind electrochemical properties of the neurophysiological state associated with 2), which was caused in the same way by 1). But this type of causation is purely blind deterministic efficient causation. That is, it is simply matter dancing to the tune of physical laws, and therefore cannot be called rational in any sense. Thus stated, on identity theory any type of “rational inference” is, upon closer inspection, only pseudo-rationality, and therefore rationality as such is only an illusion.

Note that these are not the only devastating critiques of identity theory (there are others that I've put forth here and here). But I believe they go a long way in demonstrating that this reductive materialist theory of mind—that the mind is what the brain does—is vacuous. Now let it also be noted that the arguments above are not soul-of-the-gaps arguments. That is, the arguments are not saying “this or that aspect of cognition is difficult to explain currently, so we need to posit the soul or some Cartesian substance to account for them.” Rather the above arguments are ontological and deductive. That is, they don’t argue from current scientific ignorance, but instead argue that the ontology of a materialist theory of mind cannot ever in principle solve these problems. These problems then are a sort of reductio ad absurdum of materialism, and thus findings in neuroscience are irrelevant here.

In conclusion, then, we see that identity theory rests on a conflation of correlation with causation, and faces insurmountable arguments that demonstrate that the mind cannot simply be “what the mind does.”

Is the brain a computer? Part IV

Let us conclude our series regarding the theory of computationalism. The last few posts (part I, II and III) have, at least to my thoughts, hurled some pretty devastating objections towards the view that our minds are simply some type of software program run on the hardware of the brain—different in degree but not in kind from a digital computer. These aforementioned objections have demonstrated that computers, even in principle, are simply unable to produce any of the defining characteristics that make up human mental experience. Yet, there is still one more aspect of human cognition that computationalism falls entirely short of accounting for: intentionality.

In philosophy of mind, intentionality refers to the ability of something to be about, or refer to, or represent something beyond itself. For instance, my writing this post exhibits a multitude of intentionality: I am thinking about the post and about the words I type; I am referring to many concepts beyond myself (e.g. intentionality); I have a desire to write this post etc. Seeing as how all cognition consists in a subject entertaining the presence of an object, and therefore of the subject having thoughts about said object, it should be apparent that intentionality is present in every act of cognition. In fact, intentionality has such an intimate correlation with the mental that Franz Brentano—the pioneer of addressing intentionality as such an important philosophical topic—labeled intentionality the “mark of the mental”. Now, if intentionality is such a crucial aspect of what constitutes the mental, then surely any worthwhile theory promulgated in the philosophy of mind should be able to, at the very least, account for such a phenomenon. So the question becomes whether or not computationalism can account for “the mark of the mental” (i.e. intentionality).

It might seem that computationalism can indeed account for intentionality. For computers seem to exhibit all types of intentionality. A computer might generate a picture of a lake, and surely the computer is representing something beyond itself, namely, a lake. Or, a computer could produce a game of chess, and surely the computer is now about something, namely, chess. Of course we could think of hundreds of more examples where a computer seems to demonstrate the aspects that make up intentionality. But, do these instances really exhibit intentionality on the part of a computer? No, they do not; at least, not the kind of intentionality that the mental exhibits.

For instance, picture a piece of paper with the word “bunny” written on it. We obviously know what the paper is referring to, namely, a bunny. Yet, does the paper really have intentionality as we would characterize it? Surely the paper is referring to something beyond itself, which is exactly what constitutes the intentional. However, I maintain that upon closer examination it can be demonstrated that no such intentionality is really going on. To illuminate this, let’s imagine that all humans on the face of the earth were suddenly to go extinct. Would the paper still refer to a bunny? No, the word “bunny” on the paper is now just a meaningless set of ink blots on a paper. The term “bunny” is a term that us humans invented to represent an animal. It is we who grounded the term “bunny” with its meaning and, therefore, its intentionality. Without human minds the term “bunny” is absolutely vacuous with regards to meaning.

However, it is still true that while the human mind grounds the semantics of the term “bunny”, the term does in fact still refer beyond itself. But, this intentionality is only valid if it derives its meaning from a mind—again, the term is meaningless apart from a mind that imparts such meaning to it. Therefore, the intentionality is not intrinsic to the paper or the term. What we have here is what John Searle calls “derived intentionality”.  And if we look around the world at inanimate objects that seem to exhibit intentionality, we will see that all forms of such intentionality are derived. The painting of the Mona Lisa only derives its representation, of the lady that was painted, from the painter. The number “2” that represents the mathematical concept of two only represents such a concept because we endowed it with such semantics. Thus, we see that any intentionality not constitutive of a mind is only derived intentionality; only the mind has intrinsic intentionality.

The points made above should demonstrate the obvious ramifications for computationalism. Computers only exhibit derived intentionality; any intentionality present in a computer is imparted to it by the programmers and users. The only reason a computer produces a game of chess is because the programmers have put “produce a game of chess” into its programs; moreover, a game of chess is only a game of chess relative to chess players. But this provides an insurmountable problem for computationalism. If computers can only exhibit derived intentionality, then they surely cannot produce any kind of a mind. For minds, by their very nature, exhibit intrinsic intentionality. So, a mind would already have to exist for a computer to even attempt to produce a mind. The computationalist, once again, has to presuppose what he’s trying to account for! Oh, what a tangled web we weave.

Once again, we see that computationalism is unable to account for the very characteristics that make a mind what it is, and therefore it is an unsuccessful theory in the philosophy of mind.

Lastly, as I similarly articulated in the post on rationality, the dead-end that computationalism runs into when trying to ground intentionality is no different than the dead-end that any naturalist account of the mind runs into as well. If the physical can only exhibit derived intentionality, then it cannot account for intrinsic intentionality; and therefore it cannot account for the existence of the mind.