The Philosophy of Mind
In contemporary philosophy of mind, scholars discuss above all in order to know which psychological school would have correctly formulated the program of a study of the mental life (behaviorism, psychoanalysis, cognitivism). In France, the dominating view (phenomenology, Lacanian approach to the unconscious) remains hostile to the program of the “naturalization of the mind”. But yet these discussions presuppose that we have to maintain the traditional concept of the mental life. Certainly Wittgenstein has made a revolution in 20th century concerning our conception of the mental life: in his Philosophical Investigations he has promoted a grammatical clarification of our psychological concepts. He has shown that we cannot assimilate the mental life just to a flux of (either conscious and unconscious) representations.
Sergio Benvenuto: You work on the philosophy of mind. What do you mean by that and what is the difference, in your view, between esprit as understood in the French tradition and the concept of Mind in the Anglo-Saxon tradition?
Vincent Descombes: I work on the philosophy of mind in the contemporary sense, thus with a very explicit reference to the meaning that our English-speaking colleagues intend by “philosophy of mind”. However, I think that the philosophy of mind must return to some of the traditional problems regarding the mind and the mental that were developed before the discipline ‘philosophy of mind’ was established. It was only after the Second World War that people began speaking of the philosophy of mind. In order to understand what it’s about, one must situate it in the context of twentieth-century philosophy. The philosophy of mind, in fact, deals with a very specific problem, namely, how to make psychology into one of the natural sciences: a positive science like biology or chemistry.
However, one needs to understand that this philosophy of mind is just philosophy: it is not science even though it draws on science. This is why one needs to place the current debate over the philosophy of mind in a broader context, a context that witnessed other attempts to turn psychology into a science. In the inter-war period of the twentieth century, even though people spoke of providing psychology with a genuine, scientific aspect—many would have thought of psychoanalysis, of Freud’s theory of the unconscious—today’s philosophy of mind does not deal with phenomena that were of interest to Freud. Nevertheless, the project of making psychology scientific, that is, of making it into a science that’s solidly linked to the other natural sciences, is already found in Freud’s first work, titled “Project for a Scientific Psychology”.
Benvenuto: One has the impression that the French philosophic tradition of recent years gives a rather anti-naturalist (especially anti-biologist) reading of psychoanalysis and Freud’s work. There are those who think that psychoanalysis in France was led—whether by Ricoeur, Lacan, Lyotard, or Derrida—to a kind of spiritualism; that psychoanalysis took the place of Bergsonism. What is your opinion of this very French trend, and how is it connected to the general question of a philosophy of mind?
Descombes: That’s right. In France we have one of the largest psychoanalytic bodies in Europe, much larger than in other Western countries, at the same level as Argentina or Brazil. On the other hand, the reference by French psychoanalysts to science was not well supported. Or rather, if there was a reference to science, it wasn’t in the sense that Freud himself would have first understood it, namely, to a science of the brain, to neurology. For example, Lacan’s theory is well known for referring to the sciences, but they are primarily linguistics and anthropology, not to mention mathematical topology, and do not really include neurology.
Yet psychoanalysis, in Freud’s own project, had to participate in the integration of the mental to the natural domain. But this is precisely the recurrent project of the philosophy of mind in the twentieth century. So, if one wants to understand the contemporary meaning of ‘philosophy of mind’, one needs to consider why there is this recurrent project and why it must endlessly assume new forms. First, why must we worry about making psychology scientific? And second, why must one continuously change the way in which this project is accomplished? From this standpoint, one can examine what comes out of the comparison between the psychoanalytic venture with that of the philosophy of mind.
Interpretation, which was and remains dominant in France, is understood by psychoanalysis as a therapeutic, practical enterprise, and not as a positive science: the emphasis is placed on interpretation, on signification. Of course, this view does not entirely agree with Freud’s project for a scientific psychology, but it certainly agrees with what takes place in an analytic cure. As a result, French interpretation is not at all a betrayal, but rather a way of recognizing that the project for a scientific psychology could not serve as the foundation for a psychoanalytic talking cure.
There is no doubt that psychoanalysis was a decisive event for our conception of mind, from a purely philosophical point of view. From a cultural, historical point of view, from the point of view of ideas and of the present-day culture of human beings, psychoanalysis holds an important position in our modern view of the human being. Its vocabulary is often found in everyday language; its concepts have practically become common sense. But does this make for a great philosophical enterprise? The answer is not so certain since in order to make a decisive contribution to the philosophy of mind as such, it must propose a decisive concept of the mind. The problem that subsequent waves of philosophers have tackled is that our concept of mind prevents a reintegration of our mental life with reality: a reality defined in terms of what’s studied in the natural sciences. The problem for the philosophy of mind is knowing why the mental domain has the effect of being a separate sphere. From this point of view, psychoanalysis did not propose a new concept of mind. Even though it proposed any number of ideas that perhaps changed man’s own representation of himself and mores, it proposed neither a new concept of mind nor a new philosophy of mind. As concerns the concept of the mental from the philosophical point of view, the idea of the unconscious or of conflicts within a psychological subject are not revolutionary ideas, and are easily integrated into the definition traditionally given to mental activity.
Benvenuto: But there is also a definition of mind in the modern cognitivist sense, coming from Chomsky’s work, which is completely unlike the traditional one.
Descombes: In fact, therein lies the meaning of ‘philosophy of mind’. My colleagues in France who work on the philosophy of mind in this sense always emphasize that they understand it in the English sense, and not in the old, idealist or spiritualist sense found earlier among all European philosophers. That philosophy of mind, in the post-Hegelian sense, was a philosophy of the spiritual activities of human kind, and flourished in the philosophy of art and, more generally, the philosophy of culture. The modern philosophy of mind, instead, is above all a “philosophy of psychology”, to use Wittgenstein’s expression. Yet, to speak of psychology is not enough. One could say: the philosophy of mind is as old as the philosophers’ psychology; the Greeks already had a philosophy of mind, given that Plato and Aristotle had a philosophy of perception, of knowledge, of intelligence, of memory, and so on. That philosophers had a psychology is nothing new. The goal of a ‘philosophy of mind’ is not simply to do psychology, but to tackle the paradoxes arising from the dualism between mind and body, between the mental and physical. In the end, this is the only strictly philosophical problem. In the Anglo-Saxon definition, the philosophy of mind is above all a program that at times is called a cognitivist program. It aims at overcoming this dualism that has been carried over from classical philosophy and can be traced back to the work of Descartes. The latter based his entire philosophy on an opposition between the physical domain, that is, extended matter, and the mental or intellectual domain, the domain of the mind. For Descartes, just as the body can exist without a mind, so, too, can the mind exist without being the mind of a body. The question for a philosopher of mind today is, “What is your position regarding the dualism between matter and the mental?”
What is the dominant position of philosophers in France who are more or less my age? If we continue to talk about the philosophy of mind in the English sense, it’s because in France we have simply imported these debates of philosophers working in English. Many have adopted in one form or another the cognitivist program, hoping to skirt the conceptual problem of dualism by referring to a model taken from artificial intelligence (AI), that is, by referring above all to machines we build. These machines not only carry out calculations, but can even perform tasks that we can only describe in psychological terms: in terms of memory, reasoning. For example, there are machines that play chess. The basic, ingenious idea is that a machine, as everyone would agree, is nothing other than an assembly of parts, since we know perfectly well from which pieces it was made. This is why there is an exhaustive physical description of machine: it defines all the components and how they are assembled. This description is one an engineer would give.
Benvenuto: Thus, physical as intended by an engineer and not by a biologist. Don’t you think that cognitivism marks the abandonment of a physicalist view of the mind in favor of a mechanical one, which is just as bad? Is there perhaps a conflict between the idea that the human mind is a machine and the idea according to which it’s a brain?
Descombes: That’s the heart of the problem. Everyone speaks of physics, nature, the natural sciences; my colleagues talk about naturalizing the mind. But when we talk about nature, or physics, what nature are we talking about? In fact, philosophers readily give the label ‘nature’ to the object of hard physics, and hence of mechanics. If these philosophers talk about biology, they concede that it is also a natural science, but because he hopes that biology is in the end only a special science whose principles are, in the final analysis, those of a unique, fundamental physics. Cognitivism certainly wants to mechanize the mind, and any reference to biology is to neurology, the brain, and not to the organism in its natural environment. Returning to my logical machine, that is, an artificially made machine that allows me to play chess, for example, I have to choose one of two ways of describing it: either as a machine or as a player. There are, then, two descriptions, both correct, one which borrows its vocabulary from physics and the other from psychology and, more generally, from a vocabulary of human activities. The machine thinks, remembers its moves, prepares new moves, wants to win: one attributes to it a mental (intellectual) activity as if it really did all those things. So, if we can describe the computer’s operations in two vocabularies, why not say that the same holds for human beings? Although we didn’t make the human being, we can describe it in both the terminology of neurology (and ultimately in that of chemistry and hence physics), as well as in the terminology of psychology.
The cognitivist idea is that, thanks to the model of artificial intelligence, we can overcome the mental-physical dualism through the complementarity of the two descriptions. Both descriptions are true, even though they use vocabularies that cannot be reduced to one another. According to the philosophers who follow this program, the great move forward was abandoning the vain pursuit of a direct reduction of psychological concepts to physical ones, much like the reductions imagined in nineteenth-century materialism, as when Hyppolite Taine claimed that thought is only a product of the brain, just as bile is only a product of the liver. Our contemporaries have developed a much more subtle materialist philosophy since it seeks to maintain the mental whose description cannot be reduced to that of physical phenomena, while at the same time claiming that this mental is physical.
This is why one could summarize the entire cognitivist program in a single argument: that of maintaining the concept of the mental that we had up to now—namely, that the mind consists in having and combining internal representations—while at the same time rejecting the dualism that depends on this conception, which ultimately goes back to Descartes. According to the artificial intelligence model, we can consider the mental as being physical, and here we come to the problem you raised: where is biology in all of this? In other words, what physics is involved in the naturalizing program? Are we brains that function like logical machines, or are we living beings? For cognitivists, we’re mainly brains. In fact, according to them, the representations we have in our minds are also brain states. In a certain manner, cognitivists settle for a conception that remains Cartesian since they take the brain as a system that could be isolated.
Benvenuto: In that sense, do you see any important differences between cognitivism in the classical sense and connectionism which deals with neural networks?
Descombes: Cognitivism in the classical sense emphasizes the aspect of calculation and ends up taking the analogy with artificial intelligence almost literally. For its part, connectionism pays closer attention to the reality of the neurosciences. Yet both seem to me to be two variations of the same project, which is to retain our definition of the mental (a play of representations within a system) and find some kind of trick for saying that this mental is nonetheless part of a single, natural reality. This is why the cognitivist approach is not a profound revolution in the philosophy of mind. Just as the theory of the unconscious is certainly a cultural shift, though not a philosophical one, the cognitivist program is a way of saving the Cartesian concept of the mental by promising us that we can keep this concept without having to be spiritualists, without having to believe in a disembodied, immaterial mind. But the real revolution in the philosophy of mind would be to completely reform the concept of mind, that is, to tackle the Cartesian concept of mind in order to show its incoherence and to replace it with a more satisfying concept.
Benvenuto: There is a third possible variation of the philosophy of mind: the one that takes its cue from Wittgenstein and the philosophers inspired by him.
Descombes: That’s true. I think that the great, decisive event for the philosophy of mind and, what’s more, for the whole of philosophy, is Wittgenstein’s work. And that’s where we find ourselves divided today. In France, those of us who study analytic philosophy as it’s practiced in England or the United States are a minority, compared to the traditional way of doing philosophy, which I myself learned as a student.
Benvenuto: What is the traditional way that French students learn?
Descombes: It was a mix, a sort of balance, between phenomenology in Husserl’s sense (though with a more or less strong bent towards Heidegger), and German idealism, especially Hegel. We thought that contemporary philosophy was something that had to draw on the Hegelian dialectic, and later on Husserlian phenomenology, and finally on a phenomenology that was more existential than rationalist. A minority of my contemporaries, unhappy with this orientation, turned towards analytic philosophy. Some of them were seduced by the cognitivist program—and there is an entire school in France that carries out this program and that includes, for example, my excellent colleagues at the Institut Jean-Nicod in Paris: Dan Sperber, Pierre Jacob, François Recanati, Joëlle Proust and others. They don’t all hold the same position, but they would all undoubtedly say that the great event in philosophy is the cognitive revolution. On the other hand, there are other French philosophers who also want to work in an analytic manner, but who believe that the cognitivist program is regressive, since it seeks to save the Cartesian concept of mind even though it had been overturned by Wittgenstein. Those who think in this way are most often students of Jacques Bouveresse, and I include myself among their ranks. I think that the event is Wittgenstein’s posthomous book, Philosophical Investigations (Philosophische Untersuchungen). It’s odd, but there is no good translation of this book in French: there was a very poor translation, which has not yet been replaced, perhaps because we French philosophers find it difficult to accept the considerable impact of the latter Wittgenstein on the philosophy of mind.
I could perhaps characterize Wittgenstein’s work by going back to what I said earlier regarding psychoanalysis and cognitivism. I noted that, ultimately, the theory of the unconscious was not really a complete shift from the point of view of philosophy, even if it was a shock for Freud’s contemporaries. It wasn’t a revolution since the theory of the unconscious retains the idea that the mind is a sequence of representations, and it only introduces the idea that some of these representations do not reach the conscious since they are in some way held back. We have representations without knowing it, and the original Freudian idea was that they do not reach the conscious, not because they’re too small or too unimportant, but because they come into conflict with a kind of censorship carried out by the subject, or rather, that is carried out in him. But this means that psychoanalysis maintains the idea that the mind is a kind of inner world, a set of private representations that we grasp through an inward gaze—and as a result, the idea of inner meaning, of consciousness as perception of these mental states, is not at all brought into question. In contemporary cognitivism, we see exactly the same program reappear as was found in Freud, a sort of by-product of what was called ‘associationism’ in the nineteenth century, that is, the theory of the associations of ideas. What is the mind? It consists of mental processes. But what is a mental process? It’s a series of representations that are conceived as causing bodily movements. Basically, for the cognitivist the only serious problem that Descartes’ philosophy poses is that it doesn’t explain how the mind acts upon the body, and by replacing the Cartesian mind, which had the fault of being immaterial, with the brain, they could explain how the mind causes the body to move, since the brain clearly goes through a series of physical states. There’s no more need for the pineal gland! As they explain, we can re-describe the brain as a mind, just as we can describe the computer as a chess player, for example, or as a librarian, or archivist. So we now have the brain qua mind; for them the problem therefore had to do with the causal action of the mind on the body.
But the revolution that Wittgenstein brought to this tradition was questioning, by a very careful analysis, this concept of mind: the mind is not a process; it’s not composed of processes that the subject perceives within himself. We can show the importance of this change by returning to what you were saying: one talks of “a physical nature”, but is it this nature? Is it living nature, or is it the inert nature of machines, mechanisms?
What is interesting is that, if I describe someone who is playing chess, it doesn’t matter whether this player is human or a machine; it’s someone who’s playing chess. If I want to describe this player in psychological terms, complete with intentions, assessments, desires and ultimately emotions following a victory or loss, I must in any event describe the player as playing a game of chess! This player must therefore be part of an entire world, an entire context that will include other players, as well as the game of chess as a social activity having a history, a tradition, which is an institution. I must therefore place this player in an external world in which he can play. What the player needs in order to play is not at all an inward gaze upon his inner states; rather, he needs passions, a world in which one plays chess with partners or friends. The real problem is not knowing whether or not the chess player is made of silicone or metal or if it’s a human being. The real problem is: under what conditions can a system play chess? In order to describe the brain’s activity, it suffices to describe the internal states of a system; but in order to describe a chess player, I have to place the chess player in a setting that is outside of him and in which he poses his problems regarding chess.
Benvenuto: What is Wittgenstein’s concept of mind? And above all, what is the connection between this concept and what was later called the private language argument?
Descombes: What was first read and first commented upon in Philosophical Investigations is precisely what commentators—and not Wittgenstein himself—called the Private Language Argument. Understanding the impact of this argument for philosophy depends on seeing it as part of a broader endeavor: that of completely changing our conception of the mind. The private language argument looks like a reflection upon language in general and, more specifically, upon the language we use in order to talk about our sensations and our mental events. The idea that Wittgenstein criticizes throughout all his attacks against private language is that of the empiricists who say, “We know what a sensation is because we have sensations and we name them.” This is why, for the empiricists, the language of psychology was ultimately a private language. A private language means a language that I can create by myself, privately, on the basis of my own experiences, as if I could—all by myself, reduced to my subjective experience—turn towards my experiences and give a name to each of them (‘pain’, ‘hot’, ‘red’ etc.). Wittgenstein shows that no language can be made up in this manner since it is not possible to use a name, for example the name ‘red sensation’ or the name ‘headache’, on the basis of simple experience. I cannot use a name unless I have a rule that makes it possible for me to know whether or not this name is being used correctly. For example, if I call a table ‘chair’, I’m not using the word correctly; there’s a rule for the use of the word ‘table’. The same holds for all the names in our language. In essence, the Private Language Argument is the following question posed to the empiricist theory of the meaning of words: “How could someone, on the basis of his experience, establish the rules for using the words he wants to employ for designating his experiences?” This question seems to introduce a very specialized and limited debate on a rather dry topic: How is it that we understand the names we give to sensations? But what’s at issue is, in fact, much broader since these sensations, as seen from traditional philosophy of mind, are the least contestable lived experiences: the paradigmatic example of the mental is when I experience pain, for example. It goes without saying that when I experience pain it’s an event that is absolutely beyond doubt. I feel this pain and there can be no error about it; it’s something that affects me deep down. It’s why one has the impression of being at the very core of the subject, in an area in which one could establish a completely direct link between the words used and the experience.
However, Wittgenstein succeeded in showing that, in fact, no word takes its meaning from being associated with a private experience I feel inside; and the same goes for all the words of our language, whether they be names of external or mental objects (and, a fortiori, words that are not names). But if it doesn’t hold for sensations, then a fortiori it doesn’t hold for other psychological words; for example, memory or the intention of doing something. It’s not in having memories that I give a meaning to the word ‘to remember’; it’s not in having intentions that I learn the concept ‘intention’. Language is public. It’s in learning the language and being raised by others in a language that I possess a language for external objects and also for talking about myself and what happens to me, about what I think or feel, and about my intentions.
Benvenuto: But take, for example, someone who was born blind, and who later starts talking about colors, like red, green, black. He can talk about them because language—which uses color words—is always public. But don’t we think that, after all, the color words used by the blind person have a completely different meaning from those used by someone who sees? One suspects that these words don’t mean anything for the blind person.
Descombes: One could say that the question has an empirical answer, namely, that the person who was born blind, who has never seen colors, nevertheless understands the difference between one color and another: one need only explain it to him. And to explain it to him, one need only tell him that color is a means that allows us, for example, to distinguish between apples—some are red and we eat them, others are green and we don’t eat them. This explanation doesn’t provide him with the sensation of green and red, but it gives him the concept of green and red. Now, having a private language doesn’t consist in having this or that experience; it consists in having concepts that one could have derived without that experience. Likewise, although we are not cats or bats, and so do not have the perceptions of a cat or a bat, we can still understand quite well that a cat sees differently than we do; and we can conceptually approach to some extent how it sees differently, obviously without this providing us with a cat’s eyesight. Here, Wittgenstein makes a comparison: saying that understanding the word ‘red’ is having a red image in one’s head—which obviously assumes that one can see colors—is like saying that in order to understand what pain is I have to pinch myself; or that in order to understand the operation of subtraction one needs to have, for example, the experience of losing weight, of subtracting a part of one’s volume.
It’s thus mistaken to believe that the meaning of a word is constituted by a mental datum of the type “memory of a sensation”. The reason for this is always the same: even if I have a memory, this doesn’t give me the rule for its use. And moreover, I can very well use the word without having the memory. If the empiricist theory were true, I couldn’t talk of someone’s pain without feeling a kind of slight pain myself, since I would have to have—not that person’s pain, of course—but some kind of image or echo of that pain, a kind of attenuated pain. If I didn’t feel anything, then I really wouldn’t understand what it is for someone to be in pain. This theory is clearly absurd: when I see someone bent over suffering and I say, “He’s really hurting,” I’m not the one in pain, it’s the other person, and I don’t need to evoke for myself any painful experience in order to talk in a perfectly intelligible manner about his pain.
One may have the impression that this debate is highly technical, that it only concerns the foundations of language: where language comes from and how words receive their meaning. Yet, one needs to understand that this local debate in fact completely turns all of philosophy on its head, beginning with the philosophy of mind and followed by philosophy tout court, since the concept of mind can no longer be the same after what has been called the linguistic turn. The linguistic turn is understood by different philosophers in different ways: there’s the hermeneutic meaning of Gadamer; the communication theory of meaning of Habermas, etc. For Wittgenstein, it’s not at all a question of hermeneutics; it’s not a matter of referring back to a tradition from which our concepts come. Nor is it a turn in the sense used by Habermas, i.e., rationality in conversations. It has to do with a complete reform of the philosophers’ psychology.
Indeed, Wittgenstein shows us how, ultimately, all the theories we’ve inherited regarding the philosophy of mind—beginning with the eighteenth-century empiricists up to his contemporary philosophers (and after)—have presented things in the following way: in order to develop a psychological philosophy, a philosophy of mind, I need to focus on psychological phenomena, and should therefore look for these phenomena either in myself or in others. I need to observe something. If this something is myself, one then has the classic approach of introspection which immediately leads to private language, i.e., to that illusion that I can connect the word to something I experience in an Erlebnis, without having to connect it to something that I can show others. If I gave a meaning to the word ‘headache’ on the basis of the experience of a migraine, I couldn’t then tell you what I mean by it since I couldn’t show you my migraine as I feel it. That’s the approach of introspection. The other approach would be to observe psychological phenomena in others, but then one is led towards a form of behaviorism, a theory of observable behavior, in which case one would have to reduce the psychological to external activities.
Benvenuto: It’s almost become common place to say that the work of the latter Wittgenstein is a philosophical view that justifies behaviorism. Wittgenstein’s arguments are often compared to those of behaviorists like Gilbert Ryle, for example. Do you agree with this comparison?
Descombes: Ryle is, in fact, somewhat guilty of behaviorism and so also of reductionism. Roughly speaking, the project of behaviorism is to do away with all introspection, thus reducing everything that one sought to know by introspection—and which therefore was not scientific—to the observation of the external movements of individuals. There is thus a reduction of the complexity of the mental to the simplicity of external movements. Ryle’s position, however, is not strictly behaviorist, so that in condemning Ryle’s behaviorism one’s missing something. At times Ryle comes close to behaviorism, but his case isn’t clear.
Wittgenstein, on the other hand, is by no means a behaviorist. Let me give an example. There is a question that both Ryle and Wittgenstein ask, the classic question, What is thought? In a famous article Ryle asks, “What is thinking?” According to him, bad philosophy is the philosophy of introspection that says, “I’m going to try to think and then carefully watch what happens while I think,” which amounts to a psychology that derives a concept of thought from the careful observation of what happens in me when I decide to think. On this point, Wittgenstein and Ryle entirely agree, especially since Ryle was undoubtedly more or less influenced by what he had heard about Wittgenstein’s work. Ryle says, “But this isn’t how it ought to be done; one should instead observe Rodin’s statue The Thinker.” We’ll watch Rodin’s statue and wonder: But why is it a thinker, what is it that allows us to say he’s thinking? One clearly sees that Rodin’s “thinker” isn’t doing anything; he’s seated, he’s holding his chin, we can’t say much about what he’s doing. In this article on thought, Ryle is looking for an external way of answering the question, “What is thinking?”, knowing that the introspective approach has been abandoned. Ryle’s results are not entirely without interest, but in the case of Wittgenstein things are clearer. For him, it is by no means a question of observing anything whatsoever; hence, there are no phenomena to observe that might occur in me and of which I would be the only observer—the introspective approach. But nor are there behaviors outside of us to observe, and so no basis for saying, “That behavior is what we call ‘thinking’”! Wittgenstein says in fact that one needs to consider the verb ‘to think’ itself. And it’s here that we find the Wittgensteinian linguistic turn, which consists in substituting the question over the concept or essence of thought for the question concerning the verb ‘to think’. And one subsequently needs to ask about the grammar of this verb, that is, how the verb functions in different sentences that can be made about thought; how the verb is constructed in those sentences; what the rules are for using this word in different situations or “forms of life”. Thus, this linguistic turn has nothing to do with behaviorism, since behaviorism consisted in saying, “Since I can’t observe what’s inside, I’m going to observe what’s outside.” But for Wittgenstein, one needs to abandon the idea that philosophy must observe any “phenomenon” whatsoever: the philosopher must examine what he means when he uses the verb ‘to think’.
Benvenuto: Can we say that the internal-external difference, typical of the Cartesian distinction, is not outdated thanks to Wittgenstein?
Descombes: Yes, it is outdated in the sense that there is no longer only one and only one internal-external difference. It’s exactly like the difference between body and soul: is this difference outdated? For Wittgenstein, yes, but not at all for the cognitivists. For the latter, it’s been left behind while being retained. It’s retained by them when they claim that in each of us there is a mind that is at the same time a brain. Fine, but how can it be both? They no longer try to explain it directly by reducing thought to physical attributes! Can we say that my physical state appears to me as a thought (and not a brain state), while that of some other person (examined by way of a cerebral imaging technique) only appears to me as a physical state? For Wittgenstein, however, the internal-external opposition has meaning—indeed, several meanings—in different contexts, but it is also no great mystery. For example, there’s no mystery in saying that when I have a thought, I don’t necessarily express it, and so can keep it for myself. If this is all one means in talking about having one’s thoughts being “internal”, that’s fine; there’s no mystery since this thought that I keep for myself is just a thought I don’t want to express, which means that I could express it. There is nothing private about this thought inside me. And as for the “external”, it’s quite simply what one expresses in language, or manifests in various expressions or actions. There is no mystery here either. On the other hand, the idea of something internal that is both very important and completely inexpressible because private, disappears altogether; and with it disappears the idea of an external that’s completely foreign to the vocabulary of psychology since it would consist in the purely physical movements studied by behaviorism. So, when Wittgenstein talks of the external, he does not indicate by that the movements of behaviorists; he’s quite simply referring to human actions, that is to say, actions whose meaning we understand when we see them. For example, I see someone open a door; I understand that he’s about to leave. This is not a behaviorist movement; it is not a purely physical movement; it’s an intelligible action.
In the end, is there a concept of mind that comes out of all these considerations? Yes, a concept does emerge; it’s the concept of an individual’s ability to guide himself in the world, or to conduct himself in a meaningful way. It’s thus a concept that takes someone in their activity within a context. To give an example: if we go back to thought, what do we discover if we examine more closely the grammar of the verb ‘to think’? We find that there are in fact many verbs ‘to think’. Instead of having a single verb ‘to think’ in our languages, we could have several. Grammatically speaking, thought can be expressed verbally or even adverbially. And that’s an idea that Ryle also had, which is why he isn’t fully a behaviorist. With Ryle, there’s a more subtle aspect that’s clearly due to his solid training in Greek philosophy; in his case there is an entire background that prevents him from holding such an unsophisticated position as is attributed to him. In his analysis of Rodin’s Thinker, Ryle eventually provides a conception of thought according to which it has a mode of grammatical expression similar to adverbs. The difference is the following: if I speak of thought in the sense of, “I’m thinking about someone” or, “I’m engaged in a problem,” then we’re dealing with a thought whose mode of grammatical expression is verbal since it’s an activity at issue. You’ll ask, What is it that shows it’s an activity? The answer is in the very way in which this verb is used in saying, for example, “I’m busy thinking, and if I’m mulling over a problem and the telephone rings, it bothers me; I’m interrupted.” Whence a simple grammatical rule for verbs of activity: if I can be interrupted in what the verb attributes to me, it means that I had an activity. It’s impossible to be interrupted if one isn’t engaged in an activity. Let’s compare this with our use of the verb ‘to understand’. The verb ‘to understand’, in the traditional philosophy of mind, was also considered as a verb of activity: “I understand by having an image that I associate with the words I hear.” Wittgenstein showed why understanding wasn’t and couldn’t be an activity. Every activity is capable of being interrupted. Consider this as it applies to the verb ‘to understand’: I can’t be interrupted in my understanding, which means that I can’t be busy understanding, or that I can’t be ordered to understand. Clearly, someone can order me to “think about the problem,” but I can’t be given the order to “understand the problem.” It follows that ‘to understand’ does not designate an activity; rather, ‘to understand’ is a verb indicating a capability, a skill that allows us to do certain things. Here, then, is a first grammatical form of our concept of thought.
But there is another form: the adverbial. In some notes written while he was trying to clarify his views on the “philosophy of psychology”, Wittgenstein considers craftsmen. He asks, “Can one say that a carpenter thinks? What must one understand by the word ‘thought’ in order to say that the carpenter thinks? Must he have in mind sentences or representations?” That’s not necessary, of course; one need only be able to say that the carpenter is focused on his work. If we see him working and we say, “He’s working carefully, he did that deliberately,” or, “He did that by mistake,” then at that point we apply adverbs of thought to him, since the meaning of “carefully” is that the craftsman thinks about what he’s doing. For the craftsman, thinking about what he’s doing is not an activity he would do in addition to making a piece of furniture. Not at all! In this example, thought is the way in which the craftsman works. When the activity of thought consists in paying full attention to the task before us, thought is adverbial.
In general, the concept of mind that we can now formulate is much more varied than the traditional concept, which had the great defect of being uniform. The traditional concept is that of a system in which there are representations. The image classically given to it is a river. The mind is a flow of representations or experiences, a sequence of mental events following one after the other. And so all that is mental has to fit this mold: memory must be a representation; intention must be a representation; understanding must be a representation, etc. Whence the interest in Wittgenstein. He reminds us that a good part of the vocabulary of the mental designates capabilities, and this is what lies at the heart of his critique of private language. ‘To understand’ or ‘to mean’—the two basic verbs for a theory of language—are not activities. When someone talks in order to say something, he does not engage in a first activity consisting of “coming up with words” and then in another activity of voicing them which would be to mean something with those words. Instead, ‘to mean’ is the way in which I pronounce words. ‘To understand’ is not an activity that would accompany the activity of listening to someone speak: ‘to understand’ is my status vis-à-vis what the other is saying. I have this status provided I can explain what I claim to have understood, provided I can talk about it meaningfully, react in a relevant manner. Conversely, if I don’t understand, then it’s like being in front of a strange block; I don’t know what was said.
Benvenuto: Can we then say that, for Wittgenstein, the subject is basically a capability? And if that’s the case, does it resemble the ancient Greek, Aristotelian concept of ‘potentiality’ as opposed to ‘actuality’?
Descombes: I’d like to avoid answering in too brief a manner, since every question regarding the relation between Wittgenstein and the Greek philosophers is difficult, insofar as his style is not to evoke an entire library of references, so much so that even today we don’t know what he did or didn’t read. We know that he carefully read Russell and William James, that he read some of Saint Augustine’s works, but he certainly didn’t have a clear idea of Aristotle. Thus, from an exegetical point of view, it’s always difficult to make comparisons. However, the problem that needs to be raised is completely different. It’s a purely philosophical problem: Today, in order to benefit in an interesting way from what we’ve learned from Wittgenstein, must we draw on notions such as Aristotle’s distinction between ‘potentiality’ and ‘actuality’? I think that the answer is ‘yes’, and what’s more, I myself am highly favorable to an approach often found among English philosophers—and also among American philosophers, like Hilary Putnam. The English philosophers include people like Elizabeth Anscombe, Peter Geach, Anthony Kenny, and David Wiggins. According to them, Wittgenstein helps us, first of all, better understand and subsequently better see the relevance of several Aristotelian ideas, considered outmoded since Descartes, including the idea of potentiality. But this doesn’t mean that we have to go back to Aristotle; it’s not a matter of restoring a doctrine to what it once was, and clearly what we need to draw from Wittgenstein is not a literal reading of the doctrine of power. If one followed in this direction, one would see that there are a number of relations between the potential and the actual. For example, if I take understanding, I have to explain it as being more of a capability than an activity; hence, it’s on the side of potentiality. But for Aristotle—or at least in the scholastic tradition—every time we postulate a power, we must designate an act. A power is known by an act: tell me what the act is and who performs the act, and I’ll tell you what the power exerted is and who possesses it. If someone is walking, it means he has the capability of walking. But in the case of the mental, there is not this direct relationship, this kind of strict correlation between a power and an act, since the capability consisting in understanding can be manifested through any number of acts. The same would hold for belief: belief is clearly a kind of disposition, but there is no one kind of behavior that could be identified as the act corresponding to this power to believe.
Another point needs to be made: Wittgenstein would certainly warn us against a materialist conception of power, one consisting in taking ‘power’ as a kind ofhidden entity, a kind of disguised mechanism within the subject. What’s more, he was quite skeptical about many of the theories of power, and with good reason, since when people talk of dispositions, they often mean instead devices [dispositifs], a kind of mechanism one might find in a system that’s only waiting to be triggered. But, of course, dispositions are by no means devices: a device is a real arrangement of real entities; although it may be the material basis of a disposition, it is not the disposition itself.
In the end, Wittgenstein’s ideas show us another possibility for naturalizing the mind—viz., that of cognitivism—as I mentioned above in referring to artificial intelligence and connectionism. What struck me was the position of the Harvard philosopher Hilary Putnam. He is to some extent the reluctant inventor of an entire philosophical current that’s part of the cognitivist program, which is called functionalism. Functionalism is the idea that, if I study a system from the point of view of its functions—that is say, what it can do, the role it plays in a given context—then I don’t need to know how the system is materially made. What I must understand, on the other hand, is the organization or structure that allows it to fulfill this function. And so, according to Putnam, if one considers someone who, for example, develops artificial intelligence software, there comes a moment in which he’s not concerned with what material the program will be made of, but rather with the functions that the program must fulfill. Putnam introduced this idea that was taken up again by cognitivists such as Fodor who have made it the basis of their work. Putnam later realized that this functionalist idea was essentially Aristotelian, since it amounts to the distinction between an explanation of something in terms of its matter (what it’s made of), and an explanation of it in terms of its organization, its structure, and hence its form. Whence the conclusion that Putnam draws: there’s an interpretation of functionalism that’s in fact a reinstatement of the concept of form. This means that in cognitivism there are ultimately two possible schools of thought. This first, largely dominant one, is the school according to which science will soon be able to provide a mechanistic explanation of mind, either by directly following the model of the computer, or by appealing to the brain sciences as do the connectionists. But another school of thought is also possible—one that takes up the functionalist idea as reinterpreted by Putnam after having rejected his earlier position, which was still very mechanistic—which consists in saying, “Nature exists where there are forms that provide systems with functions, functions that are performed outside of the system.”
And here we arrive at what is, perhaps, really at stake in this debate: What do we mean when we talk about naturalizing the mind? Does it mean mechanizing it, or does it mean considering that the mind only exists where there are living beings. If a living being is at least a necessary condition for the existence of the mind, then one must consider an organism in its living environment, after which one can later make a distinction between an animal psychology and an intellectual psychology. But one must first have the system, i.e. the organism, which functions in an ecological environment, or in other words, which carries out its activities in its external environment by way of its structure: what Aristotle called its form.