Philosophy of Mind
In 1949 something significant happened in the world of philosophy which changed forever our view of the concept of the mind. That something was the publication of The Concept of Mind by British philosopher Gilbert Ryle. By and large, dualism was still, centuries after Descartes, the dominant view in metaphysics. Ryle's work changed that by pointing out that the Cartesian view was not riddled by a few minor problems (like the problem of interaction) but fatally flawed by one major problem. The irony was that this problem was not a real problem at all but rather a problem with the concept of mind. Ryle called this problem the "category mistake."
The mistake was thinking that the "mind" belongs to a category of things that it doesn't belong to. To illustrate, Ryle asks us to imagine the following example. Suppose I want to take you on a tour of a local university. So we get on campus and I show you the library, the student center, the science building, and so on. Just then someone asks "when are you going to show me the university?" The mistake implicit in this question is in thinking that the university is a separate thing in addition to the other buildings I have shown you. Ryle is claiming that we make the same mistake about the mind.
The whole Cartesian dogma, which Ryle refers to as "the dogma of the Ghost in the Machine," is based on the idea that the mind is a thing. Since, according to Descartes, the mind is not a physical thing, it must be a mental thing. This belief leads us to try to solve all sorts of problems that are unsolvable. Not to worry though because we can eliminate the problems by eliminating the faulty assumption that minds are things. So what is the mind? Well, let's look at it.
When we make references to the mind are we referring to a thing? Is this the way we talk about "the mind?” We're looking at language because this is the method that Ryle uses; it’s called analytic philosophy. Consider some sentences which contain words referring to mental states.
"I believe that it’s going to rain today."
"I feel happy."
"I know how to play the piano."
Now, what do these mental state words (belief, feel, know) refer to? Another way to think about this is to ask how do we know whether these sentences are true or false? If Descartes' view is correct these mental state words must be referring to a mental substance (the mind) that is inherently private and hidden from public view. If this is true then only the person saying them can tell whether they are true or false. After all only you know what you're thinking or feeling.
The problem with this view is that it seems false. When we say something like "I believe" or "I know" we do not think of it as referring to a hidden mental substance; a ghost in the machine. Let's look at the example "I know how to play the piano." What do I mean when I say this? According to Ryle what I mean is "I can play the piano" that is, I can do something. So, mental terms refer to dispositions to behave in a certain way. If I said I knew how to play the piano but couldn't play any songs you would conclude that I couldn't play the piano. Knowing refers to behavior. And this is true for all mental state words. The mind, in Ryle's view, is a set of behavioral dispositions.
If you've ever taken a psychology course this may sound familiar to you like the theory of behaviorism. What Ryle is advocating is a variation of psychological behaviorism called Logical (or philosophical) behaviorism. Essentially what it means is that we can understand mental state words (or define them) in terms of behaviors. It’s not that the behaviors are a clue to the mental state. The behavior is the mental state. There is, in this view a strict identity between the two.
Speaking of identity the next step in the philosophy of mind story involves the concept of identity. Ryle himself did not spell out the metaphysical position underlying his view but it is a materialist view. For Ryle, there is no mental substance only physical substance. The next step was taken by the Australian philosopher J. J. C. Smart (a good name for a philosopher!) in 1963. The theory he put forward, called the Identity Theory postulates that there is a strict identity between the mind and the brain. The mind is nothing more than the brain itself. So when we refer to mental states we are referring to physical states in the brain.
To advocates of this view, it seemed nothing more than good common sense. It certainly had the benefit of solving, or eliminating, the problems inherent in dualism. But to critics, the theory was nothing short of ridiculous speculation. According to these critics, there will always be aspects of the mind that cannot be explained by appealing to the physical structure of the mind. What I would like to do now is walk you through some of the criticisms of this identity theory, the responses of the advocates of the theory, and the status of the theory as we have it now. Interestingly enough, the theory as we have it now is a completely new theory of mind called functionalism.
The first objection relates to the concept of identity. According to one definition offered by Leibniz two things can only be identified if they share all the same attributes. By this law of identity, the mind and brain cannot be identical. After all, the mind has some attributes that the brain cannot have. For example, you can have a profound thought or you can say that your mental state is happy or sad. But can the brain be profound or happy or sad? It seems not. So clearly, the identity theory is flawed.
The response offered by the advocates of the theory is ingenious if nothing else. What is needed to overcome this objection is one of two things. We could show that the criterion of identity advocated by Leibniz is wrong. The other option is to offer a counterexample that seems to violate the law of identity but in fact, does not. This is the approach taken. The example involves the evening star and the morning star. These two different stars could never be identical in a million years. After all, they have radically different attributes; one comes out in the morning, one in the evening. However, they are identical. The morning star refers to the planet Venus but so does the evening star. It only seems that they are different things. In reality, they are the same thing. So, say advocates of the identity theory, though it seems as if the mind and brain could not be identified this is not the case at all.
The second objection also deals with identity but in a different way. To illustrate this objection, think of the number pi to two decimal places. Now I will think of pi to two decimal places as well. The question is “are our thoughts identical?” Well, it’s easy to tell. I was thinking of 3.14 how about you? OK good so they are identical. Now, according to the Identity theory, there should be a corresponding physical state in the brain that our thoughts are identical to. So when you think of the number pi what happens is that a C neuron fires in your brain. So your thought just is that neuron firing. Now, would it be possible for my thought of the number pi to be identical with a different neuron firing; perhaps a B neuron? It seems so. So the problem is that although we have identical thoughts our brain states are different.
The response offered to this objection is very simple with one small problem. It leads to an entirely new theory of the mind and a startling implication. Before we get to that though, the response. The only identity that matters is the identity between my thought and my brain state and your thought and your brain state. It’s not important whether there is an identity between our brain states or not. This is not the identity that counts. We know there is an identity of mental and physical states but the underlying nature of the physical state is not important; only that there is such an identity.
This seems interesting so we should see where it leads. If the underlying physical state is not so important then it seems that as long as there is thinking going on pretty much anything that causes that would be fine. Let's think of it this way. Could you imagine aliens from another planet being composed of a different physical structure? And also being able to think? That seems reasonable. Let's say that there are aliens on Jupiter whose physical makeup is based on silicon as opposed to our physical makeup which is based on carbon. It seems that as long as the aliens are doing what we're doing when we're thinking then they would also be thinking. This may be a little self-centered to define thinking in terms of what we're doing but since we're the example of thinking beings that we're addressing we'll stick with the comparison. The idea is that we're thinking and we define this not by the underlying physical structure (of course we know there is one) but by the outcome of the structure. This is the theory of functionalism. The difference between it and the Identity Theory should now be clear. In functionalism what defines thinking is the outcome or the function that is being fulfilled. So that's a new theory. What's the startling implication?
If, as we have just conceded, it is possible for a being composed of silicon to think then we don't need to look to Jupiter for a candidate for this being. We have one right here: the computer. Functionalism implies that computers might be able to think! The irony of this is that the idea that computers might be able to think was put forward by the mathematician Alan Turing only one year after Gilbert Ryle wrote The Concept of Mind.
In an article titled "Computing Machinery and Intelligence" published in the prestigious journal, Mind Turing asks us to consider this strange possibility. To do this he constructs a thought experiment called the imitation game. There are three players in this game: a man, a woman, and an interrogator (the sex of this player doesn't matter). The idea is that the interrogator asks questions to determine which player is the woman. Well, you say, how hard can that be after all the interrogator just has to look. So we'll make it a little more difficult by placing each of the players in separate rooms. Still, it might be easy because the interrogator just has to listen to the woman's voice. So we'll make the questions and answers written. But you can always recognize a woman's handwriting so we'll make the responses typed. Now the payoff comes. Suppose we substitute for the man a computer. And suppose further that the computer could fool the interrogator a statistically significant number of times. Wouldn't we have to concede that the computer is thinking? Turing said yes. I should mention at this point that no computer, even the most advanced AI computers we have now, has ever passed this test; now called the Turing test.
Of course, not everyone agreed with this implication. In particular, the philosopher John Searle took exception to the idea that computers could think and offered his thought experiment to illustrate the problem with the imitation game. He called his thought experiment the Chinese Room Example. A man, who speaks fluent English, but not Chinese, is in a room. He is given a text written in Chinese and by using a set of rules (written in English) produces another text in Chinese and hands it to observers outside the room. The first text is a story and a set of questions written in Chinese. The text that the man in the room produces, in Chinese, is a set of answers to the questions. Again, he produces this text by following a set of rules which essentially tell him what Chinese character to write down when he sees a certain Chinese character in the original text. Now, to observers outside the room, it looks as if the man knows Chinese. The problem, of course, is that he doesn't. What this illustrates is the missing element in the Turing test: understanding. While the man can manipulate a set of symbols (just as the computer can manipulate binary code) he does not understand what he's doing.
The latest update in the philosophy of mind story is a theory offered by another Australian philosopher named David Chalmers. In his book The Conscious Mind he explains that the fundamental aspect of thinking is still unexplained. "Consciousness is the biggest mystery." To explain it he attempts to tread the thin line between materialism and dualism by advocating a version of property dualism where he claims that the mind is an emergent property of the physical structure of the brain. Like a holograph which is caused by a physical system but different from it, the mind is a property of a sufficiently complex physical structure though different from it. There's much more research to do but it is an interesting start. Given the emphasis on physical structure that these theories imply we should probably consider in more detail the metaphysics of materialism.
The mistake was thinking that the "mind" belongs to a category of things that it doesn't belong to. To illustrate, Ryle asks us to imagine the following example. Suppose I want to take you on a tour of a local university. So we get on campus and I show you the library, the student center, the science building, and so on. Just then someone asks "when are you going to show me the university?" The mistake implicit in this question is in thinking that the university is a separate thing in addition to the other buildings I have shown you. Ryle is claiming that we make the same mistake about the mind.
The whole Cartesian dogma, which Ryle refers to as "the dogma of the Ghost in the Machine," is based on the idea that the mind is a thing. Since, according to Descartes, the mind is not a physical thing, it must be a mental thing. This belief leads us to try to solve all sorts of problems that are unsolvable. Not to worry though because we can eliminate the problems by eliminating the faulty assumption that minds are things. So what is the mind? Well, let's look at it.
When we make references to the mind are we referring to a thing? Is this the way we talk about "the mind?” We're looking at language because this is the method that Ryle uses; it’s called analytic philosophy. Consider some sentences which contain words referring to mental states.
"I believe that it’s going to rain today."
"I feel happy."
"I know how to play the piano."
Now, what do these mental state words (belief, feel, know) refer to? Another way to think about this is to ask how do we know whether these sentences are true or false? If Descartes' view is correct these mental state words must be referring to a mental substance (the mind) that is inherently private and hidden from public view. If this is true then only the person saying them can tell whether they are true or false. After all only you know what you're thinking or feeling.
The problem with this view is that it seems false. When we say something like "I believe" or "I know" we do not think of it as referring to a hidden mental substance; a ghost in the machine. Let's look at the example "I know how to play the piano." What do I mean when I say this? According to Ryle what I mean is "I can play the piano" that is, I can do something. So, mental terms refer to dispositions to behave in a certain way. If I said I knew how to play the piano but couldn't play any songs you would conclude that I couldn't play the piano. Knowing refers to behavior. And this is true for all mental state words. The mind, in Ryle's view, is a set of behavioral dispositions.
If you've ever taken a psychology course this may sound familiar to you like the theory of behaviorism. What Ryle is advocating is a variation of psychological behaviorism called Logical (or philosophical) behaviorism. Essentially what it means is that we can understand mental state words (or define them) in terms of behaviors. It’s not that the behaviors are a clue to the mental state. The behavior is the mental state. There is, in this view a strict identity between the two.
Speaking of identity the next step in the philosophy of mind story involves the concept of identity. Ryle himself did not spell out the metaphysical position underlying his view but it is a materialist view. For Ryle, there is no mental substance only physical substance. The next step was taken by the Australian philosopher J. J. C. Smart (a good name for a philosopher!) in 1963. The theory he put forward, called the Identity Theory postulates that there is a strict identity between the mind and the brain. The mind is nothing more than the brain itself. So when we refer to mental states we are referring to physical states in the brain.
To advocates of this view, it seemed nothing more than good common sense. It certainly had the benefit of solving, or eliminating, the problems inherent in dualism. But to critics, the theory was nothing short of ridiculous speculation. According to these critics, there will always be aspects of the mind that cannot be explained by appealing to the physical structure of the mind. What I would like to do now is walk you through some of the criticisms of this identity theory, the responses of the advocates of the theory, and the status of the theory as we have it now. Interestingly enough, the theory as we have it now is a completely new theory of mind called functionalism.
The first objection relates to the concept of identity. According to one definition offered by Leibniz two things can only be identified if they share all the same attributes. By this law of identity, the mind and brain cannot be identical. After all, the mind has some attributes that the brain cannot have. For example, you can have a profound thought or you can say that your mental state is happy or sad. But can the brain be profound or happy or sad? It seems not. So clearly, the identity theory is flawed.
The response offered by the advocates of the theory is ingenious if nothing else. What is needed to overcome this objection is one of two things. We could show that the criterion of identity advocated by Leibniz is wrong. The other option is to offer a counterexample that seems to violate the law of identity but in fact, does not. This is the approach taken. The example involves the evening star and the morning star. These two different stars could never be identical in a million years. After all, they have radically different attributes; one comes out in the morning, one in the evening. However, they are identical. The morning star refers to the planet Venus but so does the evening star. It only seems that they are different things. In reality, they are the same thing. So, say advocates of the identity theory, though it seems as if the mind and brain could not be identified this is not the case at all.
The second objection also deals with identity but in a different way. To illustrate this objection, think of the number pi to two decimal places. Now I will think of pi to two decimal places as well. The question is “are our thoughts identical?” Well, it’s easy to tell. I was thinking of 3.14 how about you? OK good so they are identical. Now, according to the Identity theory, there should be a corresponding physical state in the brain that our thoughts are identical to. So when you think of the number pi what happens is that a C neuron fires in your brain. So your thought just is that neuron firing. Now, would it be possible for my thought of the number pi to be identical with a different neuron firing; perhaps a B neuron? It seems so. So the problem is that although we have identical thoughts our brain states are different.
The response offered to this objection is very simple with one small problem. It leads to an entirely new theory of the mind and a startling implication. Before we get to that though, the response. The only identity that matters is the identity between my thought and my brain state and your thought and your brain state. It’s not important whether there is an identity between our brain states or not. This is not the identity that counts. We know there is an identity of mental and physical states but the underlying nature of the physical state is not important; only that there is such an identity.
This seems interesting so we should see where it leads. If the underlying physical state is not so important then it seems that as long as there is thinking going on pretty much anything that causes that would be fine. Let's think of it this way. Could you imagine aliens from another planet being composed of a different physical structure? And also being able to think? That seems reasonable. Let's say that there are aliens on Jupiter whose physical makeup is based on silicon as opposed to our physical makeup which is based on carbon. It seems that as long as the aliens are doing what we're doing when we're thinking then they would also be thinking. This may be a little self-centered to define thinking in terms of what we're doing but since we're the example of thinking beings that we're addressing we'll stick with the comparison. The idea is that we're thinking and we define this not by the underlying physical structure (of course we know there is one) but by the outcome of the structure. This is the theory of functionalism. The difference between it and the Identity Theory should now be clear. In functionalism what defines thinking is the outcome or the function that is being fulfilled. So that's a new theory. What's the startling implication?
If, as we have just conceded, it is possible for a being composed of silicon to think then we don't need to look to Jupiter for a candidate for this being. We have one right here: the computer. Functionalism implies that computers might be able to think! The irony of this is that the idea that computers might be able to think was put forward by the mathematician Alan Turing only one year after Gilbert Ryle wrote The Concept of Mind.
In an article titled "Computing Machinery and Intelligence" published in the prestigious journal, Mind Turing asks us to consider this strange possibility. To do this he constructs a thought experiment called the imitation game. There are three players in this game: a man, a woman, and an interrogator (the sex of this player doesn't matter). The idea is that the interrogator asks questions to determine which player is the woman. Well, you say, how hard can that be after all the interrogator just has to look. So we'll make it a little more difficult by placing each of the players in separate rooms. Still, it might be easy because the interrogator just has to listen to the woman's voice. So we'll make the questions and answers written. But you can always recognize a woman's handwriting so we'll make the responses typed. Now the payoff comes. Suppose we substitute for the man a computer. And suppose further that the computer could fool the interrogator a statistically significant number of times. Wouldn't we have to concede that the computer is thinking? Turing said yes. I should mention at this point that no computer, even the most advanced AI computers we have now, has ever passed this test; now called the Turing test.
Of course, not everyone agreed with this implication. In particular, the philosopher John Searle took exception to the idea that computers could think and offered his thought experiment to illustrate the problem with the imitation game. He called his thought experiment the Chinese Room Example. A man, who speaks fluent English, but not Chinese, is in a room. He is given a text written in Chinese and by using a set of rules (written in English) produces another text in Chinese and hands it to observers outside the room. The first text is a story and a set of questions written in Chinese. The text that the man in the room produces, in Chinese, is a set of answers to the questions. Again, he produces this text by following a set of rules which essentially tell him what Chinese character to write down when he sees a certain Chinese character in the original text. Now, to observers outside the room, it looks as if the man knows Chinese. The problem, of course, is that he doesn't. What this illustrates is the missing element in the Turing test: understanding. While the man can manipulate a set of symbols (just as the computer can manipulate binary code) he does not understand what he's doing.
The latest update in the philosophy of mind story is a theory offered by another Australian philosopher named David Chalmers. In his book The Conscious Mind he explains that the fundamental aspect of thinking is still unexplained. "Consciousness is the biggest mystery." To explain it he attempts to tread the thin line between materialism and dualism by advocating a version of property dualism where he claims that the mind is an emergent property of the physical structure of the brain. Like a holograph which is caused by a physical system but different from it, the mind is a property of a sufficiently complex physical structure though different from it. There's much more research to do but it is an interesting start. Given the emphasis on physical structure that these theories imply we should probably consider in more detail the metaphysics of materialism.