I think that the topic of my talk can be best expressed by the following question: ‘Can we understand the human mind by studying the human brain?’ It is easy to notice that an affirmative answer to this question cannot be given on a priori grounds. We have no difficulty imagining an object floating in midair. In the same way, there is nothing logically or semantically absurd about the picture of a person thinking (or performing some other cognitive task) without using her brain. So if we wish to give an affirmative answer to our question, perhaps we must find some empirical grounds for it. What could count as empirical evidence in this case? I assume that such evidence must be accounted for in terms of a statement relating some observable processes or states of our brain to some of our mental states, processes or acts that we can also observe.
This assumption leaves out of the picture the accounts that relate our neurophysiological processes with our behaviour. Informative as they can be, they are nevertheless irrelevant to our problem.
The assumption might also seem to be exclusive with respect to the reductionist theories of mind. Well, it is not. I take it for granted that there are empirical facts related to the functioning of our mind that we have access to. Some usual examples are: visual, tactile, hearing and other sorts of sensations, feeling pain, dizziness, cold, hunger, fear and so on. To this we can add having representations, imagining things, remembering something, being in a certain mental state, like the state of recognizing something or the state of being surprised by something and so forth. I will provisionally admit, contrary to the later Wittgenstein’s private language argument, that we can make accurate observations of these states and report our observations correctly. Nevertheless, reporting that I am in pain does not amount to claiming that my pain actually is some sort of mental entity, which, in some way or another, does exist. One rudimentary example for the kind of statement we might be looking for is this. Suppose that by simultaneously observing my visual sensations and my brain I succeed to relate each point in my visual field to a point in my brain. This will perhaps enable me to say that by studying that part of my brain I will get a better understanding of the mental act of seeing. I do not need to assume, of course, that when I speak of my visual sensations or of the mental act of seeing I effectively refer to any mental entity. I could very well assume that the empirical facts that I experience inwardly either supervene on or are in other way reducible to the neurophysiological processes in my brain. One could as well leave aside any considerations about what is real and what is not and just try to look for the correlations1.
If we agree to speak of mental acts in an ordinary, unbiased manner, we might accept that there are mental acts that we can explain by studying our brain. However, we might still be wondering whether all our mental acts are in the same position.
In what follows, I want to offer some support for the thesis that there are certain (cognitively relevant) mental acts which we are in principle unable to understand by studying the functioning of our brain. In order to do this, I will focus on one mental act only, namely the act of linguistic understanding.
A preliminary distinction might be useful at this point. There are at least two different ways in which we speak of linguistic understanding. In one way, we speak of understanding a word as of an ability that we posses2. We learn to understand the word “addition”, for instance, and keep understanding it after that, even when we do not perform any additions. In this sense, understanding a word is not a mental state, which we can be in for a certain amount of time. On the other hand, we speak of understanding a word or a sentence on hearing or reading it. We can understand the word “addition” in the first sense, but still not understand the phrase “addition of geometrical surfaces” on hearing it3. For the purpose of my presentation, I want to speak of linguistic understanding in the second sense only. It is this sort of understanding that I will refer to as ‘the mental act of understanding’ from now on.
The point I wish to make is this. It is in principle impossible to explain the mental act of understanding by speaking of neurophysiological processes in our brain. In order to show this I will expand on an example given by Ludwig Wittgenstein in Blue Book pp. 11-13. Wittgenstein tries to imagine an experiment by which we might prove that “certain physiological processes correspond to our thoughts”[BB 11]. The experiment “consists in looking at the brain while the subject thinks. And now you may think that the reason why my explanation is going to go wrong is that, of course, the experimenter gets the thoughts of the subject only indirectly by being told them, the subject expressing them in some way or the other. But I will remove this difficulty by assuming that the subject is at the same time the experimenter, who is looking at his own brain, say by means of a mirror.”[BB 12] My version of this experiment is going to be the following. The subject is observing her own brain (or perhaps she records the empirical data about the functioning of her brain and consults them after the experiment) while she is reading different sentences, one at a time, some of which she understands and some of which she does not. At the same time, she is also observing her inner mental states and processes. We might perhaps imagine that she pauses after reading each sentence and makes a note, reporting her inner mental states as accurately as possible.
One thing must be noted here. The subject, who is also the experimenter, may prepare the list of sentences, being aware that she does not understand some of them. Nevertheless, the fact that she is reading a sentence that she usually understands is irrelevant for her experiment. In other words, this does not count as an empirical fact when she tries to establish some correlations between the mental act of understanding and the processes in her brain. The reason for this is obvious. Reading a sentence that you usually understand does not guarantee you that you have understood the sentence that you read on that particular occasion.
Now, what the experimenter must do is to identify the inner experience that she could call ‘the experience of understanding a sentence on reading it’. She might try to do this by finding the sufficient and necessary conditions for ‘I understand (that) p on reading “p”’. Let us look at some candidates for these conditions:
(1) I am representing R to myself.
(2) I have the feeling of saying p inwardly.
(3) I have the feeling that I recognize p.
(4) I have a particular feeling of understanding4.
With respect to (1)-(4), it seems that there are quite convincing arguments to be found in the writings of the later Wittgenstein, to the effect that they are at most necessary but not sufficient for the truth of ‘‘I understand (that) p on reading “p”’. As a matter of fact, Wittgenstein has also argued that (1) and (4) are not even necessary for understanding p. Perhaps the same can be shown to be true for (3) and even for (2). However, I do not want to re-enact Wittgenstein’s arguments here5.
Another possible candidate is:
(5) I have the thought expressed by p as a result of reading “p”.
One way to dismiss it would be to ask whether having a certain thought could be experienced by using our inner sense. Actually, according to Wittgenstein’s original example, this cannot be the case. But even if we accept that it could, there is still another problem. How could we distinguish between the case when we accidentally have the thought expressed by p while reading “p”, which we do not understand, and the case when we have the thought that p because we were reading “p”? We do not experience causal relations. Besides, what happens when, after reading and understanding ‘You shall not lie!’, the first thought which I notice to come to my mind is ‘Oh, dear, I am going to hell.’?6
To put it simply, I take it to be Wittgenstein’s claim that all the inner experiences that our subject-experimenter can observe can only accompany the act of understanding. The act of understanding as such does not consist in any of them. Neither does it consist of their simultaneous presence or some combination of them. The more general thesis is perhaps that understanding just cannot be a certain observable inner state. I think that this thesis can be given some support directly. Suppose that we could find out, by noticing that we are in the particular state of understanding, that we grasped the meaning of a particular sentence on reading it. Were this the case, we could never believe that we understood something and still not understand it. But there are surely some situations when we wrongly think that we understand a sentence. The sentence might even be nonsensical. Let us look at the following example. The question ‘How many dreams are in this room?’ seems at first to make sense. On a closer look, we will conclude that the sentence is actually meaningless. It does not make sense to ask of a non-physical object whether it is contained in a room. But if understanding were something like a state that we could have direct access to by introspection, we would recognize the absence of that state instantly, on hearing the sentence. In this case, we should have known that the sentence is meaningless from the very beginning. This argument is based on the observation that we actually distinguish between ‘understanding something’ and ‘believing that you understand something’ and consider that the last phrase has full meaning7.
One could agree to this and still hold that the phrase ‘understanding a sentence on reading or hearing it’ refers to something, namely either to a behaviour disposition, or to a disposition to form some observable mental state. With respect to behaviour dispositions, for instance, the discussion could develop along the following lines, with the following candidates being proposed and rejected one by one: the disposition to give an affirmative honest answer to the question “Did you understand the previous sentence?” is not specific to the understanding of a particular sentence; a monolingual speaker could understand sentences from his own native language without having the disposition to translate them into another language, when asked to do so; the disposition to utter a sentence that has the same meaning with the one you have read or heard, when asked to do so, is irrelevant if we take for granted Quine’s thesis that there is no perfect synonymy between any two linguistic expressions, since there is no point in repeating the same sentence and there no point in saying something else either. A similar discussion could perhaps develop with respect to dispositions to form observable mental states.
However, the bottom line is that when we speak of dispositions we actually speak of causal relations between brain processes and either behavior or mental states that are not in themselves understanding. So why would we believe that by doing this we explain understanding at all?
A last resort might be to use the phrase ‘mental act of understanding’ in a way similar to the way we use theoretical terms. But in this case the correlations we will formulate can only count as explaining the empirically observable neurophysiological processes by making appeal to the mental act of understanding and not the other way round.
To conclude, there seems to me to be reasonable arguments for the claim that understanding a sentence on hearing or reading it, regarded as a mental act, cannot be explained by studying our brain. Wittgenstein’s suggestion regarding this, if I understood him correctly, would perhaps be that we give up talking of understanding as of a mental act altogether. To this I agree. The more important thing seems to be that we must determine the extent to which a recycled version of the above argument can be used with the aim to establish similar conclusions with respect to other cognitively relevant mental acts.
1 This attitude is similar to Fine’s NOA (here I suppose that you do not interpret Fine as Musgrave does).
2 This is perhaps the same with Ryle’s ‘knowing how’.
3 We might even be unable to understand “the addition of 2 and 4” if we are not paying attention to the speaker or if we are in some other way distracted.
4 A feeling which we usually express by saying ‘I see’, or ‘I get it’, or ‘Right’, ‘Indeed’ a. s. o.
5 See, for instance, H. Putnam, Mind, language and Reality, Philosophical Papers, vol. 2, Cambridge, Cabridge University Press, 1979, pp. 1-32.
6 Wittgenstein’s example with the order to represent a red patch to yourself could be perhaps used here as well. See Blue Book, p. 4, 6.
7 Using Wittgenstein’s vocabulary, we could say that ‘understanding a sentence on hearing or reading it’ has a different grammar from ‘feeling (or experiencing) something inwardly’.