The Conscious Mind

Interview Transcripts

John Searle

Interview 4/19/06 by John Rieger

Part One

Philosophers certainly make the subject of consciousness complicated. Is there a nice simple place to get started?

Sure. Well, consciousness was not considered a respectable subject for most of the 20th century, because there didn’t seem to be any way you could get at it with the techniques of science or of logical analysis. I mean it just seemed to airy-fairy touchy-feely. Well, that has changed. Now people understand that consciousness is real, it has this remarkable set of properties, and I guess if we’re gonna nail it down immediately, let’s start with those properties.

The first one is that consciousness is always subjective. That means it exists only when it’s experienced by a human or an animal subject, and for that reason I like to say it has a first-person ontology. There always has to be some I, some person, some animal who’s having it.

Second, for that reason, there’s always a qualitative feel to every conscious state. Every conscious state feels like something or other. So it feels different, it’s just a different quality from drinking wine to listening to late Beethoven quartets, and those are different from thinking about a mathematical problem.

And then the third remarkable property of consciousness is that it always comes to us in a unified form. So I don’t just hear the sound of my voice and feel the pressure of the shirt on my neck, but I have them both as part of one great big conscious field.

And then we could go on all day listing these features, but I’ll mention just one more, and that’s intentionality. That’s a funny word we use from the German, but it just means any state that is about something or refers to something. So beliefs and desires and hopes and fears, as well as intentions and perceptions and voluntary actions all have this feature of intentionality.

Intentionality is “aboutness”?

Intentionality is just “aboutness.” The mind is always about something, and often those mental states are conscious. So those are four features of consciousness.

Now as somebody who once had to explain to his parents what they were paying for me to study here in this fine institution, I can’t stay away from the question, “Who cares?” Why is consciousness important? What does it matter?

Well, there are some people who care and other people who don’t, and I would say if you don’t care don’t waste your time on it, but I care desperately about it because I think that’s what life’s about. I think it is the single most important feature of our life, and if you think about it, what would it be like to have your life only as a total zombie, without any consciousness whatever, if you went through life like a stone? So there’s a funny thing about consciousness, if somebody says to me, what’s the importance of consciousness, I would say as a start, the importance of anything else depends on consciousness, because something is important only relative to a conscious agent. However, that doesn’t mean it’s important to study it. A lot of people are bored with studying it, no reason why they should. I’m not trying to force it on anybody.

I really liked a sentence in the beginning of your recent book where you said that all the most important and influential theories in the philosophy of mind were wrong. Y

Yeah, well that’s right. I discovered this when I started working in the philosophy of mind. It was quite different from the philosophy of language. For one thing, the intellectual level was much lower. But secondly, the standard views I thought were without exception mistaken, so I tried to present a non-standard view that would at any rate be less mistaken than the others.

Now what are they and why are they mistaken? Well, mostly they’re mistaken because they accept a certain set of categories which are inappropriate. And the source of the categories comes from dualism, the idea that reality divides into two different kinds of things—the mind and the body, the mental and the physical, the spiritual and the flesh—and that has led to all these horrible problems. So dualism makes it impossible to give a coherent account of the world by saying, well, really we live in two different worlds.

Now the standard reaction to dualism in the late 20th century was something called materialism, and the problem with materialism is it’s just as crazy as dualism, because the materialists denied the existence of the mental as something ineliminable and non-reducible. They thought, well, consciousness as some kind of first person qualitative thing: that can’t exist. Science has to be about things that have an objective existence like tectonic plates and molecules and planets and galaxies, so all this touchy-feely stuff has to be gotten rid of.

Now I try to point out, they inherit the worst mistake of the dualists. They inherit the mistake of thinking that consciousness is not part of the ordinary world that we all live in, and I emphasize that it’s biological. Consciousness is a biological function of the brain in the same way that digestion is a biological function of the stomach and the digestive tract. There’s nothing mysterious about it at all. It is unusual, in that it has this special qualitative character that I was describing earlier, but that’s just a fact of nature. It’s a fact about human evolution that we have become endowed with a capacity to have conscious experiences, and that I think is simply wonderful, not something to be denied or lamented.

Everybody seems to bring their own methodological, or epistemological or metaphysical commitments to the subject of consciousness. They all have an ax to grind, and consciousness seems to be the number one grindstone. Maybe you could give me a few of the great names?

Ok, let me tell you the names and numbers of some of the players. Dualism really goes back to the ancient Greeks, and certainly to Plato. For the modern era, dualism in the form that it’s come down to us was invented by René Descartes in the 17th century. And Descartes said there are really two kinds of things in the universe, mental substances and physical substances, and you’d be surprised how powerful that view is to this day.

Now dualism was rejected by a whole lot of people, and there is no single materialist who stands out, but there is a materialist tradition that says, no, there is no such thing as the mind as something distinct from or over and above the body, rather there is just matter, and everything is a feature of matter. And that led in the 20th century to the view that, well, the way to study the mind is just to study behavior, and this view is called behaviorism: all there is to the mind is behavior. The great names in behaviorism are, well, I guess you’d say Hempel, Karl Gustav Hempel, known to his friends as Peter Hempel, was one of the founding fathers of this view. Gilbert Ryle, though his view is not exactly behaviorism is standardly characterized as a behaviorist. So behaviorism did thrive.

But then that was replaced by something called The Identity Theory, and that was the idea that well, really mental states just are states of the brain. There’s nothing else going on in there except neuron firings, and that in the modern form was started by two Australian philosophers, Jack Smart and U. T. Place, and that is now still an influential view, but there’s something grossly implausible about both these views for any normal human being.

Was that motivated in some way by logical positivism and the attempt to make sense of our reports of inner states in logical positivistic terms?

That’s absolutely right, and if I were gonna go into more detail on this I’d have to go into more detail on the history of positivism, but you’re absolutely right, the positivists thought that the meaning of a statement is given by it’s method of verification. So if I make a statement about your mind, the only way I can verify it is by studying your behavior, and then they thought, well, that’s all there is then to the meaning of the statement, is a set of statements about how you would behave.

So then, here again is an epistemological commitment that flavors their approach to the mind?

Absolutely, one of the greatest curses of the history of western philosophy is the constant confusion between epistemology and ontology, between how we know and what it is that we know when we do know, and I’ve always insisted the question “How do we find out whether somebody is in pain?” is not the same question as the question, “What is it to be in pain?” We typically find out from their behavior, but what it is to be in pain is to have one of these qualitative subjective states, an unpleasant state. So Positivism was one of the leading forces behind the materialist rejection of dualism, and the form that it took in classical positivism was to claim well really there’s nothing to mind but behavior. Any statement about the mind can be translated into a set of statements just about behavior.

Now as I said, that was replaced by the identity theory, sometimes called physicalism, that said, no, mental states just are identical with brain states. But now then those guys have got a problem, well, what is it about this brain state that makes it mental, and makes that other brain state next door to it not mental. And the answer to that—and I’m giving you a very truncated version of this history—the answer to that was something called functionalism. You can identify the mental by it’s causal relations. So if I believe that it’s raining and I don’t want to get wet, then I’ll carry an umbrella. So my belief and desire are identified with their functional or causal role. Now that led to what everybody thought was the most wonderful theory of all: what plays this functional role? It’s a computer program…

Listen to the audio >>

top

Part Two

There was this wonderful period—I hope now it’s mercifully past because it’s ridiculously false—but there was a period when people were very enthusiastic about the idea, “The brain is just a digital computer. The mind is to the brain as the program is to the hardware.” This mantra was repeated in any number of textbooks and university courses—I like it cause you can refute it in a couple of minutes, and if you want me to I will. But that was a sort of high water mark of this kind of materialism. It was called computer functionalism, or I baptized it “strong artificial intelligence,” where the programmer doesn’t just simulate a mind. AI people have told me, the programmer sitting at his console typing out the program is creating a mind, a mind is being created right here in front of our eyes. I’ve heard otherwise sane people say this kind of nonsense to me, so it was not, it’s not just an isolated occurrence, there’s a whole movement of people who believe the computational theory of the mind. I think it’s mostly in retreat now as we learn more about brain science, and as it’s been refuted.

Well, I’ve never understood—that whole theory never seemed intuitive to me. I didn’t even understand what intuition that came from.

Ok, there were two intuitions, I think. One is, and I think this was the most powerful—what we want to know about the mind is what does it do? Well, what it does is process information. Well, now, wait a second, it’s not the only thing in the world that processes information. There’s this whole technology of information processing, and it has a marvelous type of machine, called a computer. So the natural thing is to think that, well the mind just is a computer program, and the brain just is the hardware of a computer. That was one intuition. Let’s call that the information processing intuition.

Another one is, when we don’t understand something, we tend to model it on something we think we do understand. And if you look at the whole history of this going back to the Greeks, people have always thought the brain must work like what they thought was the latest technology. So there are Greek philosophers who thought, well, the brain is like a catapult. And in the early 20th century, I think it was Sherrington who said, well, no, the brain is a Jacquard loom. And in my childhood, when the latest technology was the crossbar telephone system, I was told by any number of people of great seriousness, the brain is a telephone crossbar system—what else could it be? And so on, this kind of nonsense went on right up to the later decades of the 20th century, and the latest technology was obviously the digital computer.

Now one of the beautiful things about this conception—and I have to tell you I think it’s ridiculously false—but one of the beautiful things about it was, it made it easier for the cognitive scientists to investigate the mind, because you don’t have to know how the brain works. You see, we don’t know how the brain works. That’s very, very hard. But if the brain is just the hardware in which the real mind is implemented, if it’s just the physical substructure to what really counts, then we can understand the mind just by writing the right programs, by getting the programs which are equivalent to the programs running in the mind, and you can forget about the hardware. The sophisticated people I know in programming, when the computer breaks down, they don’t know how to fix it. That’s for some dumb plumber. You know, get in an electrician. We don’t care about that; we’re doing serious intellectual things. And one of the beauties of the strong AI project is it made it appear that you don’t have to know how the brain works in order to know how the mind works. There are still people who think that. Again, I think that’s a profoundly mistaken and non-scientific, anti-scientific view. We’re going to have to know how the brain works because that’s where the action is taking place.

Well now I spoke to [neural network researcher] Terrence Sejnowski a couple of weeks ago, and he said—actually with a note of triumphalism in his voice, I have to say—that the symbolic AI people were defeated empirically. They simply could not build a mind. It didn’t work.

Yeah, well that’s right. There are two tests. One is, can you do an adequate simulation, where, remember, simulation just means you do a model or an imitation, a picture of the mind using such and such techniques, and you can’t do a simulation of the mind using traditional artificial intelligence techniques, using the binary digital computer that operates in serial with a series of 0’s and 1’s. And that to me is less interesting than the other failure, and that is: the simulation is not a duplication. Even if you can do a simulation, that doesn’t show that you’ve created a mind.

Look, we can do a perfect simulation of rainstorms in Berkeley or digestion in the stomach, but if we do a perfect computer simulation of digestion, let’s say, that doesn’t mean somebody can stuff a pizza into the computer and it will digest this. It gives you a picture. So what Sejnowski points out correctly is that those guys never got an adequate picture, but even if they had had an adequate picture, it still wouldn’t be a mind, it would just be a picture of the mind. It would stand to the mind the way any computer simulation of anything stands to the real thing. A computer simulation of a rainstorm won’t make us wet. And a computer simulation of a four-alarm fire won’t burn the house down, and a computer simulation of thinking, for the same reason, does not think.

Your philosophical refutation of strong AI is famous, and I hesitate to ask you to repeat it, yet I feel I must.

Ok, I’ll repeat it, it’s very simple. Somebody gives you a theory of the mind, always ask yourself, how would it work for me? So if some guy tells you, all that there is to having a pain is behaving as if you had a pain, I know that’s false, my pains hurt. There’s something in addition to pain behavior. Now if somebody tells you that all there is to having a mind and having mental abilities is to have a certain program, well try it out.

Suppose I run the program for some capacity I don’t have. As it happens, I can’t speak a word of Chinese. I don’t know Chinese. I can’t tell Chinese writing from Japanese writing. So we suppose that I run a program for answering questions given to me in Chinese. I’m locked in a room, and a bunch of Chinese symbols come in. I don’t know what they mean. I look up in a book, that’s called a “program”, and I shuffle these symbols and match them with other symbols that I’ve got in baskets around the room—those are called a database—and then I give back other symbols through the slot in the room. Those are called “answers” to the questions. So you get the picture. The questions are coming in in Chinese. I don’t know any Chinese, can’t understand a word. I shuffle the words according to the rules of the program, and then I give back answers to the questions. I don’t know these are questions. I don’t know what any of them mean. I don’t know that the stuff I’m giving back are answers, but let’s suppose that the programmers get so good at writing the program, and I get so good at shuffling the symbols, that after awhile my answers look just like the answers of a native Chinese speaker. I pass the so-called Turing Test—you can’t tell my answers from any other Chinese speaker, and the guys on the outside say, boy, the machine really understands Chinese well, or he really understands Chinese, but we know perfectly well that I don’t understand a word of Chinese.

Now why not? In order to understand Chinese I’ve got to know what the words mean, and I don’t know what any of these damn words mean, I’m just a computer. Now, this is the punch line of the whole story: if I don’t understand Chinese on the basis of implementing the program for understanding Chinese, then neither does any other digital computer solely on that basis, cause no computer has anything I don’t have.

Now, one last sense. You can see the force of that by contrasting me in the “Chinese Room” as I call it, answering questions in Chinese, and me in the same room answering questions in English. They ask me questions in English, and I answer the questions in English. My answers to the questions in English are indistinguishable form any other native speaker, because I am a native English speaker. They ask me questions in Chinese, I answer the questions in Chinese, but that’s totally different. Though my answers are indistinguishable from any other Chinese speaker, I don’t answer them by understanding anything, I answer them by running the computer program.

There is a huge difference between a program for cognition and actual cognition. And the difference can be stated in four words: “Syntax is not semantics.” That is, the program is defined syntactically as a set of symbol manipulations. Semantics has something in addition to syntax, it’s got meaning, or mental content, and it’s that that’s lacking from the digital computer.

Listen to the audio >>

top

Part Three

Ok, so the brain is not a computer, the mind is not a computer program. I think we’re ready for the punch line. What are they?

Yeah, ok. The brain is a biological organ, and it functions on very specific electro-chemical principals, and it is as specific as any other biological organ. Now that doesn’t mean that you couldn’t make an artificial machine that was conscious and had thought processes. We ought to hear the question, “can you make an artificial brain that thinks out of something other than tissue?” we ought to hear that like the question, “can you make a machine that pumps blood the way the heart pumps blood by making it out of something other than muscle tissue?” We know the answer to the heart question. We don’t know the answer to the brain question, because we don’t know how the brain does it in real life. But the failure in all of these things is to neglect, is to fail to respect the specificity of human biology.

Now that’s why I call my view “biological naturalism.” Think of mind and consciousness as part of nature, and the right level of describing them is biological. Now we don’t yet know how the brain does it, but we’re getting closer, and there are three or four things we have to keep in mind. One: consciousness is real. You can’t get rid of it. It’s here to stay. Two: it is entirely caused by brain processes. There just isn’t any question that the right level for explaining it is neuronal. We’re not sure if it’s individual neurons or whole maps or maybe sub-neuronal, but it’s something going on in the plumbing. Three: consciousness is realized in the brain. It’s all going on… mine is all going on in my skull, yours is all going on in your skull, and four: it functions causally. I decide to raise my arm, and the damn thing goes up. Now that’s absolutely crucial to respect that fact, because a lot of people think, well, even if consciousness did exist it couldn’t function causally cause it’s not part of the real world. It is part of the real world, just raise your arm.

I mean this, in a way, I feel slightly guilty that I get paid for doing the philosophy of mind. It’s much easier than the philosophy of language, and most of the points I make are screamingly obvious. Now, why aren’t they obvious to everybody? I think the answer is this tradition that I was telling you about, the tradition on the one hand there’s dualism and a kind of spiritualism, and people are trying to save the immortal soul. On the other hand there’s a kind of mindless scientism that says “anything that tries to claim that it’s something that’s not reducible to microphysics must be non-scientific. You must be trying to throw out 300 years of scientific progress.” And I want to say, forget about all those old categories and just try to think about how the damn thing works, how it actually works in real life.

So the mind is caused by the brain. It’s part of the physical world that just happens to have an irreducibly first-person ontology?

You just got an A in the course. That’s exactly right. It is a part of the physical world, and it’s as much a part of the physical world as any other biological phenomenon. It obviously has enormous evolutionary advantages to organisms that evolve consciousness, and we know that it functions causally in the organism’s behavior. So I think these are the philosophical parts of the mind/body problem. Now the really hard part is the scientific part, is to figure out how does the brain do it in detail? Where exactly and how exactly are these thought processes caused in my brain?

Aren’t you really trying to have your cake and eat it too, here? I mean, you’re saying it’s part of the natural world, it just happens to be irreducibly subjective. It has an irreducibly first-person ontology, but it’s part of the natural world. It sounds like you’re expanding the definition of “natural world,” not to solve the problem, but to just make it go away.

If we have a set of definitions that prevent us from stating the facts, then we have to change the definitions. Here are the facts. One: consciousness is real. You can’t get rid of it by pretending it doesn’t exist. Now why not? Why can’t you get rid of it? Well, the way we get rid of things is by showing that they’re an illusion. But in the case of consciousness, if I have the illusion that I’m conscious, then I am conscious. You can’t make the illusion/reality distinction for the very existence of consciousness the way you can make it for rainbows. The illusion that there’s a big arch in the sky, that’s an illusion. We can make a distinction between the appearance and the reality. But where the existence of consciousness is concerned, if it actually seems to me that I’m conscious, if I have the conscious illusion that I’m conscious, then I am conscious. So you can’t get rid of it. Furthermore, you can’t reduce it to something else. because it has this first person ontology, and if you try to reduce it to something that has a third person ontology you end up leaving out it’s essential feature. Okay, that’s the first thing. Now the second thing to remember is, it’s entirely caused by neuronal processes. There isn’t any doubt about that. And the third thing is it’s going on in the brain. That’s where it is, it’s realized in the brain. And as I said earlier, the fourth thing is to remember that it actually functions causally in our behavior.

Ok, now notice I said all those four things without using any of the traditional categories. I didn’t say this is dualism, this is materialism. I simply said, this is how nature works. Respect the facts. Then people come with their categories and say, well, that can’t be right, because it doesn’t fit my dualistic model. Or they say it can’t be right because it doesn’t fit my materialistic model. And what I want to say is, if you’ve got a theory, and your theory conflicts with the facts, you’d better think about changing the theory, and that’s the problem here is that a lot of these guys are very reluctant to give up on the theory, for in large part quasi- or explicitly religious reasons. They think there has to be an immortal soul, and that can’t be just part of our biology, and a lot of other people think science has shown there’s nothing that is not third-personal; science has shown there is nothing that has a first-personal ontology. I cannot imagine what their sex life is like! If you don’t have conscious experiences, what’s the point of living?

But in any case, those are the two forces that I’m militating against, and as you can imagine, they are held with great passion. Ironically, the materialists are more passionate than the spiritualists, than the dualists. I don’t have much problem with people who are avowedly religious. I mean, I guess they’re used to arguing with guys like me. But my experience is the materialists feel deeply threatened by this view that I call biological naturalism.

OK, now Thomas Nagel famously wondered what it was like to be a bat. And David Chalmers more recently speculated as to what it might be like to be a thermostat. Now what are these people wondering about?

Ok, well, Nagel in a very important article said that even if we had a perfect functionalist account of an organism, of an animal, had functionalism succeeded, so we had a perfect description of input/output relations, all the same there’d be something left out. What’s it feel like? What’s it feel like to have these conscious states? And he gave a very picturesque example. Even if we had a perfect knowledge of how the bat works, it’s physiology and all the programs it might be running and the rest of it, we still wouldn’t know what’s it feel like to be a bat. What does it feel like to hang upside down all day sleeping and then fly around in the dark navigating by bouncing sonar off of walls. What does that feel like? I haven’t the faintest idea!

It’s a fascinating question, really!

It is a fascinating question, but now, says Nagel, correctly: but that’s the problem of consciousness, and the problem of consciousness is the problem of what things feel like. Ok, now that was Nagel’s argument, and that was motivated by his desire to refute functionalism and materialism and it succeeds. I think it’s a brilliant argument.

Chalmers’ argument was totally different. Chalmers tries to hang onto dualism, and the result is that he gets a kind of panpsychism. It turns out that anything that has the right sorts of causal relations for him would be conscious. I don’t know if he still believes this, but there was a time when he wrote a book, when he gave this example. But then that seems to allow the possibility that thermostats might be conscious, and that’s a possibility he takes seriously.

Now I think under mature reflection he would deny, of course, that thermostats can be conscious, but at one time he took the possibility quite seriously and then tried to describe what it would be like to be a thermostat. The world would come in… I don’t even want to go into it, it’s too silly. I mean, the point is that a thermostat is the wrong kind of hunk of junk to be conscious. It doesn’t have anything like the structure capable of causing consciousness. If we’re serious, we’re really serious, we’re trying to earn our keep, and we want to know what kind of mechanisms cause consciousness, you have to respect the facts, and thermostats simply do not have the right kind of mechanisms.

But Chalmers wants to say that consciousness is part of the natural physical world too. In fact, he wants to say that maybe there are new fundamental properties—you know, you have charm, you have gravity, you have electromagnetism, and you’ve got “inwardness,” some kind of proto-phenomenal property out of which consciousness would be built in the right configuration.

As I understand his view it can’t function causally in the world. It can’t make any causal difference. His view is part of a tradition called “epiphenomenalism,” that says consciousness does not play any causal role, it just kinda goes along for the ride. I think that’s bad evolution—bad evolutionary theory. That is to say, there’s no way we’re going to get a phenotype as rich and complex and as expensive biologically as consciousness if it’s just going along for the ride, if it plays no causal role whatever. So it’s just not a serious hypothesis to say that consciousness plays no role in our behavior.

You know, it might turn out, we can imagine some huge scientific revolution that really showed nobody in the history of the world ever ate because he was hungry or drank because he was thirsty or did anything because he wanted to do it, now just think of what kind of proof we would have to have for that, that nobody was ever moved by fears and hopes and desires and lust and shame and disgust. That would be just not a minor scientific revolution like Einstein or Newton, that would be the greatest upheaval in the history of human thought, and it’s not a serious possibility on what we know today.

Listen to the audio >>

top

Part Four

Don’t we have a problem of free will here that goes along with the epiphenomenalism issue?

Well, we do have a terrible problem of free will, and the problem can be put really in one or two sentences. The problem of free will arises because on the one hand we have a complete conviction that everything that happens in nature must have a causal explanation based on causally sufficient conditions, conditions that are sufficient to make it the case that what happened had to happen. At the same time, we all have an experience of our own free will every time we make up our mind to do something. And now here’s what makes it such a tough one, is you can’t give up on the conviction of free will, even if you become philosophically convinced that it’s false.

You can see this by trying to imagine what it would be like to live on the presupposition that you had no free will. You’re in a restaurant. The waiter comes and say, well what will you have? You’re looking at the menu, and the choice is between veal and steak, let’s say. And the waiter says will you have the veal or the steak? And you try to say, “Look, I am a determinist, I’ll just wait and see what I order. Que sera, sera.” You can’t do that. Why not? Well, the reason you can’t do that is if you do that, doing that is itself only intelligible to you as an exercise of your free will. That is, the refusal to exercise free will is itself only intelligible to the agent as an exercise of free will.

So there’s no getting out of the presupposition of free will as long as you’re engaged in conscious, voluntary actions. Yet we don’t know how to make free will consistent with our general theory about how the universe works.

I don’t see how to make consciousness—even if it’s caused biologically and is part of the physical world—I don’t see how to make consciousness a cause.

It’s the easiest thing in the world, you just watch me. I have decided to raise my arm. I’m sorry we’re not on television, but my arm just went up. Notice a very important fact. We don’t say, “Well, that’s the thing about the old arm, some days she goes up and some days she doesn’t go up.” It is entirely up to me. That is all the tests for causation. If I hadn’t decided to raise my arm, it wouldn’t have gone up. Once I decided and started to try to raise it, it was bound to go up. All the tests for causation work there. Now how is that possible? Well the answer is very simple. Anything that causes my arm to go up must have electrochemical properties. Why? well, to leave out a lot of details, the secretion of acetylcholine at the axon endplates of my motor neurons has got to activate the ion channels in the muscle fibers, and a whole lot of other things have got to happen. If none of that stuff goes on, the arm does not go up.

So you get a very simple derivation. Anything that causes my arm to go up must have electrochemical properties. My conscious decision caused my arm to go up. So, guess what follows: my conscious decision has electrochemical properties. To describe it as a conscious decision is to describe one event at a higher level. To describe it as having these electrochemical properties is to describe that very same event at a lower level, in the same way that when my car starts I can describe it either by saying, well, spark plugs firing caused an explosion in the cylinder, or I can describe it at a much lower level in terms of the oxidization of hydrocarbons.

Now there’s no mystery in nature about how one and the same event can have causally different levels. Each level is real. Each functions as part of the total cause, and yet the vocabulary for describing the two different levels is different.

But when one of the levels is irreducibly first person that seems to be puzzling.

Yeah but a whole lot of people had the same feeling about electromagnetism, or for that matter about life. If you’re a Newtonian physicist and you see electromagnetism, it looks mysterious. How the hell are you supposed to give a Newtonian explanation of the magnetic field? Once we understand electromagnetism it’s just not mysterious.

Now we’ve been through this argument 100 years ago about life. How could life ever be explained by electrochemical, by biological processes? You need an élan vital. And I want to say you don’t need an élan vital, you just have to understand the complexity of the biological processes. Now the same way with consciousness, is nothing mysterious about the machinery that causes consciousness, and there’s nothing mysterious about consciousness, it is a biological phenomenon like any other.

Okay, what about this problem of “downward causality”? There’s no “room at the bottom:” consciousness is the higher order phenomenon, but everything is already thoroughly caused, as caused as it needs to be down at the bottom.

Well, maybe we ought not to use this metaphor of upward and downward. I’ve used it myself, but it can be misleading. We tend to think, well, if consciousness is at the top level—I’ve used this metaphor and it seems inevitable—and if, I don’t know, quantum mechanics is at the bottom level, then it looks like the only causation goes on at the bottom level. But, of course, that’s not true. If you think of the car engine, if you really want to know how it works, you have to be able to discuss the different levels: the solidity and the shape of the piston and the cylinder function causally, and they function crucially. If you try to make a square cylinder and a round piston, or if you try to make them both out of butter, it isn’t going to work. You have to have a single causal mechanism that has different levels of description.

Now in the same way with consciousness, there is a single causal mechanism by which I raise my arm. It involves a conscious intention and action. But of course that conscious intention and action has a cellular level of description, a sub-cellular level, a molecular level, an atomic level, and a subatomic level. So what? I mean that’s just how nature is. There are different levels of description of any complex system.

I want to go back to the arm—some days she goes up, some days she doesn’t—and ask you about the work of Libet and Wegner, which seems to suggest that your conscious intention to raise the arm comes after the physical process of raising it has already begun.

Well, I don’t know if your hearers know about this. It wasn’t Libet, it was two Germans in the 1970’s named Deecke and Kornhuber. They used to send me their articles. I haven’t heard from them in a long time—I don’t know if they’re still alive. But in any case, two Germans made a very important discovery: they discovered something called the bereitspotential (sic), the “readiness potential.” And what they discovered was, if you tell an agent to do some simple act such as punching a button or wiggling your finger, but do it at random, and keep track of when you actually decide to initiate the action, what you discover is that there is an increased activity in the supplementary motor area of their brain a couple of hundred milliseconds before they are aware of having decided to move their finger.

Now what follows from that? Well, I would say, not much. I mean, that’s an interesting result, and until we know how the brain works we don’t know quite how to assimilate it. I mean, we don’t know what to make of it. But some people think, no no, this refutes free will. This shows that you don’t have free will, because your brain was active before you were aware of having made the conscious decision to do something. Two things to notice. First of all, it’s only in the context where you have already made the conscious decision that every so often you will push the button that the readiness potential takes place. It’s only for simple trivial actions where you have already formed a prior intention to perform actions of that type. Nobody says, well, when Winton Churchill decided to keep on fighting in 1940, that was due to the readiness potential 200 milliseconds before he made the decision. It’s just, nobody thinks it works for that kind of thing. So that’s the first important qualification you have to add, is this is already within the context of a pre-existing decision of which the agent is conscious. And secondly, the readiness potential doesn’t give you causally sufficient conditions. As even Libet admits, you can still veto, you can still decide not to carry out the thing that the activity in the supplementary motor area seemed to presage. As Libet puts it, we shouldn’t say it’s free will; it’s “free won’t.” That is, you can always veto and say I’m not going to do the thing that I previously decided to do.

So I would say let’s not…we always tend to overestimate the importance of neurobiological results. I think these are good results, and I like what Libet did, he designed a special experimental device, which is very helpful in seeing the importance of this. But the idea that somehow or other this is how we’re gonna refute free will—no, it’s a much tougher issue than that.

Tell me a little bit about free will and what we call “the self.”

Well, the self is another set of problems, and specifically the reason that it’s a problem for us is that we all have a conviction that we are a self, and that I am the same self today that I was yesterday, and yet as Hume reminded us, if I turn my attention inward I don’t perceive any self. All I have are particular experiences. If I clutch my forehead and try to find the self all I discover is somebody’s got his hand on my forehead. Namely me. So you can’t locate the self as if it were just another feature of our experience. I want to say, nonetheless, we do have to postulate a self to make sense of our experiences. Why? Well, one good way to see it is to consider when you do something for which you’re responsible. If you’re responsible, you can’t just regard your behavior as the result of natural forces. You have to think that there is something that performed the action on the basis of reasons, and is capable of assuming responsibility in the future. Now that something, that X—there is an X which is conscious, capable of reflection, capable of deciding and acting on its reflections, and then capable of assuming responsibility—that something I am calling the self. But you have to remember, that’s a purely formal postulation.

An analogy I like to use is this: the postulation of the self is like the postulation of a point of view in vision. There’s no way that you can make sense of your visual experiences except as coming to you from a certain point of view. If I go on the other side of the room, I’d have a different point of view. But now the point of view itself is not seen. You can’t make sense of vision without assuming a point of view, but the point of view itself is not seen. And I want to say, similarly, you can’t make sense of your experiences without postulating your self, but the self is not itself experienced. It is, so to speak, a principle of the unity of all your other experiences. It is a principle by which you’re able to reflect and act on reasons and assume responsibility for reasons, but that doesn’t mean it’s a separate entity. So in a way I don’t refute Hume. Hume says you can’t have an experience of the self; I agree. But that doesn’t mean that you can make sense of your experiences without the postulation of a self. You can’t.

Last question, professor: what is philosophy, and what kind of reasoning are you engaged in when you’re doing it?

Well, there isn’t any single definition of philosophy, but for me it is the set of questions which are at one time fundamental. They’re sort of base questions. So, for example, the question, “what causes cancer” is not a philosophical question, but the question “what is the nature of causation itself,” that is a philosophical question. So that’s the first feature. They tend to be fundamental framework questions. A second feature is that they tend to be questions that we don’t have any established method for solving. I mean, once we knew how to get understanding of the biochemical basis of life it ceased to be a philosophical question and became a scientific question. This, by the way, is why science is always right and philosophy is always wrong. As soon as we’re convinced that we’ve got the right answer, we stop calling it philosophy and we start calling it science. It’s a trivial reason. And then a third feature of philosophy is that typically, not always, it has to do with conceptual relations. It has to do with what is the logical status of the notion of the self, or the notion of free will, or the notion of causation. Those are about our concepts, and they can’t be settled just by doing more factual investigation, although factual investigation is obviously relevant, it obviously sets constraints on our conceptual analysis. But for me it is simply the most fascinating subject, and indeed all other subjects I think of as interesting because of the way they can serve philosophy.

Professor Searle, thank you very much.

Thank you, thanks for having me.

Listen to the audio >>

top